| POU N DED NGTO N I I FEBRU ~ MCM - ARY : -W XXX rfHE LIBRARY . Y of the THE HISTORY OF HUMAN MARRIAGE VOL. JI THE HISTORY OF HUMAN MARRIAGE BY EDWARD WESTERM ARCK PH.D.; HON. LL.D., ABERDEEN MARTIN WHITE PROFESSOR OF SOCIOLOGY IN THE UNIVERSITY OF LONDON PROFESSOR OF PHILOSOPHY AT THE ACADEMY OF ABO (FINLAND) AUTHOR OF " THE ORIGIN AND DEVELOPMENT OF THE MORAL IDEAS," " MARRIAGE CEREMONIES IN MOROCCO," ETC. IN THREE VOLUMES VOL. Il FIFTH EDITION REWRITTEN THE ALLERTON BOOK COMPANY NEW YORK l 9 2 2 CONTENTS CHAPTER XVII SEXUAL SELECTION IN MAN Female appreciation of manly strength and courage, pp. 1-4.-The sexual instinct stimulated by personal beauty both in men and women, savage as well as civilised, p. 4 sq.-The notions of what constitutes personal beauty vary, p. 5.-Three groups of qualities regarded as essential for personal beauty, apart from merely individual preferences, pp. 5-13.-The full and proper development of the visible character- istics belonging to the human organism in general, p. 5.-Of those peculiar to the sex, pp. 5-8.-Of those peculiar to the race, pp. 8-13.-The connection between love and beauty due to natural selection, pp. 13-23.-The biological importance of the qualities constituting the racial type of beauty, pp. 14-22.-The bodily changes involved in acclimatisation, pp. 16-18.-The connection between racial differences and external influences, pp. 19-22.- Rejection of Darwin's explanation of the connection between love and beauty, p. 22 sq.-Mental qualities as sexual stimulants, p. 23.- The connection between the sexual impulse and affection in the human race, p. 23 sq.-Conjugal affection and the influence of affection upon sexual selection, pp. 24-31.-Sexual selection influenced by the desire for offspring, p. 31 sq.-By economic considerations, PP- 32-34- ENDOGAMY Endogamous and exogamous rules, p. 35.-The aversion of animals to sexual connections with individuals of other species than their own, pp. 35-37.-The " Law of Similarity," p. 37 sq.-Intermixture of human races, p. 38 sq.-:Racial endogamy and its causes, pp. 39-43.- The influence of the crossing of races upon fertility and vitality, pp. 43-47.-Rules against marrying outside the tribe or some smaller division of it, as the clan or village, pp. 47-55.-Religious endogamy, pp. 55-59.-Caste endogamy and hypergamy, p. 59 sq.-Class endogamy, pp. 60-68.-Class endogamy parallel to the hypergamy of the Hindus, p. 64 sq.-The origin of castes and classes, pp. 65-67.- Endogamous rules referring to the marriage of cousins, pp. 68-79.- To marriage between brothers' children, pp. 69-71.-To marriage between cross-cousins, pp. 71-79.-To marriage between relatives more closely related than cousins, pp. 79-81. CHAPTER XVIII V VI CONTENTS CHAPTER XIX EXOGAMY The exogamous group, p. 82.-Marriages between parents and children, pp. 82-88.-Between brothers and sisters who are children of the same father and mother, pp. 88-95.-Between half-brothers and half- sisters, pp. 95-97.-Between uncles and nieces and between aunts and nephews, pp. 97-99.-Between cousins, pp. 99-101.-Exogamous rules referring to all the members of the clan or phratry or " class " and other exogamous rules among peoples unaffected by modern civilisation, pp. 101-149.-In America, pp. 102-113.-In Asia, pp. 113-120.-In the Indian Archipelago and the islands of the Pacific Ocean, pp. 120-131.-In Australia and Tasmania, pp. 131-136.- In Africa, pp. 136-149.-Prohibited degrees in ancient Rome, p. 149.- According to the Christian Churches, pp. 149-151.-Clan exogamy in South-Eastern Europe, p. 151.-Prohibitions applying to marriage between relatives by alliance or affinity, pp. 151-154.--Exogamous prohibitions resulting from adoption or fosterage, pp. 154-156.- From " spiritual relationship," p. 156.-From certain other relations, p. 156 sq.-Local exogamy, pp. 157-161. EXOGAMY {Concluded) Early hypotheses as regards the origin of the prohibitions of marriage between near relatives, p. 162.-Criticism of McLennan's hypothesis as to the origin of exogamy pp. 162-166.-Of Herbert Spencer's, pp. 166-168.-Of Lord Avebury's, p. 168 sq.-Of L. H. Morgan's, pp. 170-179.-Evil results ascribed by various peoples to close inter- marriage or incestuous intercourse, pp. 170-178.-Criticism of Sir James G. Frazer's suggestions as to the origin of exogamy, pp. 179-183. -The Australian class restrictions, pp. 179-181.-Criticism of Professor Durkheim's theory, pp. 183-185.-Of M. Reinach's, p. 185 sq. -Of Mr. Andrew Lang's, pp. 186-188.-Of Messrs. Hose's and McDougall's, pp. 188-190.-Of Professor Starcke's, p. 190 sq.- Objections which may be raised to all of these theories, p. 191 sq.- The fundamental cause of the exogamous prohibitions seems to be the remarkable absence of erotic feelings between persons living very closely together from childhood, leading to a positive feeling of aversion when the act is thought of, pp. 192-218.-The existence of such a feeling in mankind, pp. 193-195.-Among the lower animals, pp. 195- 197.-Reply to objections raised to the author's theory, pp. 197-207.- Objections raised^ by Dr. Havelock Ellis, p. 197.-Marriage with a half-sister, p. 199 sq.-Cases of incest and close intermarriage and how to explain them. pp. 200-203.-Objections to the author's theory raised by Sir James G. Frazer, p. 203 sq.-The connection between the aversion to sexual intercourse between persons living closely together from childhood and the prohibitions against unions between kindred, pp. 204-207.-Facts showing that the extent to which CHAPTER XX CONTENTS VII relatives are forbidden to intermarry is nearly connected with their close living together, pp. 207-214.-An exogamous clan or phratry frequently a territorial group as well as a group of kindred, pp. 210-214. -How the exogamous rules have come to include relatives who do not live together, p. 214 sq.-How to explain the prohibitions of marriage between relations by alliance, the prohibitions on the ground of " spiritual relationship," and certain other prohibitions, p. 215 sq.- Every hypothesis pretending to explain the exogamous rules in full must assume the operation of the law of association, p. 216.--The co-existence of comprehensive exogamous prohibitions and the classificatory system of relationship, p, 216 sq.-The biological founda- tion of the psychical cause of exogamy, pp. 218-239.-The effects of cross- and self-fertilisation among plants, p. 218 sq.-The effects of in-breeding among animals, pp. 219-223.-Arrangements in the animal world helping to prevent in-breeding, p. 223 sq.-The effects of marriages between the nearest relatives, p. 224 sq.-Of marriages between first cousins and other near relatives, pp. 225-229.-The effects of in-breeding in various small communities, pp. 229-234.- The bad consequences of self-fertilisation and close inter-breeding may in certain circumstances almost fail to appear, pp. 234-236.- The injurious effects of in-breeding afford a sufficient explanation of the want of inclination for, and positive aversion to, sexual inter- course between persons who have from childhood lived together in that close intimacy which characterises the mutual relations of the nearest kindred, p. 236 sq.-The question why in-breeding is injurious to the species, pp. 237-239.-The author's theory of the origin of exogamy not affected by this question, pp. 237, 239.-Certain general facts emphasised in support of his theory, p. 239. CHAPTER XXI MARRIAGE BY CAPTURE Instances of marriage by capture among the lower races, pp. 240-251.- Among Semitic and Aryan peoples, p. 251 sq.-Marriage by capture not knowm to have been the normal mode of contracting a marriage among any people, p. 253.-Chiefly found either as an incident of war or as a method of procuring a wife when it is difficult or incon- venient to get one in the ordinary manner, ibid.-No reason to suppose that it ever was the ordinary mode of contracting a marriage among uncivilised peoples, p. 254.-Customs which have been interpreted as survivals of marriage by capture in the past and how to explain them, pp. 254-277.-Cases in which sham fighting between the bride- groom or his party and the bride's family, or some other kind of resistance made by the latter, forms part of the wedding ritual, pp. 254-263.-The custom of barring the wedding procession, pp. 259- 261.-Abusive language at weddings, p. 263 sq.-Resistance made or grief expressed principally or exclusively by the bride, pp. 264-272.- The bride or newly-married wife running away from her new home, p. 270 sq.-Other sham fights at weddings, accentuating social antagonism, especially between the female friends of the bride, or the women in general, and the bridegroom, pp. 273-275.-Supposed traces of marriage by capture in the language spoken by the women, pp. 275-277.-Various other supposed survivals of marriage by capture, p. 277. VIII CONTENTS CHAPTER XXII CONSENT AS A CONDITION OF MARRIAGE Marriages arranged by the fathers, parents, or other near relatives of the parties among the lower races, pp. 278-284.-The consent of the communities to which the parties belong necessary for the conclusion of a marriage, p. 281 sq.-The permission of the chief or king required, p. 282.-The selection of mates among every people controlled by its customs or laws, p. 282 sq.-Among most of the lower races a girl is in her father's power till she marries, p. 284.-Among various peoples the consent of a mother, brother, or maternal uncle regarded as particularly essential to a girl's marriage, pp. 284-286.-Women's liberty of selecting their husbands among the lower races, pp. 286-326. -In America, pp. 286-291.-In Asia, pp. 291-296.-In the Indian Archipelago and the islands of the Pacific Ocean, pp. 296-301.- In Australia and Tasmania, p. 301 sq.-In Africa, pp. 302-308.-The woman's wishes seem to be consulted among the majority of the lower races, p. 308 sq.-If the Australian natives are excepted, the woman's liberty of choice decidedly greater among the lowest than among the more advanced savages, p. 309 sq.-Why progress in economic culture has among the lower races been unfavourable to the woman's liberty of choosing her husband, pp. 310-315.--Close connection between marriage by purchase and disregard of the woman's own wishes, pp. 310-314.-Why the woman's wishes are utterly disregarded among the Australian natives, p. 315 sq.-Every reason to believe that in primitive ages woman was much freer in her choice of partner than she is now in the Australian tribes and among many other savages, p. 316.-Various means by which a woman may prevent or break a marriage distasteful to her, pp. 316-319. -Marriage by elopement among the lower races, pp. 319-326.- Paternal or parental authority, especially with regard to the marriage of children, among the peoples of archaic culture, pp. 326-338.-In Christian Europe, pp. 338-344.-The origin of the right of fathers or parents to interfere with the marriage of their children, pp. 344-353.- The influence of magical and religious ideas, pp. 345-352.-The importance attached to the curses and blessings of parents, pp. 348-352 . --The downfall of the paternal power, p. 353. CHAPTER XXIII MARRIAGE BY CONSIDERATION AND BY THE EXCHANGE OF PRESENTS. GIFTS TO THE BRIDE AND TO THE BRIDEGROOM The exchange of bride for bride, pp. 354-359.-Marriage as compensation or a means of reconciliation in the case of homicide, p. 359 sq.~- Marriage by service, pp. 360-375.-As the regular, if not exclusive, method of acquiring a wife, pp. 364-369.-As a substitute for marriage by purchase, pp. 369-371.-As an alternative to giving no considera- tion for a wife, p. 371.-Marriage by service and purchase combined, PP- 371-373.-Explanation of marriage by service, p. 373 sq.-The consideration given for a wife consisting of property of some kind or other, pp. 375-383.-The bride price sometimes uniformly fixed by custom, p. 383.-Various factors influencing the bride price, pp. 383- 385.-To whom the consideration for a bride is given, pp. 385-388.- The paying of the bride price, partly at least, devolving on other CONTENTS IX persons than the bridegroom or his father, p. 388 sq.-The considera- tion offered by instalments, pp. 389-391.-Special payment for the offspring, p. 391 sq.-The paying of the bride price may give the husband a right to demand another woman if the wife dies or proves barren, p. 392.-Disgrace attached to being given in marriage without a price, p. 392 sq.-Explanation of the custom of giving material objects for a bride and the meaning of " marriage by purchase," PP- 393-395--The relation between marriage by purchase and marriage by capture, p. 395 sq.-The rendering of some consideration for the bride, while the normal mode of securing marriage at all grades in the uncivilised world, has assumed increasing importance with the advance of economic culture, p. 396.-Marriage by the exchange of gifts among the lower races, pp. 396-401.-The practice of giving a dowry to a daughter among them, pp. 401-403.-The bride price, or a part of it, given to the bride by her father as a dowry or looked upon as a settlement or provision for the wife, p. 403.-Presents given by suitors to their prospective wives or by bridegrooms to their brides, pp. 404-406.-Marriage by consideration among peoples of a higher culture, pp. 406-414.-The custom of offering a return gift among such peoples, p. 414 sq.-Gifts offered to the bride by the bridegroom, and the price paid for the bride becoming wholly or in part her own property among such peoples, pp. 415-423.-The marriage portion among such peoples, pp. 423-431.--The purchase of husbands, p. 431. MARRIAGE RITES The seqtience of rites connected with the conclusion of a marriage, p. 432 sq. -Their most general social object to give publicity to the union, pp. 433-438.-The union sanctioned by an official, p. 433 sq.- Contracted in the presence of witnesses, pp. 434-436.-Publicity given to the sexual consummation of marriage, p. 436 sq.-Publicity given to the union by celebrating it with feasting, p. 437 sq.-The social importance of marriage feasts, p. 438.-Rites de separation and rites d'agr&gation, ibid.-Marriage rites symbolising the union between the parties or intended to strengthen the marriage tie, pp. 438-456.- The joining of hands, p. 439 sq.-The tying together of the couple, pp. 440-442.-Other rites of contact, p. 442.-The tying of something to bride and bridegroom separately, p. 443.-Betrothal and wedding rings, p. 443 sq.-Ceremonies in which an article of dress belonging to or presented by one of the partners is worn by the other, p. 445.- The union of the parties symbolised, or effected, by means of blood, p. 445 sq.-The use of red lead and of the red colour in general in wedding rites, p. 446 sq.-The use of animal blood, p. 447 sq.-The eating together of bride and bridegroom, pp. 448-452.-The rite of drinking together, pp. 452-454.-Some other kindred rites, p. 455 sq.- Certain rites supposed to ensure the durability of the marriage, p. 456 sq.-Rites intended to ensure or facilitate the consummation of the marriage, pp. 457-467.-The breaking of eggs at weddings, pp. 457-459.-Of earthenware vessels or objects of glass, pp. 459-463. -Of a staff or rod or tree, p. 463 sq.-Of a loaf of bread or a cake, p. 464.-The red colour perhaps regarded as a means of ensuring defloration, p. 466 sq.-Rites practised, or supposed to be practised, \vith a view to making the wife fruitful or the mother of male offspring, pp. 467-486.-Little children used for this purpose, pp. 467-469.- The bridal pair or the bride being seated on the skin of an animal, p. 470.-The throwing of some kind of cereals or fruit on the bride or CHAPTER XXIV X CONTENTS on the bridal pair or on the bridegroom separately or on the wedding company, pp. 470-484.-Various ideas connected with this custom, pp. 478-484.-The use of fish, p. 484 sq.-Of eggs, p. 485 sq.-Rites intended to facilitate the delivery of the young wife, p. 486 sq.-To influence the external appearance or the behaviour of the future offspring, p. 487.-To make the couple prosperous, pp. 487-489.- The use of honey or other sweet things, p. 489.-Rites which have reference to the particular position of the husband and the wife, pp. 490-494.-The shooting of an arrow, p. 490 sq.-The bridegroom trying to gain power over his wife, p. 491 sq.-Rites intended to ■make the husband considerate or subject to his wife, pp. 492-494.- The beating of the bride or bridegroom by the other partner, p. 494 sq. CHAPTER XXV MARRIAGE RITES (Continued) The ideas that bride and bridegroom are in a state of danger and that the bride also is a source of danger to others, leading to a multitude of prophylactic and purificatory rites, p. 496 sq.-The shutting out of evil influences from the place where marriage is celebrated, p. 497.- The firing of guns and the making of noise at weddings, pp. 497-499.- Weapons of various kinds used as charms, pp. 499-502.-Other charms, such as needles and salt, p. 502.-The burning of incense, ibid.- The painting of bride and bridegroom with henna, p. 502 sq.-The bathing or washing of bride and bridegroom, pp. 503-506.-Other water rites, pp. 506-509.-Fire rites, pp. 509-514.-The circuipambula- tion of the hearth, pp. 512-514.-Other forms of circumambulation, pp. 514-516.-The beating of the bridegroom's tent, p. 516 sq.- The beating or tapping of bride and bridegroom, p. 517 sq.-Disguises at marriages, pp. 518-526.-The bride dressed in men's and the bridegroom in women's clothes, pp. 519-521.-The false bride, pp. 521-523.-Effigies substituted for the bride and bridegroom, p. 523.--Mock marriages with things or trees or animals, pp. 523-525.- The bridegroom or the bride sheltered by some person or persons dressed up to resemble him or her, p. 525 sq.-The functions of brides- maids, bridesmen, and groomsmen, p. 526 sq.-The bride shut up in a box, p. 527.-The veiling or covering of the bride, p. 527 sq.- The bridegroom covering his face, p. 528.-Bride and bridegroom sheltered by a screen, ibid.-Not allowed to leave the house for a certain period, p. 528 sq.-Protected against dangers from above, p. 529 sq.-Against dangers from below, pp. 530-542.-By sitting or walking on a carpet or on something else placed on the ground, pp. 530-534.-The carrying of bride or bridegroom, pp. 531-533-- The carrying of the bride over the-threshold, pp. 535_538--The fear of the threshold, p. 537 sq.-The throwing of an old shoe or old shoes after bride and bridegroom, pp. 539-542. MARRIAGE RITES (Concluded) Evils averted from bride and bridegroom by abstinences of various kinds, pp. 543-563.-They must not look round, p. 543.-Must not stand apart, p. 544.-Dare not extinguish the candle on the wedding night, ibid.-Any mishap an evil foreboding, ibid.-Abstinences with regard CHAPTER XXVI XI CONTENTS to eating and drinking, p. 544 sq.-Inactivity required of a bride, p. 545 sq.-Silence imposed on her, p. 546.-The bridegroom must not speak aloud, ibid.-Bride and bridegroom must keep awake, p. 546 sq. -Must observe continence for a shorter or longer time after marriage, pp. 547-564.-The prevalence of this taboo in America, pp. 547-549. •-In Africa, pp. 549-551.-In Australia, p. 551 sq.-In Melanesia, New Guinea, and the Indian Archipelago, pp. 552-554.-In Asia, PP- 554-557' 559--Europe, pp. 557-560.-Continence prescribed by the Christian Church, p. 559 sq.-The origin of this taboo, pp. 560- 564.-Why bride and bridegroom are supposed to be in a dangerous condition and the bride is considered dangerous to others, pp. 563-565. -Differences in the ritual between first and subsequent marriages, p. 564 sq.-Often difficult or impossible to draw a definite distinction between protective or purificatory rites and such as are held to result in more positive benefits, p. 565 sq.-The selection of the time when a marriage is to be celebrated, pp. 566-573.-Lucky and unlucky seasons or months, p. 566 sq.-The choice of the time when marriages are celebrated influenced by the moon, pp. 568, 572.-Certain days of the week chosen for the celebration of marriages and others avoided, PP- 569-573.-Certain dates considered unlucky or lucky, p. 572.- Marriage rites of a religious or semi-religious nature, often performed by a priest, pp. 573-577.-Civil marriage, p. 577.-The blessing of the bridal bed, p. 578.-The parson's right to kiss the bride, pp. 579, 590.-The marriage ritual influenced by the new relations into which either party enters to the other party's family or the larger social group of which he or she is a member, pp. 579-584.--By the new relations between the other members of the two families besides bride and bridegroom, p. 581 sq.-By the bride's new relations to the spiritual agents connected with the bridegroom's family or community, p. 582 sq.-By her relations to the domestic animals of her new home, p. 583 sq.-By either party passing from one social class into another, the bridegroom into the class of the married men and the bride into that of the married women, p. 584 sq.-Marriage rites supposed to influence the welfare of other persons than bride and bridegroom even independently of their relations to the principals, pp. 585-591.- Benefits expected from joining in a wedding, pp. 585-590.-A wedding looked upon as a potential cause of other weddings, pp. 586-589.- Dancing at weddings, pp. 588-590, 592.-Good luck apparently expected from contact with the bride or bridegroom or something worn by them, p. 589 sq.-The kissing of the bride, ibid.-Pre- cautionary measures against evil influences taken by persons who join in a wedding, p. 590 sq.-Marriage rites as survivals of earlier methods of concluding a marriage, p. 591 sq.-As fossilised expressions of emotional states, p. 592.--The sexual indulgence of the wedding guests, ibid.-Marriage rites said to be absent among many peoples, PP- 593-595--The connection between marriage rites and the degree of culture, p. 595. THE HISTORY OF HUMAN MARRIAGE CHAPTER XVII SEXUAL SELECTION IN MAN By sexual selection in man is meant the choice made by men or women as regards relations with the opposite sex. In the chapters on courtship and primitive means of attrac- tion we have seen how the two sexes in various ways endeavour to make themselves attractive to each other. We shall now consider certain factors influencing the choice of mates apart from courtship-woman's appreciation of manly strength and courage, the stimulating influence of beauty, the feeling of affection, and dry calculation. Darwin has shown that among the lower Vertebrates the female often shows a predilection for " the most vigorous, defiant, and mettlesome male "-a taste the origin of which is easily accounted for by the theory of natural selection. A similar instinctive appreciation of masculine strength and valour is found in women. In a song com- municated by Schoolcraft, a North American Indian girl gives the following description of her ideal:-" My love is tall and graceful as the young pine waving on the hill- And as swift in his course as the noble stately deer-His hair is flowing, and dark as the blackbird that floats through the air-And his eyes, like the eagle's, both piercing and bright-His heart, it is fearless and great-And his arm, it is strong in the fight."1 A tale from Madagascar tells us 1 Schoolcraft, Archives of Aboriginal Knowledge, v. 612. See also Russell, * Pima Indians,' in Ann. Rep. Bur. Ethn. xxvi. 183. VOL. II 1 B THE HISTORY OF HUMAN MARRIAGE chap. 2 of a princess whose beauty fascinated all men. Many princes fought to obtain possession of her ; but she refused them all, and chose a lover who was young, handsome, courageous, and strong.1 Of the Sakalava, in the same island, we are told that when a young man wishes to obtain a wife, his qualifications are tested in the following man- -ner:-" Placed at a certain distance from a clever caster of the spear, he is bidden to catch between his arm and side every spear thrown by the man opposite to him. If he displays fear or fails to catch the spear, he is ignominiously rejected; but if there be no flinching and the spears are caught, he is at once proclaimed an accepted ' lover.' "2 Among the Western Islanders of Torres Straits " in olden days the war dance, which was performed after a successful foray, would be the most powerful excitant to a marriage- able girl, especially if a young man had distinguished himself sufficiently to bring home the head of someone he had killed." And Dr. Haddon was informed that this was one of the chief reasons for head-hunting in the past, as it still is adduced on the neighbouring mainland of New Guinea.3 So also Mr. Gomes was often told by Dyaks " that the reason why the young men are so anxious to bring home a human head is because the women have so decided a preference for a man who has been able to give proof of his bravery by killing one of the enemy."4 Con- cerning the Vonums of Formosa it is said that " a savage who has refrained from head-hunting can obtain a wife among the less attractive girls, but a savage belle would look for a husband among the young braves who had proven their valour and intrepidity."5 Among the Angami Nagas, whose women in their choice of husbands judge by " thews 1 Legudvel de Lacombe, Voyage d Madagascar, ii. 121 sqq. 2 Sibree, The Great African Island, p. 251. 3 Haddon, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 222. 4 Gomes, Seventeen Years among the Sea Dyaks of Borneo, p. 73. See also Bock, Head-Hunters of Borneo, p. 216. 6 Davidson, Island of Formosa, p. 569. See also ibid. p. 566 (Atayals) ; Muller, ' Uber die Wildenstamme der Insel Formosa,' in Zeitschr. f. Ethnol. xlii. 241. XVII SEXUAL SELECTION IN MAN 3 and sinews " and by decorations, " the younger men com- plain bitterly that the British, by stopping their raids and so preventing them from gaining marks of distinction, have made it impossible for them to get wives."1 We have previously noticed instances in which a woman selects among rival suitors him who proves most courageous or skilful to be her husband.2 Speaking of the natives on the River Darling in the interior of Eastern Australia, in whose ideas of fighting the possession of gins or wives plays a prominent part, Major Mitchell remarks that " the gins have it in their power on such occasions to evince that universal characteristic of the fair, a partiality for the brave. Thus it is, that, after a battle, they do not always follow their fugitive husbands from the field, but frequently go over, as a matter of course, to the victors."3 Woman enjoys the display of manly force even when it turns against herself. It is said that among the Slavs of the lower class the wives feel hurt if they are not beaten by their husbands ; that the peasant women in some parts of Hungary do not think they are loved by their husbands until they have received the first box on the ear; that among the Italian Camorrists a wife who is not beaten by her husband regards him as a fool.4 Dr. Havelock Ellis believes that the majority of women would probably be prepared to echo the remark made by a woman in front of Rubens's ' Rape of the Sabines,' " I think the Sabine women enjoyed being carried off like that."5 We may assume that women's instinctive appreciation of manly strength and valour is due to natural selection in more than one way. A strong and courageous man is not only a likely father of strong and courageous children, but he is also better able than a weak and cowardly man to 1 Prain, ' Angami Nagas,' in Revue coloniale internationale, v. 492. See also Hodson, ' Head-Hunting among the Hill Tribes of Assam,' in Folk-Lore, xx. 141 ; Dalton, Descriptive Ethnology of Bengal, p. 40 sq. (Nagas of Upper Assam). 2 Supra, i. ch. xiii. 8 Mitchell, Expeditions into the Interior of Eastern Australia, i. 307. 4 Ellis, Studies in the Psychology of Sex, (vol. iii.) Analysis of the Sexual Impulse, &c., p. 66 sq. 6 Ibid. (vol. iii.) Analysis of the Sexual Impulse, p. 75. B 2 4 THE HISTORY OF HUMAN MARRIAGE chap. protect his offspring. At the same time women's admira- tion of manly strength may also, as Dr. Ellis suggests, be connected with the fact that the more energetic part in physical love belongs to the man, and that the spectacle of force is " the visual expression of pressure energy."1 Manly strength is closely connected with manly beauty. It has been said that beauty in the human species is above all a feminine attribute making its appeal to men, whereas the normal woman experiences no corresponding cult for the beauty of a man.2 But this seems to imply too narrow a conception of the meaning of beauty. Physical beauty may be defined as such qualities of the body as are apt to call forth aesthetic enjoyment; but at the same time these qualities, or at least some of them, have also a tendency to stimulate the sexual instinct of the opposite sex. And this is true both of men and women. If a maiden who has to choose between Adonis and Hercules turns to the latter, she may do so, not merely because she is more attracted by strength than by beauty, but because she prefers the mascu- line beauty of Hercules to the racial beauty of Adonis. It has, further, been said that personal beauty does not act as a sexual stimulant among savages.3 But this is quite contrary to numerous statements made by competent observers. Among the Thonga of South Africa, according to M. Junod, suitors prefer pretty girls to ugly ones.4 Winwood Reade says that West African negroes often discuss the beauty of their women.5 " La beaute plastique d'une femme ne laisse pas d'emouvoir un Bambara."6 In the 1 Ibid. (vol. iv.) Sexual Selection in Man, p. 191 sq. 2 Ibid. (vol. iv.) Sexual Selection in Man, p. 189 sq. 3 Finck, Primitive Love and Love-Stories, p. 270 sqq. Mr. Finck argues (ibid. p. 274) that savages can have no sense of personal beauty because they are habitually indifferent to personal clean- liness, " the most elementary and imperative of esthetic require- ments." Against this it may be said, first, that many savages are praised for their cleanliness (see my Origin and Development of the Moral Ideas, ii. 346 sqq.), and, secondly, that uncleanliness does not prevent the enjoyment of beauty when the eye is used to it and no repulsion is felt for it. * Junod, Life of a South African Tribe, i. 183. 6 Reade, quoted by Darwin, Descent of Man, ii. 373 sq. c Henry, L'dme d'un peuple africain, p. 199. xvn SEXUAL SELECTION IN MAN 5 New Hebrides " the young men have a marked taste for beauty, and there are girls that are courted by all the young fellows of the village."1 In Tahiti Cook saw several instances in which women preferred personal beauty to interest.2 The native women of Northern Queensland are fond of a handsome face in a man ; " they take particular notice of the part of the face about the eyes, and they like to see a frank and open, or perhaps more correctly a wild, expression of the countenance."3 But although we have reason to believe that no race of men are sexually indifferent to personal beauty, the ideas of what constitutes it vary greatly. Hume says, " Beauty is no quality in things them- selves ; it exists merely in the mind which contemplates them ; and each mind perceives a different beauty."4 Yet there are certain physical characteristics that are held to be beautiful by considerable groups of men, apart from individual differences of taste. We shall see what they are, and then try to explain the stimulating influence which beauty exercises upon the sexual instinct. There is, first, an ideal of beauty which is undoubtedly common to all mankind. The full and healthy develop- ment of those visible properties which are essential to the human organism may be assumed to be universally recog- nised as indispensable to perfect beauty-physical deformity, an unsymmetrical shape of the body, apparent traces of disease or old age, being regarded as unfavourable to personal appearance. A distinction is made between mascu- line and feminine beauty, and in spite of racial differences the ideas of what constitutes these forms of beauty are fundamentally similar throughout the world. To be con- sidered really beautiful a person must be a fair specimen of his or her sex. The male organism is remarkable for the development of the muscular system, the female for that of fatty elements ; and well-developed muscles are held to improve the appear- ance of a man, and rounded forms the appearance of a 1 Speiser, Two Years with the Natives in the Western Pacific, p. 235. 8 Cook, Voyage to the Pacific Ocean, ii. 161. 3 Lumholtz, Among Cannibals, p. 213. 4 Hume, Philosophical Works, iii. 268. 6 THE HISTORY OF HUMAN MARRIAGE chap. woman. Indeed, marked obesity is often admired in her. According to von Humboldt, the natives of Guiana express the beauty of a woman by saying that " she is fat and has a narrow forehead." A traveller found that a Kirghiz's estimate of female beauty was regulated by the amount of fat, for " when dilating on the beauties of his favourite wife, he laid the greatest stress on her embonpoint."1 The Tongans are of opinion that " the perfect woman must be fat-that is most imperative."2 The Makololo women, according to Livingstone, make themselves fat and pretty by drinking a peculiar drink called boydloa;3 and among the Trarsa in the Western Sahara the women take immense quantities of milk and butter to make themselves more attractive.4 Among the Ibo of Nigeria, " when a man takes a wife, his first care is to immure her in a hut, without suffering her to take exercise, until she attains the acme of beauty, according to the Ibu taste, namely, such an amount of obesity as very materially to interfere with the faculty of locomotion."5 Among the Guanches of the Canary Islands a girl was for thirty days before her marriage kept in a reclining position and fed heartily on fattening foods, since nobody wanted a lean wife.6 The most striking secondary sexual characters of women, however, are their large hips and buttocks and their breasts ; and these are consequently regarded as important features of feminine beauty. The Chukchee lover, in praising his beloved, " will begin with her strength and stature, then state that she is fat, then mention her large pelvis, and only after that call attention to her long hair, heavy eyebrows, and ruddy face."7 The hips and thighs and bust of a Tongan belle must be colossal.8 The Warega of the Belgian 1 Spencer, Descriptive Sociology, Asiatic Races, p. 29. 2 Thomson, Savage Island, p. 203. 3 Livingstone, Missionary Travels and Researches in South Africa, p. 186. 4 Chavanne, Die Sahara, p. 454. Cf. ibid. p. 340. 5 Allen and Thomson, Narrative of the Expedition to the River Niger, in 1841, i. 238. 8 Cook, ' Aborigines of the Canary Islands,' in American Anthro- pologist, "N x ii. 479. See also Bory de St. Vincent, Essais sur les Isles Fortunes, p. 99. 7 Bogoras, Chukchee, p. 36 sq. 8 Thomson, Savage Island, p. 203. xvn SEXUAL SELECTION IN MAN 7 Congo " aiment . . . chez la femme les seins pendant forte- ment et ballott6s de droite & gauche pendant la marche ; ils sont ravis d'un post^rieur fortement preeminent et de dimensions exag£r6es. "1 So also the Kafirs2 and Hottentots 8 are charmed by their women's long and pendant breasts, which in certain tribes assume such monstrous dimensions that the usual way of giving suck, when the child is carried on the back, is by throwing the breast over the shoulder; and among the Mpongwe of Gaboon even very young girls " strive to emulate the pendant beauties of their seniors."4 Among the natives near Fort Johnston, in British Central Africa, in order to secure a prominent breast, which is admired by them, " a cord is tied round the chest by girls, at the upper margin of the gland, with the idea of making it project." 5 In Papuan folk-tales the distinguishing mark of a beautiful woman is breasts that stand up.6 Among Euro- peans, as Dr. Ellis remarks, " the importance of this region is so highly esteemed that the general rule against the exposure of the body is in its favour abrogated, and the breasts are the only portion of the body, in the narrow sense, which a European lady in full dress is allowed more or less to uncover."7 The corset also furnishes her with a method of showing off both her bosom and her hips and buttocks. At the same time it makes the waist narrower, whilst the low-bodied dress displays the full length of the neck; and a narrow waist and a comparatively long neck are also feminine characteristics. The Tongans consider that a perfect woman " must have no waist," and that a defect in this respect must be disguised with draperies.8 Among Europeans women are on an average some five inches shorter 1 Delhaise, Les Warega, p. 169 sq. * v. Weber, Vier Jahre in Afrika, i. 174; ii. 200. 3 Barrow, Account of Travels into the Interior of Southern Africa, i- 390. 4 Reade, Savage Africa, p. 74. • Stannus, 'Notes on some Tribes of British Central Africa,' in Jour. Roy. Anthr. Inst. xl. 317. • Haddon, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 28. t Ellis, op. cit. (vol. iv.) Sexual Selection in Man, p. 170. 8 Thomson, Savage Island, p. 203 8 THE HISTORY OF HUMAN MARRIAGE chap. than men,1 and they are less broad in the shoulder. Women who are very tall or broad are therefore apt to be awkward, whereas a high-built and broad-shouldered figure is regarded as an ideal of manly beauty. There is, further, a racial standard of beauty, which differs in different races of men. A flat, retreating brow seems to white men to spoil what would otherwise be a pretty face ; but " the Chinook ideal of facial beauty is a straight line from the end of the nose to the crown of the head."2 A little snub-nose may embitter the life of a European girl; but the Australian natives " laugh at the sharp noses of Europeans, and call them in their language ' tomahawk noses,' much preferring their own style of flat broad noses."3 The Tahitians frequently said to the missionary John Williams, " What a pity it is that English mothers pull their children's noses so much, and make them so frightfully long."4 We admire white teeth and rosy cheeks ; but a servant of the king of Cochin China spoke with contempt of the wife of the British ambassador because she had white teeth like a dog and a rosy colour like that of a potato flower.5 In the northern parts of China, according to 1 Ellis, Man and Woman, p. 46 sq. This is not true of all races. Speaking of the natives of King George's Sound, Cook remarks (Voyage to the Pacific Ocean, ii. 303) that " the women are nearly of the same size, colour, and form, with the men ; from whom it is not easy to distinguish them." Ellis states (Polynesian Researches, i. 81) that among the Tahitians the difference between the stature of the male and female sex is not so great as that which often prevails in Europe. Diodorus Siculus says (Bibliotheca historica, v. 32. 2) that the Gallic women were as tall as the men ; and Fritsch asserts (Die Eingeborenen Sud-Afrika's, p. 398) the same with reference to the Bushman women of South Africa. Among the Californian Shastika, according to Powers (Tribes of California, p. 244), the women are even " larger and stronger-featured, and in everyway more respectable," than the men. 2 Bancroft, Native Races of the Pacific States of North America, i. 227. Cf. Sproat, Scenes and Studies of Savage Life, p. 29 (Aht) ; Heriot, Travels through the Canadas, p. 348. 3 Palmer, ' Notes on some Australian Tribes,' in Jour. Anthr. Inst. xiii. 280 note. 4 Williams, Narrative of Missionary Enterprises, p. 539. Cf. Ellis, Polynesian Researches, i. 81. s Waitz, Introduction to Anthropology, p. 305. xvn SEXUAL SELECTION IN MAN 9 Pallas, those women are preferred who are of the Manchu type-that is, who have a broad face, high cheek-bones, a very broad nose, and enormous ears.1 In Europe also the standard of beauty varies among different peoples. " To an honest Fleming, who has never studied design," says Bombet, " the forms of Rubens's women are the most beautiful in the world. Let not us, who admire slenderness of form above everything else, and to whom the figures even of Raphael's women appear rather massive, be too ready to laugh at him. If we were to consider the matter closely, it would appear that each individual, and, con- sequently, each nation, has a separate idea of beauty."2 Alexander von Humboldt observed long ago :-" Nations attach the idea of beauty to everything which particularly characterises their own physical conformation, their natural physiognomy. Thence it results that, if nature have bestowed very little beard, a narrow forehead, or a brownish- red skin, every individual thinks himself beautiful in pro- portion as his body is destitute of hair, his head flattened, his skin more covered with ' annotto,' or ' chica,' or some other coppery-red colour."3 This view has been accepted by several later writers ;4 but, as it has been disputed by others,5 it may be well to bring together some fresh evidence, as an addition to that collected by Darwin. The Sinhalese, says Davy, who are great connoisseurs of the charms of the sex, and have books on the subject and rules to aid the judgment, would not allow a woman to be perfectly beautiful unless she had the following characteris- 1 Pallas, quoted by Prichard, Researches into the Physical History of Mankind, iv. 519. 2 Bombet, Lives of Haydn and Mozart, p. 278. For other instances of different ideas of beauty see Darwin, Descent of Man, ii. 374 sqq. ; Ellis, Psychology of Sex, (vol. iv.) Sexual Selection in Man, p. 140 sqq. 3 v. Humboldt, Personal Narrative of Travels to the Equinoctial Regions of the New Continent, iii. 236 sq. 4 Geoffroy Saint-Hilaire, Histoire des anomalies, i. 268. Darwin, Descent of Man, ii. 381. Waitz-Gerland, Anthropologic der Natur- vdlker, vi. 27. Ellis, Psychology of Sex, (vol. iv.) Sexual Selection in Man, p. 175 sqq. 6 Martineau, Types of Ethical Theory, ii. 157. Delaunay, ' Sur la beauty,' in Bull. Soc. d'Anthr. Paris, ser. iii. vol. viii. 198. THE HISTORY OF HUMAN MARRIAGE chap. 10 tics :-" Her hair should be voluminous like the tail of the peacock, long, reaching to the knees, and terminating in graceful curls ; her nose should be like the bill of the hawk, and lips bright and red, like coral on the young leaf of the iron-tree. Her neck should be large and round, her chest capacious, her breasts firm and conical, like the yellow cocoa-nut, and her waist small-almost small enough to be clasped by the hand. Her hips should be wide ; her limbs tapering ; the soles of her feet without any hollow, and the surface of her body in general, soft, delicate, smooth, and rounded, without the asperities of projecting bones and sinews." Dr. Davy adds, " The preceding is the most general external character that can be given of the Sin- halese."1 The women of Indo-European peoples are remarkable for the length of their hair. " Dans nos contr6es," Geoffroy Saint-Hilaire observes, " ces developpements ajou- tent a la beautd des femmes ; dans d'autres pays, si on les y observait, ils passeraient presque pour de lagers vices de conformation."2 " A small round face," says Castr6n, " full rosy red cheeks and lips, a white forehead, black tresses, and small dark eyes are marks of a Samoyed beauty. Thus in a Samoyed song a girl is praised for her small eyes, her broad face, and its rosy colour."8 These, as we know, are the typical characteristics of the Samoyed. Concerning the Tartar women, who generally have far less prominent noses than we in Europe are accustomed to see, Father de Rubruquis states, " The less their noses, the handsomer they are esteemed."4 In Fiji the remarkably broad occiput, peculiar to its people, is looked upon as a mark of beauty.5 Mr. Curr heard an Australian native say, " when extolling the charms of a young woman, that her skin was smooth and bright as the wood of the box-tree 1 Davy, Account of the Interior of Ceylon, p. no sq. 8 Geoffroy Saint-Hilaire, op. cit. i. 268. ' Castrdn, Nordiska resor och forskningar, i. 229. 4 de Rubruquis, ' Travels into Tartary and China,' in Pinkerton, Collection of Voyages and Travels, vii. 33. • Waitz-Gerland, op. cit. vol. vi. 543. xvn SEXUAL SELECTION IN MAN 11 off which the bark has been nearly stripped ";1 and the same authority remarks that the skin of the Australian Black is " particularly velvet-like to the touch."2 Among the modern Egyptians Lane scarcely ever saw corpulent persons, and, unlike many other African peoples, they do not admire very fat women :-" In his love-songs, the Egyptian commonly describes the object of his affec- tions as of slender figure, and small waist."3 " The negroes," says v. Humboldt, " give the preference to the thickest and most prominent lips;4 the Kalmucks to tumed-up noses ; and the Greeks, in the statues of heroes, raised the facial line from 85° to ioo° beyond nature. The Aztecs, who never disfigure the heads of their children, represent their principal divinities, as their hieroglyphical manuscripts prove, with a head much more flattened than any I have ever seen among the Caribs."5 The fashion, prevalent among many peoples, of trans- forming parts of the body affords a good illustration of their ideas about personal beauty. The Indians of North America often exaggerate the peculiarly flat shape of their forehead by an artificial flattening of it.6 The same is the case with some Australian tribes, where " this marked feature of the aboriginal physiognomy is thought to be beautiful."7 In Tahiti, Samoa, and other islands of the Pacific Ocean it has been customary from time immemorial to flatten the occiputs and to press the noses of the infants, as Gerland observes, in order to increase a national characteristic 1 Curr, Recollections of Squatting in Victoria, p. 267 sq. 2 Ibid. p. 290. 3 Lane, Manners and Customs of Modern Egyptians, i. 38, 259 n.* 4 See also Monrad, Bidrag til en Skildring af Guinea-Kysten, p. 58. 6 v. Humboldt, Political Essay on the Kingdom of New Spain, i. 154 note. For other evidence for v. Humboldt's theory see- besides Darwin, Descent of Man-Waitz, op. cit. iv. 62 sq., vi. 543» 571 •' Idem, Introduction to Anthropology, p. 305 ; Zimmermann, Die Inseln des indischen und stillen Meeres, ii. ix. 6 Macfie, Vancouver Island and British Columbia, p. 441. Heriot, op. cit. p. 348. Catlin, Last Rambles among the Indians, p. 145 sq. 7 Howitt, Native Tribes of South-East Australia, p. 742 sq. THE HISTORY OF HUMAN MARRIAGE chap. 12 which is considered beautiful.1 The same practice occurs in Sumatra, and Marsden could learn no other reason for it but that it was an improvement of beauty in the estimation of the natives.2 By all this, however, I do not mean to deny that the flattening of the head may be due to other causes besides the desire to improve the personal appearance.3 Among the Chinese small feet are considered a woman's chief attraction ; hence the feet of girls are pressed from eaily childhood. Now we know, from the measurements made by Scherzer and Schwarz, that Chinese women have by nature unusually small feet-a peculiarity which has always distin- guished them from their Tartar neighbours. And as a matter of fact the Manchu Tartars never press the feet of their daughters.4 Each race considers its own colour preferable to every other. North American Indians admire a " tawny hide."5 Among the Reindeer Chukchee the colour of the face is bronze, with intermediate tints varying from brick-red to blood-red ; and their ideal of beauty in both sexes requires that the face be " as red as blood, burning like fire." But among the Maritime Chukchee, where cases of brown or even dark-brown skin are frequent, the brownish complexion is considered to be the handsomest.6 Among the Malays " the standard of perfection in colour is virgin gold, and, as a European lover compares the bosom of his mistress to the whiteness of snow, the East Insular lover compares that of his to the yellowness of the precious metal."7 The negro " estimates a woman's beauty by her corpulence and the glossy blackness of her skin."8 The Hovas, on the other 1 Ellis, Polynesian Researches, i. 81. Angas, Polynesia, p. 272. Waitz-Gerland, op. cit. vi. 27. See also Parkinson, Dreissig Jahre in der Sudsee, p. 205 sq. 2 Marsden, History of Sumatra, p. 44 sq. 3 See Stoll, Das Geschlechtsleben in der Volkerpsychologie, p. 51 sqq. 4 Weicker, ' Die Fiisse der Chinesinnen,' in Archiv f. Anthrop. v. 149. Katscher, Bilder aus dem chinesischen Leben, p. 51. 6 Darwin, Descent of Man, ii. 377. 6 Bogoras, op. cit. p. 34. T Crawfurd, History of the Indian Archipelago, i. 23. 8 Mockler-Ferryman, British Nigeria, p. 232 n. * See also Crevaux, Voyages dans l'Amdrique du Sud, p. 39 (Bonis of Dutch Guiana, who are of negro extraction). XVII SEXUAL SELECTION IN MAN 13 hand, who are probably as a rule the lightest people in Madagascar, often put a spot of dark colour on the cheeks in order to heighten the effect of their fair complexion, of which they are very proud.1 The object of the painting of the body, so commonly practised among savages, seems sometimes to be to exaggerate the natural colour of the skin. Von Humboldt believes that this is the reason why American Indians paint themselves with red ochre and earth.2 The natives of Tanna, who have the colour of an old copper coin, usually dye their bodies a few shades darker ;3 whereas the Bomabi Islanders, who have a light copper-coloured complexion, " anoint their bodies with turmeric, in order to give them- selves a whiter appearance."4 The Javanese, when in full dress, smear themselves with a yellow cosmetic.5 And speaking of the people of a place in Maabar (Coromandel Coast), Marco Polo says :-" The children that are bom here are black enough, but the blacker they be the more they are thought of ; wherefore from the day of their birth their parents do rub them every week with oil of sesam£, so that they become as black as devils. Moreover, they make their gods black and their devils white, and the images of their saints they do paint black all over."6 We have thus found three groups of qualities which are regarded as essential for personal beauty, apart from merely individual preferences, namely: the full and proper development of the visible characteristics belonging to the human organism in general; of those peculiar to the sex ; of those peculiar to the race. We shall now consider the connection between love and beauty. It may seem strange that the same qualities as are apt to call forth sesthetic feelings and therefore are styled beautiful also have a tendency to stimulate the sexual instinct, con- 1 Sibree, op. cit. pp. m, 210. 2 v. Humboldt, Personal Narrative of Travels, iii. 237. 3 Turner, Samoa, p. 307. 4 Angas, Polynesia, p. 381 sq. Cheyne, Description of Islands in the Western Pacific Ocean, p. 105. 6 Crawfurd, op. cit. i. 23. * Marco Polo, Kingdoms and Marvels of the East, ii. 355. THE HISTORY OF HUMAN MARRIAGE chap. 14 sidering that an aesthetic feeling is essentially disinterested whereas sexual love is the very reverse. Whatever be the explanation of this combination of tendencies, it is obvious that the bodily qualities which are deemed beautiful are useful to the species, and the part they play as sexual stimu- lants may consequently be traced to the influence of natural selection. They are expressions of vitality, vigour, and health, or are closely connected with propagation. The stimulating effect of a youthful appearance in a woman has undoubtedly something to do with her capacity of bearing children, and a full development of her principal secondary sexual characters is the condition of a healthy motherhood. As for the racial type of beauty, it seems necessary to discuss somewhat more fully the biological importance of the qualities which constitute it. " In barbarous nations," says v. Humboldt, " there is a physiognomy peculiar to the tribe or horde rather than to any individual. When we compare our domestic animals with those which inhabit our forests, we make the same observation."1 Similar statements have been made by later writers.2 This striking similarity is no doubt in part due to the sameness of the conditions of life to which the members of a savage tribe are subject. Villermd and Quetelet observed that there is an inequality of stature not only between the inhabitants of towns on the one hand and those of the country on the other, but also between townsfolk of different professions.3 Yet the uniformity of the tribal type must also have another origin. As Waitz pointed out, this type " is always that which harmonises with the soil and the climate, and the external relations in which the respective peoples live."4 We may assume that a certain kind of constitution is best suited for certain conditions of life, and that every considerable deviation from it must 1 v. Humboldt, Political Essay on the Kingdom of New Spain, i. 141. 2 Lawrence, Lectures on Physiology, &c., p. 474. Godron, De I'espbce et des races, ii. 310. See also ibid. ii. 175 sq. 8 Quetelet, Treatise on Man, p. 59 sq. Cf. Ranke, Der Mensch, ii. 126, 129 sq. 4 Waitz, Introduction to Anthropology, p. 86. XVII SEXUAL SELECTION IN MAN 15 perish in the struggle for existence in a state where natural selection is constantly at work and physical qualities are of the first importance. Abnormalities which may flourish in a highly civilised society are therefore doomed to perish and can never become permanent in a community of savages. Now it may seem strange that all the characteristics, however slight, in which the various races of men differ from each other should harmonise with particular surroundings and conditions of life to the exclusion of others. But it should be remembered that a fuller knowledge than that which we possess at present might teach us that qualities which appear useless are not really so. We know the utility of some special characteristics, and that of others may at least provisionally be assumed. It is certain that the physio- logical functions of persons who quit their native land and settle in a wholly different region must undergo a change if the new conditions are not to have injurious effects. More- over, many bodily structures are so intimately related that when one part varies others vary also, though we may be quite unable to assign any reason why this should be the case. Savage men are generally distinguished by relatively large jaws, which are of use in a state of nature, where food is often hard and tough, where the jaws have to perform the functions of knife and fork, and where the teeth occasionally serve as implements. This racial peculiarity, which is in fact only a mark of low civilisation, is thus easily accounted for by the law of natural selection. The less man, with advancing civilisation, was in want of large and strong jaws the greater was the chance for individuals bom with smaller jaws to survive ; and a race with comparatively small jaws gradually arose in consequence. Indeed, Virchow has shown that the prognathous type is inconsistent with the full development of the brain.1 But large jaws are accompanied by a lateral jutting-out of the cheek-bones, which is also characteristic of the lower races. And so are, as Spencer observed, certain other facial peculiarities, such as depres- 1 Virchow, Untersuchungen uber die Entwickelung des Schddel- grundes, p. 121. 16 THE HISTORY OF HUMAN MARRIAGE chap. sion of the bridge of the nose, forward opening of the nostrils, widespread alae, and a long and large mouth, which co-exist with large and protuberant jaws and great cheek-bones.1 Various peculiarities of the skull are coincident with arrested cerebral development and correlated to each other. " The characters observed in the skulls of the lower races," says Schaaffhausen, " namely, a narrow and low frontal bone, a short sagittal suture, a low temporal squama, a short occipital squama, the upper margin of which forms a flat arch, are therefore to be considered as approximations to the animal form, and they stand to each other in organic connection."2 As for the colour of the skin, which is one of the most conspicuous of racial characteristics, it is quite impossible to assign any definite reason why one race is white, another black, brown, or yellow.3 Nobody has yet been able to prove that the colour of the skin is of any direct use to man. Yet we know that there exists an intimate connection between the colour of the skin and the physiological functions of the body on the one hand, and between the latter and the conditions of life on the other. For this reason the altera- tion in the customary physiological functions called acclimatisation may be connected with some change of colour not directly depending upon the influence of the sun. Dr. Mayer observed that a European at the tropics loses his rosy complexion, the difference in colour between arterial and venous blood being strikingly diminished on account of the smaller absorption of oxygen which results from the feebler process of combustion.4 It is asserted 1 Spencer, Essays, ii. 153 sq. 2 Schaafihausen, ' On the Primitive Form of the Human Skull, in Anthropological Review, vi. 416. 3 Professor Ripley observes (Races of Europe, p. 61) that a consideration of all the races of the earth in general shows no corres- pondence whatever of the colour of the skin with the isothermal lines. There is no reason to suppose, for example, that the blackness of the negro is the direct effect of heat. See also Kohlbrugge, ' Der Einfluss des Tropenklimas auf den blonden Europaer,' in Archiv filr Rassen- und Gesellschafts-Biologie, vii, 566; Schwalbe, ' Die Hautfarbe des Menschen,' in Mitteil. Anthropol. Ciesellsch. Wien, xxxiv. 332. 4 Mayer, Die Mechanik der Wdrme, p. 9$. XVII SEXUAL SELECTION IN MAN 17 that negroes in a colder climate have undergone a change which has left them a shade lighter in complexion.1 A Baghirmi boy whom Gerhard Rohlfs brought to Germany even changed his colour from " deep black to light brown," after a two years' residence in that country.2 Klinkosch mentions the case of a negro who lost his blackness and became yellow ; and Caldani states that a negro who was a shoemaker at Venice was black when brought there as an infant, but gradually became lighter and ultimately got the hue of a person suffering from a slight jaundice.3 On the other hand, we are told of an Englishman who long led the life of a native in the jungle of Southern India and during this time acquired, even on the clothed portions of his body, a skin as brown as that of a Brahman.4 These latter statements, if accurate, certainly refer to quite exceptional cases; but as we know that certain organisms are much better able than others to undergo the change which constitutes acclimatisation, we have no positive reason to doubt that this power may sometimes be exceed- ingly great. The changes involved in acclimatisation have reference to various parts of the body which differ among different races. " In every respect except of temperature," says Professor Ripley, " it appears that the first effects of a sojourn in the tropics is to induce symptoms which point toward the peculiarities of the native type. Thus the increase in the size of the liver indicates the operation of those causes which have finally made the negro's liver normally larger than that of the European."5 It is said that the curly hair of the European tends to become straight in America-like the hair of an Indian ; that in North America and in New South Wales children of European parents are apt to become tall and lean ; and that there is a contrary tendency among European colonists at the Cape to grow fat-which reminds us of the steatopygy of the 1 Tylor, Anthropology, p. 86. 2 Rohlfs, ' Henry Noel von Bagermi,' in Zeitschr. f. Ethnol. iii. 255. 3 Reade, Savage Africa, p. 526. 4 Peschel, Races of Man, p. 92. 6 Ripley, op. cit. p. 577. VOL. Il C THE HISTORY OF HUMAN MARRIAGE chap. 18 native women.1 Dr. Boas even claims to have proved that the head form, which has always been considered as one of the most stable and permanent characteristics of human races, in course of time undergoes certain changes due to the transfer of the races of Europe to American soil.2 The bodily changes, however, which men have to undergo in the process of acclimatisation are often too great to be endurable. As Dr. Felkin remarks, Europeans are almost incapable of forming colonies in the tropics ;3 and with few exceptions they have been unable to rear a sound progeny there in marriage with white women.4 It is gener- ally believed that in India a regiment consisting of eight hundred men loses within ten years more than seven hundred.5 It is a common report that of a third generation of pure Europeans in India children are only occasionally met with, and that they never reach the age of puberty ;6 hence English parents are anxious to send their children to Europe when five or six years old, as they are otherwise much Hable to succumb.7 In Central America, according to Squier, it is the concurrent testimony of all intelligent and observing men that the pure whites are there not only relatively but absolutely decreasing in numbers, whereas the pure Indians are rapidly increasing and the Ladinos are more and more approximating to the aboriginal type.8 1 Waitz, Introduction to Anthropology, p. 53 sq. Cf. de Quatre- fages, Human Species, p. 254. 2 Boas, Changes in Bodily Form of Descendants of Immigrants, p. 7. Idem, ' Changes in the Bodily Form of Descendants of Immi- grants,' in American Anthropologist, N.S. xiv. 562. 3 Felkin, ' Introductory Address to a Course of Lectures on Diseases of the Tropics and Climatology,' in Edinburgh Medical Journal, vol. xxxi. pt. ii. 852. See Ripley, op. cit. p. 584 sqq. 4 Joest, * Reise in Afrika im Jahre 1883,' in Verhandl. Berliner Gesellsch. Anthrop. 1885, p. 475. Cf. Peschel, op. cit. p. 19 sq. 5 Sprenger, ' Acclimatisationsfahigkeit der Europaer in Asien, in Verhandl. Berliner Gesellsch. Anthrop. 1885, p. 377. 6 Pouchet, Plurality of the Human Race, p. 92. Virchow, ' Acclimatisation,' in Verhandl. Berliner Gesellsch. Anthrop. 1885, p. 213. See also Idem, ' Rassenbildung und Erblichkeit,' in Festschrift fur Adolf Bastian, p. 15. Semler, quoted by Joest, in Verhandl. Berliner Gesellsch. Anthrop. 1885, p. 475 note. 8 Squier, States of Central America, p. 56. XVII SEXUAL SELECTION IN MAN 19 We may thus take for granted that racial peculiarities stand in some connection with the external circumstances in which the various races live. It will perhaps be objected that we meet with native tribes of various types on the same degree of latitude, and under the same climatic condi- tions.1 But we must remember that it is often impossible to decide whether the conditions of life are exactly the same ; that intermixture of blood has caused a great confusion of racial types ; and that all peoples have arrived at their present localities after more or less extensive migrations. We may be sure that some characters have been preserved from earlier times when the race lived in other circumstances, and that the higher its degree of civilisation the less likely it would be to lose the stamp impressed upon it.2 It is, however, exceedingly doubtful whether racial differences are so directly the result of external influences as they have often been assumed to be-that is, whether they are the inherited effects of conditions of life to which previous generations have been subject. They cannot be so unless acquired characters may be transmitted from parent to offspring, and this, as is well known, is emphati- cally disputed by a large school of biologists. Nor does the theory here advocated presuppose the heredity of acquired characters. It seems most probable that mankind was originally restricted to a comparatively small area and was then as homogeneous as every animal and vegetable species under similar conditions. In the struggle for exist- ence the intellectual faculties of man were developed, and this made it possible for him to spread practically over the whole earth. His organism, however, had to undergo certain changes ; but we are not aware that these modifica- tions were transmitted to the descendants. All that we know is that the children bom were not exactly like each other, and that those who happened to vary most in accord- 1 Cf. Schaaffhausen, ' Darwinism and Anthropology,' in Anthro- pological Review, vol. vi. p. cviii sq. 2 M. ^lisde Reclus (quoted by de Quatrefages, op. cit. p. 255) makes a curious mistake when he asserts that, at the end of a given time, whatever be their origin, all the descendants of whites or of negroes who have immigrated to America will become redskins. C 2 20 THE HISTORY OF HUMAN MARRIAGE chap. ance with the new conditions of life as a rule survived and became ancestors of following generations. The congenital characters which enabled them to survive were of course transmitted to their offspring, and thus, through natural selection,1 races would gradually arise, the members of each of which would have as hereditary dispositions the same peculiarities as those which, to a certain extent, may be acquired through acclimatisation, but then only for the individual himself, not for his descendants. We can thus 1 Wallace (Contributions to the Theory of Natural Selection, Essay ix.) was, so far as I know, the only investigator who, before this chapter was originally written, had tried to explain the origin of human racial distinctions by the principle of natural selection. Darwin, on the other hand, maintained that they were due to sexual selection in accordance with different standards of beauty. " Let us suppose," he says (op. cit. ii. 403 sq.), " the members of a tribe, practising some form of marriage, to spread over an un- occupied continent, they would soon split up into distinct hordes, separated from each other by various barriers, and still more effec- tually by the incessant wars between all barbarous nations. The hordes would thus be exposed to slightly different conditions and habits of life, and would sooner or later come to differ in some small degree. As soon as this occurred, each isolated tribe would form for itself a slightly different standard of beauty ; and then unconscious selection would come into action through the more powerful and leading men preferring certain women to others. Thus the differences between the tribes, at first very slight, would gradually and in- evitably be more or less increased." This theory-that racial differences are due to sexual selection-obviously presupposes either that the human organism is alike well fitted to any climate and natural conditions ; or that no correlation exists between the visible parts of the body and its functions. Otherwise, of course, little effect could be produced through the preference given to certain individuals ; for in a savage state, where celibacy is an ex- ception, those men and women whose constitutions were best suited to the conditions of life would, in any case, in the end, determine the racial type. The preference for the best representatives of the race has no doubt contributed to keeping the racial type pure, but to me it seems that this preference itself, like the racial differen- tiation, owes its origin to natural selection.-Darwin also thought that the differences in external appearance between man and the lower animals are to a certain extent due to sexual selection. The chief character of the human species which he proposes to account for in this way is the general hairlessness of the body. " No one supposes," he says (op. cit. ii. 410), " that the nakedness of the XVII SEXUAL SELECTION IN MAN 21 understand how, for example, the children of a negro are black,1 as the black colour is the correlative of certain physiological, processes favourable to existence in the country of their race. They survive, whilst the children of Europeans who have emigrated to the tropics are carried off in great numbers, even though their parents have suc- ceeded in undergoing the functional modifications which accompanied the change of abode. This explanation of racial differences seems the more skin is any direct advantage to man ; his body therefore cannot have been divested of hair through natural selection." Wallace, again, who believed (op. cit. p. 359) that " a superior intelligence has guided the development of man in a definite direction," considered (ibid. p. 348 sq.) that the hairless condition of the skin comes under this head. Belt's experience in tropical countries had led him to the conclusion (The Naturalist in Nicaragua, p. 209) that, in such regions at least, there is one serious drawback to the advantage of having the skin covered with hair : "it affords cover for para- sitical insects, which, if the skin were naked, might more easily be got rid of." It is, however, impossible to believe that this advantage could have produced a hairless species. It seems to me that the real problem is not whether a hairless skin is, or has been, useful to man, but how he has been able to survive in spite of it. In con- formity with the law of variation some individuals were less hairy than others ; and when man had invented the art of making fire, and the idea of covering himself to secure protection from cold had occurred to his mind, this was no serious disadvantage in the struggle for existence. Hence natural selection ceased to operate in the matter, and a hairless race gradually arose. We find the same principle at work in various other ways. Civilised man does not need such keen vision as savages, and in consequence so many of us are short-sighted. Collins, who wrote sixty years before The Origin of Species, makes the following observation regarding the natives about Botany Bay and Port Jackson in New South Wales (Account of the English Colony in New South Wales, i. 553 sq.) : " Their sight is peculiarly fine, indeed their existence very often depends upon the accuracy of it; for a short-sighted man . . . would never be able to defend himself from their spears, which are thrown with amazing force and velocity." 1 A negro child is not born black, but becomes so after some shorter or longer time (Darwin, op. cit. ii. 342 ; Caillid, Travels through Central Africa, i. 351). The children of dark races are usually fairer than the adults (Darwin, op. cit. ii. 342 ; Moseley, ' Notes on the Inhabitants of the Admiralty Islands,' in Jour. Anthr. Inst. vi. 385). 22 THE'HISTORY OF HUMAN MARRIAGE chap. acceptable when we take into consideration the immense period which has elapsed since man began to spread over the earth, and the slow and gradual change of abodes. He was not at once moved from the tropics to the polar zones, or from the polar zones to the tropics, but had to undergo an indefinitely long chain of adaptive processes. Thus were gradually established such racial differences as those which distinguish a European from a negro and an Australian from a redskin. We can understand, then, why racial beauty, or the full development of racial characters, acts as a sexual stimulant. It indicates vitality and health. Typical beauty is thus in any case, whether referring to qualities peculiar to the species or to the sex or to the race, the outward manifesta- tion of physical perfection or fitness. And the develop- ment of the instinctive preference for it is evidently within the power of natural selection. This explanation of the connection between love and beauty is very different from that given by Darwin. " The men of each race," he says, " prefer what they are accus- tomed to ; they cannot endure any great change ; but they like variety, and admire each characteristic carried to a moderate extreme." In the fashions of our own dress, he argues, we see exactly the same principle and the same desire to carry every point to an extreme.1 Now it is true that the natural dislike of anything strange may have a repelling effect upon the sexual instinct; but this does not explain why beauty is positively attractive, nor why a face which is " unusually beautiful " is more attractive than an ordinary one. It is also true that man wants variety. Sometimes in one, sometimes in another way, he changes his dress in order to attract atten- tion or to charm. The fashions of savages are probably more permanent than ours ;2 but the extreme diversity of 1 Darwin, op. cit. ii. 384. 2 Speaking of the Rejangs of Sumatra, Marsden says (op. cit. p. 206), " The quick, and to them inexplicable, revolutions of our fashions are subject of much astonishment, and they naturally conclude that those modes can have but little intrinsic merit which we are so ready to change." XVII SEXUAL SELECTION IN MAN 23 ornaments with which many uncivilised peoples bedeck themselves shows their emulation to make themselves attractive by means of new enticements. " Each of the Outanatas (in New Guinea)," says Mr. Earl, " seemed desirous of ornamenting himself in some way different from his neighbour " j1 and with regard to the Pacific Islanders John Williams remarks that the inhabitants of almost every group " have their peculiar ideas as to what constitutes an addition to beauty."2 But it is impossible to believe that the different races' ideals of personal beauty are simply due to the capriciousness of taste. Were this the case, as Darwin suggests, it would seem that the men of each race would admire all sorts of piquant variations in the appear- ance of their women and not only each characteristic point " carried to a moderate extreme." Personal appearance may stimulate the sexual instinct not only on account of its physical beauty but as an expres- sion of mental qualities. The passion which unites the sexes is perhaps the most compound of all human feelings. Spencer thus sums up the masterly analysis he has given of it:-" Round the physical feeling forming the nucleus of the whole, are gathered the feelings produced by personal beauty, that constituting simple attachment, those of reverence, of love of approbation, of self-esteem, of property, of love of freedom, of sympathy. These, all greatly exalted, and severally tending to reflect their excitements on one another, unite to form the mental state we call love."3 Hence emotional, moral, and intellectual qualities may, by evoking affection, approbation, or admiration, indirectly act as sexual stimulants. Their influence in this respect has greatly increased in the course of mental evolution. Yet how- ever much influenced by the appreciation of mental qualities, the connection between the sexual impulse and affection in the human race seems, as has been noticed before,4 to have been an inheritance from an ape-like progenitor. 1 Earl, Papuans, p. 48. 2 Williams, Missionary Enterprises in the South Sea Islands, P- 538 sq. 3 Spencer, Principles of Psychology, i. 488. 4 Supra, i. 70. THE HISTORY OF HUMAN MARRIAGE chap. 24 Where the generative power is restricted to a certain season, it must be some degree of affection that induces male and female to remain with one another beyond that season till after the birth of the offspring. The tie which unites them must be an instinct developed through natural selection, and at the bottom of this instinct there is undoubtedly the tendency to feel some attachment to a being which has been the cause of pleasure-in this case sexual pleasure. Such a feeling would originally have induced the sexes to remain together and the male to protect the female even after the sexual desire was gratified ; and if procuring great advantage to the species in the struggle for existence, conjugal attach- ment would naturally have developed into a specific characteristic. At the lower stages of human development, however, conjugal affection seems to be much inferior to the tender feelings with which parents embrace their children T and among several peoples it is represented as distinctly feeble, if not absent. Concerning the Hovas of Madagascar Sibree says that until the spread of Christianity there was " no lack of strong affection between blood-relations- parents and children, brothers and sisters, grandparents and grandchildren," but that the idea of love between husband and wife was hardly thought of.2 Among the Kru people on the Grain Coast " a woman is always expected to retain a stronger attachment to her blood relations than to her husband and his connections."3 Of the Negroes of Angola we are told that " their passion is purely of an animal description, unaccompanied by the least sympathetic affec- tions of love or endearment";4 and among various other negro peoples love, as we understand the term, is said to be altogether unknown.5 The Beni-Amer consider it dis- 1 This statement has been quoted with approval by Grosse, Beginnings of Art, p. 245. 2 Sibree, op. cit. p. 250. 3 Wilson, Western Africa, p. 114. 4 Monteiro, Angola and the River Congo, i. 243. 5 Zoller, Forschungsreisen in der deutschen Colonie Kamerun, ii. 60, 69 (Dualla). Ellis, Tshi-speaking Peoples of the Gold Coast, p. 285. Mockler-Ferryman, British Nigeria, p. 232. Duncan, Travels in Western Africa, i. 79. XVII SEXUAL SELECTION IN MAN 25 graceful for a wife to show any affection for her husband.1 The Chittagong Hill tribes, according to Lewin, have " no idea of tenderness, nor of chivalrous devotion," and regard marriage as merely a convenient and animal connection.2 In the island of Ponape, of the Carolines, according to Finsch, love in the European sense does not exist.3 As regards the Eskimo of Newfoundland, Heriot asserts that " they treat their wives with great coldness and neglect, but their affection towards their offspring is lively and tender."4 In Greenland a man thought nothing of beating his wife, but it was a heinous offence for a mother to chastise her children.5 Almost the same is said of the Eskimo of Norton Sound 6 and the Kutchin.7 Morgan states that the refined passion of love is unknown to the North American Indians in general.8 Such statements should not mislead us. However different the love of a savage may be from that of a civilised man, we discover in it traces of the same ingredients. And we hear of many savages who are by no means strangers to conjugal affection. Concerning the South African Bushmen, Chapman says that there is love in all their marriages.8 Among the Hottentots, according to Le Vaillant, marriages are founded on reciprocal inclination, and " a lively and mutual affec- tion usually subsists between the wedded pair."10 Of the peoples of the Upper Congo we are told that love is ennobled by a certain poetry.11 Among the Warega " l'amour existe ; il porte le nom de kilanga."12 Tessmann states that genuine 1 Munzinger, Ostafrikanische Studien, p. 325. 2 Lewin, Wild Races of South-Eastern India, p. 345. 3 Finsch, ' Ueber die Bewohner von Ponap6,' in Zeitschr. f. Ethnol. xii. 317. * Heriot, op. cit. p. 25. 5 Egede, Description of Greenland, p. 144. 6 Dall, Alaska, p. 139. 7 Jones, ' Kutchin Tribes,' in Smithsonian Report, 1866, p. 326. 8 Morgan, Systems of Consanguinity and Affinity, p. 207 note. Cf. Schoolcraft, Archives of Aboriginal Knowledge, v. 272 (Creeks). 9 Chapman, Travels in the Interior of South Africa, i. 258. 10 Le Vaillant, Travels from the Cape of Good-Hope, into the Interior Parts of Africa, ii. 68. 11 Johnston, River Congo, p. 423. 12 Delhaise, cp. cit. pp. 167, 170. 26 THE HISTORY OF HUMAN MARRIAGE chap. love-matches are by no means infrequent among the West African Pangwe, and that they have love as understood by the peoples of Europe.1 Schweinfurth assures us that the Niam-Niam display an affection for their wives which is unparalleled among other natives of an equally low grade.2 Among the Touareg there is a touch of almost chivalrous sentiment in the relations between men and women.3 The Dards know and speak of the existence of " pure love," Jak Ashiqi, and their love songs show sufficiently that they are capable of a deeper than mere sexual feeling.4 The Hos are represented as good husbands and wives, and although they have no terms in their own language to express the higher emotions " they feel them all the same."5 The missionary Jellinghaus found tokens of affectionate love between married people among the Munda Kols,6 Mr. Fawcett among the Savaras,7 Mr. Man among the Anda- manese,8 Sir Spenser St. John among the Sea Dyaks.9 Concerning the Papuans of the former German New Guinea, Neuhauss observes that the question whether there is hearty affection between lovers must decidedly be answered in the affirmative, and that their tales bear testimony to the existence of the same sentiment in the past.10 Among those of Bogadjim conjugal affec- tion is very pronounced in the case of both partners.11 In Mailu, British New Guinea, mutual affection undoubtedly obtains between man and wife, although Dr. Malinowski could find no evidence of what may be called romantic 1 Tessmann, Die Pangwe, ii. 256. 2 Schweinfurth, Im Herzen von Afrika, i. 510. 8 Chavanne, Die Sahara, p. 208 sq. 4 Leitner, Results of a Tour in Dardistan, &c., iii. 35 sq. 5 Dalton, Descriptive Ethnology of Bengal, p. 206. 9 Jellinghaus, ' Sagen, Sitten und Gebrauche der Munda-Kolhs in Chota Nagpore,' in Zeitschr f. Ethnol. iii. 369. 7 Fawcett, ' Saoras of Madras,' in four. Anthr. Soc. Bombay, i. 219. 8 Man, ' Aboriginal Inhabitants of the Andaman Islands,' in Jour. Anthr. Inst. xii. 327. 9 St. John, Life in the Forests of the Far East, i. 54 sq. 10 Neuhauss, Deutsch Neu-Guinea, i. 160. 11 Hagen, Unter den Papua's, p. 243. XVII SEXUAL SELECTION IN MAN 27 sentiment j1 and among the Western Islanders of Torres Straits " at the present time much affection exists, between husband and wife."2 In Hawai "instances of love and affection were not uncommon."3 In Tonga, according to Mariner, most of the women were much attached to their husbands.4 In Samoa stories of affectionate love between husband and wife are preserved in song ;5 and the Maori have love stories not destitute of beauty.6 Among the Australian natives married people are often much attached to each other, and continue to be so even when they grow old.7 Concerning the aborigines of Victoria, Daniel Bunce says it is an error to suppose that there exists no settled love or lasting affection between the sexes ;8 and a love story among those natives is said to be very similar to a European one.9 Among the Narrinyeri Mr. Taplin has known as well-matched and loving couples as he has in Europe ;10 and among the natives of the River Darling husband and wife rarely quarrel, and " show much affection for each other in their own way."11 Among the Eskimo of Iglulik and Winter Island " young 1 Malinowski, ' Natives of Mailu,' in Trans. Roy. Soc. South Australia, xxxix. 571 sq. 3 Haddon, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 229. 3 Jarves, History of the Hawaiian Islands, p. 43. * Mariner, Account of the Natives of the Tonga Islands, ii. 171 sq. 8 Turner, Samoa, p. 102. 6 Thomson, Story of New Zealand, i. 179 sqq. 7 Brough Smyth, Aborigines of Victoria, ii. 283. Dawson, Australian Aborigines, p. 37. Bonwick, in Jour. Anthr. Inst. xvi. 205. Hill and Thornton, Notes on the Aborigines of New South Wales, p. 7. Lang, Queensland, p. 338 ; Idem, Cooksland in North- Eastern Australia, p. 394. Lumholtz, Among Cannibals, p. 213 sq. Waitz-Gerland, op. cit. vi. 775, 781. Malinowski, The Family among the Australian Aborigines, pp. 68 sqq. (particularly p. 83), 307. 8 Brough Smyth, op. cit. i. 29. 9 Oberlander, ' Die Eingeborenen der australischen Kolonie Victoria,' in Globus, iv. 278 sq. 10 Taplin, ' Narrinyeri,' in Woods, Native Tribes of South Australia, p. 12. 11 Bonney, ' Some Customs of the Aborigines of the River Darling,' in Jour. Anthr. Inst. xiii. 129. 28 THE HISTORY OF HUMAN MARRIAGE chap. couples are frequently seen rubbing noses, their favourite mark of affection, with an air of tenderness."1 The Takulli, according to Harmon, are remarkably fond of their wives.2 James says that among the Omaha marriage, on the part of the man, generally seems to be the result of love, and that " their connubial attachments are often very strong,"3 And Catlin-contrary to Morgan-goes even so far as to deny that the North American Indians are " in the least behind us in conjugal, in filial, and in paternal affection."4 Brett asserts that among the natives of Guiana instances of conjugal attachment are very frequent,5 although, according to Appun's account, onlyphysical love is known among them.6 Affection between man and wife has been found among other South American tribes,7 including those of Tierra del Fuego.8 Although we may assume that, generally speaking, affec- tion has come to play a more prominent part in the relations between the sexes in proportion as love has been influenced by mental qualities, we must not suppose that advancement in civilisation has at every step been favourable to its development. In Eastern countries with their ancient civilisation there exists even now comparatively little of that tenderness towards the woman which is the principal charm of our own family life. In China up to recent times 1 Lyon, Private Journal during the Voyage of Discovery under Captain Parry, p. 353. Cf. Nansen, First Crossing of Greenland, ii. 325 sq. 2 Harmon, Journal of Voyages and Travels in the Interior of North America, p. 292. 3 James, Account of an Expedition from Pittsburgh to the Rocky Mountains, i. 234. 4 Catlin, Illustrations of the Manners, Customs, and Condition of the North American Indians, i. 121. 6 Brett, Indian Tribes of Guiana, pp. 98, 351. 8 Appun, ' Die Indianer von Britisch-Guayana,' in Das Ausland, xliv. 832 sq. 1 Azara, Voyages dans I' Amdrique meridionals, ii. 44. Mantegazza, Rio de la Plata e Tenerife, p. 456. 8 Weddell, Voyage towards the South Pole, p. 156. Hyades and Deniker, Mission scientifique du Cap Horn, vii. 238 sq. Hyades, ' Ethnographic des Fu6giens,' in Bull. Soc. d'Anthr. Paris, ser. iii. vol. x. 334. Gallardo, Tierra del Fuego-Los Onas, p. 134. XVII SEXUAL SELECTION IN MAN 29 it was considered good form for a man to beat his wife, and we are told that if the Chinaman of humbler rank spared her a little he did so only in order not to come under the necessity of buying a successor.1 In a book containing the cream of all the moral writings of the Chinese, and intended chiefly for children, we read :-" Brothers are like hands and feet. A wife is like one's clothes. When clothes are worn out, we can substitute those that are new. When hands and feet are cut off, it is difficult to obtain substitutes for them."2 Whilst the Vedic singers know no more tender relation than that between the husband and his willing, loving wife, who is praised as " his home, the darling abode and bliss in his house,"3 it is said that sincere mutual friendship is rarely met with in the families of the modem Hindus. " It is in vain," Dubois remarks, " to expect, between husband and wife, that reciprocal confidence and kindness which consti- tute the happiness of a family. The object for which a Hindu marries is not to gain a companion to aid him in enduring the evils of life, but a slave to bear children and be subservient to his rule."4 The love of which the Persian poets sing has either a symbolic or a very profane meaning.5 Among the Arabs, says Burckhardt, " the passion of love is, indeed, much talked of by the inhabitants of towns; but I doubt whether anything is meant by them more than the grossest animal desire."6 Mr. Finck remarks that in the whole of the Bible there is not a single reference to romantic love ;7 and from the old Jewish point of view it is certainly surprising to find Sirach putting the companion- ship of a wife not only above that of a friend, but even above children.8 In Greece in the historic age the man recognised in the woman no other end than to minister to his pleasure 1 Katscher, Bilder aus dem chinesischen Leben, p. 58 sq. 2 Indo-Chinese Gleaner, i. 164. 3 Kaegi, Rigveda, p. 15. 4 Dubois, Description of the Character, &c. of the People of India, p. 109. B Polak, Persien, i. 206. 6 Burckhardt, Notes on the Bedouins and Wahdbys, p. 155, 7 Finck, Romantic Love and Personal Beauty, p. no. 8 Ecclesiasticus, xl. 19, 23. Cf. Montefiore, Hibbert Lectures on the Religion of the Ancient Hebrews, p. 491. THE HISTORY OF HUMAN MARRIAGE chap. 30 or to become the mother of his children.1 The love of women was the offspring of the common Aphrodite, who " is of the body rather than of the soul," whereas those who were inspired by the heavenly Aphrodite loved intelligent beings whose reason was beginning to be developed, much about the time at which their beards began to grow.8 Both in the East and in Greece progress in civilisation widened the gulf between the sexes and tended to alienate husband and wife, because the higher culture became almost exclusively the prerogative of the men. Another factor leading to conjugal apathy is the custom which compels the women before marriage to live strictly apart from the men. In China8 and Muhammadan countries it often happens that the parties have not even seen each other till the wedding day. In Greece Plato urged in vain that young men and women should be more frequently permitted to meet one another, so that there should be less enmity and indifference in the married life,4 whilst Plutarch hopes that love will come after marriage.5 Moreover, the little regard in which women have been held by the great religions of the world6 must also have influenced the emotions felt towards them. Conjugal affection is, of course, quite possible even where the marriage was not concluded from love. Mr. Hall's assertion as to the Eskimo visited by him, that " love- if it come at all-comes after the marriage,"7 is true of many peoples. Murdoch says of the Point Barrow Eskimo, " As far as we could learn, the marriage relation was entered upon generally from reasons of interest or convenience, with very little regard for affection, as we understand it, though there often appeared to be a warm attachment between married people."8 In Australian tribes the bride 1 Dickinson, Greek View of Life, p. 159. Dollinger, The Gentile and the Jew, u. 234. 2 Plato, Symposium, p. 181. 8 Katscher, op. cit. pp. 71, 84. 4 Plato, Leges, vi. 771 sq. 8 Plutarch, De virtute morali, ch. 3. 6 See Westermarck, Origin and Development of the Moral Ideas, i. 663 sqq. 7 Hall, Arctic Researches, p. 568. 8 Murdoch,' Ethnological Results of the Point Barrow Expedition, in Ann. Rep. Bur. Ethnol. ix. 410. XVH SEXUAL SELECTION IN MAN 31 may be dragged from her home against her will, and if fears are entertained that she will endeavour to escape a speay may be thrust through her foot or her leg; but a kind husband will " ultimately evoke affection, and fidelity and true love are not rare in Australian families."1 Among the Baganda " love did not enter into marriage contracts, though men and women became attached to one another, and a woman would cling to her husband in danger, and also mourn for him, when he died, with great sincerity."2 Parental feelings and the joint care of the offspring tend to make man and wife more attached to each other. Prescott says of the Dakota that " as children increase, the parents appear to be more affectionate " ;3 and of the father and mother in a Hottentot family we read that " the love of their children renders them every day more dear and neces- sary to each other."4 Where the person who marries has no choice in the matter or no previous knowledge of the other partner, affection or at least mutual affection cannot, of course, have anything to do with the conclusion of the marriage. But even where there is perfect liberty of choice and the unmarried of both sexes have ample opportunities to become acquainted with one another, it very often happens that the selection of a wife or a husband is influenced neither by affection nor by beauty nor by any other sexual stimulant, but by factors of a totally different character. We have hitherto dealt only with the poetry of sexual selection-love ; now something is to be said of its prose-dry calculation. We have previously noticed man's desire for offspring, which is particularly strong in the savage state and among peoples of archaic culture. A woman is valued not only as a wife but as a mother; and the respect in which she is held is often proportionate to her fecundity,5 a barren wife 1 Brough Smyth, op. cit. vol. i. p. xxiv. 2 Roscoe, Baganda, p. 87. 3 Schoolcraft, op. cit. iii. 236. 4 Le Vaillant, op. cit. ii. 68 sq. 6 Waitz, op. cit. ii. 121 (Africans). Schweinfurth, op. cit. ii. 31 (Niam-Niam). Du Chaillu, Explorations and Adventures in Equatorial Africa, p. 335. Sauer, Account of Expedition to the Northern Parts of Russia performed by Billings, p. 176 (Kaniagmiut); &c. 32 THE HISTORY OF HUMAN MARRIAGE chap. being despised as an unnatural and useless being.1 In Angola, according to Livingstone, in the native dances, when any one may wish to deride another, a line is intro- duced in the accompanying song to the effect that " So and so has no children, and never will get any " ; and the offended woman feels the insult so keenly that it is not un- common for her to rush away and commit suicide.2 Among the Creeks a man always calls his wife his son's mother ;3 and among the Todas, in addressing a man with the casual question, " Are you married ? " the ordinary way of putting it would be to say, "Is there a son ? "4 It is obvious, then, that fecundity must be one of the qualities which a man most eagerly requires of his bride. We have seen that pre-nuptial relations frequently have the character of a trial by which the lover ascertains that the woman will gratify his desire for offspring, and that in such a case marriage is not concluded before the birth of a child or until there are signs of pregnancy. In other cases, again, offspring is secured by marrying a woman who has conceived by another man. Such a woman may be preferred both because she has proved her capacity of bearing children and because she brings with her offspring already bom.5 Moreover, a wife is chosen with reference to her ability to perform the work incumbent upon her.6 Thus among the 1 King, ' Intellectual Character of the Esquimaux,' in Jour. Ethn. Soc. London, i. 150. Waitz, op. cit. iii. 100 (North American Indians). Bosman, ' New Description of the Coast of Guinea,' in Pinkerton, Collection of Voyages and Travels, xvi. 527 (Negroes of Benin). Emin Pasha in Central Africa, p. 209 (Mangbettu). Wilken, ' Over de primitieve vormen van het huwelijk en den oorsprong van het gezin,' in De Indische Gids, 1880, vol. ii. 633 (natives of the Malay Archipelago). Georgi, Beschreibung alter Nationen des russischen Reichs, pp. 10, 105, 221 (Kirghiz, Tartars of Kazan and Orenburg, Laplanders). Michaelis, Commentaries on the Laws oj Moses, i. 471 (Hebrews). Tacitus, Germania, ch. 20. 2 Livingstone, Missionary Travels and Researches in South Africa, P- 4I2< ; 3 Schoolcraft, op. cit. v. 272. 4 Marshall, A Phrenologist amongst the Todas, p. 214. 6 Supra, i. 160 sq. 6 Cardus, Las Misiones Fransiscanas entre los infieles de Bolivia, p. 254 (Matacos). Bancroft, Native Races of the Pacific States oj xvn SEXUAL SELECTION IN MAN 33 Greenlanders cleverness in sewing and skill in the manage- ment of household affairs are the most attractive qualities of a woman ;1 and both among them and other American natives a man often prefers a wife who is considerably older than himself on account of the work or comfort he expects from her.2 Among the Pawnee "the qualities most de- sired in a young woman by a suitor were that she should be of good family, and that she should be well versed in house- hold offices and in the manifold other duties of woman's life. Personal beauty, though it had its place and value, was of less consequence." 3 Among the Turkomans young widows fetch double the price of spinsters, because they are more accustomed to hard labour and more experienced in household concerns.4 A Samoyed, again, in choosing a wife cares less for beauty than for property and rank ;5 and similar considerations are undoubtedly very common among uncivilised as well as civilised peoples. As for the choice made by women we have already noticed that they instinc- tively prefer strong and brave men to feeble and cowardly ones; but reflection also makes them choose husbands who are well able to protect them and provide them with food. Of various South American Indians we are told that the women prefer men who are likely best to supply them with ample food, that is, good hunters or fishermen.6 Among the Chippewa " a distinguished and celebrated North America, i. 276 (Inland Columbians). Bogoras, op. cit. p. 571 (Chukchee). Carey and Tuck, Chin Hills, i. 189. Henry, op. cit. p. 200 (Bambara). Thurnwald, ' Ermittlungen . uber Eingebo- renenrechte der Siidsee. A. Buin auf Bougainville (Deutsche Salomo-Inseln),' in Zeitschr. f. vergl. Rechtswiss. xxiii. 337. 1 Fries, Gronland, p. in. Cf. Cranz, History of Greenland, i. 145 5?. 2 Supra, i. 346, 347, 349. 3 Dunbar, in Magazine of American History, iv. 264. 4 de Bode, ' Yamud and Goklan Tribes7 of Turkomania,' in four. Ethn. Soc. London, i. 75. 5 Pallas, Reise durch verschiedene Provinzen des Russischen Reichs, iii. 78. 6 Simson, Travels in the Wilds of Ecuador, p. 104 (Canelos). Haseman,' Some Notes on the Pawumwa Indians of South America,' in American Anthropologist, N.S. xiv. 341. Cardds, op. cit. p. 254 (Matacos). VOL. n D 34 THE HISTORY OF HUMAN MARRIAGE ch. xvh hunter has no occasion to look after wives-he can scarcely keep them at bay."1 Among the Pawnee " the con- siderations most dwelt upon by the woman were the personal prowess, rising influence, skill in hunting and fine form of her lover."2 Among the Tarahumare of Mexico, " aside from good looks, the women like best men who work well, just as in the civilised countries a woman may look out for a good ■parti."3 These are only a few instances of the part played by economic considerations in the conclusion of marriages among the lower races, even where the young people them- selves have a voice in the matter. As for modern European marriages, Nordau believes that not less than nine out of ten of them are so-called " marriages of convenience " and in no sense properly x love-marriages.4 Dr. Bloch observes that the economic question is the main determining influence among the classes who feel it their duty to keep up a par- ticular kind of appearance, namely, the aristocracy, the upper-middle classes, and the officers in the army, and that the predominance of mercenary marriages among the Jews is a well-known fact. But he also asks, " Where are money marriages more frequent than they are among our sturdy German peasants, with whom everything conventional has freest possible play ? "5 Among the peasantry both of Germany6 and other European countries7 economic equality between the parties is considered an essential condition for the conclusion of a marriage. In the West of Ireland "a man never thinks of a girl who. has not sufficient money to be his equal."8 1 Kohl, Kitchi-Gami, p. m. Cf. Hodge, Handbook of American Indians north of Mexico, i. 808 (Eskimo). 2 Dunbar, in Magazine of American History, iv. 264. 3 Lumholtz, Unknown Mexico, i. 266. 4 Nordau, Die conventionellen Lugen der Kultzirmenschheit, p. 330. 5 Bloch, Sexual Life of Our Time, p. 212 sq. 6 Meyer, Deutsche Volkskunde, p. 166. 7 For Sweden see Nicolovius (Lov6n), Folklifvet i Skytts Harad i Skdne, p. 121 ; Enestrom, Finvedsboruas seder och If, p. 78. 8 Blake, ' Matrimonial Customs in the West of Ireland,' in Folk- Lore, xviii. 78. CHAPTER XVIII ENDOGAMY Sexual selection is influenced not only by preferences but by aversions. The latter have, in fact, played a much more conspicuous part in marriage regarded as a social institution than the former, because they have led to avoid- ances enforced by custom or law, whereas no society has yet laid down any compulsory rules relating to the choice of partner which are based on the stimulating effect of beauty or on individual affection. There are endogamous rules, which forbid the members of a particular group to marry any one who is not a member of the group, and exogamous rules, which forbid the member of a particular group to marry any one who is a member of the group. These two sets of rules are by no means contradictory, in so far as they refer to different groups. Hence endogamy and exogamy occur side by side with each other among the same people. Indeed there is everywhere an outer circle-to use Sir Henry Maine's convenient expression-out of which marriage is either definitely prohibited or considered improper, and an inner circle within which no marriage is allowed. The most comprehensive endogamous group is the human species itself. The prohibition of sexual relations outside the species springs from a powerful instinct which man shares with the lower animals. " L'animal," says M. Duvernoy, " a !'instinct de se rapprocher de son espece et de s'eloigner des autres, comme il a celui de choisir ses aliments et d'eviter les poisons."1 There are hybrids even 1 Duvernoy, ' Propagation,' in Dictionnaire universel d'histoire naturelle, x. 546. 35 36 THE HISTORY OF HUMAN MARRIAGE chap. among wild species living in a state of nature, especially birds belonging to the order of Gallinae,1 but they are rare ; and their scarcity must be partly due to the aversion of animals to sexual connections with individuals of other species than their own.2 For in the vegetable kingdom, where the play of instincts is altogether out of the question, hybrids occur much more frequently; and in captivity a considerable number of animal hybrid forms are produced that are never met with in a state of nature.3 Among domesticated mammals some tricks are often required to deceive the male, and so to conquer its aversion to a female of a different species; the stallion, for instance, who is to cover a she-ass is frequently first excited by the presence of a mare, for which, at the proper moment, the she-ass is substituted.4 It is said that the aversion between allied species in the wild state is more frequently overcome by males than by females ;5 and in crosses between wild and domesticated animals the female generally belongs to the domesticated species or race.6 The origin of the instinct which helps to keep even closely allied species distinct is sufficiently clear. As a general rule, conjugation occurs most readily between gametes belonging to the same kind of organism, and even when spermatozoa of one species are capable of fertilising the ova of another the result is usually more or less abnormal. The embryo may survive for a short time and then perish, owing to some mutual incompatibility in the respective modes of growth inherited from the two parent forms; or it may grow into a mature hybrid offspring which is generally more or less infertile, if not absolutely sterile.7 If this is the case with domesticated animals it may be supposed to be so even in a higher degree with wild ones ; for, as Darwin remarks, 1 Geoffroy Saint-Hilaire, Histoire naturelie gdndrale, iii. 180. 2 Cf. Haecker, Allgemeine Vererbungslehre, p. 214 sq. 3 Geoffroy Saint-Hilaire, op. cit. iii. 169-175, 191. 4 Vogt, Lectures on Man, p. 414. 6 Ibid. p. 421. 6 Geoffroy Saint-Hilaire, op. cit. iii. 177. 7 Marshall, Physiology of Reproduction, p. 202. Darwin, Origin of Species, ii. 44, &c. Haecker, op. cit. p. 215. XVIII ENDOGAMY 37 there are good grounds for the doctrine of Pallas that the conditions to which domesticated animals and culti- vated plants have been subjected generally eliminate the tendency to mutual sterility, so that the domesticated descendants of species which in their natural state would have been in some degree sterile, when crossed, become perfectly fertile.1 It is easy to understand, then, that instincts leading to intercrossing of different species, even if appearing occasionally, could never be long-lived, as only unions between animals belonging to the same species gave birth to offspring endowed with a normal power of reproduction. The abnormal results of unions between individuals of different species is due to the fact that there is too great dissimilarity between the mating gametes. We may speak of a " Law of Similarity " according to which a certain similarity between them is necessary for reproduction, or at least for a normal reproduction. It is a general rule that the more widely different two species are the greater the difficulty in crossing them. Most animal hybrids are produced by individuals belonging to the same genus. When the species are so different that they are put into different genera the chance of their crossing is small. If they belong to different families the chance of crossing is much smaller still;2 and if they belong to different orders there is scarcely any chance at all.8 There are, however, exceptions to the rule : a multitude of closely allied species will not unite, or unite only with great difficulty, though other species, widely different from each other, can be crossed with facility. Hence Darwin inferred that the difficulty or facility in crossing " apparently depends exclusively on the sexual constitution of the species which are crossed, or on their sexual elective affinity, i.e., the ' Wahlverwandtschaft ' of Gartner." But as species rarely, or never, become modified in one character, without being at the same time 1 Darwin, Animals and Plants under Domestication, ii. 189. 2 There is in the British Museum a hybrid between a Penelope, belonging to the family Cracidae, and a pheasant (Haecker, op. cit. 211). 3 Morgan, Experimental Zoology, p. 158. 38 THE HISTORY OF HUMAN MARRIAGE chap. modified in many, and as systematic affinity includes all visible resemblances and dissimilarities, any difference in sexual constitution between two species would naturally stand in more or less close relation with their systematic position.1 Supporters of the hypothesis that mankind consists of several distinct species of the genus Homo have asserted that an instinctive aversion similar to that which keeps different animal species from intermingling exists also between the various human races.2 It may seem that the prevalence of such an aversion has been disproved by the fact that mongrels form a considerable proportion of the inhabitants of the world. It is doubtful whether there are any pure races in Europe. In North America races very different from one another have intermingled on a large scale. In the United States the infusion of white blood into the negro race has been so widespread that " the original type of the African has almost completely disappeared." 3 In Green- land, according to Dr. Nansen, in the course of a century and a half, there has been such an intermixture of races that it would now be extremely difficult, if not impossible, to find a true Eskimo throughout the whole of the west coast; and the Europeans, far from being disliked by the native women, have succeeded in inspiring them with so much respect that " the simplest European sailor is preferred to the best Eskimo seal catcher."4 In Mexico the Spanish mixed breeds are said to constitute two-thirds or three- fourths of the whole population ;5 and South America, to quote a French writer, is " le grand laboratoire des nations hybrides ou mStisses modernes."6 Even in remote Tierra 1 Darwin, Animals and Plants under Domestication, ii. 180. 2 Perier, ' Essai sur les croisements ethniques,' in Memoires Soc. d'Anthr. i. 216. Jacquinot, in Dumont d'Urville, Voyage au Pole Sud, Zoologie, ii. 92. 3 Hoffmann, Race Traits and Tendencies of the American Negro, p. 177. 4 Nansen, First Crossing of Greenland, ii. 328. 6 Topinard, Anthropology, p. 372. 6 Perier, in Mem. Soc. d'Anthr. ii. 340. See also Topinard, op. cit. p. 383 ; Rivet, ' Les Indiens Jibaros,' in L' Anthropologic, xviii. 445. xvni ENDOGAMY 39 del Fuego, according to Mr. Bridges, some mongrels of European fathers and indigenous mothers had appeared in the latter part of the nineteenth century. In Asia there are numberless instances of intermixture of breed between Tartars, Mongols, and Tungus, between Russians and Chinese, and so forth. In India there are many Eurasians. In the Malay Archipelago Chinese and Malays intermarry.1 In the islands of the Pacific Ocean the mongrels of European fathers amount to a considerable number. In Africa the Eastern Sudan is a great centre of mixed breeds between races much removed from one another. And in the southern part of Africa there is a bastard race which has originated through intercourse between Boers and Hottentot women.2 But on the other hand we also hear of many races that refrain from, disapprove of, or actually prohibit marriages or sexual relations generally with persons belonging to another race. Thus, according to Father Baucke, the Mocobis of the Gran Chaco avoided marrying Spanish captives.3 In Guiana and elsewhere Indians do not readily intermix with negroes, whom they despise/ Among the Isthmians of Central America " marriage was not contracted with strangers or people speaking a different language " ;5 and in San Salvador, according to Palacio, a man who had inter- course with a foreign woman was killed.6 Mr. Powers speaks of some Californian Indians who would put to death a woman for committing adultery with or marrying a white man ;7 and among the Havasupai of the Painted Desert region unlawful cohabitation with a white man is likewise punishable by 1 Godron, De Vesp^ce et des races dans les etres organises, ii. 360 n. 2. 8 Fischer, Die Rehobother Bastards und das Bastar dierung s- problem beim Menschen, passim. 3 Kohler, Pater Florian Baucke, ein Jesuit in Paraguay, p. 312. 4 Waitz, Introduction to Anthropology, p. 174. 5 Bancroft, Native Races of the Pacific States of North America, i. 772. 6 Spencer, Descriptive Sociology, Ancient Mexicans, &c., p. 4. 7 Powers, Tribes of California, p. 214. Cf. Mackenzie, Voyages from Montreal to the Frozen and Pacific Oceans, p. 148 (Beaver and Rocky Mountain Indians). THE HISTORY OF HUMAN MARRIAGE chap. 40 death, whilst marriage with such a man is unknown.1 Dr. Bogoras does not know of any case of marriage between the Chukchee and the Yakut, the latter of whom would consider such a union humiliating ; and so are marriages between Russian women and Chukchee considered by the Russians.2 The Chinese refuse marriage with the sur- rounding barbarous tribes, with whom, as a rule, they have no dealings, either friendly or hostile.3 The black and fairer people of the Philippines have from time immemorial dwelt in the same country without producing an inter- mediate race ;4 the Bugis of Perak have kept themselves very distinct from the people among whom they live ;5 and in Sumatra it is a rare thing for a Malay man to marry a Kubu woman.6 The Munda Kols severely punish a girl who is seduced by a Hindu, whereas intercourse with a man of their own people is regarded by most of them as quite a matter of course.7 Count de Gobineau remarks that not even a common religion and country can extinguish the hereditary aversion of the Arab to the Turk or of the Kurd to the Nestorian of Syria.8 Among the Kabyles of Algeria " le mariage avec une negresse n'est pas defendu en principe ; mais la famille s'opposerait a une pareille union."9 All the Nandi-speaking tribes freely intermarry with each other and with the Masai, but not with their Bantu neighbours, the Kavirondo.10 Among the Barolongs, a Bechuana tribe, death was formerly inflicted upon any one who had intercourse with a European.11 Speaking 1 James, Indians of the Painted Desert Region, p. 230. 2 Bogoras, Chukchee, p. 593 sqq. 3 Jamieson, ' Translations from the General Code of Laws of the Chinese Empire,' in China Review, x. 94 sq. 4 Crawfurd, ' On the Classification of the Races of Man,' in Trans. Ethn. Soc. N.S. i. 357. 6 McNair, Perak, p. 131. • Forbes, A Naturalist's Wanderings in the Eastern Archipelago, p. 241. 7 Jellinghaus, ' Sagen, Sitten und Gebrauche der Munda-Kolhs in Chota Nagpore,' in Zeitschr. f. Ethnol. iii. 366, 370 sq. 8 de Gobineau, Moral and Intellectual Diversity of Races, p. 173 sq. 9 Hanoteau and Letourneux, La Kabylie, ii. 164. 10 Johnston, Uganda Protectorate, p. 878. 11 Joest, ' Bei den Barolong,' in Das Ausland, 1884, p. 464. XVIII ENDOGAMY 41 of the Central African Pygmies, M. Hutereau observes that the natives of other tribes have a veritable repug- nance for them, and would never contract a marriage with any one of their people.1 In the extreme North of Europe, marriages between Lapps and Swedes very rarely occur, being looked upon as dishonourable by both peoples; they are equally uncommon between Lapps and Norwegians, and it hardly ever happens that a Lapp marries a Russian.2 At various times Spaniards in Central America,3 Englishmen in Mauritius,4 Frenchmen in Reunion and the Antilles,5 and Danish traders in Greenland,6 have been prevented by law from marrying natives. The Romans were prohibited from marrying barbarians; Valentinian inflicted the penalty of death for such unions.7 Tacitus thought that the ancient Germans refused marriage with foreign nations,8 and the like seems to have been the case with the Slavs.9 Indeed we may say that probably every race considers it a disgrace, if not a crime, to marry within a very different race, at least if it be an inferior one. This feeling is particu- larly strong with regard to its women. As a matter of fact, in the crossings between unequal human races the father in the vast majority of instances belongs to the superior race. " Woman," says M. de Quatrefages, " refuses to lower herself; man is less delicate."10 In North America, for example, the cases in which white women have married coloured men are very rare, and in the Southern States such marriages are prohibited not only by custom but very often by law ; hence the enormous infusion of white blood into the 1 Hutereau, Notes sur la Vie familiale et juridique de quelques populations du Congo Beige, p. 4. 2 v. Duben, Om Lappland ooh Lapparne, p. 200 sq. 3 Morelet, Reisen in Central-Amerika, p. 100. 4 Montgomery, Journal of Voyages and Travels by Tyerman and Bennet, ii. 491. 6 Godron, op. cit. ii. 360. 6 Fries, Gronland, p. 159. 7 Rossbach, Untersuchzmgen uber die rdmische Ehe, p. 465. 8 Tacitus, Germania, ch. 4. v Macieiowski, Slavische Rechtsgeschichte, ii. 191. 10 de Quatrefages, Human Species, p. 267. 42 THE HISTORY OF HUMAN MARRIAGE chap. black race is almost exclusively due to intercourse between white men and negresses.1 In New Zealand it occurs that a European man marries a Maori woman ; but Mr. Kerry- Nicholls never came across an instance where a European woman had married a Maori man.2 Racial endogamy is no doubt chiefly due to racial or national pride and lack of sympathy with, or positive anti- pathy to, individuals of another race ; and for this reason it is particularly common in the case of races who greatly differ from each other in ideas, habits, and civilisation generally. But at the same time I believe that some sexual aversion akin to the instinctive feeling which deters animals of distinct species from pairing with each other is in many cases really felt against sexual intercourse with persons of a race whose appearance is very different from that of one's own, and that such aversion is particularly common in women, in whom the sexual instinct is generally more discriminating than in men. This, however, would not prove that mankind consists of several species. Certain races of domestic or semi-domesticated animals also refuse to mingle with others. The dark and pale coloured herds of fallow-deer which have long been kept together in the Forest of Dean and two other places have never been known to mingle. On one of the Faroe Islands the half-wild native black sheep are said not to have readily mixed with the imported white sheep. And in Circassia the horses of three different sub-races of the species, whilst living a free life, almost always refuse to mingle and cross, and will even attack each other.3 It seems that the instinctive aversion which owes its origin to the physiological law of similarity may in some degree make itself felt also in rela- tions between members of the same species who are very different from each other. It may do so even in cases where that law does not act and intercrossing is not 1 Hoffmann, op. cit. p. 177. Herve, ' Noirs et blancs,' in Revue dz I'&cole d'Anthr. Paris, xvi. 338. See also Nott and Gliddon, Types of Mankind, p. 401. 2 Kerry-Nicholls, ' Maori Race,' in Jour. Anthr. Inst. xv. 195. 3 Darwin, Animals and Plants under Domestication, ii. 102 sq. XVIII ENDOGAMY 43 accompanied by lessened fertility; but the rule that different varieties or races of the same animal or vegetable species are prolific when crossed is not a rule which is altogether without exceptions.1 As for man, it has been said that the crossing of races that greatly differ from each other has a depressing effect upon fertility. Thus it was asserted by Broca that the con- nections of Europeans with Australian women have proved very slightly prolific and that the mongrels resulting from them are almost sterile ;2 but since then it has been fully proved that if a mixed race is almost wanting in certain parts of Australia this does not depend upon physiological conditions of the kind suggested by Broca.3 We may con- 1 See Darwin, Animals and Plants under Domestication, ii. 105, 190; Sebright, Art of improving the Breeds of Domestic Animals, p. 17 sq. ; Vogt, op. cit. p. 421. 2 Broca, Phenomena of Hybridity, p. 48. 3 Broca argued that although it is known that cohabitation be- tween white men and native women is very common in Australia, the number of mongrels there is nevertheless exceedingly small. As I supposed that this statement referred chiefly to the eastern and southern parts of the Australian continent, I asked Bishop R. Salvado and the Rev. Joseph Johnston, living in West Australia, to inform me whether any mixed race existed there, and, if so, whether it was fruitful or not. From the former, who had lived among the West Australian aborigines for more than forty years, and had published a well-known work on their life and customs, I had the pleasure of receiving the following answer, dated New Norcia, October 17, 1888 :-" With regard to the sterility of the half-caste natives, of which I had no experience when I wrote my book, I am able now to deny it altogether, except in cases simflar to those among the Europeans. I know several cases of husband and wife, half- caste natives, having at present six and seven and even eight children, and they may in time have more ; and I know a good many Europeans who, having married native women, have several children. In fact, in the case of one of those marriages there were six children, and in another seven, and I could give the name of each of them." The Rev. J. Johnston wrote:-" There is a school for half-caste boys and girls at Perth, and they seem bright and intelli- gent children, not unlike Polynesian children. As they grow up, they go out to service, and some of the youths are employed as post and telegraph messengers. . . . At the.New Norcia mission, there are several half-caste families, as well as blacks, and they all have 44 THE HISTORY OF HUMAN MARRIAGE chap. sider it certain that the diversities even between the races which least resemble each other are not so great but that under favourable conditions a mixed race may easily be children." The following statement of Mr. Taplin (in Curr, Aus- tralian Race, ii. 264 ; cf. Topinard, ' Note sur les m6tis d'Australiens et d'Europ^ens,' in Revue d'Anthropologic, iv. 243 sqq.), referring to the aborigines of the Lower Murray, goes in the same direction :- " The pure blacks," he says, " are not so healthy as the half-castes. Always the children of two half-castes will be healthier and stronger than either the children of blacks or the children of a black and a half-caste. When a half-caste man and woman marry, they generally have a large and vigorous family. I could point to half a dozen such." Mr. Mathew (Two Representative Tribes of Queensland, PP- 75» 8o) also speaks of the greater vitality of Australian half- castes, and mentions one half-caste woman who had been twice married and had nine children and thirteen grandchildren all living. In Mr. Gillen's ' Notes on some Manners and Customs of the Abori- gines of the McDonell Ranges belonging to the Arunta tribe ' (in Report on the Work of the Horn Scientific Expedition to Central Aus- tralia, iv. 166) we are told of twins of half-caste. As to the scarcity of mongrels in certain parts of Australia it should be noticed that half-caste children are very frequently destroyed by the natives (Eyre, fournals of Expeditions of Discovery fnto Central Australia, ii. 324 ; Westgarth, Australia Felix, p. 58 ; Henderson, Excursions and Adventures in New South Wales, p. 114 sq. ; Curr, Recollections of Squatting in Victoria, p. 252 ; Lumholtz, Among Cannibals, p. 273 ; Roth, Ethnological Studies among the North-West-Central Queensland Aborigines, p. 184; Idem, North Queensland Ethno- graphy : Bulletin No. 8. Notes on Government, &c., p. 6; Darwin, Descent of Man, i. 264 ; Peschel, Races of Man, p. g). The Rev. A. Meyer (' Manners and Customs of the Encounter Bay Tribe,* in Woods, Native Tribes of South Australia, p. 186) states that in the Encounter Bay tribe " nearly all the children of European fathers used to be put to death " ; whilst among the Narrinyeri about one- half of the half-caste infants fell victims to the jealousy of their mothers' husbands (Taplin, ' Narrinyeri,' ibid. p. 14). On the other hand, with regard to the West Australian natives in the neigh- bourhood of Fremantle the Rev. J. Johnston writes that he does not think it has been the custom there to destroy the half-caste ille- gitimate offspring of black women, as he never heard of such a thing ; and this may account for the comparatively large number of mongrels in that part of the continent. It should moreover be remembered that the sexual intercourse of Europeans with savage women is most commonly transitory and accidental, and frequently takes place with prostitutes or licentious women, who are generally known to be xvni ENDOGAMY 45 produced; but at the same time it is quite possible that crossing in some cases leads to a somewhat lessened fertility of both first crosses and mongrels.1 Our knowledge on t^e subject is yet extremely defective. The offsprirg of Boers and Hottentot women have proved to be very fruitful.2 According to v. Gortz the children of Dutchmen and Malay women in Java (Lipplapps) are only productive to the third generation ;3 this statement has been called in ques- tion,4 but there 1 .ay be some truth in it.5 Senor Guevara believes that unions between South American Indians and whites are less fertile than unions between Indians ;6 but according to Dr. Boas statistics of the number of children sterile. And even when the white settler takes a native's daughter to live with him under his own roof as a wife or a concubine, and accustoms her to a half-civilised manner of living, her unfruitfulness (see Thomson, ' On the reported Incompetency of the " Gins," ' in Jour. Ethn. Soc. London, iii. 244 sq.) may be owing to quite another cause than the mixture of blood. Darwin (Animals and Plants under Domestication, ii. 148 sqq.) has shown that changed conditions of life have an especial power of acting injuriously on the reproduc- tive system. Thus animals, as also plants, when removed from their natural conditions, are often rendered in some degree infertile or completely barren, even when the conditions have not been greatly changed. And this failure of animals to breed under confinement cannot, at least to any considerable extent, depend upon a failure in their sexual instincts. " Numerous cases," says Darwin, " have been given of various animals which couple freely under confinement, but never conceive ; or, if they conceive and produce young, these are fewer in number than is natural to the species." It is reasonable to suppose that savage man, when he moves into more civilised conditions, is subject to the same law. Indeed, statements have been reported to me which tend to show that the indigenous women at the Polynesian missionary stations have become less fruitful than they were in their native state. 1 Cf. Fehlinger, ' Kreuzungen beim Menscben,' in Archiv f. Rassen- und Gesellschafts-Biologie, viii. 448 sqq. 2 Fischer, op. cit. p. 125 sqq. Francois, Nama und Damara Deutsch-Sud-West-Afrika, p. 242. 3 v. Gortz, Reise um die Welt, iii. 288. 4 Hensen, ' Die Physiologic der Zeugung,' in Hermann, Handbuch der Physiologic, vol. vi. pt. ii. 191. fi Fischer, op. cit. p. 181. 0 Guevara, Psicolojia del pueblo Araucano, reviewed by Rivet, in L'Anthropologic, xxi. 591. 46 THE HISTORY OF HUMAN MARRIAGE chap. of Indian women and of half-blood women show on the contrary that the mixed race is more fertile than the pure stock, although both groups live practically under the same conditions.1 The statements relating to the fertility of mulattos are also conflicting. According to some writers they are but partially prolific, and become eventually extinct when kept apart from the parent stocks ;2 whereas, according to others, mulatto families are known to have intermarried for several generations and have continued on the average as fertile as either pure whites or pure blacks.3 Gould speaks of their inferior vitality as a well-known fact ;* and Mr. Hoffmann has more recently come to a similar conclusion, namely, that they are physically inferior to the whites and the negroes, and possess a lesser degree of resistance to disease and death.5 Half-breeds among the Eskimo of Cumberland Sound are said to be less able to bear exposure and fatigue than the full-blooded Eskimo.6 In New Zealand, according to Mr. Best, " half-castes are, physically, a fine people, though not long-lived as a rule " ; their moral looseness,, however, may have something to do with this-in one half-caste family at least four members have been guilty of incest.7 The mixed race produced by Boers and Hottentot women are described as strong and healthy.8 The intercrossing of races may yield very different results in different cases, and no general rule can be laid down at present. However, the question whether there is a genuine sexual aversion to races very different from one's own is by no means affected by the influence 1 Boas, ' The Half-blood Indian,' in Popular Science Monthly, xlv. 561 sq. 2 Nott and Gliddon, op. cit. p. 307 sq. Broca, op. cit. p. 37. Berenger-Feraud, 1 Note sur la fecondite des mula.res au S6n£gal,' in Revue d'Anthropologic, ser. ii. vol. ii. 577 sqq. Cf. Fischer, op. cit. p. 181. 3 Darwin, Descent of Man, i. 264 sq. de Quatrefages, op. cit. p. 264. 4 Go aid, Investigations in the Military and Anthropological Statistics of American Soldiers, p. 319. 6 Hoffmann, op. cit. p. 182 sqq. 6 Knmlien, Contributions to the Natural History of Arctic America, p. 17. 7 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New Zealand Institute, xxxvi. 31. 8 Fischer, op. cit. p. 305. XVIII ENDOGAMY 47 which the crossing of races may have upon fertility or vitality. As already indicated, such an aversion may be easily explained as a result of the law of similarity whether this law applies to the crossing of races or not. Among many peoples marriage very seldom or never takes place even outside the tribe or some smaller division of it, as the clan or village. This is the case among the Aht,1 Navaho,2 Pueblos,3 and various tribes of Guatemala.4 Of the Mohave Indians, who live along both sides of the Colo- rado River where the stream forms the boundary between Arizona and California, Professor Kroeber writes:-"Marriages with other tribes were few. Not only sexual connection but ordinary intercourse with other races were regarded with disfavor, as being a specific cause of sickness. Among the races thus to be shunned were included not only the whites, and all tribes of other linguistic relations, but some of the tribes speaking kindred Yuman languages.''5 Among the Seri Indians there is absolute tribal endogamy.6 In the village of Schawill, in Southern Mexico, " every member must marry within the rancho, and no such thing as a marriage out of it had ever occurred. They said it was impossible, it could not happen. . . . This was a thing so little apprehended that the punishment for it was not defined in their penal code ; but being questioned, after some consultations, they said that the offender, whether man or woman, would be expelled.''7 The Indians of British Guiana " do not generally marry out of their tribe, for the doing so is looked upon with disapprobation, as they consider it derogatory to the dignity of their respective tribes.''8 Speaking of the Chaymas of New Andalusia, among whom marriages are contracted between the in- habitants of the same hamlet only, v. Humboldt says, 1 Sproat, Scenes and Studies of Savage Life, p. 98. 2 Bancroft, op. cit. i. 512 n. 120. 3 Davis, El Gringo, p. 146. 4 Bancroft, op. cit. i. 703. Froeber, Preliminary Sketch of the Mohave,' in American Anthropologist, N.S. iv. 279. 8 McGee, ' Seri Indians,' iij Ann. Rep. Bur. Ethn. vol. xvii. pt. i. 283. 7 Stephens, quoted by Bancroft, op. cit. i. 663. 8 Bernau, Missionary Labours in British Guiana, p. 60. 48 THE HISTORY OF HUMAN MARRIAGE chap. " Savage nations are subdivided into an infinity of tribes, which, bearing a cruel hatred toward each other, form no intermarriages, even when their languages spring from the same root, and when only a small arm of a river, or a group of hills, separates their habitations."1 This holds true of several of the Brazilian tribes.2 In ancient Peru it was not lawful for the natives of one province or village to marry those of another.3 The Ainu not only despise the Japanese as much as the Japanese despise them, but are not very sociable even among themselves : they may be said never to marry outside their own tribe, and one village does not like to marry into another.4 A Yukaghir " seldom marries into a strange settlement or clan."5 In India, according to Mr. Gait, the Animistic tribes, Mongolian as well as Dravidian, as a general rule marry only within the tribal limits, al- though there is usually no objection to a man taking a girl from outside the tribe if he is able to obtain one.6 Of the Oraons we are told that they are so strictly endoga- mous that " anyone marrying out of the tribe would be at once ejected from it, and could not be readmitted before leaving his foreign wife."7 There may also be endogamous groups inside the tribes. Thus the Todas have not only a 1 v. Humboldt, Personal Narrative of Travels to the Equinoctial Regions of the New Continent, iii. 226 sq. 2 v. Martius, Beitrage zur Ethnographic Amerika's, i. 106. 3 Garcilasso de la Vega, First Part of the Royal Commentaries of the Yncas, i. 308. 4 Batchelor, ' Notes on the Ainu,' in Trans. Asiatic Soc. Japan, x. 211 sq. v. Siebold, Ethnologische Studien uber die Aino auf der Insel Yesso, p. 30 sq. Howard, Life with Trans-Siberian Savages, p. 182 sq. 5 Jochelson, Yukaghir, p. 86. 6 Gait, Census of India, 1911, vol. i. (India) Report, p. 253. See also Sarat Chandra Roy, Mundas, p. 400; Dalton, Descriptive Ethnology of Bengal, p. 278 (Kolams) ; Hodson, Naga Tribes of Manipur, p. 89 ; Hodgson, * On the Origin, &c. of the K6cch, Bodo, and Dhimal People,' in Jour. Asiatic Soc. Bengal, xviii. 708 ; Mason, * On Dwellings, &c. of the Karens,' ibid. vol. xxxvii. pt. ii. 151 ; Carey and Tuck, Chin Hills, i. 189. 7 Dehon, ' Religion and Customs of the Uraons,' in Memoirs Asiatic Soc. Bengal, i. 159 sq. xvin ENDOGAMY 49 strong prejudice against marriages outside their own com- munity but allow no marriage between the Tartharol and the Teivaliol, who are divisions of the Toda people. Yet a Teivali man may take a Tarthar woman to live with him at his village, the man being known as the mokhthodvaiol of the woman ; and the same kind of connection may be formed between a Tarthar man and a Teivali woman, although in this case the woman is not allowed to live at the village of the mokhthodvaiol, who may either visit her occasionally or go to live at her village.1 Of the Malayalis, a Tamil-speaking tribe in North Arcot, it is said that a man must marry within his katchi, a division which seems to have the character of a phratry,2 and marriages between near relatives are preferred.3 We are told that among the Kacharis each sub-tribe was in early times strictly endoga- mous, although members of different sub-tribes might freely eat and drink together.4 Among the Lushais and Old Kukis marriage is generally restricted to the clan, though alliances within the family are prohibited.5 Among the Palaungs, a Mongolian hill community in the Shan States and a portion of the north-east of Upper Burma, the young men of some villages as a rule decline to look outside the limits of the village for their spouses.6 Among the Sermatta Islanders one village does not like to marry into another.7 The Orang Kubu of Sumatra,8 the Minahassers of Celebes,9 the Dyaks of Borneo,10 the natives 1 Rivers, Todas, pp. 504, 505, 508 sq. Cf. King, Aboriginal Tribes of the Nilgiri Hills, p. 25. 2 See Gait, op. cit. p. 253. 3 Cox, Madras District Manuals : North Arcot, i. 205 sq. 4 Endle, Kacharis, p. 29. 5 Shakespear, Lushei Kuki Clans, pp. 50, 153. 6 Lewis, Note on the Palaungs of Hsipaw and Tawnpeng, p. 10 sq. 7 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, P 325. s Hagen, Die Orang Kubu auf Sumatra, p. 130. 9 Hickson, A Naturalist in North Celebes, p. 277. Wilken, Over de verwantschap en het huwelijks- en erfrecht bij de volken van het maleische ras, p. 21 sq. 10 Brooke Low, quoted by Ling Roth, Natives of Sarawak, i. 123. Wilken, op. cit. p. 23. VOL. II E P 325- 50 THE HISTORY OF HUMAN MARRIAGE chap. of New Britain,1 and the Papuans of Dorey in Dutch New Guinea,2 generally, if not always, marry within their own tribe. The Maori, says Mr. Best, " were endogamous in re- gard to the tribe and the sub-tribe or clan (haju), and also to some extent in regard to the sub-Aa/>«-i.e., the gens, or family group. Not that marriages did not take place as between members of one hapu and those of another, or even between those of one tribe and another. Such marriages did occur, the latter, however, much more rarely than the former. Still, it was considered desirable to marry within the haju, or clan, for social and political reasons. Marriages between members of one tribe and those of another were generally of a political nature, as to cement a peace-making. We are speaking of pre-European days now, for latterly extra- tribal marriages have become more frequent."3 In Australia there are " associated tribes," generally speaking the same dialect, which are in the habit of uniting for common defence and other purposes, and marriages between members of such tribes are very frequent ;4 but there are also many tribes which are mostly endogamous.5 Howitt says of the Dieri that " in the tribe itself there is always a hot opposition to a marriage which takes a girl out of it, and the fathers in it who have unmarried and eligible sons offer every ob- jection to the arrangement."6 1 Romilly, ' Islands of the New Britain Group,' in Proceed. Roy. Geograph. Soc. N.S. ix. 9. 2 Bink, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. xi. 396. 3 Best, ' Maori Marriage Customs,' in Trans, and Proceed New Zealand Institute, xxxvi. 20. Cf. Yate, Account of New Zealand, pp. 96, 99 ; Thomson, Story of New Zealand, i. 176. 4 Curr, Australian Race, i. 63, 67. Mathew, ' Australian Abori- gines,' in Jour, and Proceed. Roy. Soc. New South Wales, xxiii. 398. 5 Curr, op. cit. i. 298, 303, 330, 343, 377 ; ii. 21, 179, 197, 307 ; iii. 252, 272. 6 Howitt, Native Tribes of South-East Australia, p. 185. In Australia, as will be shown in the next chapter (p. 132 sqq.}, there is also a kind of endogamy which is not of a tribal character, a man being bound to marry into a certain class, sub-class, or clan other than his own. This is endogamy in so far as that class, sub-class, or clan may be regarded as part of a larger group of which he himself is a member ; but, as Howitt remarks {op. cit. p. 137), " the boundaries of a class system are usually wider than those of a tribe." xvni ENDOGAMY 51 In Madagascar clan endogamy was the rule, a marriage outside the clan being considered a crime, a kind of " social adultery " j1 and among the Hovas, according to Sibree, not only clans but " families " as a rule refrain from inter- marrying.2 Barrow states that the Hottentots always marry within their own kraal.3 Among the Thonga marriages outside the tribe are avoided and disapproved of by old native opinion, and marriages into a too far distant clan are not recommended.4 Mr. A. J. Swann informed me that among the Waguha of West Tanganyika, also, marriages into another tribe are avoided, though not prohibited; and Archdeacon Hodgson wrote that this is very often the case in Eastern Central Africa.5 Among the Warega of the Belgian Congo marriages outside the tribe occur only in frontier districts.6 The non-cannibal tribes of Equatorial Africa did not marry with their cannibal neighbours, whose peculiar practices were held in abhorrence.7 Among the Bavili " intertribal marriages were once totally prohibited, but to-day marriages take place although the offspring of such unions are looked upon much in the same prejudiced light by the Bavili as the offspring of black and white races are looked upon by the Europeans."8 Strict endogamy is not unheard of among the Berbers of Morocco. Thus the Ath Zihri (Zkara) near the Algerian frontier, who are known for their exclusiveness, are reported to marry only between themselves and also to refrain from all sexual intercourse with strangers ;9 and among the Ait Hassan, 1 .Grandidier, Ethnographic de Madagascar, ii. 168. 2 Sibree, The Great African Island, pp. 109, 256. 3 Barrow, Account of Travels into the Interior of Southern Africa, i. 144. 4 Junod, Life of a South African Tribe, pp. 240, 246 sq. 5 Cf. Baumann, Durch Massailand zur Nilquelle, p. 193 (Wassan- daui) : Stannus, ' Notes on some Tribes of British Central Africa,' in Jour. Roy. Anthr. Inst. xl. 309. 6 Delhaise, Les Warega, p. 175. Cf. Johnston, George Grenfell and the Congo, ii. 684. 7 Du Chaillu, Explorations and Adventures in Equatorial Africa, p. 97. 8 Dennett, At the Back of the Black Mans Mind, p. 38. 9 Cf. Moulieras, Une tribu Zentte anti-musulmane au Maroc (les Zkara), p. 82. E 2 52 THE HISTORY OF HUMAN MARRIAGE chap. a fraction {tdqbilt) of the Ait Warain, to the south of Fez, no stranger is even allowed to be present at a wedding.1 Concerning the ancient Semites Professor Barton observes that, so far as their feeling on this point can be historically traced, it was in favour of endogamy, Semitic parents always being grieved if their children married outside their tribe.2 The descendants of Abraham were expressly forbidden to intermarry with the Canaanites.3 It has generally been supposed that at Athens, at least from the time of Pericles, no valid marriage could be concluded between a citizen and an alien, unless Athenian citizenship had been bestowed upon the latter or he or she belonged to a community to which the Athenian Assembly had granted rights of inter- marriage.4 This view, however, has been opposed by some recent writers, according to whom there is no evidence that intermarriages of citizens and aliens really were prohibited at Athens; and we know as a matter of fact that such marriages, whether legal or not, did occur there.5 On the other hand, there is no doubt that at Rome any marriage of a citizen with a woman who was not herself a Roman citizen, or did not belong to a community possessing the privilege of connubium with Rome-which was always expressly conferred-was invalid; no legitimate children could be bom of such a marriage.6 And in early times it was even customary for a father to seek for his daughter a husband from his own gens, marriage out of it being mentioned as an extraordinary thing.7 In ancient Wales, 1 Westermarck, Marriage Ceremonies in Morocco, p. 57. 2 Barton, Sketch of Semitic Origins, p. 62. See Genesis, xxvi. 34 sq., xxvii. 46, xxviii. 8 ; Judges, xiv. 3 ; &c. 8 Genesis, xxiv. 3 ; xxviii. 1. 4 In his ' Life of Pericles ' (xxxvii. 3) Plutarch speaks of a law- according to which both parents of an Athenian citizen must be citi- zens ; and this law is said to have been re-enacted in the archonship of Eucleides (b.c. 403) with the reserve that the re-enacted law was not to be retrospective (Demosthenes, Oratio contra Eubulidem, p. 1307). 6 See particularly Beauchet, Histoire du droit privi de la Rdpub- lique Ath&nienne, i. 179 sqq. ; Hruza, Beitrage zur Geschichte des griechischen und romischen Familienrechtes, ii. 102 sqq. 6 Gaius, Institutiones, i. 56. 7 Marquardt, Das Privatleben der Rbmer, p. 30. Cf. Mommsen, Bomische Forschungen, i. 9 sq. XVIII ENDOGAMY 53 according to Lewis, marriage was to be within the clan.1 Marrying within the tribe or clan is often particularly insisted upon in the case of women. In Southern Guinea, " while men of the maritime tribes may marry any number of wives among the bush tribes, nothing would be more abhorrent to their ideas of propriety than that a bushman should marry one of their women."2 Among the Chukmas of the Chittagong Hills it is not obligatory for a man to marry within the ifibe,' but a woman marrying outside her tribe is unknown.3 The Tipperahs and Abors view with abhorrence the idea of their girls marrying out of their own clan ;4 and Colonel Dalton was gravely assured that " when one of the daughters of Padam so demeans herself, the sun and the moon refuse to shine, and there is such a strife in the elements that all labour is necessarily suspended, till by sacrifice and oblation the stain is washed away."5 In the Patani States, according to Messrs. Annandale and Robinson, " Malay society is divided into numerous families (kaum), which are so far endogamous that they prefer that their girls should marry within the limits of the family."6 Where there is little intercourse between different tribes or subdivisions of tribes marriages between their members are naturally rare if not unknown. Hagen observes that as the different tribes of the Orang Kubu hardly ever come into contact with one another, intermarriage between them is altogether out of the question ;7 and Dr. Jochelson traces the endogamous tendencies of the Yukaghir to the fact that, as a hunting-tribe, they frequently have to scatter in separate families or groups of related families in search of food, which compels them to satisfy their sexual desires 1 Lewis, Ancient Laws of Wales, p. 196. a Wilson, Western Africa, p. 266. See also Nassau, Fetichism in West Africa, p. 8. 3 Hutchinson, Account of the Chittagong Hill Tracts, p. 96. 4 Lewin, Wild Races of South-Eastern India, p. 201. 5 Dalton, op. cit. p. 28. 6 Annandale and Robinson, Fasciculi Malayenses, ii. 74. 7 Hagen, Die Orang Kubu auf Sumatra, p. 130. THE HISTORY OF HUMAN MARRIAGE chap. 54 vidthin the group.1 Habitual isolation, combined with antipathy felt against people with customs and manners and a language different from one's own, readily leads to disapproval or actual prohibition oaf intermarriage ; and so does the desire of a tribe or clan not to part with any of its members. Among the Masai, " if a woman were taken to wife by any except a Masai or if a boy went into service outside the tribe, they could never return, and, in the event of their venturing to do so, they were beaten to death."2 Speaking of the endogamy of the Maori, Mr. Best remarks that in the warlike days of old it was doubtless deemed desirable to keep as many fighting-men as possible within the clan-limits ;3 and concerning the Cheyenne Indians the oldest authorities state most positively that as by tribal custom a young man on marrying usually, went to live with his wife's people, " it was the policy of the chiefs to have their young men marry within the division, in order to keep up the strength of their own band."4 Where endogamy exists side by side with patrilocal marriage, a reason for it is undoubtedly reluctance on the part of the group to lose any of its women. Although the Todas are strongly opposed to any of them marrying outside their own community, they seem to have no strong objection to a woman's relations with a stranger so long as she remains in the community.5 The endogamy of the Maori is also partly due to the fact that " when a woman marries into another tribe or clan she and her mana (that is, ' power ') are lost to her people and clan " ;6 whilst another reason for it is that when a husband and a wife who are relatives quarrel and proceed to kai upoko, or curse each other, the remarks are not so serious as they would be if they were uttered by a non- relative.7 Among the ancient Arabs marrying within the village served the purpose of strengthening the kinship 1 Jochelson, Yukaghir, p. 86. 2 Hinde, Last of the Masai, p. 76. 3 Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 21. 4 Mooney, ' Cheyenne Indians,' in Memoirs American Anthrop. Association, i. 410. 6 Rivers, Todas, p. 508 sq. 6 Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 21. 7 Ibid. p. 23. XVIII ENDOGAMY 55 tie.1 The Berbers of the Rif in Morocco encourage marriages between members of the same village in order to keep away alien elements from the community, and therefore deny the right of inheritance to a woman who leaves her village.2 Among the Hovas of Madagascar clans and " families " re- frained from intermarrying "in order to keep landed property together, as well as from a strong clannish feeling.''3 In ancient Wales the restriction on marriage outside the clan was intended to prevent strangers and their progeny from acquiring positions of property and rule within the group.4 Moses ordered the daughters of Zelophehad to marry men of their father's tribe so that their inheritance should remain "in the tribe of the family of their father.''5 But the reason given in Deuteronomy for the prohibition of intermarriage with seven Canaanitish nations was a religious one :-" They will turn away thy son from following me, that they may serve other gods.''6 The Mosaic prohibition of marriages between Israelites and Canaanites was, at any rate from Ezra onward, extended so as to include all the pagan nations of the country ;7 and in the spirit of Ezra's ordinance, later religious authorities in the time of the Maccabeans and in the time of the wars against the Romans interdicted matrimonial connections between Israelites and all Gentiles. This prohibition is the established law of the Talmud and the Rabbinical Code ; and although in the latter no special provision was made concerning intermarriage with Christians, these were actually included in the general prohibition of intermarriage with Gentiles.8 The Great Jewish Synod convened by Napoleon in 1807 declared that marriages between Israelites and Christians are valid when contracted in accordance with the Civil Code, and that they shall not entail any 1 Wellhausen, ' Die Ehe bei den Arabem,' in Nachrichten von dev Koniglichen Gesellschaft dev Wissenschaften zu Gottingen, 1893, no. 11, p. 437 sq. 2 Westermarck, Marriage Ceremonies in Morocco, p. 56. 3 Sibree, op. cit. p. 109. 4 Lewis, op. cit. p. 196. 5 Numbers, xxxvi. 12. 0 Deuteronomy, vii. 3 sq. 7 Ezra, ix. 1 sq.; x. 10 sq. Nehemiah, x. 30; xiii. 23 sqq. 8 Mielziner, Jewish Law of Marriage and Divorce, p. 45 sq. 56 THE HISTORY OF HUMAN MARRIAGE chap. disciplinary punishment (anathema), " although such mar- riages cannot be invested with the religious forms," that is, be solemnised by the religious rites of Judaism.1 The Rabbinical Conference held at Braunschweig in 1844 went a step further by resolving that the intermarriage of Jews and Christians, and in general the intermarriage of Jews with adherents of any of the monotheistic religions, " is not forbidden," provided that the parent? are permitted by the law of the State to bring up the offspring in the Jewish faith.2 This decision, however, which decidedly abandoned the Talmudic standpoint, has been strongly criticised even by some of the most pronounced advocates of reformed Judaism.3 No section of Jewish opinion favours marriages between parties who are not of the same religion.4. In some European countries, such as Russia, marriages between Jews and Christians are still very rare. In Vienna there were, in 1898, no mixed marriages as against 847 purely Jewish marriages, while in Prague there were only 6 as against 354. In Berlin, on the other hand, there were, in 1899, as many as 229 mixed marriages as against 621 purely Jewish marriages ; while in Prussia generally there were, during 1900, 474 Jews and Jewesses who married out- side their faith and 4,799 Jews who married Jewesses.5 In all these cases it is of course necessary to double the number of purely Jewish marriages in order to determine the proportion of persons married within or without the faith. Marriages between Christians and Jews were also pro- hibited by the Christians-by Constantine and later emperors 1 Halphen, Recueil des lois &c. concernant les Israelites depuis la Revolution de 1789, p. 25. 2 Mielziner, op. cit. p. 47 sq. 8 Ibid. p. 48 sqq. Even the author of the Braunschweig resolution. Rabbi Dr. L. Philippson, afterwards modified his views concerning the subject. He says in one of his later works (Die Israelitische Religionslehre, iii. 250) that religion must pronounce against mixed marriages, as it must be conceded that they contribute toward the weakening of true religiousness and sincerity in matters of faith. ♦ Abrahams, '/Marriage (Jewish),' in Hastings, Encyclopedia of Religion and Etltics, viii. 461. 6 Jacobs, ' Iiftermarriage,' in Jewish Encyclopedia, vi. 612. XVIII ENDOGAMY 57 and by various Councils ; and during the Middle Ages they were universally avoided.1 " The folklore of Europe," says Mr. Jacobs, " regarded the Jews as something infra- human, and it would require an almost impossible amount of large toleration for a Christian maiden of the Middle Ages to regard union with a Jew as anything other than unnatural."2 Indeed, owing to the intense Jewish hatred for the sacred name of Christ, the early Church was more opposed to wedlock with Jews than with pagans. Al- though St. Paul indicates that a Christian must not marry a heathen,3 and Tertullian calls such an alliance fornication,4 the Church, in early times, often even encouraged marriages of this sort as a means of propagating Christianity ; and it was only when its success was beyond doubt that it actually prohibited them.5 When the ' Decretum ' of Gratian was published, in the twelfth century, the impediment disparitas cultus became part of the canon law of the Church,6 and from that time forward all marriages contracted between Catholics and infidels were held to be invalid unless a dis- pensation from such marriages had been obtained from the ecclesiastical authority. Marriages between Catholics and heretics, on the other hand, were considered valid, though illicit if a dispensation mixta religionis had not been ob- tained ; but there had been much opposition to such unions already in early times, and various Councils had legislated against them. The Council of Trent declared all matri- monial unions between Catholics and non-Catholics null and void, unless entered into before the ecclesiastical authority; but by degrees the Popes felt constrained to make various concessions for mixed marriages. " The Church," says Taunton, " has always abhorred these 1 Mielziner, op. cit. p. 46 n. 6. Kohler, ' Intermarriage,' in Jewish Encyclopedia, vi. 611. Low, Gesammelte Schriften, iii. 175. Andree, Zur Volkskunde der Juden, p. 48. Neubauer, 'Notes on the Race-Types of the Jews,' in Jour. Anthr. Inst. xv. 19. 2 Jacobs, ' On the Racial Characteristics of Modern Jews,' in Jour. Anthr. Inst. xv. 52. 3 1 Corinthians, vii. 39. 4 Tertullian, Ad uxorem, ii. 3 (Migne, Patrologies cursus, i. 1292 sqj. 5 Winroth, Offentlig Rcitt. Familjerdtt : Aktenskapshindren, p. 212. 6 Gratian, Decretum, ii. 28. 1. 1. 58 THE HISTORY OF HUMAN MARRIAGE chap. marriages both on account of the danger of perversion and the difficulty of educating the offspring as well as on account of the communicatio in sacris."1 The Protestants also originally forbade such marriages.2 But mixed marriages are not now contrary to the civil law either in Roman Catholic or Protestant countries. The case is different in countries belonging to the Greek Church, where the ecclesiastical restrictions have been adopted by the State. In the Eastern Church the Council in Trullo, in the seventh century, declared marriages between Catholics and heretics null and void, and this enactment has since remained in force. The Greek Church has also shown itself opposed to marriages with members of the Roman Church ; and in Russia various laws were passed ordering that such marriages be not permitted unless the children of the union are to be brought up in the Orthodox faith.3 But even in countries which are partly Roman Catholic, partly Protestant, mixed marriages form only a comparatively small percentage of the whole number of marriages.4 Their number, however, has been increasing. In Bavaria, for instance, they amounted in 1835-1850 to 2'8 per cent, of the whole number of marriages, in 1876-77 to 66 per cent.5 Catholic men seem to be the most liberal in contracting mixed marriages, especially with Protestant women.6 Islam has also made religion a bar to intermarriage. The Koran said in explicit terms, " Marry not a woman of the Polytheists until she embraces Islam " ; but it also declared that such women as are of chaste reputation and belong to the Scriptural sects or believe in a revealed or moral religion are lawful to Moslems. From these and similar directions two somewhat divergent conclusions have 1 Taunton, Law of the Church, p. 439. 2 (Scheurl and) Sehling, ' Eherecht,' in Herzog-Hauck, Realen- cyklopadie fiir protestantische Theologie und Kirche, v. 211. 3 Zhishman, Das Eherecht der orientalischen Kirche, p. 519 sqq. Fanning, * Marriage, Mixed,' in Catholic Encyclopedia, ix. 698 sq. Winroth, op. cit. p. 213 sqq. 4 v. Oettingen, Die Moralstatistik in ihrer Bedeutung fiir eine Socialethik, p. 128 sqq. Mayo-Smith, Science of Statistics, i. no. 6 v. Oettingen, op. cit. p. 130 sq. 6 Mayo-Smith, op. cit. i. no. XVIII ENDOGAMY 59 been drawn by Sunni and Shi'ah lawyers. The Sunnis recognise as legal and valid a marriage contracted between a Moslem on one side and a Hebrew or a Christian woman on the other, but hold that a marriage between a Moslem and a Magian or a Hindu woman is invalid. Some of the Shi'ahs, on the other hand, do not recognise as legal a permanent contract of marriage between a Moslem and a woman follow- ing any other creed, but allow a temporary contract (mut'd) extending over a certain specified period with a Christian, Jewish, or Magian female. Both schools prohibit a Moslem from marrying an idolatrous female, or one who worships the stars or any kind of fetish whatsoever. In no circumstances is a Muhammadan woman permitted to marry a man who is not a Moslem.1 Among the Hindus religion even forbids the intermarriage of persons belonging to different castes. Endogamy is the essence of the caste system. Nay, not only must a Hindu refrain from marrying outside the limits of his caste, but where, as is usually the case, the caste is divided into sub-castes, he must ordinafily not marry outside his sub- caste, although he may sometimes marry in certain sub- castes but not in others ; and there may be some sub-castes from whom he may take a girl in marriage but to whom he may not give one.2 Where a caste is divided into sections of different status, there may be hypergamy, or the rule whereby parents are obliged to marry their daughters into an equal or higher section, and if they fail to do so, are themselves reduced to the status of the section in which their daughter marries ; whereas the men may marry girls of their own or any inferior section. Hypergamy prevails over a wide area at the present day, but is almost unknown in the south of India and in Assam.3 It is a modern form 1 Ameer Ali, Mahommedan Law, ii. 320 sq. Lane, Account of the Manners and Customs of the Modern Egyptians, i. 137. 2 Gait, op. cit. p. 249. Jogendra Nath Battacharya, Hindu Castes and Sects, p. 11. Monier Williams, Indian Wisdom, p. 218 note. 3 Gait, op. cit. p. 254 sq. Risley, People of India, p. 163 sqq. On the Coast of Malabar, however, women of castes equal to or higher than the Nayars are prohibited from forming unions with 60 THE HISTORY OF HUMAN MARRIAGE chap. of a very ancient custom. Manu recognised the possi- bility of marriage with the next lower caste as producing legitimate children, but condemned the marriage of an Aryan with a woman of lower caste.1 The Paraskara Grihya- sutra allows the marriage of a Kshatriya with a wife of his own caste or of the lower caste, of a Brahman with a wife of his own caste or of the two lower classes, and of a Vaisya with a Vaisya wife only. But it quotes the opinion of certain authorities that all of them can marry a &udra wife, while other authorities condemn the marriage with a £udra wife in certain circumstances, which implies that in other cases it might be justified.2 Among the Muhammadans of India it is considered desirable that a man should take as his first wife a virgin bride of the same social standing as himself and preferably of the same main division or tribe.3 Class endogamy prevails among a large number of peoples in different parts of the world. Among the Algonkin, according to Charlevoix, there were " some considerable families " who could not intermarry but among themselves.4 Among the Aht, who take great pride in honourable birth, a patrician loses caste unless he marries a woman of corre- sponding rank, in his own or another tribe.5 In the Salish tribes the sons and daughters of chiefs customarily inter- married only with persons of their own class, and in like manner nobles married exclusively with other nobles.8 men of castes below them in rank, though the men of these castes are not similarly restricted (Francis, Census of India, 1901, vol. xv. [Madras] Report, p. 131). But in the extreme south of British Malabar there are instances of marriages contracted between a high-caste bride and a low-caste bridegroom (Gopal Panikkar, Malabar and its Folk, p. 22). 1 Laws of Manu, x. 5; iii. 15. 2 Pdraskara-Grihya-SHtra, i. 4. 8 sqq. See Macdonell and Keith, Vedic Index of Names and Subjects, ii. 258. 8 Gait, op. cit. p. 252. 4 Charlevoix, Voyage to North-America, ii. 37. " Sproat, Scenes and Studies of Savage Life, p. 91 sqq. 0 Hill Tout, ' Report on the Ethnology of the South-Eastern Tribes of Vancouver Island, British Columbia,' in Jour. Roy. A nt hr. Inst, xxxvii. 308. XVIII ENDOGAMY 61 Among the Isthmians of Central America the lords married only the daughters of noble blood; whilst in Guatemala marriage with a slave reduced the free-man to a slave's con- dition.1 The tribes of Brazilian Indians also consider such alliances highly disgraceful.2 Among the Bayaka of Congo " slaves must marry slaves and free men must marry free women ; even slave concubines are forbidden to free men."3 In Madagascar there was not only clan but class endogamy.4 Thus among the Hovas the three great divisions-the nobles, the commoners, and the slaves-with few exceptions, could not intermarry ; nor did the three different classes of slaves marry into one another.5 The different orders of the Marea in North-Eastern Africa are likewise endoga- mous.6 Among the S6guela on the Ivory Coast the arti- sans .marry only between themselves.7 Among the Teda the smiths, who form an hereditary and utterly despised caste by themselves, are obliged to marry solely with members of their own caste.8 So also among the Masai the families of smiths are not allowed to marry into the families of persons who have another occupation.9 The Aeneze of Northern Arabia never intermarry with the szona, handicraftsmen or artisans; nor do they ever marry their daughters to Fellahs, or to inhabitants of towns.10 Among the Kalmucks no marriages are contracted between members of different classes.11 In the Chittagong district a slave can only marry a slave.12 In the Malay Archipelago 1 Bancroft, op. cit. ii. 659. 2 v. Martius, op. cit. i. 71. v. Spix and v. Martius, Travels in Brazil, ii. 74. 3 Torday, Camp and Tramp in African Wilds, p. 135. 4 Grandidier, op. cit. ii. 168. 5 Sibree, op. cit. pp. 185, 256. 0 Munzinger, Ostafrikanische Studien, p. 240. 7 Clozel and •Villamur, Les coutumes indigenes de la Cote d'Ivoire, p. 328. 8 Nachtigal, Sahara und Sudan, i. 443 sq. 9 Merker, Die Masai, p. 47. On the position of smiths among uncivilised and semi-civilised peoples see Landtman, Primary Causes of Social Inequality, p. no sqq. 11 Burckhardt, Notes on the Bedouins and Wahdbys, p. 63. Cf. Burton, Narrative of a Pilgrimage to Meccah and Medinah, ii. 10. 11 Bergmann, Nomadische Streifereien unter den Kalmuken, iii. 146. 12 Lewin, Wild Races of South-Eastern India, p. 86 note. 62 THE HISTORY OF HUMAN MARRIAGE chap. marriages between persons of different rank are as a rule disapproved of and in some places prohibited.1 In the Philippine Islands, at the period when they were discovered, the different classes of the natives did not intermarry, and the nobles in particular took care to avoid misalliances.2 In the Ladrone Islands, where only the nobles were supposed to have an immortal soul, a nobleman who married a girl of the people was punished with death.3 In Polynesia, also, the commoners were looked upon by the nobility almost as a different species of beings ;4 and in the higher ranks marriages between nobles and commoners were strongly opposed by the former.5 In Tahiti, if a woman of condition chose an inferior person as a husband, the children he had by her were killed.6 Among the Line Islanders of the Gilbert Group, if an aomata, or person belonging to the gentry, married a rau, or commoner, his or her land was forfeited.7 In China play-actors, policemen, boatmen, and slaves are not allowed to marry women of any other class than that to which they respectively belong.8 Class endogamy pre- vails in Corea.9 In Japan, in the age of the Taiho-ryo (a.d. 701-1192), there was a humble class which was not allowed to intermarry with others ; subsequently, during 1 Wilken, ' Plechtigheden en gebruiken bij verlovingen en huwelij- ken bij de volken van den Indischen Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, ser. v. vol. i. 153. St. John, ' Wild Tribes of the North-West Coast of Borneo,' in Trans. Ethn. Soc. N.S. ii. 234 sq. (Sea Dyaks). 'Hickson, op. cit. p. 278 (Minahassers). Matthes, Bijdragen tot de Ethnologic van Zuid-Celebes, p. 13 (Bugis and Macassars). Hollander-Eck, Hand- leiding bij de beoefening der land- en volkenkunde van Nederlandsch Oost-Indie, i. 600 (Niase). Riedel, op. cit. pp. 302, 434 (natives of Timorlaut and Wetter). 2 Mallat, Les Philippines, i. 57. de Morga, Philippine Islands at the close of the Sixteenth Century, p. 300. 3 Waitz-Gerland, Anthropologic der Naturvblker, vol. v. pt. ii. 112. 4 Ibid. vi. 165, 186. 5 Moerenhout, Voyages aux lies du Grand Ocean, ii. 67. 6 Ellis, Polynesian Researches, i. 256. 7 Tutuila, ' Line Islanders,' in Jour. Polynesian Soc. i. 267. 8 Gray, China, i. 187. 9 Ross, History of Corea, p. 311. XVIII ENDOGAMY 63 the feudal age (from the end of the twelfth century to the Meiji era) the people were divided into lords, knights, and commoners, including farmers, artisans, and merchants, and intermarrying required special permission.1 In Europe there have been similar prohibitions. In Rome plebeians and patricians could not intermarry till the year 445 b.c.,2 nor were marriages allowed between patricians and clients. Cicero himself disapproved of intermarriages of ingenui and freedmen, and although such alliances were generally permitted under the Emperors, a senator could not marry a freed-woman nor a patroness her liberated slave. Between freemen and slaves con- tubernium could take place, but not marriage.3 The marriage laws of Augustus also forbade the marriage of free-born men with women of tarnished reputation, such as courtesans, procuresses, and adulteresses; while senators were further prohibited from marrying, not only emanci- pated women, but the daughters of parents of ill repute. Yet although matrimonium with such women was forbidden, concubinatus was allowed.4 Among the Teutonic peoples any freeman who had intercourse with a slave was punished with slavery, and a woman guilty of such a crime might.be killed. In the Scandinavian countries slavery came to an end at a com- paratively early period, whilst in Germany it was suc- ceeded by serfdom ; and equality of birth continued to be regarded as an indispensable condition of lawful marriage. As late as the thirteenth century any German woman who had intercourse with a serf lost her liberty.5 From the class of freemen, both in Germany and Scandinavia, the nobility gradually emerged as a distinct order, and marriages between persons of noble birth and persons who, although 1 Nakajima, ' Marriage (Japanese and Korean),' in Hastings, op. cit. viii. 459. 2 Mommsen, History of Rome, i. 371. 3 Rossbach, Untersuchungen uber die romische Ehe, pp. 249, 456, 457 sq. 4 Meyer, Der romische Konkubinat, p. 23 sqq. 0 Winroth, op. cit. pp. 227, 230, 233. Weinhold, D;e deutschen Frauen in dem Mittelalter, i. 349, 353 sq. 64 THE HISTORY OF HUMAN MARRIAGE chap. free, were not noble came to be considered misalliances.1 In Sweden such marriages formerly entailed serious economic consequences.2 Even now there are traces of the former class endogamy in Europe. According to German Civil Law, the marriage of a man belonging to the high nobility with a woman of inferior birth is still regarded as a dis- paragium ; and the woman is not entitled to the rank of her husband, nor is the full right of inheritance possessed by her or by her children.3 But, as Sir Henry Maine remarks, the outer or endogamous limit within which a man or woman must marry has been mostly taken under the shelter of fashion or prejudice. " In France, in spite of all formal institutions, marriages between a person belonging to the noblesse and a person belonging to the bourgeoisie (distinguished roughly from one another by the particle ' de ') are wonderfully rare, though they are not unknown."4 Customs parallel to the hypergamy of the Hindus are found among other peoples. In Madagascar the Anti- merina or patrician caste was divided into six classes, each of which generally claimed descent from a royal ancestor and regarded itself as a group of blood-relations. These groups were endogamous in theory, but a man of a higher class might marry a woman of a lower class. On the other hand, a woman of higher rank was prohibited by strict taboo from marrying beneath her rank ; and if she so far demeaned herself as to marry a commoner, she lost her title of nobility and was disowned by her family.5 The Bushongo consider the son of a slave, even if he is himself a freeman, to be of too humble extraction to become the husband of a Bushongo girl, whereas they allow a freeman to marry the daughter of a slave.6 Among the Seguela on the Ivory Coast a free- woman cannot marry a captive, although a man can marry 1 Weinhold, op. cit. i. 349 sq. 2 Fahlbeck, Sveriges adel, ii. 154. 3 v. Gierke, ' Grundziige des deutschen Privatrechts,' in v. Holt- zendorff, Enzyclopadie der Rechtswissenschaft, i. 203. 4 Maine, Dissertations on Early Law and Custom, p. 224 sq. 5 van Gennep, Tabou et totemisme d Madagascar, p. 125 sq. Cf. Sibree, op. cit. p. 256 ; Grandidier, op. cit. ii. 170 sqq. 0 Torday and Joyce, Les Bushongo, p. 116. XVIII ENDOGAMY 65 a woman whom he has captured.1 The Beni-Amer allow a nobleman to marry a commoner's daughter, but forbid a commoner to marry the daughter of a nobleman.2 Among the Maori " chiefs married slaves, but it was disgraceful for chiefs' daughters to wed slaves." 3 Among the Line Islanders a niki-raurora, or harlot, belonging to the gentry cannot co- habit with a commoner without disgrace, whereas a man of the gentry can cohabit with a harlot of the class of the commoners without disgrace.4 In many cases caste or class endogamy is no doubt due to racial or national differences.5 Social differentiation may be the result of foreign conquest and subjugation, the conquerors becoming the nobility and the subjugated the commonalty or slaves. The Incas of Peru were known as a conquering race. In England, before the Norman conquest the aristocracy was Saxon ; after it, Norman. The descend- ants of the German conquerors of Gaul were for a thousand years the dominant race in France, and until the fifteenth century all the higher nobility were of Frankish or Bur- gundian origin.6 Sir William Ridgeway suggests that the Roman patricians were originally a body of Sabines who had become masters of Rome, while the plebeians were the aboriginal Ligurians; and in Sparta there was an aristo- cracy of conquerors sharply distinguished from the subju- gated Helots.7 It has been argued that the Sanskrit word for caste, varna (lit. " colour "), indicates how the distinction of high and low caste arose in India. This word, which in the later Samhitas and Brahmanas is used in speaking of the four castes, is in the Rig-Veda applied to denote two classes of men, the Dasa and the Aryan Varna, who are contrasted on 1 Clozel and Villamur, op. cit. p. 328. 2 Munzinger, op. cit. p. 313. 3 Thomson, Story of New Zealand, i. 176. 4 Tutuila, in Jour. Polynesian Soc. i. 267. 5 Cf. Landtman, op. cit. ch. vi. (on social inequality caused by the amalgamation of tribes), p. 126 sqq. 6 Hotz, in de Gobineau, op. cit. p. 239. 7 Ridgeway, ' Who were the Rom ms ? ' in Proceedings of the British Academy, 1907-1908, p. 30. VOL II F 66 THE HISTORY OF HUMAN MARRIAGE chap. account of their colour. India was inhabited by a dark people before the fairer Aryans took possession of it, entering from beyond the North-West frontier and bringing with them the Sanskrit language and the religious ideas which are expressed in the Vedas and Upanishads; and the domineering spirit of the conquerors, their bitter contempt for foreign tribes, and their strong antipathies of race and religion, found vent in the sharp distinction which they drew between themselves and the conquered population, the Audras. Having at first too few women of their own, they were often obliged to take aboriginal girls as their wives ; but later on, when this scarcity no longer existed, they closed their ranks to any further intermixture, and when they did this each group became a caste like those of the present day. It is said that even now caste largely corresponds to race, and that, in Northern India at least, the social status of a caste is indicated by its physical type, those at the top having Aryan and those at the bottom an aboriginal physiognomy.1 It is true that these views have not been universally accepted; according to some writers the origin of the caste system can only be found in community of function and occupation.2 But the theory of the ethnological basis of caste may be correct although social rank is nowadays closely connected with occupation. Racial differences would from the beginning have been accompanied by functional differences ; for the conquerors would naturally have reserved for themselves the higher occupations, leaving the more primitive ones to the people of non-Aryan descent, the Sudras.3 The differences of type between nobles and commoners often suggest origin from different stocks. Among the Shans 1 Zimmer, Altindisches Leben, p. 113 sq. Muller, Chips from a German Workshop, i. 322 sq. Risley, Tribes and Castes of Bengal- Ethnographic Glossary, vol. i. p. xxxviii. sqq. Gait, op. cit. p. 380 sq. Rhys Davids, Buddhist India, p. 53 sq. Keith and Macdonell, op. cit. ii. 267. 2 See Nesfield, Brief View of the Caste System of the North-Western Provinces and Oudh, p. 88 sqq.', Crooke, Tribes and Castes of the North-Western Provinces and Oudh, vol. i. p. cxxxix. ' Cf. Gait, -op. cit. p. 381 ; Rhys Davids, op. cit. p. 54. Xvni ENDOGAMY 67 " the majority of the higher classes seemed to be distinguished from the common people by more elongated oval faces and a decidedly Tartar type of countenance."1 Among the Beni-Amer, the nobles are mostly light coloured while the commoners are blackish.2 The Polynesian nobility have a comparatively fair complexion,3 and " are, almost without exception, as much superior to the peasantry or common people, in stateliness, dignified deportment, and physical strength, as they are in rank and circumstances; although they are not elected to their station on account of their personal endowments, but derive their rank and elevation from their ancestry. This is the case with most of the groups of the Pacific, but particularly so in Tahiti and the adjacent islands."4 As descendants of different ancestors, members of noble, families tend to keep up their separate position and remain almost as foreigners to the people among whom they live. Speculating on the want of sympathy between the various classes in societies where such distinctions are recognised, Count de Tocqueville observes :-" Each caste has its own opinions, feelings, rights, manners, and modes of living. Thus, the men of whom each caste is composed do not re- semble the mass of their fellow-citizens ; they do not think or feel in the same manner, and they scarcely believe that they belong to the same human race. . . . When the chroniclers of the Middle Ages, who all belonged to the aristocracy by birth or education, relate the tragical end of a noble, their grief flows apace ; whereas they tell you at a breath, and without wincing, of massacres and tortures inflicted on the common sort of people. Not that these writers felt habitual hatred or systematic disdain for the people ; war between the several classes of the community was not yet declared. They were impelled by an instinct 1 Anderson, Mandalay to Momien, p. 289. 2 Munzinger, Ostafrikanische Studien, p. 336. 3 Waitz-Gerland, op. cit. vi. 6. Lisiansky, Voyage round the World, p. 85 (Nukahivans). 4 Ellis, Polynesian Researches, i. 82. Cf. Beechey, Narrative of a Voyage to the Pacific and Beering's Strait, i; 205 sq. ; Seemann, Viti, p. 79. F 2 68 THE HISTORY OF HUMAN MARRIAGE chap. rather than by a passion ; as they had formed no clear notion of a poor man's sufferings, they cared but little for his fate."1 It is to this exclusiveness, with all that it carries with it, that the prohibition of marriages out of the class, or the general avoidance of such marriages, owes its origin. Ir- regular connections outside the endogamous circle are often looked upon with less intolerance than marriage, which places the parties on a more equal footing. This is true not only in the case of class endogamy. A traveller relates that at Djidda, where sexual morality is held in little respect, a Bedouin woman may yield herself for money to a Turk or European, but would think herself for ever dishonoured if she were joined to him in lawful wedlock.2 Modern civilisation tends more or less to lower or pull down the barriers which separate races, nations, the adherents of different religions, and the various classes of society. It has therefore made the endogamous rules less stringent and less restricted, it has widened the limit within which a man or woman may marry and generally marries. This process has been one of vast importance in man's history. Largely originating in race- or class-pride or in religious intolerance, the endogamous rules have in their turn helped to keep up and strengthen these feelings, whereas frequent intermarriages must have the very opposite effect. There are, finally, endogamous rules referring to the marriage of certain near relatives. Various peoples con- sider that marriage with a cousin is the most proper marriage, or that a man has a right to his cousin's hand, or even that it is his duty to marry her. This, however, nearly always applies to certain cousins only, not to all cousins indiscriminately.3 1 de Tocqueville, Democracy in America, ii. 149 sqq. 2 de Gobineau, op. cit. p. 174 n. 1. • The subject of cousin marriage has been very fully discussed by Sir James G. Frazer in his book Folk-Lore in the Old Testament (ii. 98 sqq.), which was published only after the present chapter was written. If this chapter had not already been in type when I had an opportunity to peruse Sir James's book I should have availed myself of several of its facts, but my general views on the subject have only been strengthened by his discussion. Like myself, he xvm ENDOGAMY 69 The ancient Arabs held that a man had a right to wed his bint 'amm, that is, the daughter of his father's brother j1 such a marriage strengthened the kinship tie,2 and kept the property in the family.8 The right to the bint 'amm's hand is still frequently recognised in the Muhammadan world. According to Burckhardt, " all Arabian Bedouins acknowledge the first cousin's prior right to a girl; whose father cannot refuse to bestow her on him in marriage, should he pay a reasonable price ; and that price is always some- thing less than would be demanded from a stranger. He is not obliged to marry her, but she cannot, without his consent, become the wife of any other person. If a man permits his cousin to marry her lover, or if a husband divorces his runaway wife, he usually says, ' She was my slipper, I have cast her off.' "4 Among the peasants of Palestine, if a girl is given in marriage to another man, her cousin even considers himself entitled to carry her away by force from the bridal procession ;5 and I met with a similar assigns much importance to exogamic prohibitions and economic motives. He maintains {ibid. ii. 263 sq.) that marriages with the daughter either of a mother's brother or of a father's sister " owe their popularity in large measure to their cheapness, a man having a claim on the hands of such cousins and getting them to wife, either for nothing, or at a lower rate than he would have had to pay for wives who were not so related to him." Sir James also believes {ibid. ii. 245 ; see also ibid. ii. 210) that the preference for the marriage of cross-cousins was " a direct consequence " of an earlier practice of interchanging sisters ; but, although the two practices may have partly sprung from similar motives, I can find no reason for regarding one of them as a preceding cause of the other. ^Robertson Smith, Kinship and Marriage in Early Arabia, pp. 82, 138, 164. Wellhausen, in Nachrichten Konigl. Gesellsch. Wissensch. Gottingen, 1893, no. n, p. 436 sq. Wilken, Das Matri- archat bei den alten Arabern, p. 59. 2 Wellhausen, loc. cit. p. 437 sq. 3 Kitab al-agani, ed. Bulak, viii. 113, quoted by Goldziher, ' Endogamy and Polygamy among the Arabs,' in Academy, xviii. 26. 4 Burckhardt, Notes on the Bedouins and Wahdbys, pp. 154, 64 sq. Cf. Burton, Pilgrimage to Meccah and Medinah, ii. 84. 6 Klein, ' Mittheilungen fiber Leben, Sitten und Gebrauche der Fellachen in Palastina, ' in Zeitschr. d. Deutschen Palaestina-Vereins, vi. 84. See also Robinson Lees, Witness of the Wilderness, p. 121 ; Wilson, Peasant Life in the Holy Land, p. 107 sq. ; Jaussen, Coutumes des Arabes au pays de Moab, p. 45 sqq. THE HISTORY OF HUMAN MARRIAGE chap. 70 idea in Morocco.1 In the Rif instances are known in which an uncle who has married his daughter to another man has been killed by his nephew. In Morocco, too, the bride price {sddq) paid for a paternal uncle's daughter is often smaller than usual, although it also happens that a man tries to pre- vent his nephew from marrying his daughter by making his claims excessive*. Marriages between such cousins are, moreover, popular because they keep the property in the family, as also, especially in shereefian families, because they preserve the blood pure. They are said to be conducive to domestic happiness : whilst marrying a strange woman is like drinking water from an earthenware bottle, marriage with a cousin is like a drink from a dish-you are aware of what you drink.2 Such a marriage gives the husband greater power over his wife, since, if she runs away, her father or brother will bring her back. It has the further advantage that she cannot curse her husband by cursing his ancestors without implicating herself in the curse. It confers religious merit on a man to marry his cousin-by doing so he will not be punished on the day of Resurrection ; and at the same time it is a kind of duty. " He who carries away the dunghills of the people carries away his own "-it is not right of a man to leave his own cousin unmarried by taking another woman for wife.3 Among the Muhammadans of India marriage of first cousins, whether the children of 1 Westermarck, Marriage Ceremonies in Morocco, p. 53. For cousin marriages in modem Egypt see Lane, Modern Egyptians, p. 170 sq. •, Idem, Arabian Society in the Middle Ages, p. 227; Klunzinger, Upper Egypt, p. 196. For similar marriages in Algeria see Gaudefroy-Demombynes, Les ceremonies du mariage chez les indigenes de I'Algtrie, p. 7. See also on this subject Burckhardt, A rabic Proverbs, p. 181 ; Snouck Hurgronje, Mekkanische Sprich- worter und Redensarten, p. 17. 2 On the other hand, I was told at Fez that cousin marriages easily lead to quarrels between the husband's and wife's families, both of whom want to interfere in the married life of the couple. So also the poet of the Mu'allakat gives in his testament the follow- ing advice to his children : "Do not marry in your own family, for domestic enmity arises therefrom '' {Kitab al-agani, ix. 185, quoted by Goldziher, in Academy, xviii. 26. See also Wellhausen, loc. cit. p. 4~7). 3 Westermarck, Marriage Ceremonies in Morocco, p. 53 sq. XVHI ENDOGAMY 71 two brothers or of two sisters, or of a brother and sister, is considered very suitable, and failing them an alliance is preferred with some family with which there have already been marriage relations ; yet no marriage is more common than that of the children of two brothers. It is sometimes said by these Muhammadans that the object of cousin mar- riage is to keep the family as free as possible from foreign blood, and to retain in the family the property inherited by the young couple.1 In Madagascar marriage between brothers' children is exceedingly common, and is looked upon as the most proper kind of connection as keeping property together in the same family; the marriage of two persons nearly re- lated to each other is called Idvatsi-miftndra, that is, " in- heritance not removing." Marriage between brothers' and sisters' children is also allowable on the performance of a slight prescribed ceremony, supposed to remove any impediment from consanguinity; but that of sisters' children, when the sisters have the same mother, is regarded with horror as incest.2 Speaking of the Bantu tribes of South Africa, Mr. McCall Theal states that the native of the mountains in the interior almost as a rule married the daughter of his father's brother, in order to keep property from being lost to his family.3 In many cases the most proper marriage is considered to be that between a man and his father's sister's daughter or his mother's brother's daughter, a so-called " cross- cousin marriage." This custom is very prevalent in the south of India and is also found in certain communities in other parts of that country.4 Although a man may seldom 1 Gait, op. cit. pp. 252, 256. Matin-uz-Zaman Khan, Census of India, 1911, vol. xx. (Kashmir) Report, p. 139. 2 Sibree, op. cit. p. 248. Grandidier, op. cit. ii. 167. 8 Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 209. Idem, History of the Boers in South Africa, p. 16 sq. 4 See Rivers, ' Marriage of Cousins in India,' in Jour. Roy. Asiatic Soc. 1907, p. 625 sqq. ; Gait, op. cit. p. 255 sq. ; Frazer, Totemism and Exogamy, ii. 224 sqq.; Richards, ' Cross Cousin Marriage in South India,' in Man, xiv. 194 sq. In Dr. Rivers's list no distinction is made between cases in which cross-cousin marriage is ordained and cases in which it is merely allowed. 72 THE HISTORY OF HUMAN MARRIAGE chap. marry the daughter of his father's brother or of his mother's sister, he has a right or may even be obliged to marry the daughter of his mother's brother1 or the daughter of his father's sister,2 or one or the other of these relatives.3 The 1 E.g., in the Telugu country, where it is called menarikam (Padfield, The Hindu at Home, p. 97), especially among the Komatis of North Arcot (Cox, Madras District Manuals: North Arcot, i. 2O5 SI- >' Jogendra Nath Bhattacharya, op. cit. p. 222 ; Thurston, Ethnographic Notes in Southern India, p. 54) and also among the Ayyar Brahmans who have adopted the custom (ibid. p. 54) ; among the Muduvars or Mudugars in Coimbatore, Madura, Malabar, and Travancore (Thurston, Castes and Tribes of Southern India, v. 92), the Mikirs in the Province of Assam (formerly ; Stack, Mikirs, p. 18), and the Chins or Khyengs living in the mountains between Assam and the Chinese province Junnan (Maung Tet Pyo, Customary Law of the Chin Tribe, pp. 4, 5, 8 sq.). Among the Yeru- kalas, a nomad tribe in the Telugu country, the first two daughters of a family may be claimed by the maternal uncle as wives for his sons (Thurston, Ethnographic Notes, p. 56). According to Nayar ideas the most desirable bride for a young man is the daughter of his maternal uncle (Anantha Krishna Iyer, Cochin Tribes and Castes, ii. 22). 2 E.g., among the Kappiliyans (Canarese-speaking farmers ; Thurston, Ethnographic Notes, p. 52), Nattamans (Tamil cultivators ; ibid. p. 54), Bottadas and Omanaito (Uriya cultivators or labourers ; Thurston, Castes and Tribes, i. 265, v. 444), Nattukottai Chettis (ibid. v. 265), Kunnuvans and other castes of the Palni Hills (Dahmen, ' Kunnuvans or Mannadis,' in Anthropos, v. 325), Kasubas of the Nilgiris (Hayavadana Rao, ' Kasubas,' ibid. iv. 179), the wild Gonds of the Madras Presidency, &c. (Idem,'' Gonds of the Eastern Ghauts,' ibid. v. 794). Among the Kalians of the Madura district, according to Mr. Thurston (Ethnographic Notes, p. .53), the most proper alliance is one between a man and the daughter of his father's sister, but it is also said that if his father's brother has a daughter, and insists upon his marrying her, he cannot refuse, whatever may be the woman's age. 3 E.g., among the Tottiyans (Telugu cultivators ; Thurston, Ethno- graphic Notes, p. 52), Vallambans (Tamil cultivators; ibid. p. 53 sq.), Todas (Rivers, Todas, pp. 483, 512), Irulans of the Gingee Hills (Hayavadana Rao, ' Irulans of the Gingee Hills,' in Anthropos, vi. 811), Gonds of the Central Provinces (Grant, Gazetteer of the Central Provinces, p. 276), Rajput Bhotias in the Darma pargana, Almora District (Sherring, ' Notes on the Bhotias of Almora and British Garhwal,' in Memoirs Asiatic Soc. Bengal, i. 105). A Nayadi, if possible, marries his maternal uncle's daughter or his father's sister's daughter (Madras Government Museum's Bulletin, iv. 75). XVIII ENDOGAMY 73 strictness with which these rules are followed and the rigour with which they are enforced vary in different tribes or castes. Among the Kasubas of the Nilgiris a man is not bound to marry his father's sister's daughter if she is older than himself-he may then marry either his sister's daughter or his mother's brother's daughter j1 but in other communi- ties no disparity between their respective ages is an excuse for not marrying her.2 According to the customary law of the Chins or Khyengs-who live in the mountains between Assam and the Chinese province Junnan-the daughters of brothers and the sons of sisters have the right of inter- marriage, " but if one party is not come of age while the other party is, the party who has come to puberty shall not be bound to wait for the minor till he or she comes of age." Otherwise the party who breaks the law " and marries elsewhere " shall pay a fine, except that if, during the life- time of a sister and a brother, the son of the sister breaks his engagement, no fine shall be taken, provided a promise be made for the marriage of the grandchildren.3 Among the Mikirs of the Province of Assam, in former days, if a boy did not marry his first cousin on the mother's side, the maternal uncle could beat him as much as he liked.4 In some cases the parents of a girl will be excommunicated from the caste if they marry her to a man other than the relative who has a claim on her; but in other cases it is a mere matter of form to offer the cousin the first refusal.5 Among the Sinhalese of Ceylon " two sisters' or brothers' children are rarely allowed to marry, but the solicitation of a mother's brother's or a father's sister's son is always preferred to that of any other." 6 Among the He-miao, one of the aboriginal tribes of China, a girl must marry a 1 Hayavadana Rao, in Anthropos, iv. 179. 1 See, e.g., Thurston, Ethnographic Notes, p. 52 sq. (Tottiyans, Kalians); Dahmen, in Anthropos, v. 325 (Kunnuvans, &c.). 8 Maung Tet Pyo, op. cit. pp. 4, 5, 8 sq. 4 Stack, op. cit. p. 18. 6 Richards, in Man, xiv. 194 sq. 6 Perera, Glimpses of Singhalese Social Life, p. 3. See also Lewis, ' On the Terms of Relationship in Sinhalese and Tamil,' in Orientalist, i. 217 sq. THE HISTORY OF HUMAN MARRIAGE chap. 74 son of her maternal uncle,1 that is, a man has a right to the hand of his father's sister's daughter. According to Gilyak custom, a man shall marry his mother's brother's daughter and the girl her father's sister's son.2 In certain parts of Sumatra, whilst marriage with the daughter of a father's sister is forbidden, marriage with the daughter of a mother's brother is regarded as desirable and normal; and if a man refrains from contracting such a marriage his uncle may take offence, nay some people even say that the gods are angry.3 In New Caledonia, according to Lambert, marriages be- tween first cousins who are the children of a brother and a sister respectively are thought particularly proper, whereas the children of two brothers or of two sisters must strictly avoid each other, as if they were brothers and sisters.4 Sir Basil Thomson states that the young Fijian is from his birth regarded as the natural husband of the daughters of his father's sister and of his mother's brother. " In theory the field of choice is very large, for the concubitant relationship might include third or even fifth cousins, but in practice the tendency is to marry the concubitant who is next in degree-generally a first cousin-the daughter of a maternal uncle." The natives assured our informant that the practice of concubitancy has greatly decreased since the introduction of Christianity and settled govern- ment, but even now thirty per cent, of the people are said to marry their concubitants. Marriages between cousins who are the children of two brothers or of two sisters are prohibited ; and in some places the offspring of a brother and sister respectively do not become concubitant until the second generation.5 Among the Australian Urabunna a man's proper wife is always either the daughter of his mother's elder brother or of his father's elder sister, whereas he is strictly forbidden to marry the daughter of his mother's sister 1 Schotter, ' Notes ethnographiques sur les tribus de Kouy- tcheou (Chine),' in Anthropos, vi. 320. See also Neumann, Asiatische Studien, i. 74. 2 Sternberg, in Verhandl. Berliner Gesellsch. Anthr. 1901, p. 38. 3 Frazer, Totemism and Exogamy, ii. 188, 191. 4 Lambert, Mceurs et superstitions des Ndo-Calddoniens, pp. 94, 1145^. 6 Thomson, Fijians, p. 183 sqq. XVIII ENDOGAMY 75 and of his father's brother, because these first cousins belong to the same exogamous class as himself.1 In Africa cross- cousin marriage is the general custom among the Herero, who also strictly prohibit marriages between the children of two brothers and between the children of two sisters.2 Among many other peoples cross-cousin marriage is said to be allowed, but this, of course, does not imply that it is regarded as the most proper marriage. In Savage Island, for example, the children of a brother and sister, as also the children of two brothers, may marry without shocking the sentiments of the community, but such marriages are not popular.3 The Fanti of West Africa allow a man to marry his father's sister's daughter, though not his mother's sister's daughter, but the union is not encouraged.4 Among the East African Sango cross-cousins are the only cousins who are permitted to intermarry, but marriages into non- related families are preferred.5 That cross-cousins so often are allowed to marry each other is no doubt due to the fact that they belong to different groups. On the other hand, where the clan system prevails the children of two brothers belong to the same clan if descent is traced through the father, and the children of two sisters belong to the same clan if descent is traced through the mother; and in the former case the children of two sisters also belong to the same clan if the sisters have married into that clan, and in the latter case the children of two brothers belong to the same clan if the brothers have married into that clan. But even when the children of brothers or the children of sisters belong to different clans they may stand to each 1 Spencer and Gillen, Native Tribes of Central Australia, pp. 61, 64. I idem, Northern Tribes of Central Australia, p. 71 sq. Cf. Frazer, Totemism and Exogamy, i. 180 sq. 2 Schinz, Deutsch-Siid-West-Afrika, p. 177. Bensen, quoted by Kohler, ' Das Recht der Herero,' in Zeitschr. vergl. Rechtswiss. xiv. 300. Dannert, Zum Rechte der Herero, pp. 33, 37. v. Frangois, Nama und Damara Deutsch-Siid-West-Afrika, p. 195. 3 Thomson, Savage Island, p. 135 sq. 4 Sarbah, Fanti Customary Laws, p. 46. 5 Heese, * Sitte und Branch der Sango,' in Archiv f. Anthrop. N.S. xii. 134. THE HISTORY OF HUMAN MARRIAGE chap. 76 other in a closer social relationship than the children of a brother and sister, and marriages between them may on that account be regarded as incest. It has also been sug- gested that where a change has been made from maternal to paternal descent and it consequently became wrong to marry the daughter of the father's brother (who would now be of the same clan), the old prohibition against marriage with the daughter of the mother's sister would probably still persist; whereas so far as clan restrictions are concerned there would be no necessary prohibition of marriage with the daughter of the mother's brother and the father's sister.1 But the problem which we have here to solve is not why cross-cousin marriages are merely allowed, but why they are prescribed by custom. The solution of this problem is probably to be iound in circumstances very similar to those which among other peoples have led to customary marriages between brothers' children. Cross-cousin marriages may serve the object of keeping together related families, and this object may be the more important since the families in the present case would belong to different clans. Of the Maori we are told that " among brothers and relatives, in order to cement and perpetuate friendship, it is often arranged that if the one party has a girl and the other a male child, their marriage shall take place at an early period " ;2 and even in civilised society it is not unusual to find friends, endeavouring to strengthen the bond between then! by encouraging marriages between their respective children.3 And not only does cross-cousin marriage keep the family together, but it also prevents dispersion of the family property.4 Hence in a society where inheritance runs through females a father who wishes to provide for his son out of the ancestral property would marry him to his sister's daughter if not to his own sister, and a father who wishes to provide for 1 Rivers, in Jour. Roy. Asiatic Soc. 1907, p. 622. Gait, op. cit. p. 256. 2 Polack, Manners and Customs of the New Zealanders, i. 136. 3 Cf. Gait, op. cit. p. 256 sq. 4 Cf. Molony, Census of India, 1911, vol. xii. (Madras) Report, 107 ; Dannert, op. cit. p. 34 (Herero). xvin ENDOGAMY 77 his daughter would marry her to his sister's son, who is his own heir.1 Thus among the Carrier Indians of the Western Dene, who count descent exclusively on the mother's side, a man frequently arranged that his legal inheritor and' successor, his sister's son, should marry one or more of his daughters in order " that his own offspring might share in his property and not be wholly deprived under the clan rule of his possessions, as under their laws no hereditary property or rights could be alienated or passed over to the members of another clan even though the recipients were the donor's own children. These laws and regulations were very strictly observed among the Carriers, and hence this practice of marrying a girl to her first cousin on her father's side."2 With reference to cross-cousin marriage in South India, again, Mr. Richards observes that the rule which gives a man the first refusal of his maternal uncle's or his paternal aunt's daughter in marriage may be interpreted as a sort of compromise between matrilineal succession and Brah- manic law, which lays emphasis on inheritance through males ; for it preserves inviolate the principles of matrilineal inheritance under patrilineal forms.3 Cross-cousin marriage may also secure other advantages. Mr. Hayavadana Rao maintains that one reason for a man's claim to the hand of his father's sister's daughter among the Gonds of the Eastern Ghauts, as also in the southern districts of Tinnevelly and Madura and in Salem and Coimbatore, may be " the incur- ring of less marriage expenses, a bride amongst these tribes and castes being rated at very heavy prices."4 Among various peoples a marriage with a cross-cousin is expressly said to be cheaper than one with a non-related person ;5 1 Cf. Richards, in Man, xiv. 196 sqq. ; Gait, op. cit. p. 257. 2 Hill-Tout, The Far West, the Home of the Salish and p. 145 sq. 3 Richards,in Man,xiv. 198. 4HayavadanaRao,in24M^ro/>05,v.794. 5 E.g., the Yerukalas in the Telugu country (Thurston, Ethno- graphic Notes, p. 56), Gonds of the Central Provinces (Grant, op. cit. p. 277 ; Kohler, ' Indische Gewohnheitsrechte,' in Zeitschr. vergl. Rechtswiss. viii. 144), He-miao in China (Schotter, in Anthropos, vi. 320). In Sierra Leone a man marrying another member of his family does not pay so large a bride price as he would if marrying into another family (Vergette, Certain Marriage Customs of some of the Tribes in the Protectorate of Sierra Leone, p. 9). 78 THE HISTORY OF HUMAN MARRIAGE chap. among the Irulans of the Gingee Hills1 and in Hiw, of Torres Islands,2 no bride price at all is paid for the daughter of a maternal uncle. Again, of a man's claim to his paternal aunt's daughter in the Madura country it is said that the motive for this custom " appears to be the feeling that a woman is bound to replace the loss to her father's family occasioned by her marrying out of it by returning one of her daughters to that family."3 The idea of restitution is clearly implied in the expression dudh lautana, " giving back the milk," by which the giving of a daughter in marriage to the son of her mother's brother is known among certain tribes of the Central Provinces.4 The He-miao of China say that such a marriage is "to restore the seed furnished to another family."5 We may thus trace the custom of cross-cousin marriage to various motives without assuming, as certain writers have done, that it is merely a survival which has no meaning except as the vestige of an old social order. Dr. Rivers finds it very difficult " to conceive any motive which should make the marriage of the children of brother and sister desirable, while the marriage of the children of two brothers or of two sisters is so strictly forbidden."6 But as we have seen, there may be reasons why marriage with a cousin is held desirable, and these may lead to cross-cousin marriage in cases where other kinds of cousin marriage are regarded as incestuous; and there are, further, circumstances in which cross-cousin marriage may be considered the most desirable kind of marriage, whether other cousin marriages be allowable or not. Dr. Rivers has suggested that in India and elsewhere cross-cousin marriage is derived from the bisection of the community into two exogamous moieties or classes, such as is found in some Australian tribes ;7 but apart from the significant fact that no such organisation 1 Hayavadana Rao, in Anthropos, vi. 811. 2 Rivers, History of Melanesian Society, i. 184 sq. 3 Francis, quoted by Dahmen, in Anthropos, :. 325. 4 Marten, quoted by Gait, op. cit. p. 257. fi Schotter, xn Anthvopos, vi. 320. 0 Rivers, in Jour. Roy. Asiatic Soc. 1907, p. 623 sq 7 Ibid. p. 622 sq. XVIII ENDOGAMY 79 of society is known to have existed in India, it is difficult to understand why it should have led to the marriage of first- cousins to the exclusion of marriages between other members of the two exogamous moieties. In Melanesia, on the other hand, Dr. Rivers believes that cross-cousin marriage has not been a direct and immediate consequence of the dual organisa- tion of society, but has been the ultimate effect of a series of events which have resulted from a condition of dominance of the old men within such a dual organisation. The old men monopolised all the young women of the community ; as young men grew up their only chance of obtaining wives was that women who had already been the wives of their elders might be given to them ; the old men first gave their wives to their sisters' sons, but subsequently they gave their daughters instead of their wives, and this produced the cross-cousin marriage.1 This whole scheme I consider to be a baseless conjecture ; for reasons stated in a previous chapter, I cannot recognise as valid any attempt to support it by conclusions drawn from terms of relationship. Here again I must protest against the method of trying to explain customs or institutions which may be satisfactorily accounted for by known facts as survivals of unknown and entirely hypothetical conditions in the past. In some cases marriages between relatives even more closely related than cousins are prescribed by custom. In the Koracha caste in Mysore " there is a special obligation for a man marrying his sister's daughter, even though older than himself " ;2 and a similar custom obtains among other castes in the south of India.3 The Kappiliyans-Canarese- speaking farmers-recognise a strong right in a man to marry his sister's or aunt's daughter.4 Among the Kasubas of the Nilgiris, if a man is younger than his paternal aunt's daughter he may marry either his sister's daughter or his 1 Rivers, History of Melanesian Society, ii. 59, m, 122. 2 Thyagaraja Aiyar, Census of India, 1911, vol. xxi. (Mysore) Report, p. 99. 3 Mandlik, Vyavahdra Mayukha, ii. 425. Ranga Rao, ' Yanadis of the Nellore District,' in the Madras Government Museum's Bulletin, iv. 99. 4 Thurston, Ethnographic Notes, p. 52. 80 THE HISTORY OF HUMAN MARRIAGE chap. maternal uncle's daughter.1 Among the wild Miao of China a girl must marry the son of her mother's brother, but if the uncle has no son he may himself marry her.2 Among the Irulans of the Gingee Hills one of the most proper kinds of marriage is the marriage between a man and his younger brother's daughter.3 If the paternal aunt of a Kalian has no daughter he must marry his aunt or his niece or some other near relative.4 Among the Garos marriage with one's maternal aunt is very common and, in default of such a relative, marriage with one's maternal uncle's daughter; if a widowed maternal aunt is available, marriage with any other female is deemed improper, if not impossible.5 In Hiw " it was said that if the father's sister was young enough a man would marry her, but if not, he would marry her daughter. The cross-cousin marriage was spoken of as a kind of substitute for marriage with the father's sister."6 The origin of these kinds of marriages seems to be very similar to that of cross-cousin marriages, with which they alternate. Among the Irulans no bride price need be paid by a man who marries his younger brother's sister.7 In the Meda caste in Mysore, if a girl is married to her maternal uncle the amount of the bride price is either reduced or excused altogether.8 With reference to similar marriages in the same part of India it is said that when people are rich the maternal uncles take their nieces in marriage out of affection towards them, and that when they are poor such marriages take place out of necessity to both parties. The maternal uncles cannot get brides without much expense, and the nieces' parents have no dowry to give to a strange bride- groom. If the parents of the nieces are poor and the maternal uncles are rich, the former insist upon the latter's taking 1 Hayavadana Rao, in Anthropos, iv. 179. 2 Neumann, Asiatische Studien, i. 74. 3 Hayavadana Rao, in Anthropos, vi. 811. 4 Thurston, Ethnographic Notes, p. 53. 5 Chunder Dey, ' Account of the Garos,' in Calcutta Review, cxxviii. 158. c Rivers, History of Melanesian Society, i. 184. 7 Hayavadana Rao, in Anthropos, vi. 811. 8 Thyagaraja Aiyar, op. cit. p. 100. XVIII ENDOGAMY 81 the nieces in marriage, as the mother of the niece cannot bear to see her brother's riches go to a stranger.1 Among some peoples it has even been required of brothers to marry their sisters, namely, in the case of kings in order to maintain the blood royal pure from admixture. Instances ol this will be given in the following chapter. 1 Mandlik, op. cit. ii. 425. VOL. II CHAPTER XIX EXOGAMY From the endogamous rules, which forbid the members of a particular group to marry any one who is not a member of the group, we shall pass to the exogamous rules, which forbid the members of a particular group to marry any other member of it.1 The exogamous group is in most cases composed of persons who are, or consider themselves to be, related by blood or of the same kin ; and the nearer the relationship, the more frequently it is a bar to inter- marriage, at least within the same line of descent. The most frequent of all exogamous rules are those which prohibit a son from marrying his mother and a father from marrying his daughter. These rules seem, in fact, to be universally prevalent in mankind. We hear, it is true, of " marriages " between parents and children among certain peoples. But in several of these cases they are expressly said to be disapproved of; and it is exceedingly doubtful whether such unions have ever been sanctioned by the customs of any people. Du Tertre states that the Caribs " have no prohibited degree of consanguinity among them," and that fathers have been known to marry their own daughters and mothers 1 The term " exogamy " is generally used for a prohibition of marrying inside a larger group than one consisting of members of the same family, particularly the clan ; but neither in the etymology of the word nor in the nature of the prohibitions can I find any reason for such a restriction. On the contrary, it has the dis- advantage of disassociating rules which intrinsically belong to the same class and, in my opinion, have the same foundation. ch. xix EXOGAMY 83 to marry their sons, " though that is very rare " j1 and de la Borde says that he has himself seen some Caribs who had their daughters for wives.2 But de Poincy denies the accuracy of the statement that fathers marry their daughters among them, and adds that they have a horror of such a crime.3 Among the Piojes of Ecuador, according to Mr. Simson, " a widow often takes her son to replace the deceased husband; and a widower his daughter, upon the death of his first wife."4 Ross states that among the Eastern Tinne of North-West America instances of men united to their mothers, their sisters, or their daughters, though not common, are far from rare; he heard of two sons. keeping their mother as a common wife, of another wedded to his daughter, and of several married to their sisters, but he remarks that such unions are not considered correct by general opinion.5 Speaking of the Southern Indians of the Tinne stock, Hearne observes that "it is notoriously known, that many of them cohabit occasionally with their own mothers, and frequently espouse their sisters and daughters " ; and that he has known several of them who, " after having lived in that state for some time with their daughters, have given them to their sons, and all parties been perfectly reconciled to it."6 But he admits that the Northern Indians hold incestuous unions in ab- horrence ; and Father Morice affirms that his own experience goes to confirm the latter observation.7 We hear of a case of an Eskimo from the Ungava District, Hudson's Bay territory, who took his mother for wife ; but " the sentiment of the community compelled him to discard her." 8 Dalager 1 Du Tertre, Histoire generate des Antilles, ii. 377 sq. 2 de la Borde, Relation de I'origine, mceurs, coustumes &c. des Caraibes, p. 19. 3 de Poincy, Histoire naturelle et morale des Iles Antilles de I'Amerique, p. 549. 4 Simson, Travels in the Wilds of Ecuador, p. 196. 5 Ross, ' Eastern Tinneh,' in Smithsonian Report, 1866, p. 310. 6 Hearne, Journey from Prince of Wales's Fort to the Northern Ocean, p. 130 n.* 7 Morice, ' The Great D6n6 Race,' in Anthropos, ii. 33. 8 Turner, ' Ethnology of the Ungava District,' in Ann. Rep. Bur. Ethn. xi. 180. 84 THE HISTORY OF HUMAN MARRIAGE chap. knew a Greenlander who married his own daughter; but the natives themselves said that the very thought of such a crime gave them palpitation.1 Among the Banyoro of Central Africa, according to Wilson and Felkin, brothers may marry their sisters, and even fathers their daughters, although a son does not marry his own mother ;2 but this does not agree with the Rev. J. Roscoe's statement that even in the case of royalty a father refrained from marrying his daughter.3 Speaking of the king of the Warua, Cameron observes that in his harem are to be found not only his sisters and nieces, but even his own daughters.4 Among the Bushmen, according to Stow, marriage seems to have been barred by no degrees of relationship except those of parent and child and brother and sister, " although it has been declared that among some of the isolated clans even this restriction did not exist." 5 Turner states that among the New Caledonians no laws of consanguinity were observed in their marriages, and that even the nearest relatives united ; 6 but this statement differs from other more authoritative descriptions of the same people.7 We are told that in Kiwai, an island off New Guinea, a father may take his own daughter to wife, although marriage is not allowed between brother and sister nor between cousins ;8 but this statement must also be wrong, since it is a well-established fact that among the Kiwai Papuans people of the same totem cannot marry and 1 Dalager, Gr^nlandske Relationer, p. 68. a Wilson and Felkin, Uganda and the Egyptian Soudan, ii. 49. 3 Roscoe, Northern Bantu, p. 36 sq. 4 Cameron, Across Africa, ii. 70. 6 Stow, Native Races of South Africa, p. 95. The existence of brother-and-sister marriages among the Bushmen is also denied by other authorities (Kaufmann, ' Die Auin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 156; Trenk, 'Die Buschleute der Namib,' ibid. xxiii. 168). 6 Turner, Samoa, p. 341. 7 de Rochas, La Nouvelle Caledonie, p. 232. Lambert, Moeurs et superstitions des Neo-Calddoniens, p. 94. 8 Chalmers, ' Notes on the Natives of Kiwai Island,' in four. Anthr. Inst, xxxiii. 124. XIX EXOGAMY 85 children inherit their father's totem.1 In Bougainville and Buka, in the Solomon Group, it not infrequently happens that a father consorts with his own daughter and begets children by her; this, we are told, is not regarded as ille- gitimate, whereas the union of two persons of the same crest is viewed as a crime.2 In Hawai marriage between father and daughter is said to have occurred among chiefs,3 whilst marriage between brother and sister in certain cases, as will be seen presently, was a well-known institution among them. In the Marshall Islands incestuous unions or marriages between brother and sister and between father and daughter are said to be known, especially in the families of chiefs, but they are looked upon as disgraceful and are supposed to be punished in the future life.4 Kubary speaks of a Pelew Islander who married his own daughter, and in consequence became an object of scorn and ridicule among his compatriots, who came from all parts of the group to look at the curious couple ; but not infrequently among those islanders fathers have connection with their young daughters, though public opinion disapproves of it.5 In the Malay Archipelago marriages between brothers and sisters and between parents and children are said to occur in certain tribes.6 Thus it is reported by the Javanese that among the Kalangs, who have claims to be considered 1 Landtman, Nya Guinea farden, p. 80. Hely, quoted by Haddon and Rivers, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 189. Haddon, Headhunters, p. 103. 2 Parkinson, Zur Ethnographic der nordwestlichen Salomo Inseln, p. 6. 3 Poepoe, quoted by Rivers, History of Melanesian Society, i. 381 sq. 4 Kohler, ' Das Recht der Marschallinsulaner,' in Zeitschr. f. vergl. Rechtswiss. xiv. 419. Cf. infra, ii. 130, 177. 6 Kubary, ' Die Verbrechen und das Strafverfahren auf den Pelau-Inseln,' in Original-Mittheil. ethnol. Abtheil. konigl. Muzeen zu Berlin, i. 81. 6 Wilken, Over de verwantschap en het huwelijks- en erfrecht bij de volken van het maleische ras, p. 22. Idem, ' Plechtigheden en gebruiken bij verlovingen en huwelijken bij de volken van den Indischen Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, ser. v. vol. i. 151. Schwaner, Borneo. Beschrij- ving van het stroomgebied van den Barito, i. 198. 86 THE HISTORY OF HUMAN MARRIAGE chap. the aborigines of Java, mother and son often live together as man and wife, and that they even consider such a union to lead to prosperity and riches ;1 and we are told that in the southern districts of Minahassa, in olden times, father and daughter, mother and son, brother and sister, frequently lived together in bonds of matrimony.2 Among the Karens of Tenasserim, according to Helfer, " matrimonial alliances between brother and sister, or father and daughter, are not uncommon " ;3 but Dr. Alonzo Bunker, who had been a resident among the Karens of Burma for more than twenty years, assured me that marriages between parents and children and between brothers and sisters are prohibited everywhere among the Karens, and that even first cousins very seldom marry, though there is no law against such connections. Among the Kukis, as described by Rennel, marriages were generally contracted without regard to blood-relationship, with the exception that a mother could not wed her child ; 4 whereas Stewart asserts of the same people that they have the strictest rules forbidding too close intermarriage in families and even marriage between cousins.5 Greek writers state that the ancient Persians married near relatives, even their mothers, sisters, and daughters.6 But these statements have been severely criticised by modern Parsi scholars, who maintain that the so-called xvaetvadatha 1 Ketjen, * De Kalanger^/in Tijdschrift voor Indische taal-, land- en volkenkunde, xxiv. 427.' 2 Hickson, A Naturalist in North Celebes, p. 277 sq. 3 Helfer, 'Animal Productions of the Tenasserim Provinces, in Jour. Asiatic Soc. Bengal, vii. 856. 4 Lewin, Wild Races of South-Eastern India, p. 276. 5 Stewart, ' Notes on Northern Cachar,' in Jour. Asiatic Soc. Bengal, xxiv. 640. 6 See Rapp, ' Die Religion und Sitte der Perser und iibrigen Iranier nach den griechischen und romischen Quellen,' in Zeitschr. Deutsch. Morgenland. Gesellsch. xx. 112 sq. ; Hiibschmann, ' Ueber die persische Verwandtenheirath,' ibid, xliii. 308 sqq. ; Gray, ' Marriage (Iranian).-Next-of-kin marriage,' in Hastings, Encyclo- pedia of Religion and Ethics, viii. 457. According to Herodotus (iii. 31), it was not the custom of the Persians before the time of Cambyses to marry their sisters. XIX EXOGAMY 87 or (in Pahlavi) xvetokdas, which the foreign writers regarded as referring to marriages between parents and children and between uterine brothers and sisters, was a marriage between first cousins, as is certainly its present connotation.1 And West observes that, so far as can be ascertained from the extant fragments of the Avesta-the only internal authority regarding the ancient practices of Mazda-worship -the Parsis are perfectly justified in believing that their religion did not originally sanction marriages between those who are next of kin. When we descend to the Pahlavi translations and writings of the better class, which in their present form probably range from the sixth to the ninth century, we find many allusions to xvetokdas between the nearest relatives. Such marriages are defended and ad- vocated " with all the warmth and vehemence that usually indicate much difficulty in convincing the laity." The priests, in the later years of the Sasanian dynasty and for some centuries subsequently, strongly advocated such marriages, though probably with little success. As we come to the modern writings of the Persian ' Rivayats,' which may have commenced about the fifteenth century, we find the present form of xvetokdas, the marriage of first cousins, the only form in use, though obscure allusions are made to the other forms as being long extinct.2 The Chinese historian Ma-touan-lin, who wrote in the thirteenth century, says that the A-si, or Parthians, married theii sisters and mothers " in the same manner as the brutes."3 The old Prussians are said to have married blood-relatives with the exception of " almost nobody " but their own mothers.4 Among the Irish close kinship ties were not always an obstacle to marriage : Lugaid, supreme king of Ireland, married his mother, and a king of Leinster had his two 1 Darab Dastur Peshotan Sunjana, N ext-of-kin Marriages in Old Irdn, p. 16 sqq. and passim. 2 West, ' The Meaning of Khvetdk-das or Khv^tudad,' in Sacred Books of the East, xviii. 427 sq. See also Moulton, Early Zoro- astrianism, p. 207. 3 Rdmusat, Nouveaux melanges asiatiques, i. 217 sqq. 4 Hartknoch, Alt- und Neues Preussen, p. 176. 88 THE HISTORY OF HUMAN MARRIAGE chap. sisters as wives.1 But Strabo's statement that the Irish were in the habit of marrying their mothers and sisters 2 must be received with suspicion.3 Strabo also says that in South Arabia a man had sexual intercourse with his mother;4 but, as Winckler remarks, we need not suppose that the latter was his mother in our sense of the term.5 Hardly less universal than the prohibition of marriages between parents and children is the exogamous rule which forbids marriages between brothers and sisters who are children of the same father and mother. We have already considered some reported exceptions to this rule, and others may be added. But most of these statements are either obviously erroneous or of a more or less doubtful character, if for no other reason, because it is uncertain whether they refer to full brothers and sisters or to such as have only one parent in common. In some Brazilian tribes which live dispersed in small isolated hordes or families it is said to happen quite fre- quently that a brother and a sister dwell together as man and wife ; but of the ancient Tupinambase (ancestors of the Tupis) we are told that they allowed nothing of the kind openly.6 Du Chaillu states that among the Obongo in West Africa sisters marry with brothers ; but this state- ment is only based on information derived from another people, the Ashango, who have a strong antipathy to them.7 Among the Wateita in East Africa, according to Thomson, marriage with a sister is not infrequent, although highly reprobated.8 Paulitschke says that though marriages be- tween near relations are unusual among the Galla, they 1 d'Arbois de Jubainville, Cours de Utt Mature celtique, vi. 296. Dottin, ' Marriage (Celtic),' in Hastings, op. cit. vii. 433. 2 Strabo, Geographica, iv. 5. 4. 3 Cf. d'Arbois de Jubainville, op. cit. vi. 296. ♦ Strabo, op. cit. xvi. 4. 25. 5 Winckler, ' Polyandrie bei Semiten,' in Verhandl. Berliner Gesellsch. Anthrop. 1898, p. 29 n.1 8 v. Martius, Beitrage zur Ethnographic Amerika's, i. 116. 7 Du Chaillu, Journey to Ashango-Land, p. 320. 8 Thomson, Through Masai Land, p. 51. XIX EXOGAMY 89 nevertheless in some cases permit Schwesterehe; 1 but what he means by this expression is not quite clear.2 Felkin asserts that among the Baganda " marriages between relations are not forbidden, and brothers sometimes marry their sisters " ; 3 but according to the Rev. J. Roscoe a man among them might neither marry into his own clan nor into his mother's clan, and the punishment for incest was death.4 The Marquesas Islanders have also been said to marry their sisters; but Dr. Tautain believes that this statement is due to a misunderstanding of their terms of relation- ship.5 Ignorance of the true meaning of terms may be a frequent cause of alleged marriages between the nearest relatives. Speaking of the next-of-kin marriage imputed to the ancient Persians, Darab Dastur Peshotan Sunjana observes that great care is required to avoid the confusion arising from the indiscriminate use of the words " sister," " daughter," and " mother " among some Oriental peoples ;6 and a similar caution is needed with regard to other peoples as well. Some sort of misunderstanding is evidently at the bottom of Meinicke's statement that the Caroline Islanders could marry their sisters.7 His authority is Don Luis de Torres, as quoted by Arago, who " understood that brothers and sisters may intermarry ; and from the replies which were made to his questions on this subject, was led to think that such marriages are preferred to others." But Arago adds, " He is, however, by no means certain that his repre- sentation is correct " ;8 and Kubary has no doubt that he was mistaken.9 1 Paulitschke, Ethnographic Nordost-Afrikas, die materielle Cultur der Dandkil, Galla und Somdl, p. 196. 2 Cf. Frazer, Totemism and Exogamy, ii. 542 note. 8 Felkin, ' Notes on the Waganda Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xiii. 756. 4 Roscoe, Baganda, p. 128 sq. 5 Tautain, ' Stride sur le manage chez les Polyn&iens des ties Marquises,' in L'Anthropologic, vi. 644. 8 Darab Dastur Peshotan Sunjana, op. cit. p. 45. 7 Meinicke, Die Inseln des Stillen Oceans, ii. 383. 8 Arago, Narrative of a Voyage round the World, ii. 17 sq. 9 Kubary, * Die Bewohner der Mortlock Inseln,' in Mittheil. Geograph, Gesellsch, Hamburg, 1878-79, p. 251. THE HISTORY OF HUMAN MARRIAGE chap. 90 What ignorance of the meaning of native terms may lead to is well shown by the case of the Veddas of Ceylon, who for a long time have been the standing example of a savage people practising brother-and-sister marriage. Bailey wrote that not long ago the wild Veddas regarded the marriage of a man with his younger sister as not only proper and natural, but, in fact, as the proper marriage, though marriage with an elder sister or aunt would have been as incestuous and revolting to them as to us.1 But a later and more authoritative writer on the Veddas, Mr. Nevill, both denies the accuracy of this statement and explains how the mistake arose. " Such incest," he says, " never was allowed, and never would be, while the Vaedda customs lingered. Incest is regarded as worse than murder. So positive is this feeling, that the Tamils have based a legend upon the instant murder of his sister by a Vaedda to whom she had made undue advances. The mistake arose from gross ignorance of Vaedda usages. The title of a cousin with whom marriage ought to be contracted, that is, mother's brother's daughter, or father's sister's daughter, is nagd or nangi. This, in Sinhalese, is applied to a younger sister. Hence if you ask a Vaedda, ' Do you marry your sisters ? ' the Sinhalese interpreter is apt to say, ' Do you marry your naga ? ' The reply is (I have often tested it), ' Yes-we always did formerly, but now it is not always observed.' You say then, 'What? marry your own- sister-naga ? ' and the reply is an angry and insulted denial, the very question appearing a gross insult." The same writer adds :-" In no case did a person marry one of the same family, even though the relationship was lost in remote antiquity. Such a marriage is incest. The penalty for incest was death."2 The latest students of the Veddas, Professor and Mrs. Seligman, say that their own conclusions on the subject thoroughly agree with Nevill's.3 In his ' Journey to Lhasa and Central Tibet ' Sarat 1 Bailey, ' Wild Tribes of the Veddahs of Ceylon,' in Trans. Ethn. Soc. N.S. ii. 294 sq. 8 Nevill, ' Vaeddas of Ceylon,' in Taprobanian, i. 178. 3 Seligman, Veddas, p. 66 n. 1. XIX EXOGAMY 91 Chandra Das writes :-" Among the Pobos and Khamba marriage is promiscuously contracted, the brother marrying his sister, the nephew his aunt. Among the common Tibetans, so long as the parties do not claim a common father, there is no objection to the marriage ; the uterine brother and sister may be united, and a man may marry his step-mother or aunts."1 But the editor of the book, Mr. Rockhill, observes that, so far as the Khamba are concerned, the author is not correct in his statement, although it is highly probable that it may have once been as he says.2 The Lycian Tachtadshys, who call themselves Muhammadans but are regarded by all true believers as infidels, are said by their neighbours not only to indulge in terrible orgies but also to marry their sisters, and our informant adds that he knows himself two certain cases of such marriages ; but he does not tell us whether the con- tracting parties were full brothers and sisters.3 Liebich states that the German Gypsies allow a brother to marry his sister, though unions of this sort are generally avoided by them ;4 and speaking of their English and Welsh kindred, Mr. Thompson remarks that " it is almost certain that there never have been any degrees of kinship within which marriage was prohibited, except the direct line." 5 This, however, is certainly not true of the Gypsies of Finland, who greatly abhor incest.6 From ancient Egypt there is plenty of evidence that later Pharaohs married their sisters or half-sisters, and the Ptolemies followed the precedent of the Pharaohs.7 In 1 Sarat Chandra Das, Journey to Lhasa and Central Tibet, p. 326 sq. 2 Ibid. p. 326 sq. n. 2. 3 Petersen and Luschan, Reisen in Lykien, Milyas ztnd Kibyratis, p. 19g. 4 Liebich, Die Zigeuner, p. 4g. 6 Thompson, ' Ceremonial Customs of the British Gipsies,' in Folk Lore, xxiv. 331. 8 Thesleff, ' Zigenarlif i Finland,' in Nya Pressen, iSgy, no. 331 B. 7 Griffith, ' Marriage (Egyptian),' in Hastings, op. cit. viii. 444. Erman, Life in Ancient Egypt, p. 153. Pausanias, Descriptio Graeciae, i. 7. 1. King Amdsis I. (18th dynasty) speaks of a certain princess as " the mother of my mother " and " the mother of my father " (Ayrton, Currelly, and Weigall, Abydos, iii. 44 ; I am indebted to Dr. Alan Gardiner for this reference). THE HISTORY OF HUMAN MARRIAGE chap. 92 the Roman age marriage of half-sisters and full sisters occurred frequently in the families of cultivators of the soil and artisans J but that such marriages were equally or even more common in earlier periods2 is a mere suggestion, for which no evidence has been produced. A case of a less exalted person than a Pharaoh marrying his sister in the twenty-second dynasty is noted by Breasted,3 but one or a few isolated cases make no rule. It has been argued that Osiris' and Set's marriages with their sisters Isis and Nephthys reflected an ancient social custom.4 But here we are on dangerous ground. Myths and legends of marriages between divine or human brothers and sisters are not uncommon among peoples who have never been known to allow or practise marriage between full brothers and sisters. The Greeks had their stories of the marriages of Cronus with Rhea, of Zeus with Hera,5 and of Aeolus' six sons with their sisters.6 Stories of the same sort are found among the Niase, Javanese, Santals, and other peoples.7 In a song of the ' Rig-Veda ' Yami appears in support of the marriage of brother and sister, while the opposition is personified in Yama.8 Buddhist legends mention various cases of such unions.8 In the Scandinavian ' Ynglinga Saga ' it is said that " while Niord was with the Vans he had taken his own sister in marriage, for that he was allowed by their law " ; but " among the Asas it was forbidden for such near relatives to come together."10 1 Erman, op. cit. p. 153. Nietzold, Die Ehe in Agypten zur ptolemaisch-romischen Zeit, p. 13 sq. Griffith, loc. cit. p. 444. 2 Nietzold, op. cit. p. 14. Frazer, Adonis Attis Osiris, ii. 215. 3 Breasted, Ancient Records of Egypt, iv. 388. Griffith, loc. cit.p. 444. 4 Nietzold, op. cit. p. 12 sq. Frazer, Adonis Attis Osiris, ii. 216. 6 Ilias, xvi. 432 ; xviii. 356. 6 Odyssea, x. 5 sqq. 7 Frazer, Pausanias's Description of Greece, ii. 85. Percy Smith, ' Niue Island,' in Jour. Polynesian Soc. xi. 170. For a prominent incest-motive in Eskimo mythology see Gilbertson, Some Ethical Phases of Eskimo Culture, p. 63. 8 Rig-Veda, x. 10. See also v. Schroeder, Mysterium und Mimus im Rigveda, p. 275 sqq. 0 Schrader, Prehistoric Antiquities of the Aryan Peoples, p. 392 note. 10 Ynglinga Saga, 4 ; in Heimskringla (edited by Finnur Jdnsson), i. 13. XIX EXOGAMY 93 Brother-and-sister marriages are mentioned in the epic poems of the Russian peasants.1 But whatever be the origin of these and similar stories, they can certainly not be taken as evidence of customs existing in the past.2 The best authenticated cases of customary brother-and- sister marriage are, generally, found in the families of kings or ruling chiefs, although in many of these instances also it is uncertain whether the parties are full brothers and sisters.3 The most definite statements relating to such marriages come from Hawai. " A suitable partner for a chief of the highest rank," says Malo, " was his own sister, begotten by the same father and mother as himself. Such a pairing was called a pi' 0 (a bow, a loop, a thing bent on itself) ; and if the union bore fruit, the child would be a chief of the highest rank, a ninau pi'o, so sacred that all who came into his presence must prostrate themselves. He was called divine, akua. . . . Another suitable partner for a great chief was his half-sister, born, it might be of the same mother, but of a different father, or of the same father but of a different mother. Such a union was called a naha. The child would be a great chief, niau-pio ; but it would have only the kapu-a-noho (sitting tabu). ... If such unions as these could not be obtained for a great chief, he would then be paired with the daughter of an elder or younger brother, or of a sister. Such a union was called a hoi (return)." But " afterwards, when the couple had begotten children of their own, if the man wished to take another woman-or the woman another man-even though this second partner were not of such choice blood as the first, it was permitted them to do so."4 Consanguineous marriages, however, were the special privilege of the chiefs ; 1 Kovalewsky, Modern Customs and Ancient Laws of Russia, p. 14. 2 Cf. Macdonell and Keith, Vedic Index of Names and Subjects, i. 475 :-" The dialogue of Yama and Yami in the Rigveda seems clearly to point to a prohibition of the marriage of brother and sister. It can hardly be said, as Weber thinks, to point to a practice which was once in use and later became antiquated." 3 See Frazer, Pausanias's Description of Greece, ii. 85 ; Idem, The Dying God, p. 193 n. 3. 4 Malo, Hawaiian Antiquities, p. 80 sq. THE HISTORY OF HUMAN MARRIAGE chap. 94 among the ordinary people they were not allowed, and incest was considered disgraceful.1 We are even told that in Hawai " the fact that the first Christian teachers allowed marriage between cousins was a distinct stumbling- block when the new religion was introduced." 2 It is said that the Sinhalese kings " frequently married their sisters." 3 In Iboina in Madagascar the kings were occasionally united with their sisters, but such marriages were preceded by a ceremony in which the woman was sprinkled with consecrated water and prayers were recited asking for her happiness and fecundity, as if there was a fear that the union might call down divine anger upon the parties.4 Among the Bahima of Ankole the exogamous rule which prohibits a man from marrying into his father's clan does not apply to royalty ; "for in the royal family endogamy obtains, in so far as princesses may only marry their brothers, or members of the royal family."5 So also among the Banyoro, the only exception to the rule of clan exogamy was in the case of the royal family, where brothers frequently married their sisters. Yet we are told that when princes did so, " there was no binding marriage contract, and a princess was free to leave her brother to go to some other prince, if she elected to do so. Such marriages being more of the nature of love-matches, the couple came together for a time and their union was rather of a secret than of a public nature " ; and if children were born of these unions they were usually killed at birth.6 Among the Guanches of the Canary Islands some kings were married to their sisters, but nobody else was allowed to contract such a marriage.7 Of the brother-and-sister marriages of the Egyptian kings we have spoken above. And Cambyses and other Persian kings contracted similar marriages.8 1 Jarves, History of the Hawaiian Islands, p. 43. Ellis, Hawaii, p. 414 sq. Rivers, History of Melanesian Society, i. 382. 2 Rivers, op. cit. i. 382. 3 Tennent, Ceylon, ii. 459. 4 Sibree, The Great African Island, p. 252. 5 Roscoe, Northern Bantu, p, 119. 0 Ibid. pp. 27, 36 sq. 7 Bory de St. Vincent, Essais sur les Isles Fortunees, p. 99. 8 Herodotus, iii. 31. Spiegel, Erdnische Alterthumskunde, iii. 678 sq. XIX EXOGAMY 95 Garcilasso de la Vega states that the Incas of Peru from the first established it as a very stringent law that the heir to the kingdom should marry his eldest sister, legitimate both on the side of the father and that of the mother.1 But according to Acosta and Ondegardo it had always been held unlawful by the Peruvians to contract marriage in the first degree until Tupac Inca Yupanqui, at the close of the fifteenth century, married his sister on the father's side, and decreed " that the Incas might marry with their sisters by the father's side, and no other." 2 Nobody but the sovereign, however, was authorised to dispense with the law of nature so far as to marry his own sister.3 Marriage with a half-sister is not infrequent, and it seems that in nearly all such cases the brother and sister have the same father. In Dahomey unions of this kind were allowed in the royal family.4 Among the Baganda the queen was elected from the new king's half-sisters by the same father, and was the daughter of a woman who had no sons ; but she was absolutely forbidden to have children.5 In Burma the sovereign always married at least one half-sister, though nobody else was allowed to do so ;6 and in Siam there were, in 1900, two queens, the First and the Second Queen, both of whom were half-sisters of the reigning sovereign.7 But marriage with a half-sister is also found outside royal families. Among some of the pagan Fulani of Northern Nigeria such marriage is allowed, with the condition that the common parent is the father ; 8 and the same is the case at Gwaton in the Edo country and at Ososo.9 Among the Yoruba of the Slave Coast " half-brother cannot marry 1 Garcilasso de la Vega, First Part of the Royal Commentaries of the Yncas, i. 308. 2 Acosta, Natural and Moral H istory of the Indies, ii. 425. Prescott, History of the Conquest of Peru, p. 9 n. 3. 3 Fernandez, quoted by Prescott, op. cit. p. 54. 4 Le Herisse, L'ancient royaume du Dahomey, p. 214 sq. 5 Roscoe, Baganda, p. 84. 6 Shway Yoe, The Burman, p. 60. Forbes, British Burma, p. 48. 7 Young, Kingdom of the Yellow Robe, p. 99 sq. 8 Palmer, quoted by Frazer, Totemism and Exogamy, ii. 602. 9 Thomas, Anthropological Report on the Edo-speaking Peoples of Nigeria, i. 62. 96 THE HISTORY OF HUMAN MARRIAGE chap. half-sister ; but on the Gold Coast such marriages are permitted, provided the pair have not the same mother." 1 Among the Bayaka of Congo " marriage between children of the same mother is prohibited ; between children of the same father by different mothers it occurs, but is considered unseemly." 2 Among the Maori, according to Mr. Tregear, incest was almost unknown and reckoned sinful; instances of a brother and sister living together as man and wife, which were extremely rare, only occurred when the man's wife failed to give him issue and his sister would come as an extra wife, but " it was generally not a sister by both sides, not a daughter of the same father and mother, but a half-sister, or first cousin."3 Shortland never knew of more than one instance of a brother and sister living thus together among the Maori, and they were children of different mothers ;4 and with special reference to the Tuhoe tribe Mr. Best says that half-sisters and brothers are not allowed to marry.5 Among the natives of Mowat of British New Guinea6 and among the Aleut7 the children of one father by different mothers are permitted to intermarry, whereas those by one mother and different fathers are forbidden to do so. The same was in ancient times the case in Japan 8 and among Semitic peoples. Abraham married his half-sister Sarah, the daughter of his father ;9 Tamar might have been legally married to her half-brother Amnon ; 10 and such unions were still known in Judah at the time of Ezekiel, 1 Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 188. 2 Torday and Joyce, ' Notes on the Ethnography of the Ba-Yaka, in Jour. Anthr. Inst, xxxvi. 45. 3 Tregear, The Maori Race, p. 298. Cf. infra, ii. 129 sq. 4 Shortland, Traditions and Superstitions of the New Zealanders, p. 119. 5 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New Zealand Inst. xxxvi. 21. 0 Haddon, ' Notes on Mr. Beardmore's Paper,' in Jour. Anthr. Inst. xix. 467. 7 Sarytschew, ' Account of a Voyage of Discovery to the North- East of Siberia,' in Collection of Modern and Contemporary Voyages and Travels, vi. 77. 8 Griffis, Religions of Japan, p. 93. 9 Genesis, xx. 12. 10 2 Samuel, xiii. 13. XIX EXOGAMY 97 although he condemned them as abominations.1 Among the Phoenicians king Tabnith married his father's daughter Am'ashtoreth, and at Tyre a man might marry his father's daughter down to the time of Achilles Tatius. The same thing occurred at Mecca ; and a trace of this kind of marriage has survived to modern times at Mirbat.2 The Muham- madan Southern Slavs, also, allow marriages between half- brothers and half-sisters who have different mothers, although the seduction of a full sister is regarded in their songs as a crime punishable with death, or rather as some- thing which cannot occur.3 The ancient Athenians per- mitted similar marriages,4 but it seems that public opinion looked upon them with disfavour.5 In Guatemala and Yucatan on the other hand, contrary to the general rule, relationship on the mother's side was no bar to marriage : a man could marry his sister provided she was by another father.6 It has been asserted that where clan exogamy prevails a man is allowed to marry his sister either on the father's or on the mother's side, according as descent is traced through the mother or through the father.7 But this assertion is based on the perfectly erroneous supposition that exogamous rules depend exclusively upon the mode of tracing descent. We have previously noticed that among certain peoples a man has a special right, or is even expected or obliged, to marry his sister's daughter,, or his brother's daughter, 1 Ezekiel, xxii. ii. 2 Robertson Smith, Kinship and Marriage in Early Arabia, p. 163. 3 Krauss, Sitte und Brauch der Sudslaven, p. 221 sq. 4 Philo Judaeus, ' De specialibus legibus,' in Opera, ii. 303. Plu- tarch, Themistocles, xxxii. 3. Cornelius Nepos, Cimon, i. 2. Beau- chet, Histoire du droit prive de la Republique Athenienne, i. 165 sqq. Hruza (Beitrage zur Geschichte des griechischen und rdmischen Fami- lienrechtes, ii. 165 sqq.) even denies the accuracy of Philo's statement that marriage with a uterine sister was forbidden by law at Athens. 5 Beauchet, op. cit. i. 174. Hruza, op. cit. ii. 170 sq. 6 Bancroft, Native Races of the Pacific States of North America, ii. 664 sq. J McLennan, Studies in Ancient History, p. 160. Wilken, Huwelij- ken tusschen bloedverwanten, p. 31. Robertson Smith, op. cit. p. 162. VOL II H 98 THE HISTORY OF HUMAN MARRIAGE chap. or his father's sister, or his mother's sister ; 1 and among various other peoples such marriages are at any rate allowed. The Ainu, says Mr. Batchelor, " marry their cousins very often, and in some cases their nieces even."2 Marriage between a man and his brother's daughter occurs in Northern Australia and in some parts of Melanesia.3 Among the Macusis of British Guiana, again, although a. paternal uncle may never marry his niece, everybody is permitted to marry his sister's daughter.4 Castren tells us that his Tartar host, although prohibited by custom from marrying any woman belonging to his own clan, had married his sister's daughter, " following not his Greek Orthodox but his Tartar conscience."5 The Ossetes of the Caucasus, who likewise are divided into patrilineal clans, consider a marriage with a mother's sister quite a proper thing, though a marriage with a father's sister would be regarded as highly incestuous.6 Yet we must not assume that wherever clan exogamy prevails a man is allowed by custom to marry a niece or an aunt belonging to another clan than his own. The Rev. B. Danks observes that in New Britain, where upon theoretical grounds a man may without law-breaking marry his niece, as belonging to another clan, there is nevertheless a great repugnance to such unions among the natives ; and in one case where such a union was brought about it was utterly condemned by them.7 Of the Guanches of Tenerife we are told that " in the intercourse of the sexes they respected their mothers and sisters, but all other relations were eligible, aunts, cousins, nieces, or sisters-in-law."8 Dr. Gardiner has pointed out 1 Supra, ii. 68 sqq. 2 Batchelor, Ainu and their Folk-Lore, p. 228. 3 Rivers, ' Marriage (Introductory and Primitive),' in Hastings, op. cit. viii. 426. 4 Schomburgk, Reisen in Britisch-Guiana, ii. 318. Appun, ' Die Indianer von Britisch-Guayana, in Das Ausland, xliv. 447. 5 Castrdn, Nordiska resor och forskningar, ii. 298. v. Haxthausen, Transcaucasia, p. 406. 1 Danks, ' Marriage Customs of the New Britain Group,' in Jo w. Anthr. Inst, xviii. 283/ 8 de Espinosa, Guanches of Tenerife, p. 35. XIX EXOGAMY 99 two marriages of paternal uncle and niece in one family among the Egyptians of the twenty-first dynasty.1 In Rome the Emperor Claudius, wishing to marry his niece Agrippina, obtained from the Senate a decree that marriage with a brother's daughter should be legal, although marriage with a sister's daughter and with an aunt remained illegal as before ; but in the fourth century a man was again prohibited from marrying his brother's daughter.2 By Jewish law an uncle may marry his niece, though an aunt may not marry her nephew ; but no such marriage would be performed by Jewish rites in England, because English law forbids marriage between uncle and niece.3 Unions of this kind, as also marriages between aunts and nephews, are on the other hand permitted by law in Germany,4 the State of New York,5 Peru,6 and Uruguay ;7 whilst in various other countries, such as France, Italy, Belgium, Holland,8 Sweden,9 and Mexico,10 the legal prohibition of them may be dispensed with. Marriages between uncles and nieces and between aunts and nephews are prohibited in all those Christian countries in which cousins are forbidden to intermarry ; but the reverse is by no means the case. On the contrary, the large majority of modern law-books which prohibit marriages of 1 Gardiner, ' Goddess Nekhbet at the Jubilee Festival of Rameses III.,' in Zeitschr. f. agyptische Sprache und Altertumskunde, xlviii. 50. 2 Marquardt, Das Privatleben der Romer, i. 31. 3 Keil, Manual of Biblical Archeology, ii. 156. Abrahams, ' Marriage (Jewish),' in Hastings, op. cit. viii. 461. According to Austrian law, Jews are allowed to marry their nieces, although other persons are not [Das allgemeine burgerliche Gesetzbuch fur das Kaisertum Oesterreich, §§ 65, 125). 4 Biirgerliches Gesetzbuch, § 1310. 6 Wright, Report on Marriage and Divorce in the United States, P- 34- 8 Winroth, Aktenskapshindren, p. 205 n. 3. 7 Roguin, Traits de droit civil compart. Le mariage, p. 84. 8 Code civil, art. 164. Codice civile, art. 68. Roguin, op. cit. p. 84. 9 Lag om dktenskaps ingdende och upplosning av den 12 november 1915, ii. 8. 10 Roguin, op. cit. p. 84. H 2 100 THE HISTORY OF HUMAN MARRIAGE chap. the two former kinds have no objection to marriages between cousins ; and this is true even of the laws in such countries as Switzerland, England, and most of the North American States, where the prohibition to marry a niece or a nephew is absolute.1 Generally speaking, there can be little doubt that the relationships of uncle and niece and of aunt and nephew are more frequent bars to intermarriage than the relationship of cousins. The Burmese, for instance, have no restrictions on marriage beyond the rule that a man may not marry his mother, daughter, sister, aunt, grandmother, or granddaughter ; and the marriage of cousins of all kinds is very common among them.2 Among some of the Dyaks marriages are apparently only prohibited between parents and children, brothers and sisters, uncles and nieces, and aunts and nephews.3 Among the Subanu of Mindanao kinship nearer than first cousins constitutes a bar to inter- marriage, but usually this degree of relationship is also prohibitive.4 Among vthe Copper Indians, according to Franklin, there was no prohibition of the intermarriage of cousins, but a man was forbidden to marry his niece.6 The marriage of cousins is allowed among many peoples ;6 but some of the statements to this effect also imply the absence of any prohibition of marriage between uncle and 1 Schweizerisches Zivilgesetzbuch vom io. Dezeniber 1907, art. 100. Roguin, op. cit. p. 84. Wright, op. cit. p. 31 sqq. 2 Gait, Census of India, 1911, vol. i. (India) Report, p. 252. 3 Perelaer, Ethnographische beschrijving der Dajaks, p. 59. 4 Finley and Churchill, Subanu, p. 40. 6 Franklin, Narrative of a Journey to the Shores of the Polar Sea, p. 289. Cf. v. Martius, op. cit. i. 116, 393 (certain Brazilian tribes). 8 E.g., among the Chan6 (Nordenskiold, Indianliv i El Gran Chaco [Syd-Amerikaf p. 196), the KarayA (Krause, In den Wildnissen Brasiliens, p. 324), the Caribs of the Antilles (de Poincy, op. cit. p. 544), some Eskimo (Waldmann, quoted by Gilbertson, op. cit. p. 62 [Labrador Eskimo] ; Lyon, Private Journal during the Voyage of Discovery under Captain Parry, p. 353 [Eskimo in the neighbourhood of Iglulik] ; Nelson,' Eskimo about Bering Strait,' in Ann. Rep. Bur. Ethn. xviii. 291), the Nicobarese (Lowis, Census of India, 1911, vol. ii. [Andaman and Nicobar Islands], p. 100), the Marquesas Islanders (Tautain, in L'Anthropologic, vi. 644), and various Bantu tribes (Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 219 n.*). XIX EXOGAMY 101 niece and between aunt and nephew.1 Among the ancient Teutons marriage seems to have been prohibited only between ascendants and descendants and between brothers and sisters.2 Cousins are permitted to marry both by Jewish and Muhammadan law. In Europe they are pro- hibited from doing so only in a few countries-under the influence of ecclesiastical law,-notably Russia,3 Austria (except among Jews),4 Hungary, and Spain ; and in the two latter countries the prohibition may be dispensed with.5 Among peoples who favour the marriage of cousins it is usually between certain definite cousins only, whereas marriage between other cousins may be prohibited.6 Where clan exogamy prevails a man cannot, of course, marry a cousin belonging to his own clan. This does not imply that he is necessarily allowed to marry a cousin belonging to another clan, but very frequently he may do so. Among peoples unaffected by modern civilisation the exogamic rules are probably in the large majority of cases more extensive than among ourselves, very often referring to all the members of the clan or phratry or " class " (in the sense used with reference to Australian and Melanesian exogamy) or local community.7 1 Hagen, Die Orang Kubu auf Sumatra, p. 130. Kaufmann, ' Die Auin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 156; Trenk, ' Die Buschleute der Namib,' ibid, xxiii. 168 ; Stow, op. cit. p. 95 (Bushmen). Hutereau, Notes sur la Vie familiale et juridique de quelquespopulations du Congo Beige, p. 4 (Batua). Torday and Joyce, ' Notes on the Ethnography of the Ba-Huana,' in Jour. Anthr. Inst. xxxvi. 285. Monrad, Bidrag til en Skildring af Guinea-Kysten, p. 59 (Negroes of Accra). 2 Grimm, Deutsche Rechtsalterthiimer, p. 435. Nordstrom, Bidrag till den svenska samhalls-forfattningens historia, ii. 18. Hoops, Reallexikon der Germanischen Altertumskunde, iv. 417. 3 Codex des Civilrechts (Russisches Civilgesetzbuch), § 23. . 4 Das allgemeine biirgerliche Gesetzbuch fiir das Kaisertum Oester- reich, §§ 65, 125. 5 Roguin, op. cit. p. 83. Cddigo civil de Espana, art 64 sq. 6 Supra, ii. 68 sqq. 7 In compiling the following list of facts I have in various ways derived much benefit from Sir James G. Frazer's very learned work Totemism and Exogamy. 102 THE HISTORY OF HUMAN MARRIAGE chap. Clan exogamy has been, or still is, very frequent among the North American Indians.1 Each of the six tribes which composed the Iroquois confederacy was subdivided into a number of totemic and exogamous clans with descent in the female line. Moreover, in four of the tribes--the Onondaga, the Cayuga, the Seneca, and the Tuscarora,- each of which contained eight clans, these clans were dis- tributed in two groups or phratries, and the rule was that no man might marry a woman of any of the clans in his own group, although he might marry a woman of any clan in the other group. " Whoever violated these laws of marriage," says Morgan, " incurred the deepest detestation and disgrace." But in process of time the rigour of the system was relaxed, until finally the prohibition was confined to the clan of the individual, which, among the residue of the Iroquois, is still religiously observed.2 The Hurons or Wyandots, who inhabited the country to the north of Lake Ontario and the Saint Lawrence River and were closely related by blood and language to the Iroquois, were likewise divided into totemic and exogamous clans with descent in the female line.3 According to one account, these clans were grouped in four phratries or exogamous classes; 4 whereas, according to another account, there were never more than two phratries in the Wyandot tribe, one of the clans-the Wolf clan-belonging to neither, but bearing the relation of cousin to both of them and acting as mediator or umpire both between the phratries and between the clans.5 In former times marriage was prohibited within the phratry as well as within the totem clan ; but at a later time the rule prohibiting marriage within the phratry was abolished, and the prohibition was restricted to the clan.6 1 For the names of North American tribes I have adopted the spelling used in Hodge's Handbook of American Indians north of Mexico. 2 Morgan, League of the Iroquois, p. 79 sqq. Idem, Ancient Society, p. 70. Hale, Iroquois Book of Rites, p. 52 sq. 3 Morgan, Ancient Society, p. 153. Powell, 'Wyandot Govern- ment,' in Ann. Rep. Bur. Ethn. i. 59, 60, 63. Connelly, quoted by Frazer, op. cit. iii. 30 sqq. 4 Powell, loc. cit. p. 60. 5 Connelly, quoted by Frazer, op. cit. iii. 33 sq. 6 Ibid. iii. 34. XIX EXOGAMY 103 Passing to the widely-spread Algonkin family, we notice the division into exogamous clans or phratries among the Delaware Indians orLenapeJthe Mohegan,2 and the Abnaki;3 among the two former tribes the descent of the clan was in the female line, whereas among the Abnaki it is, now at least, in the male line. The Ojibway or Chippewa, another Algonkin tribe, who ranged over a vast territory about the Great Lakes and westward, were divided into at least forty totemic and exogamous clans4 with descent in the male line.5 Warren, who wrote his history of this tribe in the middle of the last century, states that marriage between two persons of the same totem " is one of the greatest sins that can be committed in the Ojibway code of moral laws, and tradition says that in former times it was punishable with death. In the present somewhat degenerate times, when persons of the same Totem intermarry (which even now very seldom occurs), they become objects of reproach. It is an offence equivalent among the whites to the sin of a man marrying his own sister." 6 Inter- marriage should not occur even between persons of the same totem who belong to different tribes.7 The Pota- watomi, whose country also was in the region of the Great Lakes,8 and various Algonkin tribes of the Mississippi-the Miami,9 the Shawnee,10 the Sauk and 1 Loskiel, History of the Mission of the United Brethren among the Indians in North America, i. 56. Morgan, Ancient Society, p. 171 sq. Hodge, Handbook of American Indians north of Mexico, i. 386. Harrington, ' Preliminary Sketch of Len4pe Culture,' in American Anthropologist, N.S. xv. 211. 2 Morgan, Ancient Society, p. 174. 3 Ibid. p. 174 sq. 4 This estimate has been made by Frazer (op. cit. iii. 52 s^.) from the accounts given by Warren (' History of the Ojibways,' in Collections of the Minnesota Historical Society, v. 44 sq.), Morgan (Ancient Society, p. 165 sq.}, Edwin James (Narrative of the Captivity and Adventures of John Tanner, p. 314 sq.}, and Peter Jones (History of the Ojebway Indians, p. 138). 5 Warren, loc. cit. p. 42. Morgan, Ancient Society, p. 166. • Warren, loc. cit. p. 42. Cf. Jones, op. cit. p. 138 ; Morgan, Ancient Society, p. 167. 7 Warren, loc. cit. p. 35. 8 Morgan, Ancient Society, p. 167. 9 Ibid. p. 168. 10 Ibid. p. 169. THE HISTORY OF HUMAN MARRIAGE chap. 104 Foxes,1 and the Kickapoo 2-have likewise been organised in exogamous totem clans with descent in the male line. The three tribes of the Blackfeet, the western- most of the Algonkin, who range over the eastern slopes of the Rocky Mountains and the prairies at their foot, are subdivided into a number of clans tracing their descent in the male line, and in former days a man was strictly prohibited from marrying within his clan. But although exogamous, the clans were apparently not totemic.3 Various tribes belonging to the Siouan or Dakotan family are, or were till lately, divided into totemic and exogamous clans, although this is not the case with the largest tribe of the family, the Sioux or Dakota, whose numerous bands, at least in modern times, have been neither totemic nor exo- gamous.4 The Omaha, a Siouan tribe of the Missouri, have been found organised in totemic clans, according to Dorsey ten in number, which are equally distributed between two great tribal divisions, and each clan is subdivided into sub-clans.5 Children belong to the clan of their father ;8 and, according to Morgan, no man is allowed to marry a woman of his own clan.7 But from Dorsey we learn that a man is forbidden to marry any woman either of his father's or of his mother's clan, and that the same prohibition extends to the sub-clan of his father's mother, to the sub-clan of his father's mother's mother, to the sub-clan of his mother's mother, and to the sub clan of his mother's mother's mother. Nor may he marry a woman of the sub-clan to which the wife of his son, his nephew, or his grandson belongs ; and 1 Ibid. p. 170. Forsyth says (' Account of the Manners and Customs of the Sauk and Fox Nations of Indians Tradition,' in Emma Helen Blair's Indian Tribes of the Upper Mississippi Valley and Region of the Great Lakes, ii. 215) that " if the least relationship exists between an Indian and a girl it will prevent them from being married." 2 Hodge, op. cit. i. 685. 8 Frazer, op. cit. iii. 84. 4 Ibid. iii. 86 sq. 8 Dorsey, 'Omaha Sociology,' in Ann. Rep. Bur. Ethn. iii. 215. Idem, ' Siouan Sociology,' ibid. xv. 226. See also James, Account of an Expedition from Pittsburgh to the Rocky Mountains, i. 234 ; Morgan, Ancient Society, p. 155 ; Frazer, op. cit. iii. 91 sqq. • Dorsey, in Ann. Rep. Bur. Ethn. xv. 225. 7 Morgan, Ancient Society, p. 155. XIX EXOGAMY 105 he is also prohibited from marrying a woman of the sub -clan to which the husband of his daughter, his niece, or his granddaughter belongs.1 With regard to the importance of the sub-clans or subgentes, as he calls them, Dorsey observes that " were it not for the institution of subgentes a man would be compelled to marry outside of his tribe, as all the women would be his kindred, owing to previous intermarriages between the ten gentes. But in any gens those on the other side of the gentile unethe, or fireplace, are not reckoned as full kindred, though they cannot inter- marry."2 On the other hand, a man is allowed to marry a woman of his own totem, provided that she is a member of another tribe.3 Various other Dakotan tribes of the Missouri are divided into exogamous totem clans: the Ponca,4 the Iowa,5 the Kansa,6 and the Winnebago,7 whose clans are hereditary in the male line ; and the Oto and the Missouri,8 who have maternal descent. An exogamous clan organisation with descent in the female line has also been found among the Mandan,9 the Hidatsa,10 and the Crows,11 although their clans do not seem to have been totemic.12 Of the Osage, who by language and blood are closely allied to the Kansa, La Flesche says that the wife must be chosen from a gens other than the gentes to which the father and the mother of the youth belong.13 The so-called Gulf nations, inhabiting the south-eastern portion of what is now the United States-the Creeks,14 the Seminole,15 the Choctaw,16 the Chickasaw,17 and the Cherokee18-were divided into exogamous clans with descent in the female line. The'eight clans of the Choctaw, which 1 Dorsey, in Ann. Rep. Bur. Ethn. iii. 256 sq. 2 Ibid. p. 258. 3 Ibid. p. 257. 4 Morgan, Ancient Society, p. 155. 6 Ibid. p. 156. 9 Ibid. p. 156 sq. 7 Ibid. p. 157. 8 Ibid. p. 156. 9 Ibid. p. 158- 1(1 Ibid. p. 159. 11 Ibid. p. 159. 12 Cf. Frazer, op. cit. iii. 136 sqq. 13 La Flesche, ' Osage Marriage Customs,' in American Anthropo- logist, N.S. xiv. 127. 14 Morgan, Ancient Society, p. 161. Gatschet, Migration Legend of the Creek Indians, i. 154 sqq. Speck, ' Creek Indians of Taskigi Town,' in Memoirs American Anthrop. Association, ii. 114 sq. 15 MacCauley, 'Seminole Indians of Florida,' in Ann. Rep. Bur Ethn. v. 507 sq. 19 Morgan, Ancient Society, p. 162. 17 Ibid. p. 163. 18 Ibid. p. 164. 106 THE HISTORY OF HUMAN MARRIAGE chap. were totemic, were equally distributed between two phratries, and no man might marry a woman of any clan in his own phratry.1 Of the Yuchi Indians, who after fruitless efforts to resist the pressure of the Creek confederacy finally joined the league, Mr. Speck says that marriage between members of the same totem clan is strictly tabooed as a form of incest, whereas a man is free to marry a woman of any clan but his own.2 Of the Natchez of Louisiana Le Page du Pratz wrote that they did not marry within the third degree.3 The Pueblo Indians of Arizona and New Mexico are divided into a large number of totemic and exogamous clans with matrilineal descent, and the clans are further grouped in phratries. Formerly marriage within the clan, and some- times within the phratry, was rigorously forbidden ; but nowadays the old rule of exogamy has to some extent broken down, and in many villages a man may marry any woman without regard to her clan and phratry.4 Among the Sia, a Pueblo tribe of the Keresan stock, a man was forbidden to marry a woman either of his father's or of his mother's clan. But " at present there is nothing for the Sia to do but to break these law's, if they would preserve the remnant of their people, and while such marriages are looked upon with disfavor, it is ' the inevitable.' " 5 The Apache and the Navaho, the most southerly representatives of the Atha- pascan family, are also organised in a large number of exogamous clans with descent in the female line, and the clans are grouped together in phratries. At the present day no Navaho may marry a woman of his own clan or phratry, neither may he marry a woman of his father's clan or phratry.6 The tribes of California, so far as is 1 Ibid. p. 162. 2 Speck, Ethnology of the Yuchi Indians, p. 95. 3 Le Page du Pratz, History of Louisiana, p. 344. 4 Frazer, op. cit. iii. 208. Morgan, Ancient Society, p. 180. Mrs. Stevenson, ' Sia,' in Ann. Rep. Bur. Ethn. xi. 19. 5 Mrs. Stevenson, loc. cit. p. 19. 6 Bourke, ' Notes upon the Gentile Organization of the Apaches of Arizona,' in four. American Folk-Lore, iii. in sqq. Matthews, ' Gentile System of the Navajo Indians/ ibid. iii. 103 sqq. Idem, Navaho Legends, p. 29 sqq. Bourke, Snake-Dance of the Moquis of Arizona, p. 279. Ostermann, ' Navajo Indians of New Mexico and Arizona/ in Anthropos, iii. 861. XIX EXOGAMY 107 known, had no system of exogamous clans.1 But we are told that among the Luiseno Indians, belonging to the Shoshonian linguistic family and living on the coast of California, marriages with even distant relatives were looked upon with extreme disfavour,2 and that the Gualala, who live on the creek called by their name in the north- west corner of Sonoma County, account it " poison " for a person to marry a cousin or an avuncular relation.3 Among the Takelma Indians of South-Western Oregon cousins, but no more distant relations, are forbidden to intermarry.4 Many but not all of the Indian tribes of North-West America are organised in totemic and exogamous clans.5 The Tlingit are divided into two exogamous classes or phratries with matrilineal descent, which bear the name of the Raven and the Wolf (or Eagle) respectively, and each of these classes is subdivided into a number of exogamous clans which are named after various animals.8 Of the Haida, who inhabit the Queen Charlotte Islands, Mr. Swanton writes :-" The whole people is divided into two strictly exogamic clans-the Raven and the Eagle-with descent in the female line. ... So close was relationship held to be between persons of the same clan, that marriage within it was viewed by them almost as incest is by us."7 The Tsimshian, a small stock of Indians inhabiting the coast of 1 Kroeber, ' Preliminary Sketch of the Mohave Indians,' in American Anthropologist, N.S. iv. 278. Frazer, op. cit. iii. 247. 2 Sparkman, ' Culture of the Luiseno Indians,' in University of California Publications in American Archceology and Ethnology, viii. 214. 3 Powers, Tribes of California, p. 192. * Sapir, ' Notes on the Takelma Indians of Southwestern Oregon,' in American Anthropologist, N.S. ix. 267. 5 See Frazer, op. cit. iii. 263 sqq. • Holmberg, ' Uber die Volker des russischen Amerika,' in Acta Societatis Scientiarum Fennie a, iv. 292, 293, 313. Petroff, Report on the Population, Tndzistries, and Resources of Alaska, p. 165 sq. Boas, ' First General Report on the Indians of British Columbia,' in Fifth Report of the Committee on the North-Western Tribes of Canada, pp. 23, 25. Swanton, ' Social Condition, Beliefs, and Linguistic Relationship of the Tlingit Indians,' in Ann. Rep. Bur. Ethn. xxvi. 398, 423 sq. 7 Swanton, Haida, p. 62. See also Boas, loc. cit. pp. 23, 26. 108 THE HISTORY OF HUMAN MARRIAGE chap. the mainland of British Columbia from the Nass River on the north to Millbank Sound on the south, have four exoga- mous clans with matrilineal descent, and these clans are further divided into smaller groups.1 The type of social organisation which prevails among the Northern Kwakiutl, who speak the Haisla and Heiltsuk dialects, closely resembles that of their neighbours to the north-the Tsimshian, the Haida, and the Tlingit-although among them the parents in certain cases are free to assign their children to the father's clan instead of to the mother's.2 And not only are all these peoples subdivided into totemic and exogamous clans, but those exogamous divisions which possess the same crest or totem are regarded as equivalent to each other though found in different tribes. Dr. Boas observes - " One of the main facts is, that the phratries, viz. gentes, of the Tlingit, Haida, Tsimshian, and Heiltsuk, are exoga- mous, not only among each tribe, but throughout the whole region. A member of the eagle gens of the Heiltsuk, for instance, cannot marry a member of the eagle phratry of the Tlingit. Those gentes are considered identical which have the same crest." 3 Among the Salish tribes in the interior of British Columbia there is no division into exogamous clans.4 The Bella Coola Indians, the most northern tribe of the Salish stock, prefer marriages of people who are distantly related, or not related at all, to marriages among near relatives; but " it even happens that cousins marry, or that an uncle marries his niece, in order to keep the clan tradition from being acquired by another village community. It seems, however, that, 1 Boas, loc. cit. pp. 23, 27. Sapir, Sketch of the Social Organization of the Nass River Indians, p. 3 sq. 2 Boas, loc. cit. pp. 23, 29. Idem, ' Mythology of the Bella Coola Indians,' in -Publications of the Jesup North Pacific Expedition, i. 121. Frazer, op. cit. iii. 318 sq. 3 Boas, in Fifth Report of the Committee on the North-Western Tribes of Canada, p. 32. Swanton, Haida, pp. 65, 66, 112 sq. 4 Teit, ' Indian Tribes of the Interior,' in Canada and its Provinces, vol. xxi. The Pacific Province, i. 306. Idem, ' Lillooet Indians,' in Publications of the Jesup North Pacific Expedition, ii. 269. Frazer, op. cit. iii. 338. XIX EXOGAMY 109 owing to the influence of the Coast tribes, the endogamic system has begun to give way to an exogamic system." 1 Among the Stlatlumh " the old people, expressed astonish- ment that first cousins, who with them are regarded as ' brothers ' and ' sisters,' should be permitted by us to intermarry " ; but nowadays they generally follow our customs and permit such marriages.2 Mr. Teit states that among the Thompson Indians " cousins were forbidden to marry, because they were of one blood, similar to sister and brother; and the union of distant blood relations was discountenanced. Even if second-cousins married, they were laughed at and talked about." 3 Among the Shuswap, another tribe of British Columbia, according to the same authority, " blood-relations did not marry, not even second- cousins " ; 4 whereas Dr. Boas says of the same tribe that " marriages between cousins were not forbidden." 5 Among the Tinne or Dene-the widespread Athapascan nation who inhabit for the most part the interior of Alaska and a great extent of the Canadian territory which stretches from there to the Arctic Ocean on the north, towards Hudson's Bay on the east, and to the Lillooet mountains on the south-we again find totemic and exogamous clans, with descent in the female line, namely in the western tribes.6 But among the Eastern Tinne, as has been noticed above, even unions which we should regard as incestuous are known to have occurred.7 Among the Eskimo, who range over the immense region 1 Boas, in Publications of the Jesup North Pacific Expedition, i. 116. 2 Hill Tout, * Report on the Ethnology of the Stlatlumh of British Columbia,' in Jour. Anthr. Inst. xxxv. 133. 3 Teit, ' Thompson Indians of British Columbia,' in Publications of the Jesup North Pacific Expedition, i. 325, 4 Idem, ' Shuswap,' in Publications of the Jesup North Pacific Ex- pedition, ii. 591. 5 Boas, in ' Sixth Report of the Committee on the North-Western Tribes of Canada,' in Report of the Sixtieth Meeting of the British Association held at Leeds in 1890, p. 643. 6 Hardisty, ' Loucheux Indians,' in Smithsonian Report, 1866, p. 315. Dall, Alaska, p. 196 sqq. Frazer, op. cit. iii. 348 sqq. 7 Supra, ii. 83. THE HISTORY OF HUMAN MARRIAGE chap. 110 from Bering Strait on the west to Greenland on the east, the clan organisation seems to be wholly lacking. Nelson claims to have found among certain Alaskan Eskimo " a regular system of totem marks and the accompanying subdivision of the people into gentes " j1 but his evidence does not seem sufficient.2 The choice of a mate, however, is, as usual, barred by specified degrees of kinship. Among some Eskimo tribes marriages between cousins are said to be permitted,3 among others to be prohibited. Dr. Rink asserts that the Eskimo disapproves of such marriages.4 Among the Davis Strait tribes of the Central Eskimo, according to Dr. Boas, " marriages between relatives are forbidden : cousins, nephew and niece, aunt and uncle, are not allowed to intermarry."5 Hans Egede wrote that the Greenlanders refrained from marrying their nearest relatives, even in the third degree, considering such matches to be " unwarrantable and quite unnatural."6 Cranz says of the same people that they seldom marry first cousins.7 It is quite possible, of course, that in these and other cases when marriages between cousins are said to be prohibited, avoided, or disapproved of, the statements are true with reference to certain cousins only and not to all.8 Among the Pipiles of Salvador an ancestral tree with seven main branches denoting degrees of kindred was painted upon cloth, and within these seven branches or degrees no one was allowed to marry, except as a recompense for some great public or warlike service rendered ; but within four degrees of consanguinity none, under any pretext, might marry.9 In Yucatan there was a strong prejudice against a man wedding a woman who bore the same name as himself, and^o far was this fancy carried that he who broke the rule was looked upon as a renegade and an outcast; nor could a man marry 1 Nelson, loc. cit. p. 322. 2 Cf. Frazer, op. cit. iii. 368 sq. ; Gilbertson, op. cit. pp. 9, 62. 3 Supra, ii. 100 n. 6. 4 Rink, Eskimo Tribes, p. 23. 5 Boas, ' Central Eskimo,' in Ann. Rep. Bur. Ethn. vi. 579. • Egede, Description of Greenland, p. 141. 7 Cranz, History of Greenland^ i. 147. 8 Cf. Frazer, Folk-Lore in the Old Testament, ii. 142 sqq. 9 Bancroft, op. cit. ii. 665. XIX EXOGAMY 111 his mother's sister.1 Among the Azteks, too, marriages between blood-relations or those descended from a common ancestor were not allowed.2 / The three tribes Cabecar, Bri-bri, and Tiribi, inhabiting the Atlantic slope of Costa Rica, are, according to Mr. Gabb, divided into exogamous clans. " I could not ascertain exactly," he says, " how the question is settled as to which clan a person belongs, whether he inherits from father or mother, but so far as I could gather, I think from the father. Cousins, even to a remote degree, are called brother and sister, and are most strictly prohibited from intermarriage. The law, or custom, is not an introduced one, but one handed down from remote times. The penalty for its violation was originally very severe ; nothing less than the burial alive of both parties. This penalty was not only enforced against improper marriage, but even against illicit intercourse on the part of persons within the forbidden limits."3 According to a later account the Bri-bri are divided into two exogamous classes or phratries, each of which is subdivided into a number of apparently totemic clans.4 From Mr. Simons' account of the totemic clans of the Goajiro, a South American tribe inhabiting the Goajira peninsula in Colombia,5 we may be justified in inferring that these clans are exogamous with descent in the female line.6 A similar organisation prevails among the Arawaks of British Guiana.7 The Mundrucus seem to be divided into 1 Ibid. ii. 665. de Herrera, General History of the West Indies, iv. 171. 2 Bancroft, op. cit. ii. 251. 3 Gabb, ' Indian Tribes and Languages of Costa Rica,' in Proceed. American Philosophical Soc. Philadelphia, xiv., quoted by Frazer, Totemism and Exogamy, iii. 552. 4 Pittier de Fdbrega, Die Sprache der Bribri-Indianer in Costa Rica, p. 22 sq. (in Sitzungsberichte der philosophischen-historischen Classe der Kaiserl. Akademie d. Wissenschaften, vol. cxxxviii.). Cf. Frazer, Totemism and Exogamy, iii. 553. 6 Simons, ' Exploration of the Goajira Peninsula,' in Proceed. Roy. Geograph. Soc. N.S. vii. 786 sq. • See Frazer, Totemism and Exogamy, iii. 561. 7 Hilhouse, ' Notices of the Indians settled in the Interior of British Guiana,' in Jour. Roy. Geograph. Soc. ii. 228. Im Thurn, Among the Indians of Guiana, p. 175 sqq. THE HISTORY OF HUMAN MARRIAGE chap. 112 exogamous clans. " A Mundrucu Indian," says Agassiz, " treats a woman of the same order (clan) with himself as a sister ; any nearer relation between them is impossible."1 Of the Indians on the Uaupes River in North-West Brazil we are told that they do not often marry with relations, or even neighbours, preferring wives from a distance, or from other tribes.2 M. Coudreau writes of them :-" L'exogamie est la coutume generale. On prend sa femme dans la tribu voisine, mais sans rapt ni simulacre de rapt."3 The Arau- canians of Chili scrupulously avoided in their marriages " the more immediate degrees of relationship " ;4 and certain facts have been supposed to indicate that they once had a totemic system with exogamy.5 Dobrizhoffer states that the Guarani and Abipones abhorred marriages with even the remotest relations ;6 and according to Father Florian Baucke, who was living in Paraguay in the middle of the eighteenth century, any man among the Abipones or the Mocobis would have been horrified if the suggestion had been made to him that he should marry a woman related to him in the fifth or sixth degree.7 Marriage with a cousin is said to be prohibited by or repugnant to various tribes of the Gran Chaco.8 Concerning the Onas, who inhabit especially the eastern part of the island of Tierra del Fuego and are a branch of the Tehuelches of Southern Patagonia, we are told that they refrain from marrying even the remotest relatives and look upon such marriages with the greatest 1 Agassiz, Journey in Brazil, p. 320. 2 Wallace, Travels on the Amazon, p. 497. v. Martius, op. cit. k 594- 3 Coudreau, La France equinoxiale, ii. 176. 4 Molina, Geographical, Natural, and Civil History of Chili, ii. 115. 6 Latcham, 1 Ethnology of the Araucanos,' in Jour. Roy. Anthr. Inst, xxxix. 359 sq. Frazer, Totemism and Exogamy, iii. 581 sqq. G Dobrizhoffer, Account of the Abipones, i. 63 ; ii. 212. Charlevoix says {History of Paraguay, i. 202) that among the Guarani the men who had embraced the Christian religion never married any of their relations, " even within those degrees with which the Church readily dispenses." 7 Kohler, Pater Florian Baucke, p. 311. 8 Nordenskiold, op. cit. pp. 83, 196 (Choroti, Ashluslay, Matacos). Church, Aborigines of South America, p. 238 (Chiriguanos). XIX EXOGAMY 113 disgust.1 Of the Yahgans of Tierra del Fuego Mr. Bridges wrote to me that " no marriage, no intercourse, ever takes place among blood-relations even to second cousins." Such intercourse is held in utter abomination. Passing to Asiatic peoples, we notice that among the Koryak, according to Dr. Jochelson, a man is forbidden to marry his mother and her sisters, his father's sisters, his own sisters, his cousins, his daughters, his brother's daughters, and his sister's daughters ; but in their legendary tales marriages between cousins are quite frequently men- tioned and meet with no reproaches from anybody. In answer to Dr. Jochelson's questions concerning second cousins, some Koryak replied that they did not consider them relatives.2 The Yukaghir assured the same ethnolo- gist that, " according to their former customs, marriages between relatives were forbidden only to the generation of first-cousins inclusive " ; but nowadays such marriages do occur among them.3 Exogamous clans are found among many peoples in Siberia or in other parts of the former Russian empire, such as the Tungus,4 Yakut,5 Gold (a tribe settled in the middle course of the Amoor River),6 Samoyed,7 Ostyak,8 Tartars,9 and Ossetes.10 Well-to-do Yakut not only refrain from marrying within their own clan, but will not even marry a woman of their own nasleg, which 1 Gallardo, Tierra del Fuego-Los Onas, p. 215. See also Cojazzi, Los indios del Archipielago Fueguino, p. 16. 2 Jochelson, Koryak, pp. 736, 738. 3 Idem, Yukaghir, pp. 82, 84. 4 Ibid. p. 86 sq. Bogoras, Chukchee, p. 570. 5 Sumner, ' Yakuts, abridged from the Russian of Sieroshevski,' in Jour. Anthr. Inst. xxxi. 87. Miss Czaplicka, Aboriginal Siberia, p. in. Jochelson, Yukaghir, pp. 80, 86. -C-Laufer, ' Preliminary Notes-on Explorations among the Amoor Tribes,' in American Anthropologist, N.S. ii. 318. " Georgi, Beschreibung alter Nationen des russischen Reichs, p. 282. Pallas, Reise durch verschiedene Provimen des russischen Reichs, iii. 78. Castren, op. cit. i. 298 sq. ; ii. 168. Finsch, Reise nach West- Sibirien, p. 543. Donner, Bland samojeder i Sibirien, p. 116. 8 Castr6n, op. cit. i. 298 sq. 9 Ibid. ii. 298. 10 v. Haxthausen, op. cit. p. 406 note. VOL. II I 114 THE HISTORY OF HUMAN MARRIAGE chap. is another and usually larger division comprising from one to five aga-ussas, or clans.1 Among the Kalmucks the common people enter into no unions in which the parties are not distant from one another by three or four degrees ; and they have a proverb that " Princes and dogs know no relationship," alluding to the fact that the great folk some- times marry sisters-in-law.2 Among the Circassians mem- bers of the same fraternity " are not only themselves inter- dicted from intermarrying, but their serfs too must wed with the serfs of another fraternity ; and where, as is generally the case, many fraternities enter into one general bond, this law, in regard to marriage, must be observed by all." Formerly a marriage within the fraternity was punished by drowning, but now a fine of two hundred oxen, and restitution of the wife to her parents, are only exacted.3 In his description of a visit made to the Sard and Shera Yogurs-two small tribes which under the common name of Huang-fan (" the yellow barbarians ") inhabit the northern reaches of the Nanshan mountains in the district of Kanchow-Hsuchow in Northern China-Baron Manner- heim states that these tribes consist of sections which are called " bones," and that persons belonging to the same " bone " are not allowed to intermarry ; nor is marriage allowed between cousins, between uncle and niece, between aunt and nephew, and between step-parents and step- children.4 " In Tibet," says Mr. Sarat Chandra Das, " members of the same family are forbidden by law to contract matrimonial alliances with their kindred within seven degrees. This rule is, however, nowadays disregarded by the people, who are known to make alliances with their kinsmen who are distant only three or four degrees of consanguinity." 5 It is obvious that these prohibitions refer 1 Sumner, in Jour. Anthr. Inst. xxxi. 87. Jochelson, Yukaghir, p. 80. 2 Bergmann, Nomadische Streifereien unter den Kalmuken, iii. 146. 3 Bell, Journal of a Residence in Circassia, i. 347 sq. 4 Mannerheim, Visit to the Sard and Shera Yogurs, pp. 14, 33, 35. 5 Sarat Chandra Das, Journey to I.hasa and Central Tibet, p. 326. XIX- EXOGAMY 115 to relatives on the father's side.1 The Chins of Burma are divided into a large number of exogamous clans ; 2 whilst among the Chingpaw or Kachins, inhabiting the country on the north, north-east, and north-west of Upper Burma, all persons bearing the same family name regard themselves as of one blood and will not marry each other, even though they may belong to different tribes.3 The people of Corea are divided into exogamous clans, each of which traces its descent by primogeniture from a single male ancestor. The prohibition of marriage between persons who bear the same clan name " is more than a law with penalty for infraction ; it is a traditional custom of which the negative is inconceivable."4 In China, also, there is a rule of exogamy attaching to family names. Large bodies of persons in that country bear the same surname-indeed, among the entire native population there are hardly more than 530 family names : and according to the old Chinese penal code a penalty of sixty blows is inflicted on any one who marries a person' with his own name, and the marriage shall be null and void.6 But the punishment attached to the intermarriage of nearer relations on the father's side is much more severe : marriage or incestuous intercourse with a grand-uncle, a father's first cousin, a brother, or a nephew is punishable by death.6 And besides these prohibitions there are others applying within a narrower range to relatives on the female side. A man who marries his mother's sister or his sister's daughter is strangled. Less severe punishment is inflicted on a person who marries a uterine half-sister, and still less 1 Cf. Puini, Il Tibet secondo la relazione del viaggio del P. Ippolito Desideri, p. 129. 2 Whitehead, ' Notes on the Chins of Burma,' in Indian Antiquary, xxxvi. 206. 8 Scott and Hardiman, Gazetteer of Upper Burma and the Shan States, pt. i. vol. i. 402, 404. 4 Hough, ' Korean Clan Organization,' in American Anthropologist, N.S. i. 150, 152. 5 Medhurst, * Marriage, Affinity, and Inheritance in China,' in Trans. Roy. Asiatic Soc. China Branch, iv. 21 sq. Ta Tsing Leu.Lee, sec. cvii. p. 114. 8 Medhurst, loc. cit. p. 24. Ta Tsing Leu Lee, sec. cix. p. 116. r 2 116 THE HISTORY OF HUMAN MARRIAGE chap. severe-eighty blows-on any one who marries his father's sister's daughter, his mother's brother's daughter, or his mother's sister's daughter. An after-clause abrogates this prohibition, and permits marriage between children of brother and sister, or of sisters, but intermarriage between those of brothers is of course inadmissible.1 The Chinese code also interdicts occasional intercourse with any of those relatives with whom marriage is prohibited, the punishment in both cases being the same.2 The division into exogamous clans or septs is found all over India.3 In many cases the clans are totemic ; but Sir James G. Frazer thinks it is doubtful whether totemism proper is practised by any race of India except the Dravidian, although there are some resemblances to it in combination with exogamy among the Mongoloid peoples of Assam.4 Descent is nearly always traced through the father ; almost the only exceptions to this rule are found among the Khasis and Garos in Assam and in some parts of Southern India, chiefly on the Malabar coast, where children belong to their mother's clan.5 How stringent the rule of exogamy may be, appears from the following statement referring to the Khasis :-" As the clans are strictly exogamous, a Khasi cannot take a wife from his own clan ; to do this would entail the most disastrous religious, as well as social con- sequences. For to marry within the clan is the greatest sin a Khasi can commit, and would cause excommunication by his kinsfolk and the refusal of funeral ceremonies at death, and his bones would not be allowed a resting-place 1 Medhurst, loc. cit. p. 23. Jamieson, ' Translations from the General Code of Laws of the Chinese Empire,' in China Review, x. 82 sq. Cf. Gray, China, i. 186. 2 Medhurst, loc. cit. p. 27. 3 Russell, Tribes and Castes of the Central Provinces of India, vols. ii.-iv. passim. Thurston, Castes and Tribes of Southern India, passim. Nanjundayya, Ethnographical Survey of Mysore, passim. Crooke, Tribes and Castes of the North-Western Provinces and Oudh, passim. Dalton, Descriptive Ethnology of Bengal, passim. Risley, Tribes and Castes of Bengal, passim. Idem, People of India, p. 161 sqq. Gait, op. cit. p. 249 sqq. Frazer, Totemism and Exogamy, ch. x. 4 Frazer, op. cit. ii. 329 sq. 6 Gait, op. cit. pp. 237, 238, 249. XIX EXOGAMY 117 in the sepulchre of the clan."1 Most Dravidian tribes, which allow great freedom between the sexes prior to marriage, also prohibit extra-matrimonial intercourse be- tween persons of the same exogamous group, but it never- theless occasionally takes place. The Mongoloid tribes are more often indifferent to this consideration ;2 and Dr. Rivers doubts whether there is any strict prohibition of sexual inter- course within the clan among the Todas. He was, in fact, told the contrary.3 In various tribes in India a man is not only forbidden to marry a woman of his own clan, but must also refrain from marrying in some other clan or clans. Thus, among the Bants in South Canara the taboo extends to certain allied clans.4 Among the Meitheis in Manipur some of the clans are or were formerly forbidden to intermarry ;5 and although among the Nagas marriage as a rule is free between all the clans in a village or group, we find at Liyai that the four component clans are arranged in pairs which mutually forbid marriage.6 Not infrequently a man is said to be prohibited from taking a wife not only from his own clan, but from that of his mother7 or from that of his grandmother or either of his grandmothers as well.8 Among the Kandhs of Orissa and the Santals, who are divided into exogamous clans and sub-clans, a man may marry into his mother's clan but not into her sub-clan.9 The Banias-an occupa- tional caste of bankers, moneylenders, and dealers in grain, butter, and groceries, who form rather over one per cent, of the population in the Central Provinces-consist of endogamous sub-castes, most of which have an elaborate 1 Gurdon, Khasis, p. 77. 2 Gait, op. cit. p. 243. 3 Rivers, Todas, pp. 505, 530. * Sturrock, Madras District Manuals : South Canara, i. 160. 5 Hodson, Meitheis, pp. 73, 75. 8 Idem, Naga Tribes of Manipur, p. 89. 7 Idem, Meitheis, p. 76. Russell, op. cit. ii. 295 (Bhilalas), 481 Dhangars) ; iv. 116 (Lodhis), 203 (Marathas). In some Naga villages Hodson (Naga Tribes of Manipur, p. 89) heard that marriages with girls of the mother's clan are not liked. 8 Russell, op. cit. ii. 287 (Bhils), 409 (Chamars) ; iii. 233 (Jats), 386 Katias), 418 (Kayasths) ; iv. 22 (Kunbis). 9 Risley, Tribes and Castes of Bengal, i. 400 sq. 118 THE HISTORY OF HUMAN MARRIAGE chap. system of exogamy. " They are either divided into a large number of sections, or into a few gotras (clans), usually twelve, each of which is further split up into subsections. Marriage can then be regulated by forbidding a man to take a wife from the whole of his own section or from the subsection of his mother, grandmothers and even great- grandmothers. By this means the union of persons within five or more degrees of relationship either through males or females is avoided, and most Banias prohibit inter- marriage, at any rate nominally, up to five degrees."1 In many castes marriage is prohibited within three degrees of relationship between persons connected through females ;2 and very frequently we are told that a man is allowed to marry neither a woman of his own clan nor a first cousin.3 The patrilineal Todas forbid marriage with the mother's sister's daughter,4 and the matrilineal Bants of South Canara with the father's brother's daughter.5 In addition to the prohibition of marriage within the gotra, or clan, a Brahman must not, according to Hindu law, marry a girl of his mother's or maternal grandfather's gotra, or one who is a sapinda, of his father or maternal grandfather. The rule barring the marriage of sapindas is that two persons cannot marry if they are both as near as fourth in descent from a common ancestor, and the relation- ship is derived through the father of either party ; whereas if the relationship of the couple is through their mothers in each case, then they cannot marry if they are third in descent from the same ancestor. But in practice marriages are held to be valid between persons fourth in descent from a common ancestor in the case of male relationship, and third in the case of female relationship, that is, persons having a common great-grandparent in the male line or a 1 Russell, op. cit. ii. 121. 2 Ibid. ii. 194, 245 ; iii. 286 ; iv. 60, 152, 166. Risley, op. cit. i. 38, 122 sq. ; ii. 45, 47, 49, and Appendix, p. 97. Dehon, ' Religion and Customs of the Uraons,' in Memoirs Asiatic Soc. Bengal, i. 160. Sarat Chandra Roy, Or dons of Chota Nagpur, p. 351. 3 Russell, op. cit. ii. 10, 26, 210, 295, 302, 332, 401, 481, 491, 504, 520 ; iii. 5, 215, 282, 303, 371, 424, 441, 546, 556 ; iv. 251, 302, 326, 406, 547. Gordon, Indian Folk Tales, p. 7... 4 Rivers, Todas, p. 509. B Sturrock, op. cit. i. 160. XIX EXOGAMY 119 common grandparent in the female line can marry.1 In the ' Laws of Manu ' we read that a damsel " who is neither a Sapinda on the mother's side, nor belongs to the same family on the father's side, is recommended to twice-born men for wedlock and conjugal union."2 In the sacred law-books the sapinda relationship is said to extend to six degrees where the' common ancestor is a male, and in those of Manu and Apastamba also when the common ancestor is a female ; whereas Gautama, Vishnu, Narada, and other authorities in the latter case limit it to four degrees.3 Of the exogamous rules among the natives of the Malay Peninsula little is known. Messrs. Annandale and Robinson assert that among the Malays of the Patani States the union of first cousins is not permitted, cousins being regarded as brothers and sisters (suddraY* Vaughan Stevens states that the Orang Laut were originally divided into families recognising a special locality or district as their home, and that marriage was not allowed between men and women belonging to the same locality. This rule fell into disuse when the domain of the Orang Laut became restricted to its present area, but nevertheless the spouse is still chosen from as distant a locality as possible.5 Of the Beduanda Kallang in Johor it is said that " persons of the same family cannot intermarry however remote the degree," though " the traces of relationship must soon be lost."6 With reference to some other tribes of the Orang Laut we are told that the children of brothers might not intermarry.7 Hale 1 Russell, op. cit. ii. 362 sq. 8 Laws of Manu, iii. 5. 3 Ibid. iii. 5 ;. v. 60. Apastamba, ii. 5. 11. 16. Gautama, iv. 3, 5, Institutes of Vishnu, xxiv. 9 sq. Ndrada, xii. 7. Mayne, Treatise on Hindu Law and Usage, p. 104. See also Mandlik, Vyavahdra Mayukha, p. 345 sq. ; Macdonell and Keith, op. cit. i. 475. 4 Annandale and Robinson, Fasciculi Malayenses, ii. 72. 5 Stevens, ' Mittheilungen aus dem Frauenleben der Orang Belendas, der Orang Djakun. und der Orang Laut,' in Zeitschr. f. Ethnol. xxviii. 174. 6 Logan, ' Biduanda Kallang of the River Pulai in Johore,' in Jour. Indian Archipelago, i. 300. 7 Idem, ' Orang Sabimba of the Extremity of the Malay Peninsula, ' in Jour. Indian Archipelago, i. 297. Idem, ' Orang Muka Kuning,' ibid. i. 339 h.* THE HISTORY OF HUMAN MARRIAGE chap. 120 states that among the Kinta Sakai of Perak " a man goes to a considerable distance for a wife, generally to a tribe who speak quite a different dialect." 1 Among the people of Kuala Kernan, who belong to the Sakai, " it was not usual to marry a woman brought up in the same camp."2 According to one account, the Sinhalese of Ceylon were formerly, under the native kings, forbidden " to contract matrimonial alliances with any relation who could claim a nearer degree of affinity than that of second-cousin."3 But from another source we learn that the most proper marriage among them is that with a father's sister's son or a mother's brother's son, though two brothers' or sisters' children are rarely allowed to marry.4 Concerning the Andamanese Mr. Man states that " their customs do not permit of the union of any who are known to be even distantly related ; the fact of our allowing first cousins to marry seems to them highly objectionable and immoral."5 According to Mr. Portman, incestuous marriages never take place among them, and a man prefers to marry into another sept, or tribe, of the same group.6 Speaking of the natives at the settlement Port Blair, Mr. Lowis, on the other hand, asserts that there are no exogamous groups or prohibited degrees of relationship among them ; 7 but this statement cannot be supposed to be literally correct. The Battas in the interior of Sumatra are divided into a number of exogamous totem clans, called mar gas, with descent in the male line ; and not only is marriage forbidden between members of the same clan, but sexual intercourse 1 Hale, ' Sakais,' in Jour. Anthr. Inst. xv. 291. See also infra, ii. 209. 2 Knocker, 'Notes on the Wild Tribes of the Ulu Plus, Perak,' in Jour. Roy. Anthr. Inst, xxxix. 146. 3 Sirr, Ceylon and the Cingalese, ii. 167. 4 Supra, ii. 73. 5 Man, ' Aboriginal Inhabitants of the Andaman Islands,' in Jour. Anthr. Inst. xii. 135 sq. Cf. 'Kloss, In the Andamans and Nicobar s, p. 188. 8 Portman, History of Our Relations with the Andamanese, i. 39. 7 Lewis, op. cit. p. 67. XIX EXOGAMY 121 between such persons is viewed as incest in the highest degree and severely punished.1 They, moreover, prohibit marriages between members of different margas who are closely related by blood.2 Clan exogamy, apparently combined with totemism, is also found in Mandailing, a district on the west coast of Sumatra ; and in some cases the prohibition of marriage extends over a group of clans, all of which regard themselves as related in the male line. Marriage between two persons of the same clan has occasion- ally happened ; but such an event is considered a disgrace to the community in which it takes place, the guilty couple are banished from it, and the husband is obliged to sacrifice a buffalo, a cow, or a goat for the common weal.3 Clan exogamy, either with paternal or maternal descent, prevails in various other parts of Sumatra,4 as also in Nias, to the 1 van Hoevell, ' lets over 't oorlogvoeren der Batta's,' in Tijd- schrift voor Nederlandsch Indie, N.S. vol. vii. pt. ii. 434 n. 1. Wilken, Over de verwantschap en het huwelijks- en erfrecht bij de volken van het maleische ras, pp. 20, 21, 36. Neumann,' Het Pane-en Bila-stroom- gebied op het eiland Sumatra,' in Tijdschrift van het Kon. Nederlandsch Aardrijkskundig Genootschap, ser. ii. vol. iii. 473, 491. v. Brenner, Besuch bei den Kannibalen Sumatras, pp. 197, 250. van Dijk, ' Eenige aanteekeningen omtrent de verschillende stammen (margas) en de stamverdeeling bij de Battaks,' in Tijdschrift voor Indische taal-, land- en volkenkunde, xxxviii. 296 sq. Joustra, ' Het leven, de zeden en gewoonten der Bataks,' in Mededeelingen van wege het Nederlandsche Zendelinggenootschap, xlvi. 388 sq. Frazer, Totemism and Exogamy, ii. 186 sqq. 2 Neumann, loc. cit. p. 492. 3 Ris, ' De onderafdeeling Klein Mandailing Oeloe en Pahantan en hare bevolking met uitzondering van de Oeloe's,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, xlvi. 472, 473, 490, 492, 508 sq. Heyting. ' Beschrijving der onder-afdeeling Groot-Mandeling en Batang-Natal,' in Tijdschrift van het Kon. Neder- landsch Aardrijkskundig Genootschap, ser. ii. vol. xiv. 245, 246, 257. Frazer, op. cit. ii. 190 sq. 4 van Hasselt, Volksbeschrijving van Midden-Sumatra, p. 214 sq. (Malays of Lebong in Palembang). Wilken, ' Over het huwelijks- en erfrecht bij de volken van Zuid-Sumatra,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, xl. 154 sqq. ; Idem, Over de verwantschap en het huwelijks- en erfrecht bij de volken van het maleische ras, p. 13 (Pasemahers of Southern Sumatra). Idem, quoted by Frazer, Totemism and Exogamy, ii. 194 (Malays of Mapat-Toenggoel and Rantau-Binoewang in the north-east of the 122 THE HISTORY OF HUMAN MARRIAGE chap. west of that island, where descent is patrilineal.1 There are also exogamous clans with descent in the male line in Ceram,2 Burn,3 and the northern part of Halmahera. In the last-mentioned island sexual intercourse between the members of the same clan is deemed very culpable, although it is not a crime. But the clan has there no name of its own and is not reckoned after the fourth generation, hence a great-great-grandson can marry the great-great-grand- daughter of the same great-great-grandfather.4 I am not aware of the existence of clan exogamy in other parts of the Indian Archipelago. The pagan tribes of Borneo are expressly said to have no such thing.5 But most of them seem to prohibit or to have an objection to the marriage of first cousins, though in some places such marriages are tolerated on condition of the payment of a fine or the performance of an expiatory -ceremony.6 Among the Land Dyaks of Sarawak, according to Sir Spenser St. John, not even second cousins are allowed to intermarry except after paying a fine of two jars, one of which is given by the woman to the relations of her lover and the other one by the lover to her relations.7 All marriages between cousins are prohibited in Minahassa in Celebes 8 and among the Watubela Islanders ; 9 but in many parts of the Archipelago island). Graafland, ' De verbreiding van het matriarchaat in het landschap Indragiri,'in Bijdragen totdetaal-, land- en volkenkunde van Nederlandsch-Indie, xxxix. 42 sq. (Orang Mamaq), 44 (Tiga Loeroeng). Moszkowski, Auf neuen Wegen durch Sumatra, p. 104 sq. (Sakai). 1 v. Rosenberg, Der malayische Archipel, pp. 155, 160. Wilken, Verwantschap, pp. 21, 46 sq. 2 Wilier, Het eiland Boeroe, p. 7. Wilken, Verwantschap, p. 21. 3 Wilier, op. cit. p. 100. Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, pp. 20, 22. Martin, Reisen in den Molukken. p. 289. 4 Wilier, op. cit. pp. 44, 45, 53. 5 Hose and McDougall, Pagan Tribes of Borneo, i. 73. 6 St. John, Life in the Forests of the Far East, i. 85. Low, Sarawak, p. 300. Brooke, Ten Years in Sarawak, ii. 336 sq. Hose and McDougall, op. cit. i. 73. Ling Roth, Natives of Sarawak and British Borneo, i. 123. Hornaday, Ten Years in the Jungle, p. 453. Frazer, Folk-Lore in the Old Testament, ii. 172 sqq. Cf. Gomes, Seventeen Years among the Sea Dyaks of Borneo, p. 128. See, how- ever, supra, ii. 100. 7 St. John, op. cit. i. 198. 8 Hickson, A Naturalist in North Celebes, p. 277. Wilken, Verwantschap, p, sq. 9 Riedel, op. cit. p. 206. XIX EXOGAMY 123 cross-cousins are allowed to marry, although the children of two brothers or of two sisters are forbidden to do so,1 or man may marry the daughter of his mother's brother but not the daughter of his father's sister.2 Among the Bagobo of Mindanao " blood relationship is traced as far as the second cousin and is a bar to marriage. . . . Incest should be punished by the death of the culprits, for should such a crime go unpunished the spirits would cause the sea to rise and cover the land."3 It is a frequent notion in the Indian Archipelago that incest will blight the fruits of the earth,4 and it is often punished with the utmost severity. In many places the guilty parties are drowned,5 or burned,6 or pinned to the earth with a stake driven through their 1 Roos, ' lets over Endeh,' in Tijdschrift voor Indische taal-, land- en volkenkunde, xxiv. 523 (Endeh in the island of Flores). In the Aru Islands cross-cousins and also the children of two sisters can intermarry, but not the children of two brothers (Riedel, op. cit. pp. 250, 474 ; Wilken, in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, ser. v. vol. i. 146) ; wherets in the islands of Leti, Moa, and Lakor marriage may. take place between cross- cousins and between the children of two brothers, but not between the children of two sisters (Riedel, op. cit. pp. 385, 474 ; Wilken, in Bijdragen (Sc.,-ser. v. vol. i. 146). 2 Neumann, loc. cit. pp. 243, 492 ; Joustra, loc. cit. p. 390 (Battas). Ris, loc. cit. p. 508 (Lubus of Mandailing). Marsden, History of Sumatra, p. 228 (Rejangs). Pleyte, ' Ethnographische beschrijving der Kei-Eilanden,' in Tijdschrift van het Nederlandsch Aardrijks- kundig Genootschap, ser. ii. vol. x. 808. Wilken, in Bijdragen &c., ser. v. vol. i. 148 (natives of Amboyna). van Schmid, ' Aantee- keningen nopens de zeden, gewoonten en gebruiken, benevens de vooroordeelen en bijgeloovigheden der bevolking van de eilanden Saparoea, Haroekoe, Noessa Laut, en van een gedeelte van de zuid-kust van Ceram,' in Tijdschrift voor Norlands Indie, vol. v. pt. ii. 596 sq. Schadee, ' Heirats- und andere Gebrauche bei den Mansela und Nusawele Alfuren in der Unterabteilung Wahaai der Insel Seram (Ceram),' in Internationales Archiv fiir Ethnographic, xxii. 134. 3 Cole, ' Wild Tribes of Davao District, Mindanao,' in Field Museum of Natural History, Anthropological Series, xii. 95, 98. 4 Frazer, Magic Art, ii. 107 sqq. Idem, Psyche's Task, p. 46 sqq. Idem, Folk-Lore in the Old Testament, ii. 171, 173 sq. 4 Wilken, Huwelijken tusschen bloedverwanten, p. 26 sq. Perelaer, op. cit. p. 59 sq. ; Hose and McDougall, op. cit. ii. 196 (Borneo).. Jacobs, Eenigen Tijd onder de Baliers, p. 126. Riedel, op. cit. p. 460 (Romang). Finley and Churchill, Subanu, p. 41. 6 Jacobs, op, cit. p. 126 (Bali). 124 THE HISTORY OF HUMAN MARRIAGE chap. bodies,1 or speared to death,2 or-among the Battas-killed and eaten.3 The Western Islanders of Torres Straits are divided into exogamous totem clans with descent in the male line, and not only marriage but sexual intercourse was prohibited between members of the same clan. However, marriages might take place between clans that had the same chief totem but different subsidiary totems-most of the clans having more than one totem, a chief totem and in addition other ones of less importance ; and the prohibition to marry a woman of the same totem clan did not extend to women of other islands. On the other hand, a man is forbidden to marry not only women of his own totem clan, but also women of other totem clans if they are connected with him by certain ties of kinship.4 The genealogical record taken by the Cambridge Anthropological Expedition to Torres Straits shows that marriages between people nearly related to one another never occurred, while marriages between people related to one another even remotely were rare ; there is not on record a single definite case of marriage between first cousins.5 Dr. Haddon writes of these islanders, " So far as we could see the act which was held by the natives to be above all others morally reprehensible was incest, i.e. marriage or connection within the clan or between those whom they considered as too nearly related to one another."6 In New Guinea and the little islands in its immediate neighbourhood clan exogamy combined with totemism is of common occurrence,7 though evidently not universal, 1 Hose and McDougall, op. cit. ii. 196 (pagan tribes of Borneo). 2 Cole, loc. cit. p. 193 (Mandaya in Southern Mindanao). 3 Wilken, V erwantschap, p. 18. 4 Haddon and Rivers, in Reports of the Cambridge Anthropological Expedition io Torres Straits, v. 159 sqq. Rivers, ibid. v. 235 sqq. 5 Rivers, ibid. v. 239. * Haddon, ibid. v. 274. 7 Seligman, Melanesians of British New Guinea, pp. 10, 16, 271, 447, 507, 683, 714. Williamson, ' Some unrecorded Customs of the Mekeo People of British New Guinea,' in four. Roy. Anthr. Inst, xliii. 275. Hely, quoted by Haddon and Rivers, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 189; Landtman, op. cit. p. 80 (Kiwai). Malinowski, ' Natives of Mailu/ in Trans. XIX EXOGAMY 125 whatever may have been the case in former days ; we know that in some places it was once more rigidly observed than it is, now. Among the Massim, occupying the south-eastern extremity of British New Guinea and adjacent islands, a man is, or was, forbidden to marry not only within his own but into his father's clan.1 Of the limitation imposed by. consanguinity among the Southern Massim Professor Selig- man says that it was not clear how far it extended, " but it certainly seemed that third cousins might not marry."2 Among the Koita, too, the forbidden degrees extend to third cousins.3 The Mekeo tribes, which reckon descent in the male line, allow a man to marry a girl related to him in the female line only on condition that their parents are suffi- ciently removed in relationship from each other; for example, marriage between the children of two sisters is not allowed, and marriage between their children (second cousins) is not strictly regular.4 In Kaiser Wilhelm Land it is the rule that cousins must not marry, and in Tami even second cousins are forbidden to do so.5 Among the Mekeo tribes sexual intercourse is not forbidden between members of the same clan, though marriage is ;6 whereas among the natives of Mailu clan exogamy regulates both marriage and extra-connubial intercourse, though not with strict rigidity.7 It seems that in nearly all Melanesian islands, with the exception of New Caledonia and Fiji, the natives are divided into two or more exogamous classes, not infrequently of a more or less totemic character, with descent in the female Roy. Soc. South Australia, xxxix. 560. Poch, ' Vierter Bericht iiber meine Reise nach Neuguinea (Niederlandisch-Neuguinea),' in Sitzungsberichte der mathematisch-natiirwissenschaftlichen Klasse der Kaiserl. Akademie der Wissenschaften, vol. cxv. pt. i. 900 (Kaya- Kaya, on the southern coast of Dutch New Guinea). Neuhauss, Deutsch Neu-Guinea, i. 161. 1 Seligman, op. cit. pp. 10, 447, 507, 683, 714. 2 Ibid. p. 507. 8 Ibid. p. 82. 4 Williamson, in Jour. Roy. Anthr. Inst, xliii. 275. 5 Neuhauss, op. cit. i. 161. Vetter, ' Bericht iiber papuanische Rechtsverhaltnisse bei den Jabim,' in Nachrichten uber Kaiser Wilhelms-Land und den Bismarck-Archipel, 1897, P- 89. 8 Seligmann, op. cit. p. 364. 7 Malinowski, loc. cit. p. 560. THE HISTORY OF HUMAN MARRIAGE CHAf. 126 line.1 And the rule that' a man may not marry a woman of his own class is, as usual, supplemented by a further prohi- bition of marrying women of another class who are nearly related to him.2 He may be forbidden to marry his father's sister 3 or his brother's daughter,4 or a cousin belonging to another class than his own ; in Mota, one of the Banks Islands, for example,-the children of a brother and sister are 1 Codrington, Melanesians, p. 21 sqq. Macdonald, Oceania, p. 181 sq. (Efate). Speiser, Two Years with the Natives in the Western Pacific, p. 233 (Pentecost). Rivers, ' Totemism in Polynesia and Melanesia,' in Jour. Roy. Anthr. Inst, xxxix. 164-167 (Santa Cruz Island), 168-170 (some of the Solomon Islands), 172 (Efate, Pente- cost). Idem, History of Melanesian Society, i. 20, 24 (Banks Islands). Ribbe, Zivei Jahre unter den Kannibalen der Salomo-Inseln, p. 140 sq. Williamson, Ways of the South Sea Savage, p. 60 (some of the Solomon Islands). Thurnwald, ' Ermittlungen uber Eingeborenenrechte der Siidsee,' in Zeitschr. vergl. Rechtswiss. xxiii. 335 ; Idem, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 15, 61 sq. (Buin, in Bougainville). Parkinson, Dreissig Jahre in der Sudsee, pp. 61, 67 (Gazelle Peninsula of New Britain), 267 (New Ireland, New Hanover), 393 (Admiralty Islands), 481' (Bougainville), 649, 650 (New Ireland), 652, 653 (Tanga, Aneri, and the district Siara on the south-eastern coast of New Ireland), 660 sq.(islands of the Solomon group which formerly belonged to Germany, generally). Brown, ' Notes on the Duke of York Group, New Britain, and New Ireland,' in Jour. Roy. Geograph. Soc. xlvii. 149. Danks, ' Marriage Customs of the New Britain Group,' in Jour. Anthr. Inst, xviii. 282 sq. Pfeil, Studien und Beobachtungen aus der Sudsee, p. 27 (Gazelle Peninsula, Duke of York Group). Burger, Die Kilsten- und Bergvblker der Gazellehalbinsel, p. 24 sq. (coast people). Hahl, ' Ueber die Rechts- anschauurigen der Eingeborenen eines Theiles der Blanchebucht und des Innern der Gazelle Halbinsel,' in Nachrichten uber Kaiser Wilhelms-Land und den Bismarck-Archipel, 1897, p. 80. Peckel, ' Die Verwandtschaftsnamen des mittleren Neumecklenburg,' in Anthropos, iii. 458; Stephan and Graebner, Neu-Mecklenburg (Bis- marck-Archipel), p. 106 (New Ireland). Schnee, Bilder aus der Sudsee. Unter den kannibalischen Stammen des Bismarck-Ar chipels, p. 38. Frazer, Totemism and Exogamy, ch. vii. 2 Codrington, op. cit. p. 29. Brown, in Jour. Roy. Geograph. Soc. xlvii. 149 (Duke of York Islands, New Ireland, and New Britain). Rivers, History of Melanesian Society, i. 47 (Banks Islands). Speiser, op. cit. p. 234 (Pentecost). 3 Rivers, op. cit. i. 47 (Banks Islands). 4 Danks, loc. cit. p. 283 (New Britain). Speiser, op. cit. p. 234 (Pentecost). XIX EXOGAMY 127 thought " too near " to marry.1 Marriage between persons of the same class, should it ever occur, is in many islands an offence which is punished by death.2 Among the Batiks Islanders, according to Dr. Rivers, " when in the old days a man married a woman of his own vevc (exogamous class) the offending couple were killed in public by their relatives, usually by shooting with bows and arrows, less frequently by clubbing to death."3 The Rev. Benjamin Danks says that in New Britain a marriage within the exogamous class would bring instant destruction upon the woman, and if not immediate death to the man, his life would never be secure. The relatives of the woman would be so ashamed that only her death could satisfy them. But such a marriage never occurs in a thickly populated district. If a man should be accused of adultery or fornication with a woman, he would at once be acquitted by the public voice if he could say, " She is one of us," that is, she belongs to my totem, which in itself precludes the possibility of any sexual intercourse between us.4 Not only marriage but also extra-matrimonial intercourse within the class is prohibited, and is generally punished with the same severity.5 In Florida a man who had been guilty of it would formerly have been killed and the woman would have been made a harlot.6 In Fiji, society is supposed to have been formerly organised on the dual plan with two exogamous moieties and matrilineal descent ;7 and Mr. Hocart even states that this system " exists in that part of Vanua Levu which appears to have been least affected by the invasion."8 In New Caledonia, the Loyalty Islands, and the whole of Polynesia clan 1 Codrington, op. cit. p. 29. 2 Speiser, op. cit. p. 233. Parkinson, op. cit. pp. 61, 267, 649, 650, 653. Hahl, loc. cit. p. 80. Pfeil, op. cit. p. 27. Peckel, loc. cit. p. 458. 3 Rivers, op. cit. i. 50. 4 Danks, loc. cit. p. 282 sq. 5 Codrington, op. cit. p. 23 sq. Parkinson, op. cit. pp. 61, 267, 649, 650, 653. Thurnwald, op. cit. iii. 15. Schnee, op. cit. p. 38. Hahl, loc. cit. p. 80. Burger, op. cit. p. 25. 6 Codrington, op. cit. p. 23. 7 Rivers, in Jour. Roy. Anthr. Inst, xxxix. 158. Hocart, * Early Fijians,' ibid. xlix. 47. 8 Hocart, loc. cit. p. 47. 128 THE HISTORY OF HUMAN MARRIAGE chap. exogamy is not known to exist.1 There are, however, every- where prohibited degrees of relationship. Mr. Hadfield wrote to me from Lifu, one of the Loyalty Islands, that marriages are there forbidden between first, though not second, cousins both on the mother's and father's side, as well as between uncles and nieces and between nephews and aunts ; but according to another account marriages between second cousins were also prohibited.2 In New Caledonia3 and Fiji4 first cousins who are the children of two sisters or of two brothers may not intermarry, whereas the marriage of first cousins who are the children of a brother and sister respectively is not only allowed but favoured. In some parts of Fiji, however, cross-cousins " do not become concubitant until the second generation. In the first generation they are called tabu, but marriage is not actually prohibited."6 In Polynesia all first cousins seem generally to be for- bidden to intermarry. Sir Basil Thomson writes:- " Inquiries that have been made among the natives of Samoa, Futuna, Rotuma, Uea, and Malanta (Solomon Group), have satisfied me that the practice of concubitant marriage is unknown in those islands ; indeed, in Samoa and Rotuma, not only is the marriage of cousins-german for- bidden, but the descendants of a brother and sister respec- tively, who in Fiji would be expected to marry, are there regarded as being within the forbidden degrees as long as their common origin can be remembered. This rule is also recognised throughout the Gilbert Islands, with the exception of Apemama and Makin, and is there only violated by the high chiefs. In Tonga, it is true, a trace of the custom can be detected. The union of the grandchildren (and occasion- ally even of the children) of a brother and sister is there regarded as a fit and proper custom for the superior chiefs, 1 Cf. Frazer, Totemism and Exogamy, ii. 63, and ch. viii. Pere Lambert (op. cit. p. 82) speaks of families souches, or family stocks, in New Caledonia, but does not expressly say that they are exogamous. 2 Ray ' People and Language of Lifu, Loyalty Islands,' in Jour. Roy. Anthr. Inst, xlvii. 286. 3 Lambert, op. cit. pp. 94, 114 sq. 4 Thomson, Fijians, p. 183 sqq. 5 Ibid. p. 190 sq. See supra, ii. 74. XIX EXOGAMY 129 but not for the common people."1 Other writers tell us that in Samoa marriages are forbidden between first cousins, as also between all those who are subject to the same fater familias ; 2 or between the members of all families which have the same name, between persons who are related on the father's side, and between the relatives of the father's wives.3 But the Samoan prohibited degrees in the collateral line are not fixed : if the common origin lies so far back that the relationship is almost forgotten, the marriage is no longer regarded as illegal.4 Concerning the natives of Rotuma Professor Stanley Gardiner states that second cousins were allowed to intermarry if they were descended from a brother and a sister respectively, but not if they were the grand- children of two brothers or of two sisters.5 In Penrhyn Island or Tongarewa " the laws of the island forbid marriage with any relative as near as a second cousin." 6 In Mangaia, one of the Hervey Islands, according to the Rev. W. Wyatt Gill, distant cousins sometimes, though rarely, marry, but they " must be of the same generation, i.e., be descended in the same degree (fourth or fifth, or even more remotely) from the common ancestor." 7 Of the Easter Islanders it is simply said that " marriages between relations were unknown."8 Bishop Colenso states that among the Maori of New Zealand the marriage of first cousins has always been viewed with disgust.9 Among the Tuhoe tribe, according to Mr. Best, the members of the third generation of two divergent lines from a common ancestor might intermarry, but to marry any one of closer kinship than a third cousin was 1 Ibid. p. 191. 2 Kramer, Die Samoa-Inseln, ii. 58 n. 3. 3 v. Billow, ' Die Ehegesetze der Samoaner,' in Globus, Ixxiii. 185. 4 Schultz, ' The most important Principles of Samoan Family Law,' in Jour. Polynesian Soc. xx. 49. 6 Gardiner, ' Natives of Rotuma,' in Jour. Anthr. Inst, xxvii. 478. 0 Lamont, Wild Life among the Pacific Islanders, p. 136. 7 Gill, ' Mangaia (Hervey Islands),' in Report of the Second Meeting of the Australasian Association for the Advancement of Science, held at Melbourne, in 1890, p. 330. 8 Tregear, ' Easter Island,' in Jour. Polynesian Soc. i. 99. 9 Colenso, Maori Races of New Zealand, p. 48. VOL. II K THE HISTORY OF HUMAN MARRIAGE chap. 130 deemed incest and severely condemned.1 It does not appear that incest was common among the natives, though it is sometimes given as an explanation of peculiarities in genealogies.2 Colenso says that it was scarcely known, even by name ;3 but Mr. Best mentions two cases of it, which were punished with expulsion from the tribe. He observes that three of the terms for incest are connected with dogs : those who commit incest are compared to a dog which turns and bites its own tail.4 Among the Moriori of the Chatham Islands-a group of islands situated at a distance of 480 miles in a south-easterly direction from Wellington- " the marriage of close connections such as first cousins was much disapproved of, and even when not so closely related, as in the case of second and third cousins, the others, to show their disapprobation of marriages between close relatives, sung a song by way of contempt, calling it tivare (incestuous)."5 In Micronesia we again meet with clan exogamy. The people of Nauru, one of the Marshall Islands, who in 1894 consisted of less than fifteen hundred individuals, are divided into twelve clans which are strictly exogamous. Marriage within the clan is looked upon as the greatest crime which can be committed, and very rarely occurs. The culprits become objects of general contempt and are expelled from the clan, and no child born of such a union is allowed to live.6 Exogamous clans with descent in the mater- nal line are found in Ponape 7 and the Mortlock Islands,8 1 Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 15, 20. 2 Ibid. p. 31. 3 Colenso, op. cit. p. 48. 4 Best, loc. cit. p. 31. Cf. supra, ii. 96. 6 Shand, ' Moriori People of the Chatham Islands,' in Jour. Polynesian Soc. vi. 148. Mr. Shand of the Chatham Islands, Native Interpreter, is the only recognised authority on the subject of the history, legends, and customs of the Moriori (Tregear, The Maori Race, p. 581 n.*). 6 Jung> ' Aufzeichnungen fiber die Rechtsanschauungen der Eingeborenen von Nauru,' in Mittheil. Deutsch. Schutzgeb. x. 65. See also Senftt, ' Die Insel Nauru,' ibid. ix. 107. 7 Christian, Caroline Islands, p. 74. 8 Kubary, in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, pp. 245, 251, 252, 260. XIX EXOGAMY 131 belonging to the Caroline Group, and in the Pelew Islands.1 The Mortlock Islanders regard sexual intercourse between persons of the same clan as incest of the most heinous sort, and any member of the clan would be ready to avenge such an outrage.2 The great majority of Australian tribes about which we possess accurate information are divided into two matri- monial classes (phratries or moieties), the members of which are not allowed to marry within their own class. In many tribes each of these classes is composed of two exogamous sections, and each section may again be divided into two exogamous subsections; so that the tribes consist of two, four, or eight exogamous classes or sub-classes, the members of each of which are bound to seek their husbands or wives in a class or sub-class different from their own.3 The two-class system is found among the Urabunna, one of the southern-central tribes ; 4 among the tribes about Lake Eyre, in the interior of South Australia, such as the Dieri, who form the largest and most important of them,5 the Yendakarangu,6 who are a southern branch of the Urabunna, and others;7 among the tribes on the Darling River and other tribes extending for some distance up the Murray River from its junction with the Darling River, for instance, the Wiimbaio;8 among the tribes inhabiting the high plateaux extending from about Woodspoint in Victoria to New South Wales ;9 and among the Kulin people in Central and Southern Victoria,10 the Wotjobaluk in North-Western 1 Idem, Die socialen Einrichtungen der Pelauer, p. 35. Idem, ' Die Religion der Pelauer,' in Bastian, Allerlei aus Volks- und Menschenkunde, i. 16. 2 Kubary, in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, p.251. 8 An exhaustive .and very lucid account of Australian exogamy has been given by Frazer, in his Totemism and Exogamy, i. chs. i.-iv., and a briefer one by Thomas, in his Kinship Organisations and Group Marriage in Australia. 4 Spencer and Gillen, Native Tribes of Central Australia, p. 59 sq. I idem, Northern Tribes of Central Australia, p. 70 sq. 5 Howitt, Native Tribes of South-East Australia, p. 90 sq. 8 Ibid. pp. 93, 187 sq. 7 Ibid. p. 91 sqq. 8 Ibid. pp. 49-52, 97> Io°. 194- 9 Ibid. pp. 77, 101, 196 sq. 10 Ibid. p. 252 sq. K 2 THE HISTORY OF HUMAN MARRIAGE chap. 132 Victoria,1 the Gournditch-mara in South-Western Victoria,2 and the Buandik tribe about Mount Gambier in the extreme south-eastern corner of South Australia.3 Among all these tribes children take their class from their mother, with the exception of the Kulin tribes, where the descent of the class is in the male line. The four-class system with female descent prevails among the Kamilaroi, a large nation of North-Eastern New South Wales ; 4 among the Wiradjuri, a powerful tribe or nation of tribes in Central New South Wales ; 6 among the Wonghibon tribe to the north of the Wiradjuri, on the other side of the Lachlan River ;6 among the Kuinmurbura on the coast of Queensland, to the north of Rockhampton,7 the Kongulu, to the south-west of them,8 and the Wakelbura farther west;• among the Buntamurra in South-Western Queensland;10 and among the tribes throughout North-West-Central Queens- land.11 In South-Eastern Queensland, round about Mary- borough, there is a group of tribes with four sub-classes and male descent.12 Tribes with four exogamous classes or sub-classes also live on the coast of West Australia.13 The eight-class system is found in all the tribes from the Arunta at the centre of Australia to the Mara and Anula on the Gulf of Carpentaria. Every tribe is divided into two exogamous classes, each of which is composed of two sub- classes ; and each of these sub-classes is again divided into two sections, although in some of the tribes only four of the exogamous sections have separate names. Descent is in the male line.14 In the tribes which have the four-class system a man is, 1 Ibid. p. 241. 2 Ibid. pp. 69, 124. 3 Ibid. p. 123. 4 Ibid. p. 104 sq. 5 Ibid. p. 105 sq. 6 Ibid. pp. 108, 214. 7I&id.p.2i8. 8 Ibid. pp. in, 220. 9 Ibid. p. 112. 10 Ibid. p. 11^ sq. 11 Roth, Ethnological Studies among the North-West-Central Queensland Aborigines, p. 56 sqq. 12 Howitt, op. cit. p. 58 sqq. 13 Mrs. Bates, ' Marriage Laws and some Customs of the West Australian Aborigines,' in Victorian Geographical Journal, xxiii.-xxiv., quoted by Frazer, Totemism and Exogamy, i. 560 sqq. Brown, ' Three Tribes of Western Australia,' in Jour. Roy. Anthr. Inst. xliii. 147, 148, 172, 176 sq. 14 Spencer and Gillen, Native Tribes, p. 70 sqq. lidem, Northern Tribes, pp. 74, 75, 96-103, 116 sqq. XIX EXOGAMY 133 as a general rule, not merely prohibited from marrying a woman of his own moiety, but can only marry into one of the sections of the other moiety.1 The children belong to the sub-class neither of their father nor of their mother, but to the other sub-class of their mother's moiety if the descent of the moiety or class is in the female line, and to the other sub-class of their father's moiety if the descent of the moiety is in the male line ; what has been previously said about descent has thus reference to the moiety or primary class only, not to the sub-class. To take an instance. Among the Kamilaroi the moieties or classes, which bear the names of Kupathin and Dilbi, are subdivided each into two sub- classes, of which those of Kupathin bear the names of Ipai and Kumbo and those of Dilbi the names of Murri and Kubbi. An Ipai man marries a Kubbi woman and their children are Murri; a Kumbo man marries a Murri woman and their children are Kubbi; a Murri man marries a Kumbo woman and their children are Ipai; a Kubbi man marries an Ipai woman and their children are Kumbo.2 So also in the tribes which have the eight-class system a man always marries a woman of a certain sub-class in the moiety which is not his own, and the children belong to another sub-class of the father's moiety.3 In most of the tribes known to us there is not only class exogamy but clan exogamy as well, each moiety or class being subdivided into a number of exogamous totem clans. In some of these tribes, for example the Dieri 4 and the Wotjobaluk,6 a man of one class is free to marry a woman of any totem of the other class, whereas in other tribes, for instance, the Urabunna,6 Yendakarangu,7 and Kuin- 1 For a few exceptions to this rule see Howitt, op. cit. pp. 203, 204, 212 sqq. ; Mrs. Bates, loc. cit., quoted by Frazer, Totemism and Exogamy, i. 561. 2 Howitt, op. cit. pp. 62, 68 sq. 3 Spencer and Gillen, Native Tribes, p. 70 sqq. lidem, Northern Tribes, pp. 96-101, 116 sqq. 4 Howitt, op. cit. p. 175 sq. 6 Ibid. p. 241 sq. • Spencer and Gillen, Native Tribes, p. 114. lidem, Northern Tribes, p. 144. These writers admit, however {Native Tribes, p. 60 n. 1), that further investigation may reveal the fact that among the Urabunna a man is allowed to marry women of more than one totem. 7 Howitt, op. cit. p. 187 sq. THE HISTORY OF HUMAN MARRIAGE chap. 134 murbura,1 a man may only many a wortian of some particular totem. Among the Urabunna he may not even marry a woman of the proper totem unless she is a daughter of his mother's elder brother or (what comes to the same thing) of his father's elder sister, these terms being used in their classificatory sense.2 But there are tribes where the totems have no influence whatever on marriage, a man being free to marry a woman of his own or any other totem, provided that she belongs to the class and sub-class into which he is bound to marry. This is the case among the Arunta in the centre of Australia.3 Besides class and clan exogamy there are in Australia additional rules which prohibit the intermarriage of certain relatives who otherwise would be allowed to marry each other. Whilst the two-class system with paternal descent prevents a father from marrying his daughter, as she is of his own class, it does not prevent a son from marrying his mother, as they belong to different classes; and whilst the two-class system with maternal descent prevents a son from marrying his mother, it does not prevent a father from marrying his daughter. But in no Australian tribe is a father permitted to marry his daughter, or a son permitted to marry his mother. Moreover, whilst the two-class and four-class systems bar the marriage between cousins who are the children of two brothers or of two sisters, since these cousins necessarily belong to the same exogamous class, they do not bar the marriage between cross-cousins, who are the children of a brother and a sister respectively, because they belong to two different intermarrying classes.4 Yet in many tribes with the two- or four-class system such cousins also are prohibited from marrying each other. This is, for instance, the case among the Dieri5 and the Kulin 1 Ibid. p. 218. 2 Spencer and Gillen, Native Tribes, pp. 6x, 64. lidem, Northern Tribes, p. 71 sq. 3 lidem, Native Tribes, pp. 73, 116 sq. lidem, Northern Tribes, p. 152. 4 Cf. Frazer, Totemism and Exogamy, i. x8x, 449 ; Idem, Folk-Lore in the Old Testament, ii. 187 sqq. 6 Howitt, op. cit. p. 164 sq. XIX EXOGAMY 135 tribes,1 which are divided into two classes, and among the North-West-Central Queensland tribes 2 and some South- Eastern Queensland tribes,3 which are divided into four. The Bangerang4 and Wotjobaluk5 of Victoria not only forbid cross-cousins to marry each other, but also prohibit the marriage between the descendants of these cousins, so far as the relationship can be traced, because they are "too near " and only a little removed from " brother" and " sister," or because " their flesh is too near." The Narrinyeri, a tribe of South Australia, who are divided into exogamous totem clans with paternal descent, have a very great aversion to the marriage of second cousins.6 Among the Kurnai of Gippsland, who have neither exogamous classes nor totem clans, even third cousins are said to be within the prohibited degrees of relationship.7 It should be added that in tribes with the eight-class system the marriage of a man's children with the children of his sister, blood or tribal, is prevented by the rule of class exogamy.8 Mr. Curr even goes so far as to affirm, in general terms, that among the Australian aborigines " a man may not marry his mother, sister, half- sister, daughter, grand-daughter, aunt, niece, first or second cousin."9 This statement, although obviously an exaggera- tion, indicates that the disapprobation of cousin marriage in aboriginal Australia is very widespread.10 Not only are the exogamous rules among the Australian natives comprehensive and complicated, but they are, or were until lately, enforced with the greatest rigidity, the penalty of death being commonly inflicted on the transgressors.11 This is true both in the case of marriage 1 Ibid. p. 254. 2 Roth, op. cit. pp. 57, 182. 3 Howitt, op. cit. p. 232. 4 Ibid. p. 257. 8 Ibid. pp. 241, 243. 6 Taplin, ' Narrinyeri,' in Woods, Native Tribes of South Australia, p. 12. 7 Brough Smyth, Aborigines of Victoria, ii. 386. 8 Spencer and Gillen, Northern Tribes,, p. 117. Frazer, Totemism and Exogamy, i. 277 sqq. This had previously been pointed out by Crawley, Mystic Rose, p. 473. 9 Curr, op. cit. i. 106. 10 Cf. Frazer, Folk-Lore in the Old Testament, ii. 193. 11 Curr, op. cit. i. 118. Cameron, ' Notes on some Tribes of New South Wales,' in Jour. Anihr. 'Inst. xiv. 351 sq. Mathew, ' Australian Aborigines,' in Jour. Roy. Soc. New South Wales, xxiii. 399. Idem, Eaglehawk and Crowfp. 100. Howitt, op. cit. pp. 208, 232, 246, 247, THE HISTORY OF HUMAN MARRIAGE chap. 136 and in the case of irregular sexual intercourse, except that in certain tribes, on occasions when ceremonial licence takes place, the prohibitory class rules are disregarded ; yet even then intercourse between the nearest relatives-between parents and children and between brothers and listers- is strictly forbidden.1 In recent times contact with white people has brought about many changes, and not a few of the statements made above have reference to the past rather than to the present. Concerning the Tasmanians, a race which is now extinct, we are told that " it was rarely the custom amongst them to select wives from their own tribes, but rather to take them furtively, or by open force, from neighbouring clans."2 Clan exogamy is found in vogue among a large number of African peoples.3 Among the Herero every person belongs at once to two different clans, one of which, called an eanda, he inherits from his mother, and the other, called an oruzo, he inherits from his father, and both sets of clans appear to be totemic.4 As to the rules which regulate marriage in the Herero clans we have very little information, but from one of our authorities5 we hear that the maternal clans are exogamous, whatever may be the case with the paternal clans.6 The marriage of cousins who are the children of a brother and of a sister respectively is favoured, whereas marriage or sexual intercourse between the children of two brothers or of two sisters is regarded with horror as incest 255, 264, 266. Spencer and Gillen, Native Tribes, pp. 15, no. lidem, Northern Tribes, pp. 136, 140. Roth, op. cit. p. 182. Idem, North Queensland Ethnography : Bulletin No. 8. Notes on Govern- ment, Morals, and Crime, p. 5. Grey, Journals of Two Expeditions of Discovery in North-West and Western Australia, ii. 242. 1 Spencer and Gillen, Native Tribes, p. 96 'sqq. lidem, Northern Tribes, p. 136 sqq. 2 Milligan, quoted by Nixon, Cruise of the Beacon, p. 29. See also Brough Smyth, op. cit. ii. 386. 3 See Frazer, Totemism and Exogamy, ii. chs. xii.-xiv. 4 Ibid. ii. 357. 5 Dannert, ' Soziale Verhaltnisse der Ovaherero,' in Mitteilungen d. Geographischen Gesellschaft zu Jena, vi. 117. G See Frazer, Totemism and Exogamy, ii. 364. XIX EXOGAMY 137 and subjects the culprits to the consequences of the blood- feud.1 The Bechuana tribes are evidently totemic, but tyhether they are also exogamous must be regarded as an open question,2 in spite of Captain Conder's vague statement that exogamy seems to be the common practice among them.3 Casalis says that " the Bechuanas and the Caffres acknowledge and respect the same degrees of consanguinity as we do. They do not reckon relationship beyond the degree of second cousin. Marriages between brothers and sisters, uncles and nieces, nephews and aunts, are disapproved of. Those between cousins frequently take place, but there are some tribes that condemn them as incestuous."4 With reference to the Eastern Bantu of South Africa Mr. McCall Theal writes :-" No man of any coast tribe would marry a girl whose relationship by blood to himself on his father's side could be traced, no matter how distantly connected they might be. So scrupulous was he in this respect that he would not even marry a girl who belonged to another tribe, if she had the same family name as himself, though the relationship could not be traced. ... In some tribes, as at present in the Pondos, Tembus, and Xosas, the same rule was applied to relatives by blood on the mother's side also. Children take the family title of the father, and can thus marry those of the same family title as the mother, provided their blood relationship cannot be traced. Every man of a coast tribe regarded himself as the protector of those females whom we would call his cousins, second cousins, third cousins, and so forth, on the father's side, while some had a similar feeling towards the same relatives on the mother's side as well, and classified them all as sisters. Immorality with one of them would have been considered incestuous, something horrible, something un- 1 Dannert, Zum Rechte der Herero, p. 33 sq. Schinz, Deutsch-Sild- West-Afrika, p. 177. v. Francois, Nama und Damara Deutsch-Sud- West-Afrika, p. 195. Bensen, quoted by Kohler, in Zeitschr. vergl. Rechtswiss. xiv. 300 sq. Supra, ii. 75. 2 See Frazer, Totemism and Exogamy, ii. 378 ; iv. 304. 3 Conder, ' Present Condition of the Native Tribes in Bechuana- land,' in Jour. Anthr. Inst. xvi. 85. 4 Casalis, Basutos, p. 191. 138 THE HISTORY OF HUMAN MARRIAGE chap. utterably disgraceful. Of old it was punished by the death of the male, and even now a heavy fine is inflicted upon him, while the guilt of the female must be atoned by a sacrifice performed with due ceremony by the tribal priest, or it is believed a curse will rest upon her and her issue." In contrast to this prohibition the man of the interior, as we have seen, almost as a rule married the daughter of his father's brother, in order, as he said, to keep property from being lost to his family.1 Mr. Cousins, to whom I am indebted for a paper on the Cis-Natalian Kafirs, writes that they generally endeavour to marry out of their own " tribe," though marriages within the tribe and village often take place. He says that there is among them some kind of " class " division, with which he is not fully acquainted, and that members of the same class do not seem to intermarry. At any rate, near relatives, paternal and maternal, avoid marriage with each other ; and although no penalty is attached to such a marriage, custom is so strong on this point that the general rule is seldom broken. So also, according to Warner, unions which are regarded as incestuous are not punished by law among the Kafirs, " but they have a far more powerful preventative in their superstitious fears, which teach them to dread that some supernatural evil will befall the parties committing such acts ; they lose caste as it were, and are considered in the light of sorcerers : hence such crimes are seldom committed. Consanguineous marriages are prohibited by custom rather than by law."2 According to Shooter 3 and Dugmore,4 a marriage is considered in- cestuous if the man and woman are of any known or remem- bered degree of relationship by common descent, and if a man were to take a wife within the degrees prohibited by custom he would be denounced as an " evildoer " ; and Brownlee states that intercourse in such cases is punished, 1 Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 217 sqq. Supra, ii. 71. 2 Warner, in Maclean, Compendium of Kafir Laws and Customs, p. 62 sq. 3 Shooter, Kafirs of Natal and the Zulu Country, p. 45 sq. 4 Dugmore, in Maclean, op. cit. p. 163. XIX EXOGAMY 139 whether it be by marriage or without marriage.1 Mr. Eyles wrote to me with regard to the Zulus that there is no Intermarriage between the inhabitants of the village, the members of which are, as a rule, related. All intermarrying of relations is prohibited by custom, and such a thing is neither heard nor thought of. Even if the relationship is only traditional, the custom holds good. Among the Matabele, who are an offshoot of the Zulus, according to Decle, " people of the same class bearing the same family name cannot marry; the relationship, however, is only considered in the male line."2 Concerning the Hottentots Mr. McCall Theal says that " it was customary with them to take their wives not from their own but from another clan."3 According to an old writer, Kolbe, they punished alliances between first and second cousins with death.4 A modern traveller states that marriages between parents and children, between brothers and sisters, and between cousins are pro- hibited by the Hottentots, and that any allusion to such marriages is rejected with indignation.5 Among the Baronga of Delagoa Bay, a portion of the Thonga, two persons are forbidden to marry each other if it can be shown that they have a single common ancestor in the paternal line ; whereas it seems that this rule is not so stringent in the northern sections of the same people. According to one of M. Junod's informants, " marriage is absolutely prohibited between all the descendants of a grandfather, viz., be.tween first cousins. Between second cousins it is permitted conditionally, ' by killing the family tie,' and between third cousins it is allowed. ... On the mother's side, this absolute prohibition extends to first cousins when mothers are sisters."6 The Mananja, Yao, Makololo, and Chikunda tribes, at the southern end of Lake Nyasa, have clans with maternal 1 Brownlee, in Maclean, op. cit. p. 115. 2 Decle, Three Years in Savage Africa, p. 158. 3 Theal, op. cit. p. 85. 4 Kolbe, Present State of the Cape of Good-Hope, i. 155 sq. 5 v. Fran 901s, op. cit. p. 212. 6 Junod, Life of a South African Tribe, i. 241. See also Idem Les Ba-Ronga, p. 84 sqq. THE HISTORY OF HUMAN MARRIAGE chap. 140 descent, and members of the same clan are forbidden to intermarry.1 The Nyanja-speaking tribes of Central Angoni- land are likewise divided into exogamous and totemic clans, most of which have maternal descent. The clans are totemic, and no man may marry a woman who bears his own clan name, though she may be of another race and live in a distant country ; but first cousins who are the children of a brother and a sister respectively may marry each other, and the same is the case with the children of two sisters whenever they have different totems.2 Among the Awemba, another Bantu tribe of North-Eastern Rhodesia, a man may not marry a woman of his mother's totem, nor any first cousin who is the daughter either of his mother's sister or of his father's brother ;8 and among the neighbouring Winamwanga marriage with first cousins on either the father's or the mother's side is absolutely prohibited,4 as " marriage with a cousin would be looked on as marriage with a sister."5 In the latter tribe a man is prohibited from marrying any female of his own family of the same totem. . . . Within the totem no sexual intercourse is allowed. If a brother and sister, or two cousins descended from males of the same totem, had intercourse, they were burnt to death in the olden time." 8 Several Bantu tribes of the former German East Africa, such as the Makua and the Makonde, are divided into exogamous clans with descent in the female line,7 though 1 Stannus, ' Notes on some Tribes of British Central Africa,' in Jour. Roy. Anthr. Inst. xl. 307. 2 Rattray, Some Folk-Lore Stories and Songs in Chinyanja, pp. 174, 177, 202. Cf. Miss Werner, Natives of British Central Africa, p. 252 sq. 3 Gouldsbury and Sheane, The Great Plateau of Northern Rhodesia, p. 172. Gouldsbury, ' Notes on the Customary Law of the Awemba and Kindred Tribes,' in Jour. African Soc. xv. 157. 4 Gouldsbury and Sheane, op. cit. p. 173. Coxhead, Native Tribes of North-Eastern Rhodesia, p. 51. 5 Chisholm, ' Manners and Customs of the Winamwanga and Wiwa,' in Jour. African Soc. ix. 383. 6 Coxhead, op. cit. p. 51. 7 Weule, Wissenschaftliche- Ergebnisse meiner ethnographischen Forschungsreise in den Sildosten Deutsch-0 stafrikas, pp. 96, 97, 104. XIX EXOGAMY 141 they appear not to practise totemism.1 In Bukoba, on the other hand, we find exogamous patrilineal clans which are tptemic ;2 whilst the Wahehe in Iringa have totemism but not combined with exogamy, a man being free to marry a woman of his own totem.3 Among the last-mentioned tribe,4 as also among some other tribes of the same region, such as the Wagogo,5 and Sango,6 the marriage of cousins who are the children of two brothers or of two sisters is not allowed ; and the Sango, though they permit the marriage of cross-cousins, prefer to take their wives from families with which they are not related.7 The Masai in Eastern Equatorial Africa consist of five clans with descent in the male line, which are not exogamous ; but these clans are divided into exogamous sub-clans. Moreover, " first cousins and second cousins may not marry, but there is no objection to third cousins marrying if the relationship is no nearer than ol-le-'sotwa (or en-e-sotwa). Thus, a man's son's son's son may not marry the man's brother's son's son's daughter, nor may a man's son's son's son marry the sister's son's son's daughter, but there would be no objection to a man's son's son's son marrying the brother's daughter's daughter's daughter or the sister's daughter's daughter's daughter. Likewise, though a man's son's son may not marry the man's maternal uncle's son's son's daughter, he may marry the maternal uncle's son's daughter's daughter. These unions are always contingent on the two parties not belonging to the same sub-clan. The rules of consanguinity and affinity which regulate marriage also apply to the sexual intercourse of warriors with immature girls before marriage and to the rights of hospitality after marriage." If a man is knowingly guilty of incest, or has sexual intercourse with a daughter of his own sub-clan, he is punished by his relations, who flog him and slaughter some of his cattle. If a man unintentionally 1 Frazer, Totemism and Exogamy, ii. 406. 2 Richter, ' Der Bezirk Bukoba,' in Mittheil. Deutsch. Schutzgeb. xii. 83 sq. 3 Nigmann, Die Wahehe, 42, 43, 60. 4 Ibid. p. 60. 6 Claus, Die Wagogo, p. 58. 9 Heese, ' Sitte und Brauch der Sango,' in Archiv f. Anthropologic, N.S. xii. 134. 7 Ibid. p. 134. 142 THE HISTORY OF HUMAN MARRIAGE chap. commits incest-and it is quite conceivable that a man might not know his fourth or fifth cousin, for instance, should the two live in different districts-he has to present a cow to the girl's relations in order to " kill the relationship." 1 Among the Wataveta, also, the clans, though not themselves exogamous, are subdivided into exogamous sub-clans. Their clans are totemic,2 whereas the Masai, apparently, are not a totemic people.3 The Akamba are composed of at least twenty-five chief clans, and many of these fall into sub-clans.4 The large majority of the clans are known to be totemic, and the others are presumably so.5 Mr. Hobley states that the members of the same sub-clan originally were prohibited from marrying each other, although, curiously enough, " they could marry back into the original stock." This custom, however, he adds, is not rigidly enforced nowadays ; for they say that the numbers of each clan are now becoming so great that the intermarriage referred to is no longer regarded as a serious matter.6 According to Dr. Lindblom, on the other hand, " marriage between individuals of the same clan is strictly forbidden, even if the parties live in different parts of the country and have never heard of each other before. If such a forbidden union between members of the same clan is entered into, it is a very grave crime, and the culprits must submit to a process of purifica- tion, which is carried out by an old man who is specially versed in such matters." 7 The Akitutu, however, constitute an exception to the general rule and marry within the same clan.8 Clan exogamy, whether combined with totemism or not, is found in many other tribes of British East Africa, such as the Mweru,9 the Suk, the Turkana, and some of 1 Hollis, ' Note on the Masai System of Relationship and other matters connected therewith,' in Jour. Roy. Anthr. Inst. xl. 479 sq. 2 Frazer, Totemism and Exogamy, ii. 418 sq. 3 Ibid. ii. 408. 4 Hobley, Ethnology of the A-Kamba and other East African Tribes, p. 64. Lindblom, Akamba, p. 106. 5 Lindblom, op. cit. p. 107. 6 Hobley, op. cit. p. 64 sq. 7 Lindblom, op. cit. p. 113. 8 Hofmann, quoted ibid. p. 113 n. I. 9 Hobley, quoted by Frazer, Totemism and Exogamy, ii. 425. 143 THE HISTORY OF HUMAN MARRIAGE chap. the Bahima of Ankole.1 And among these tribes, with the exception perhaps of the Bateso, Banyoro, and Batoro, a man is not only prohibited from marrying a woman of his own clan, but he must also avoid marrying into his mother's clan.2 The Bateso,3 the Basoga,4 and the Banyoro,5 who prohibit the marriage of all first cousins, allow marriage between second cousins in certain cases, namely, between a man's son's children and his sister's daughter's children. Among the Baganda marriage and sexual intercourse be- tween persons belonging to the same clan used to be punish- able by death,6 and so also the Banyoro formerly in- flicted this penalty on a man who married a woman of his own clan.7' But among the latter tribe, as well as among the Bahima, the exogamous rules did not apply to the royal family, kings or princes being allowed to marry or have sexual intercourse with their own sisters.8 The Basoga are said to abhor incest even in their cattle. If a bull covers his mother-cow or his sister-cow, the bull and the cow are sent by night to a fetish tree and tied to it, and the next morning the chief appropriates them to his own use.9 The Dinka tribes of the White Nile, who are totemistic, are divided into a number of exogamous clans.10 Descent is in the male line; but marriage is prohibited between persons who are connected by blood on the mother's side as well, so far as the genealogies can be traced in tradition.11 The Galla, who dwell to the south of Abyssinia, are divided into two tribes or classes, the Baretuma or Baretu and the Harusi or Arusi, and the men of each tribe have to select 1 Roscoe, Northern Bantu, p. 116. 2 As for the Bahima, see ibid. p. 118. In an earlier essay on this tribe (' Bahima,' in Jour. Roy. Anthr. Inst, xxxvii. 105) Mr. Roscoe says that the rule of exogamy does not apply to the mother's clan, but that it is not usual for a lad to marry into it. 3 Roscoe, Northern Bantu, p. 263. 4 Ibid. p. 209. 8 Ibid. p. .38. 6 Roscoe, Baganda, p. 129. 7 Idem, Northern Bantu, p. 28. 8 See supra, ii. 94. 9 Johnston, Uganda Protectorate, p. 719. 10 Seligman, Report on Totemism and Religion of the Dinka of the White Nile, p. 2 sq. 11 O'Sullivan, ' Dinka Laws and Customs,' in Jour. Roy. Anthr. Inst. xl. 176, 187. XIX EXOGAMY 144 the Kamasia in the Baringo district,1 the Wagiriama,2 and the natives of Kavirondo, both of the Bantu and of the Nilotic stock.3 Among the Akikuyu, who are divided into thirteen clans with descent in the male line, a man may not marry a member of the clan of either his father or his mother, though he may return to that of his grandmother ; and there are said to be certain other restrictions as to marriage between particular clans.4 The marriage of first and second cousins, the children and grandchildren of brothers and sisters, are regarded by them as a very grave sin.5 Among the Nandi " a man may not marry a woman of the same family as himself, though there is no objection to his marrying into his own clan " ; and this rule also applies to warriors having sexual intercourse with immature girls before marriage.6 Incestuous intercourse with a step-mother, step-daughter, cousin, or other near relation is punished severely : the guilty man is flogged by the women, his houses and crops are destroyed, and some of his stock is confiscated.7 Exogamous totem clans with paternal descent are found among the Bagesu of Mount Elgon on the eastern boundary of the Uganda Protectorate,8 the Bakene of Busoga,9 the Basoga,10 the Bateso (whose territory borders on Northern Busoga),11 the Baganda,12 the Banyoro,13 the Batoro,14 and 1 Dundas, ' Notes on the Tribes inhabiting the Baringo District,' in Jour. Roy. Anthr. Inst. xl. 59, 67, 65. 2 Barrett, ' Notes on the Customs and Beliefs of the Wa-Giriama, etc., British East Africa,' in Jour. Roy. Anthr. Inst. xli. 22. 3 Hobley, Eastern Uganda, pp. 13, 29. Johnston, Uganda Protectorate, p. 791. 4 Routledge, With a Prehistoric People, p. 20. 6 Hobley, ' Kikuyu Customs and Beliefs,' in Jour. Roy. Anthr. Inst. xl. 438. 6 Hollis, Nandi, p. 6. 7 Ibid. p. 76. 8 Roscoe, Northern Bantu, pp. 163, 164, 172. Mr. Roscoe says (' Notes on the Bageshu,' in Jour. Roy. Anthr. Inst, xxxix. 182) that his limited knowledge of the language made it impossible for him to get at their totems, though there was every indication of totemism. 9 Idem, Northern Bantu, p. 148. 10 Ibid. p. 207 sq. 11 Ibid. p. 262. 12 Roscoe, Baganda, pp. 82, 128, 133 sqq. 13 Idem, Northern Bantu, p. 27. 14 Maddox, quoted by Frazer, Totemism and Exogamy, ii. 531. XIX EXOGAMY 145 their, wives from the other; 1 and the Southern Galla, at least, have another law which prohibits a man from marrying in his own family, however remote.2 Among the Bogos of Eritrea persons related within the seventh degree may not intermarry, whether the relationship be on the paternal or maternal side.3 The Tshi-speaking tribes of the Gold Coast are divided into exogamous totem clans with maternal descent.4 In old days, if members of the same clan married or had sexual intercourse, they were both either beheaded or sold into slavery. At the present time, if such a thing occurred, the matter would be investigated by the head of the clan or the chief of the town, the parties would be divorced and the man fined, and a sheep would be killed and the male culprit compelled to walk in its blood ; whilst if the act has been committed unwittingly, the guilty parties must kill a sheep and protest their innocence. If a chief had any- thing to do with a woman of the same tribe as himself he would be deposed. On the coast, however/these marriage laws are not strictly enforced, though there is a prejudice against marriage within the clan.5 A man may marry a woman of any clan except his own.6 The Ewhe-speaking tribes of the Slave Coast are likewise divided into exogamous totem clans with descent in the maternal line at least among the common people ; but the exogamous restriction is not now always scrupulously observed by the seaboard tribes.7 1 New, Life, Wanderings, and Labours in Eastern Africa, p. 274. Wakefield, ' Marriage Customs*of the Southern Gallas,' in Folk-Lore, xviii. 323. 2 Wakefield, loc. cit. p. 323. 3 Munzinger, Ueber die Sitten und das Recht der Bogos, p. 58. 4 Connolly, ' Social Life in Fanti-land,' in Jour. Anthr. Inst. xxvi. 132 sq. Harper and others, ' Notes on the Totemism of the Gold Coast,' ibid, xxxvi. 179, 180, 182 sq. Arthur ffoulkes, ' Fanti Family System,' in Jour. African Soc. vii. 394 sqq. 5 Harper and others, loc. cit. pp. 182, 183, 186, 188. According to Mr. Arthur ffoulkes (loc. cit. p. 399), marriages within the clan do sometimes occur but are not countenanced. 6 Harper and others, loc. cit. p. 186. 7 Ellis, Ewe-speaking Peoples of the Slave Coast, p. 207. Among the upper classes of Dahomey descent is traced through the males (ibid. p. 209). VOL II L 146 THE HISTORY OF HUMAN MARRIAGE chap. First cousins who are the children either of two brothers or of two sisters may not marry each other, whereas marriage is allowed between two first cousins who are the children of a brother and a sister respectively.1 Concerning the Yoruba-speaking peoples of the Slave Coast Major Ellis writes :-" Marriage is forbidden in the same blood ; and, as descent is traced on both sides of the house, it is con- sequently forbidden both in the father's and mother's families, as far as relationship can be traced. This, however, is not far, as there is no longer the clan-name. ... As a rule relationship does not seem to be traced further than second- cousins."2 Among the Ibo-speaking peoples of Nigeria a man is always forbidden to marry in his own sept (a body of people claiming descent from a single ancestor), and marriage into the father's or mother's family is never allowed.3 So also over the whole area occupied by the Edo-speaking peoples of Nigeria a man is as a rule prohibited from marrying a woman who belongs either to his father's or his mother's family ; but Mr. Thomas says that it is difficult to ascertain precisely how far the limits of the family extend. At Gwaton in the Edo-country and at Ososo, however, as said above, marriage with a half-sister on the father's side is permitted.4 In some of the tribes in Sierra Leone a man is not allowed to marry his cousin, either german or uterine, no matter how many times removed ; and " speaking generally, all in- dividuals who are the descendants of one woman, irrespective of the father of that woman's children, are within the prohibited degree of consanguinity." 5 Among the Hausa of Northern Nigeria, who have totemism, the men in some clans prefer to marry women who have the same totems as their 1 Zundel, ' Land und Volk der Eweer auf der Sclavenkiiste in Westafrika,' in Zeitschr. d. Gesellsch. f. Erdkunde zu Berlin, xii. 390. 2 Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 188. 3 Thomas, Anthropological Report on Ibo-speaking Peoples of Nigeria, i. 69. 4 Idem, Anthropological Report on the Edo-speaking Peoples op Nigeria, i. 61 sq. 5 Vergette, Certain Marriage Customs of some of the Tribes in the Protectorate of Sierra Leone, p. 5. XIX EXOGAMY 147 mothers, " but usually they are content so long as the women have not the same totems as they themselves have, i.e., they are exogamous." 1 Some of them, however, now marry within the clan, but say that formerly they did not do so.2 Children of sisters or of half-sisters may not marry each other, nor may those of brothers or of half-brothers inter- marry, but the child of a brother or of a half-brother may wed the child of a sister or of a half-sister.3 Of the Bakalai and various other tribes of Western Equa- torial Africa which inhabit the country between two degrees north and two degrees south of the equator and to a distance of about four hundred miles into the interior Du Chaillu states that each of them is divided into exogamous clans, most of which trace descent in the female line.4 Among the Fans, whose village communities are essentially groups of kindred, a man may not marry a woman of his own village ; it happens that such marriages sometimes take place when the village is large and the relationship between the couple is distant, but in such a case the persons contracting the union are looked at askance.5 The marriage of cousins both on the father's and the mother's side seems to be forbidden among them ;6 and the Baya, another tribe in the French Congo, also forbid the marriage of second cousins.7 Among the Bavili, a Bantu people of Loango, a man may not marry a woman whose totem is the same as that of his mother ; nor may he marry the daughters of his father's brothers, although he may marry the daughters of his father's sisters.8 Among the Bantu people of the Lower Congo, who likewise trace descent through the mother, marriage is prohibited 1 Tremearne, Hausa Superstitions and Customs, p. 80. 2 Palmer, quoted by Frazer, Totemism and Exogamy, ii. 606. 3 Ibid. ii. 607. Tremearne, op. cit. p. 79. 4 Du Chaillu, Explorations and Adventures in Equatorial Africa, p. 388. Idem, ' Observations on the People of Western Equatorial Africa,' in Trans. Ethn. Soc. London, N.S. i. 307. Idem, Journey to Ashango-Land, pp. 425, 427. 5 Martrou, ' Les " Eki " des Fang,' in Anthropos, i. 754. 6 Miss Kingsley, Travels in West Africa, p. 321. 7 Poupon, ' £tude ethnographique des Baya de la circonscription du M'Bimou,' in L' Anthropologic, xxvi. 127. 8 Dennett, At the Back of the Black Man's Mind, p. 36. L 2 148 THE HISTORY OF HUMAN MARRIAGE chap. between maternal cousins, and it seems to be the general custom for a man to marry into another clan than his own.1 Mr. Ch. E. Ingham wrote to me from Banza Manteka that the Bakongo hold all unions between near relatives, eitner on the father's or on the mother's side, in utter abomination. Dr. Sims, writing from Stanley Pool, informed me that among the Bateke marriages are forbidden between brothers and sisters of the same mother or father, between first cousins, and between uncle and niece or aunt and nephew. Among the Bushongo, who occupy the valley of the Sankuru River, a tributary to the Congo, " aucun homme ne peut epouser une femme avec laquelle il a quelques liens de parente, de telles unions etant considerees comme incestueuses." 2 Thus, a man is forbidden to marry his cousin, and formerly he might not marry any woman who had his own ikina bari, or totem.3 A large number of tribes inhabiting the region of the Upper Congo and its northern tributaries have been found to be divided into exogamous totem clans.4 Among those investigated by the Duke of Mecklenburg's Expedition, who all trace descent in the paternal line, the only exception to the rule of exogamy is presented by some of the Azande clans, including the royal clan of the reigning dynasty.5 Among the Warega, " 1'empechement au rnariage s'etend aux cousins les plus eloign^s, tant du cote maiernel que dti cote paternel." 6 In Madagascar marriage between children and descendants of two brothers was exceedingly common and considered as desirable, and marriage between children of a brother and of a sister was allowable on the performance of a pre- scribed ceremony supposed to remove any impediment from consanguinity.7 This ceremony, which in some places 1 Weeks, Among the Primitive. Bakongo, p. 141. 2 Torday and Joyce, Les Bushongo, p. 116. 3 Ibid. p. 113. 4 de Calonne Beaufaict, ' ZooHtrie et Totdmisme chez les peu- plades septentrionales du Congo Beige,' in Revue des etudes ethno- graphiques et sociologiques, ii. 193 sqq. 8 Czekanowski, ' Die anthropologisch-ethnographischen Arbeiten der Expedition S.H. des Herzogs Adolf Friedrich zu Mecklenburg,' in Zeitschr. f. Ethnol. xli. 597. e Delhaise, Les Warega, p. 177. 7 Sibree, op. cit. p. 248. Grandidier, Ethnographic de Madagascar, ii. 167. Supra, ii. 71. XIX EXOGAMY 149 seems to have been required also in the case of a marriage between brothers' children,1 consisted in the sacrifice of an ox, a sheep, or a fowl, according to the degree of relationship between the couple and their wealth or poverty, or in the sprinkling of the newly-wedded- pair with cow's dung mixed with boiled rice.2 On the other hand, marriage between children and descendants of two sisters, at least when the sisters had the same mother, was regarded with horror as incest, as a crime against nature, and was not allowable down to the fifth or sixth generation.3 It was generally punished with a fine of two oxen ; whilst incest of the highest degree was punished with death.4 In ancient Rome marriages between persons under the same patria potestas, that is cognati related within the sixth degree (the degree of second cousins), were considered immoral and unlawful-they were nefariae et incestuae nuptiae.5 These prohibitions were gradually relaxed : from the time of the Second Punic War, at least, first cousins were allowed to intermarry,6 and subsequently marriage with a brother's daughter was declared legal.7 But in later times, under the influence of the ascetic ideas prevalent in the Church, the prohibited degrees were again extended. In the Eastern Church marriage was forbidden within the seventh degree according to the Roman method of computing degrees of relationship, which was to count from one of the parties up to a common ancestor and then down to the other party, so that, for example, first cousins were held to be related in the fourth degree and uncle and niece in the third. This rule is still in force in the Eastern Church.8 The Western Church went still farther in her prohibitions. 1 Grandidier, op. cit. ii. 167. 2 Ibid. ii. 149 sq. 3 Ibid. ii. 167. Sibree, op. cit. p. 248 sq. 4 Grandidier, op. cit. ii. 148 sq. 5 Marquardt, op. cit. p. 30 sq. Rossbach, op. cit. p. 422 sqq. 6 Livy, Historian, xlii. 34. For later instances of such marriages see Rossbach, op. cit. p. 431 sq. r Supra, ii. 99. 8 Zhishman, Das Eherecht der Orientalischen Kirche, p. 241 sqq. For the reckoning of degrees see ibid. p. 217 sqq. 150 THE HISTORY OF HUMAN MARRIAGE chap. The forbidden degrees became gradually as many as seven according to the new Western reckoning, or " canonical computation," by which seven degrees were practically equivalent to seven generations ; brother and sister were related in the first degree, first cousins in the second degree, second cousins in the third degree, and similarly beyond.1 The seventh degree seems to have been chosen by rigorous theorists who would have forbidden a marriage between kinsfolk however remote ; for it seems to have been a common rule among the Teutonic peoples that for the pur- poses of inheritance kinship could not be traced beyond the seventh generation, and so to prohibit marriage within seven degrees was to prohibit it among all persons who for any legal purpose could claim blood-relationship with each other.2 The fourth Lateran Council, held in a.d. 1215 under Innocent III., reduced the prohibited degrees from seven to four, that is, marriage was permitted beyond the degree of third cousins ;3 and since then there has been no change.4 The forbidden degrees of the Western Church thus almost coincide with those of the Eastern Church, the fourth degree of canonical computation corresponding to the seventh and eighth degrees of the Roman reckoning.5 But there is this important difference between the legislation of the two Churches, that in the Eastern Church no dis- pensation is held possible from any one of the prohibited degrees,6 whereas in the West dispensation is not only allowed but has since early times been practised on a very large scale. It does not seem, however, that the field of the Levitical prohibitions was entered upon by the Papal dispensing claims till the fifteenth century.7 The Reformers went in principle back to the prohibited degrees of the 1 Freisen, Geschichte des Canonischen Eherechts, p. 393 sqq. Watkins, Holy Matrimony, p. 702 sq. For the canonical compu- tation see Freisen, p. 423 sqq. 2 Pollock and Maitland, History of the English Law before the Time of Edward I., ii. 387 sq. Winroth, op. cit. p. 181. 3 Concilium Lateranense IV., ch. 50 (Labbe-Mansi, Sacrorum Conciliorum collect™, xxii. 1037 sq.). 4 Freisen, op. cit. p. 405. 5 Cf. Zhishman, op. cit. p. 253. * Ibid. p. 713. 7 Watkins, op. cit. p. 704. XIX EXOGAMY 151 Mosaic law.1 Henry VIII. declared in 1540 that nothing, " God's law except, shall trouble or impeach any marriage without the Levitical degrees " ; as the farthest of which was considered that between uncle and niece.2 Among the Southern Slavs marriage is or has been as a rule avoided or prohibited within the bratstvo, or patrilineal clan.3 In High Albania, says Miss Durham, " the men and women descending from a common male ancestor, though very remote, regard one another as brother and sister, and marriage between them is forbidden as incestuous. Though the relationship be such that the Catholic Church permits marriage, it is regarded with such genuine horror that I have heard of but one instance where it was attempted or desired, when against tribal law. Even a native priest told me that a marriage between cousins separated by twelve generations was to him a horrible idea, though the Church permitted it, ' for really they are brothers and sisters.' "4 Besides prohibitions of marriage between blood-relatives there are other prohibitions applying to marriage between relatives by alliance or affinity. The rules relating to the latter kind of marriage vary extremely in different countries. While many uncivilised peoples allow a man to marry his wife's sisters and even give him a prior claim to their hands,5 there are others that either prohibit or disappove of such marriages altogether or-which is very common-only permit marriage with a deceased wife's younger sister.6 1 Scheurl-Sehling, ' Eherecht,' in Herzog-Hauck, Realencyklopddie fur protestantische Theologie und Kirche, v. 210. Winroth, op. cit. p. 186. 2 Stephen, Commentaries on the Laws of England, ii. 386. 3 Schrader, Reallexikon der indogermanischen Altertumskunde, p. 910. Miss Durham, ' Some Montenegrin Manners and Customs,' in Jour. Roy. Anthr. Inst, xxxix. 88. 4 Miss Durham, High Albania, p. 20 sq. See also Idem, 'High Albania and its Customs in 1908,' in Jour. Roy. Anthr. Inst. xl. 458, 6 See infra, on Monogamy and Polygyny. 0 E.g., the Eskimo in the neighbourhood of Iglulik (Lyon, op. cit. P- 353)i the Greenlanders (Dalager, op. cit. p. 68 ; Holm,' Ethnologisk Skizze af Angmagsalikerne,' in Meddelelscr om Gr (inland, x. 96), 152 THE HISTORY OF HUMAN MARRIAGE chap. Among the Thonga of Mozambique a man can marry neither the elder sister of his wife nor the daughters of his wife's sisters j1 among the Negroes of Accra on the Gold Coast2 and the Koryak of Siberia3 he cannot even marry his wife's cousin, at any rate as long as the wife is alive. The Levitical law4 and Islam5 forbid marriage with two sisters simul- taneously, and the Catholic Church forbids marriage with a deceased wife's sister, though the prohibition may be dispensed with.6 In England such marriages were con- demned by the canon law of the English Church, and their illegality was confirmed in 1835 ;7 and, as is well known, it was only after many futile attempts and in face of very strong opposition that an Act legalising marriage with a deceased wife's sister in the United Kingdom was passed in 1907.8 The Eastern Church even prohibits two brothers from marrying two sisters ;9 and a similar prohibition is found in some uncivilised tribes.10 Again, while very many peoples permit a man to marry the Koryak (Jochelson, Koryak, p. 737), the Kacharis of Assam (Endle, Kacharis, p. 29), the Southern Massim of British New Guinea (Seligman, Melanesians of British New Guinea, p. 507), the natives of Bambatana and Tambatamba in Choiseul Island of the Solomon Group (Thumwald, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 17), the natives of the Belep Islands to the north of New Caledonia (Lambert, op. cit. p. 95), the Polynesians of Tikopia (Rivers, History of Melanesian Society, i. 309), the Bantu tribes of South Africa (Theal, op. cit. p. 220 ; Kropf, Das Volk der Xosa-Kaffern im ostlichen Siidafrika, p. 137), the tribes near Fort Johnston in British Central Africa (Stannus, in Jour. Roy. Anthr. Inst. xl. 309), the Bateso in the Uganda Protectorate (Roscoe, Northern Bantu, p. 262), the Akamba of British East Africa (Hobley, Ethnology of A-Kamba, p. 65). Frazer, Folk-Lore in the Old Testament, ii. 264 sq. 1 Junod, Life of a South African Tribe, i. 243. 2 Daniell, ' Ethnography of Akkrah and Adampe,' in Jour. Ethn. Soc. London, iv. 14. 3 Jochelson, Koryak, p. 737. 4 Leviticus, xviii. 18. 5 Koran, iv. 27. 6 Roguin, op. cit. p. 87 sq. r Ibid. p. 88. 8 Earl of Halsbury, Laws of England, xvi. 284. 8 Zhishman, op. cit. p. 298. 10 Sapir, ' Takelma Indians of Southwestern Oregon,' in American Anthropologist, N.S. ix. 267 sq. Batchelor, Ainu and their Folk-Lore, p. 228. Jochelson, Koryak, p. 748. XIX EXOGAMY 153 the widow of his deceased brother, or elder brother, and even regard it as his duty to do so,1 there are others that prohibit all such marriages.2 Some of the Queensland tribes near Brisbane considered it monstrous that a man should marry his brother's widow.1 The Chinese penal code punished such a union with strangulation, although marriage with a deceased wife's sister has always been regarded as particularly honourable in China.4 Marriage with a deceased brother's widow, as well as with a deceased wife's sister, was prohibited by Canon Law,5 and is also prohibited by the laws of many, especially Latin, countries, although dispensation is easily obtained.6 In many uncivilised tribes the eldest son, or all the sons, inherit the father's widows, with the exception of the mother ;7 but in others marriage with a step-mother is looked upon as incestuous.8 Thus, the Herero would consider it highly objectionable if a son who inherits his 1 See infra, on Polyandry. 2 E.g., the Hajongs and Morans of Assam (Endle, op. cit. pp. 87, 89), the Meches of the Bengal Terai, the Muduvars and Udayas of Madras, and the Arayans of Cochin (Gait, op. cit. p. 246), the Koita of British New Guinea (Seligman, op. cit. p. 79), the natives of Tikopia (Rivers, History of Melanesian Society, i. 436) and of Bambatana and Tambatamba in Choiseul Island of the Solomon Group (Thurnwald, op. cit. iii. 17), the Ngarigo tribe of New South Wales (Howitt, Native Tribes of South-East Australia, p. 199), the Masai (Hollis, in Jour. Roy. Anthr. Inst. xl. 481). 3 Howitt, op. cit. p. 237. 4 Medhurst, in Trans. Roy. Asiatic Soc. China Branch, iv. 24 sq. note. 5 Roguin, op. cit. p. 87 sq. 8 Ibid. p. 88. Winroth, op. cit. p. 206. ' See infra, on Polyandry. 8 E.g., the Koryak (Jochelson, Koryak, p. 737), the Subanu (Finley and Churchill, op. cit. p. 40), the natives of Bambatana and Tambatamba in Choiseul (Thurnwald, op. cit. iii. 17), the Bayaka (Torday and Joyce, ' Notes on the Ethnography of the Ba-Yaka,' in Jour. Anthr. Inst, xxxvi. 45), the Bangongo (Jidem, Les Bushongo, p. 116). In his description of the Bantu tribes south of the Zambesi Mr. McCall Theal states (op. cit. p. 234 ; see also Shooter, op. cit. p. 86) that in some tribes along the coast a son did not marry his father's widows but nevertheless claimed the children borne by them in unions with his paternal uncles, among whom they were dis- tributed, " as they were held to be the children of his deceased father." 154 THE HISTORY OF HUMAN MARRIAGE chap. father's property also were to inherit his widows;1 and the Winamwanga custom of a son taking over his father's widows, with the natural exception of his own mother, fills the Awemba with disgust.2 According to the Levitical law,3 the Koran,4, the Roman Law,5 Canon Law, and the law-books of all Christian countries,6 marriage is prohibited with a step-mother and a step-daughter, as well as with a mother-in-law and a daughter-in-law-according to the Koran, however, with a step-daughter only if she is under the guardianship of her mother's husband. A very frequent bar to marriage is the artificial relationship created by adoption or fosterage. Thus among the Luiseno Indians, when a child is adopted by a family, "it is looked upon as one of their own children, and its marriage with one of its foster relations is regarded as incestuous.''7 Concerning the Hurons and the Iroquois, Charlevoix wrote that there " must be no manner of relation between the parties to be married, and even adoption is comprehended in this law.''8 Egede says that among the Greenlanders it would be reckoned uncouth and blamable if a lad and a girl who had served and been educated in one family desired to be married to one another.9 The Bontoc Igorot prohibit a man from marrying his adopted daughter or sister or even his first cousin by adoption.10 The Kayans of Borneo very strongly reprobate the connection of a man with an adopted daughter, although marriage with a sister by adoption is allowed.11 In New Caledonia, on the other hand, 1 Dannert, op. cit. p. 39. 2 Gouldsbury and Sheane, op. cit. p. 173. 3 Leviticus, xviii. 8. 4 Koran, iv. 26 sq. 5 Institutiones, i. 10. 6 sq. 0 Roguin, op. cit. p. 85 sq. Winroth, op. cit. p. 205. As a curious exception to the rule which prohibits marriage with a step- mother may be mentioned the custom, apparently in vogue among the Anglo-Saxons, that a new king married his step-mother (Roeder, Die Familie bei den Angelsachsen, p. 40). 7 Sparkman, in University of California Publications in American Archaeology and Ethnology, viii. 214. 8 Charlevoix, Voyage to North-America, ii. 36. B Egede, op. cit. p. 141. See also Cranz, op. cit. i. 147. 10 Jenks, Bontoc Igorot, p. 68. 11 Hose and McDougall, op, cit. ii. 197. XIX EXOGAMY 155 such a marriage is, as it seems, prohibited / and in the Belep Islands adopted children are subject to the same exogamous rules as actual children.2 In Samoa marriage with an adopted son or daughter or sister is forbidden.3 Speaking of the marriage rules of the Adjahdurah tribe of Yorke's Peninsula in South Australia, Mr. Sutton states that " foster children were treated as their own."4 In Madagascar an adoptive father or mother could not marry his or her adopted daughter or son.5 In his report on Ibo-speaking tribes in Nigeria Mr.'Thomas writes that at Enugu he was told " that a foster child could not marry an actual child of a foster parent."6 Among the Fanti of the Gold Coast " an adopted son or daughter falls within the same rules both in his adop- tive and natural families, and the same rule applies to their issue."7 The Koran prohibits a man from marrying his foster- mother and his foster-sister.8 In Dardistan it would even be impossible for a man to marry the widow of his foster-son.9 In ancient Greece an adoptive father was not allowed to marry his adopted daughter, but his son might marry her.10 According to Roman law, marriage with an adopted daughter or granddaughter was prohibited even after the adoption had been dissolved, and marriage with a sister or brother by adoption was prohibited as long as the adoption subsisted.11 These rules passed into Canon Law, but the Church never laid much weight upon adoption as an obstacle to marriage-- an obstacle not known to the Mosaic law,-and it has even been questioned whether adoption in its present form should be regarded by the Roman Catholic Church as a bar to 1 Brainne, La Nouvelle Caledonie, p. 240. 2 Lambert, op. cit. p. 94. 3 Schultz, in Jour. Polynesian Soc. xx. 50. Kramer, op. cit. p. 58 n. 3. 4 Sutton, in Proceed. Roy. Geograph. Soc. Australasia : South Australian Branch, ii. 17. 5 Grandidier, op. cit. ii. 149. 6 Thomas, Anthropological Report on Ibu-speaking Peoples of Nigeria, i. 73. 7 Sarbah, op. cit. p. 46. 8 Koran, iv. 27. 9 Biddulph, Tribes of the Hindoo Koosh, p. 83. 10 Beauchet, op. cit. i. 176. 11 Institutiones, i.10.1 sq. 156 THE HISTORY OF HUMAN MARRIAGE chap. matrimony.1 In many European countries the law pro- hibits marriage between an adoptive parent and child, and in Latin countries also between the child of the former and the adopted child ; whilst in France, Belgium, and Italy even two adoptive children are forbidden by law to intermarry.2 According to the customary law of the Southern Slavs, among whom entire sub-families are en- grafted on the house communities and unrelated individuals are taken into the sub-families, " the adopted member of a family or house community is assimilated to the naturally born kinsman for all purposes indiscriminately."3 Christianity introduced a new obstacle to marriage by establishing the so-called cogndtio spiritualis, or " spiritual relationship." The Emperor Justinian passed a law for- bidding a man to marry a woman for whom he had stood as godfather in baptism, the tie of the godfather and godchild being so analogous to that of father and child as to make such a marriage appear improper ;4 and to this law the Church added various other prohibitions on account of spiritual relationship, for instance, against marriage between the minister of the sacrament and the person baptised and that person's parents, between a godfather and a sister of the godchild, between two sponsors, and between a sponsor and the child of another sponsor born after the act of baptism. Similar prohibitions arose from relationships created by confirmation.5 According to Russian legal practice spiritual relationship is still to some extent a bar to intermarriage.6 In Montenegro " the godchild ranks as own child and is therefore not intermarriageable with ■ the godfather's children."7 In some uncivilised tribes there are certain prohibitions . 1 Winroth, op. cit. pp. 182, 209. 2 Ibid. p. 209. 3 Maine, Dissertations on Early Law and Custom, p. 256. 4 Codex Justinianus, v. 4. 26. 5 For the prohibitions on account of spiritual relationship see Freisen, op. cit. p. 508 sqq.; Zhishman, op. cit. p. 265 sqq.; Watkins, op. cit. p. 700 sqq.; Boudinhon, ' Impediments, Canonical,' in Catholic Encyclopedia, vii. 697; Winroth, op. cit. p. 183. 6 Klibanski, Handbuch des gesamten russischen Zivilrechts, i. 7. 7 Miss Durham, in Jour. Roy. Anthr. Inst. xxxix. 91. ■ XIX EXOGAMY 157 which may be mentioned in the present connection. Among the Southern Massim in British New Guinea the members of a so-called eriam fellowship, who have marital rights over the wives of their fellow eriam, do not marry or even sleep with each other's sisters.1 Among the Western Islanders of Torres Straits, according to Dr. Haddon, marriage was forbidden, " with a remarkable delicacy of feeling, to the sister of a man's particular friend."2. Among the Batamba of Busoga blood-brotherhood constitutes an impediment to marriage between members of the two families.3 Some tribes of Sierra Leone strictly forbid a man to have connec- tion with a woman at whose breast he has been.4 The Masai 5 and the Merina of Madagascar 6 prohibit marriage between a milk-brother and a milk-sister ; and the Boloki on the Upper Congo at any rate regard such a union as very irregular.7 Among the Bohindu, a tribe in the Belgian Congo belonging to the Basongo Meno, children who are born in the same village on the same day are said to be ishoke to one another and regarded as twins, and consequently they cannot intermarry.8 Among various peoples marriage between persons belong- ing to the same village or other local group is said to be prohibited or as a rule avoided.9 In many of these cases 1 Seligman, Melanesians of British New Guinea, p. 470. 2 Haddon, ' Ethnology of the Western Tribe of Torres Straits,' in Jour. Anthr. Tnst. xix. 315. 3 Condon, ' Contribution to the Ethnography of the Basoga- Batamba, Uganda Protectorate,' in Anthropos, vi. 379. 4 Vergette, op. cit. p. 10. 5 Merker, Die Masai, p. 47. 6 Grandidier, op. cit. ii. 149. 7 Weeks, Among Congo Cannibals, p. 132. 8 Torday and Joyce, Les Bushongo, p. 271. 9 Nordenskiold, op. cit. p. 82 (Choroti of the Gran Chaco). Pie- drahita, quoted by Tylor, 'On a Method of investigating the Deve- lopment of Institutions,' in Jour. Anthr. Inst, xviii. 268 (Panches of Bogota). Farrand, ' Notes on the Alsea Indians of Oregon,' in American Anthropologist, N.S. iii. 242. Teit, in Publications of the Jesup North Pacific Expedition, i. 325 (Lower Thompson Indians of British Columbia). Golder, ' Songs and Stories of the Aleuts,' in Jour. American Folk-Lore, xx. 134. Krasheninnikoff, op. cit. p. 212 (Kamchadal). Jochelson, Koryak, p. 750. Prain, ' Angami 158 THE HISTORY OF HUMAN MARRIAGE chap. we are not told whether all the members of the local group are related by blood or not, but in other cases they are certainly not so. The Yameos on the Amazon will not suffer an intermarriage between members of the same community, " as being friends in blood, though no real affinity between them can be proved."1 The Australian tribe, as Dr. Hjwitt points out, is organised in two ways. On the one hand, it is divided socially into phratries and clans; and on the other hand, it is divided geographically into hordes. The two organisations are co-existent, but the divisions of the one do not correspond with those of the other For while all the people who belong to any given local group are found in one locality alone, those who belong to any given social group are very frequently to be found distributed among many, if not among all, of the local groups. Now, in many tribes local proximity by birth is quite an insuperable obstacle to marriage, a man being absolutely forbidden to marry, or to have sexual intercourse with, a woman of the same horde or sub-horde. " However eligible she may be in other respects, the fact that both parties belong to the same locality is held by certain tribes, the Kurnai for example, to make them ' too near each other.' " It is chiefly in tribes where the clan system has been weakened, or has become almost extinct, that the local organisation has assumed such overwhelming preponderance, but even in some of the tribes that have a vigorous clan Nagas,' in Revue coloniale internationale, v. 492. Metz, Tribes inhabiting the Neilgherry Hills, p. 131 (Kotas). Hose and Mc- Dougall, op. cit. ii. 183 (Punans of Borneo). Riedel, ' Galela und Tobeloresen,' in Zeitschr. f. Ethnol. xvii. 77. Passarge, ' Die Busch- manner der Kalahari,' in Mittheil. Deutsch. Schutzgeb. xviii. 265 J Kaufmann, ' Die Auin,' ibid, xxiii. 156. Mr. Eyles, in a letter (Zulus). Hildebrandt, in Zeztschr. f. Ethnol. x. 401 (Akamba). Felkin, ' Notes on the Madi or Moru Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xii. 320. Parkinson, * Note on the Asaba People (Ibos),' in Jour. Anthr. Inst, xxxvi. 316. Johnston, George Grenfell and the Congo, ii. 676 (certain tribes along the Upper Wele and Bomokandi and Upper Rubi rivets). Delhaise, Les Warega, p. 175. For natives of the Malay Peninsula, see supra, ii. 119 sq., and infra, ii. 209. 1 v. Martius, op. cit. i. 117. XIX EXOGAMY 159 system local restraints upon marriage are strictly enforced.1 Thus in the Wotjobaluk tribe of North-Western Victoria which traced descent in the female line, there was besides the class restriction another rule prohibiting a man from marrying a woman of the same place as his mother, because his " flesh " was considered to be too near to that of the women who lived there. Hence he had to go for a wife to some place where there was no " flesh " near to his ; and the same applied to the woman.2 So also among the Gournditch-mara " wives were obtained from distant places as not being so ' close in flesh ' as those in or near to the same localities."3 Indeed, in this and various other tribes of South-Western Victoria a man was forbidden to marry into his father's tribe, into his mother's tribe, into his grandmother's (evidently his mother's mother's)4 tribe, into an adjoining tribe, and into any tribe that spoke his own dialect.5 In the Murring tribes on the south coast of New South Wales it was said that a man not only must take a woman of a different name to his own, but besides " must go for a wife to a place as far as possible from his own place."6 Of the aborigines in the neighbourhood of King George's Sound at the extreme south-west point of Australia, who had a very stringent rule of class exogamy, we are also told that they considered it " best to procure a wife from the greatest distance possible."7 In India, according to Mr. Gait, " apart from the restric- tions based on the exogamous group and the prohibited degrees of relationship, there is often a rule that a man should not marry a girl of his own village. Thus the Rajputs and 1 Howitt, ' Australian Group Relations,' in Smithsonian Report, 1883, pp. 8oo, 810, 819 sq. Idem, Native Tribes of South-East Australia, pp. 43, 134, 168, 269, 271, 282 sq. Cf. Mathew, in Jour. Roy. Soc. N.S. Wales, xxiii. 399. 2 Howitt, Native Tribes of South-East Australia, p. 241. 3 Ibid. p. 249. 4 Cf. ibid. p. 250 sq. 5 Dawson, Australian Aborigines, p. 26 sq. 6 Howitt, Native Tribes of South-East Australia, p. 262. 7 Nind, ' Description of the Natives of King George's Sound (Swan River Colony) and adjoining Country,' in Jour. Roy. Geograph. Soc. i. 37, 38, 44. 160 THE HISTORY OF HUMAN MARRIAGE chap. Lewa Kunbis of Baroda regard all the caste people living in the same village as related to each other, and marriages must therefore be arranged with persons living elsewhere. A similar rule obtains amongst the Mundas and other tribes of Chota Nagpur. It is also observed in the eastern Punjab and the Himalayan area of the United Provinces, especially among communities that have no exogamous system based on the gotra ; and Mr. Marten tells us . . . that it is observed also in parts of the Central Provinces and Berar."1 In his work on the tribes and castes of the Central Provinces of India Mr. Russell states that a feeling of kinship subsists between Hindus living in the same village, even though they may belong to different castes and clans. " It is commonly found that all the households of a village believe themselves in a manner related. A man will address all the men of the generation above his own as uncle, though they may be of different castes, and the children of the generation below his own as niece and nephew. When a girl is married, all the old men of the village call her husband ' son-in-law.' This extends even to the impure castes who cannot be touched. Yet owing to the fact that they live together they are considered by fiction to be related."2 The Oraons of Chota Nagpur " have an aversion to marriages between a young man and a girl of the same village ; and such a marriage is generally believed to bode ill for one or both of the married pair."3 Mr. Crooke mentions various instances of local exogamy in the North-Western Provinces, and speaks of the fiction by which a bride is supposed to be brought from a distance.4 Professor Kovalewsky states that in some parts of Russia the bride is always taken from another village than the bridegroom's ; and even in provinces in which no similar custom is known to exist, " the bride- 1 Gait, op. cit. p. 252. 2 Russell, Tribes and Castes of the Central Provinces of India, i- 155- 3 Sarat Chandra Roy, Ordons of Chota Nagpur, p. 396. 4 Crooke, Tribes and Castes of the North-Western Provinces and Oudh, i. p. ccii. ; iii. 139 ; iv. 42, 279. xix EXOGAMY 161 groom is constantly spoken of as a foreigner (' choujoy,' * choujaninin and his friends and attendants are repre- sented as coming with him from a distant country, in order to take away the future spouse."1 In Bulgarian songs brides are invariably said to be brought from other villages.2 1 Kovalewsky, ' Marriage'among the Early Slavs,' in Folk-Lore, *• 475- 2 Piprek, Slawische Brautwerbungs- und Hochzeitsgebrauche, p. 139. VOL. II CHAPTER XX EXOGAMY (Concluded} Many attempts have been made to account for the rules of exogamy. The prohibitions of marriage between near relatives have been ascribed to such motives as fear lest relationship should become too involved or fear lest affection should be concentrated within too narrow a circle. Or their object is said to have been to prevent marriage from taking place too early or to prevent people from marrying each other in order that property might be kept in the family. Or such marriages are said to be prohibited because they outrage " natural modesty " or because God has forbidden them.1 Explanations of this sort need not be considered. I shall only discuss some more recent theories which take into account not only the prohibition of marriage between members of the same family but exogamous rules of a more comprehensive character, and shall then state my own views as to the origin of all these rules. McLennan, who first coined the term " exogamy " for the rule forbidding marriage within the " tribe or group of kindred," tried to show that exogamy has arisen from female infanticide, which he assumed to be " common among savages everywhere." He argued that to tribes surrounded by enemies and, unaided by art, contending with the difficulties of subsistence, sons were a source of strength, both for defence and in the quest for food, whereas 1 See Huth, Marriage of Near Kin considered with respect to the Laws of Nations, p. 24. ch. xx EXOGAMY 163 daughters were a source of weakness. Hence the cruel cus- tom which left the primitive human hordes with very few young women of their own, thus seriously disturbing the balance of the sexes within the hordes and forcing them to prey upon one another for wives. Usage, induced by necessity, would then in time establish a prejudice among the tribes observing it against marrying women of their own stock.1 The name of " wife " would become synony- mous with a subject and enslaved woman in the power of her captor or captors, and the name of marriage would be applied to a man's relation to such a woman as possessor of her. And since a subject or enslaved wife would, in the circumstances of the time, be attainable only through capture, marriage would be possible only through capture, and marriage with a woman of the same stock would be a crime and a sin.2 This theory is untenable for various reasons. First, McLennan grossly exaggerated the prevalence of female infanticide in the savage world. Among a large number of uncivilised tribes it is said to be unheard of or almost so,3 and to these belong peoples of so low a type as the Andaman 1 McLennan, Studies in Ancient History, pp. 50, 75, 76, 160. 2 Idem, Studies in Ancient History. Second Series comprising an Inquiry into the Origin of Exogamy, p. 64 sq. 3 See, e.g., Hooper, Ten Months among the Tents of the Tuski, p. 201 ; Seemann, Narrative of the Voyage of Herald, ii. 66 (Western Eskimo) ; Mackenzie, Voyages from Montreal to the Frozen and Pacific Oceans, p. xcviii. (Cree) ; Sproat, Scenes and Studies of Savage Life, p. 94 (Nootka) ; Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethn. iii. 369 ; Kirke, Twenty-five Years in British Guiana, p. 160 ; Turner, Samoa, p. 79; West, Ten Years in South-Central Polynesia, p. 270 (Tongans) ; Williams, Narrative of Missionary Enterprises in the South Sea Islands, p. 558 (natives of the Mitchell's and Hervey Groups) ; Chalmers, Pioneering in New Guinea, p. 163, and Bink, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. xi. 392 (some natives of New Guinea) ; Hodgson, Miscellaneous Essays, i. 123 (B6do and Dhim&ls) ; Baumann, Durch Massailand zur Nilquelle, p. 161 (Masai). The peoples of the African continent are not generally addicted to infanticide, except in some particular circum- stances. Among many other uncivilised peoples infanticide is said to be looked upon with abhorrence or punished with death (Westermarck, Origin and Development of the Moral Ideas, i. 403 sq.). 164 THE HISTORY OF HUMAN MARRIAGE chap. Islanders,1 the Botocudos,2 and certain Californian tribes.3 The Veddas of Ceylon have never been known to practise it.4 Among the Yahgans of Tierra del Fuego, as Mr. Bridges informed me, it occurred only occasionally, and then it was almost always the deed of the mother, who acted from " jealousy, or hatred of her husband, or because of desertion and wretchedness."5 Mr. Fison, who lived for a long time among uncivilised races, thought it would be found that female infanticide is far less common among the lower savages than it is among the more advanced tribes.® Con- sidering further that the custom of infanticide, being opposed to the instinct of parental love, presupposes a certain amount of reasoning or forethought, it seems probable that where it occurs it is not a survival of earliest savagery but has grown up under specific conditions in later stages of development.7 It is, for instance, very generally asserted that certain Indians of California never committed infanticide before the arrival of the whites ;8 and Ellis thinks there is every reason to suppose that this custom was practised less extensively by the Polynesians during the early periods of their history than it was afterwards.9 In many tribes it is usual to kill an infant only in certain particular circum- stances-if it is a bastard, if its mother dies, if it is deformed or diseased, if there is anything unusual or uncanny about it, 1 Man, * On the Aboriginal Inhabitants of the Andaman Islands,' in Jour. Anthr. Inst. xii. 329. 2 Wied-Neuwied, Reise nach Brasilien, ii. 39. Keane, * Boto- cudos,' in Jour. Anthr. Inst. xiii. 206. 3 Powers, Tribes oj California, pp. 192, 271, 382. 4 Sarasin, Ergebnisse naturwissenschaftlicher Forschungen auf Ceylon, iii. 469, 539. B Bridges, in a letter dated Downeast, Tierra del Fuego, August 28th, 1888. Cf. Idem, ' Manners and Customs of the Firelanders,' in A Voice for South America, xiii. 181 ; Hyades, ' Ethnographic des Fu6giens,' in Bull. Soc. d'Anthr. Paris, ser. iii. vol. x. 331. 8 Fison, in Fison and Howitt, Kamilaroi and Kumai, p. 134 sqq. Cf. Farrer, Primitive Manners and Customs, p. 224 ; Sutherland, Origin and Growth of the Moral Instinct, i. 114 sq. 7 Cf. Darwin, Descent of Man, ii. 400 sq. 8 Powers, op. cit. p. 207. Cf. ibid. p. 183. 9 Ellis, Polynesian Researches, i. 249. XX EXOGAMY 165 if it for some reason or other is regarded as an unlucky child, or, very frequently, in the case of twins. But in such circumstances boys are killed as well as girls ; and the same is often the case among peoples who practise infanticide on a larger scale.1 In many Australian tribes, for instance, no distinction is said to be made between the sexes in this respect.8 The chief motives for infanticide3 are, in fact, such as to make us doubt whether even among savages who do not practise marriage by purchase and treat their girls as a marketable commodity girls are as a rule killed more frequently than boys.4 Moreover, McLennan exaggerated not only the prevalence of female infanticide, but also the prevalence of marriage by capture. This method of obtaining a wife has been found in various parts of the world, chiefly as an incident of war or as a means of procuring a mate when it is difficult or inconvenient to get one in a peaceful manner; but among no people is marriage by capture known to be or to have been the usual or normal mode of contracting a marriage. Nor is there any reason to suppose that in earlier times a man, in order to procure a wife, was in ordinary circum- stances compelled to take her by force from her relatives ; for savages do not usually live at odds with all their neigh- bours, and among many of them wars are quite exceptional if not unknown.5 But even if we assume that there was 1 Westermarck, op. cit. i. 394 sqq. 8 Howitt, Native Tribes of South-East Australia, p. 749 (Wotjoba- luk). Eyre, Journals of Expeditions of Discovery into Central Australia, ii. 324. Spencer and Gillen, Northern Tribes of Central Australia, p. 608 :-" In all of the tribes infanticide is practised. There is no difference made in respect of either sex." On the other hand, Curr {The Australian Race, i. 70) supposed that the Australian woman, as a rule, reared only one girl and two boys, the rest of her children being destroyed ; and West garth says in his account of the settlement of Port Phillip in New South Wales {Australia Felix, p. 59) that " the female infant appears to be much more frequently despatched than the male." We must not suppose that the custom of infanticide is uniform in all the Australian tribes. 3 See Westermarck, op. cit. i. 398 sqq. 4 Cf. Frazer, Totemism and Exogamy, iv. 80 sq. 5 See infra, on Marriage by Capture. 166 THE HISTORY OF HUMAN MARRIAGE chap. a time when each man did his best to capture a woman from another community, it is obvious that where fepiale in- fanticide was universally practised many men could never have succeeded in their endeavours for the simple reason that there was not a sufficient number of women to be had. Herbert Spencer justly asks, " If each tribe had fewer women than men, how could the tribes get wived by taking one another's women ? "1 And why should many men remain celibates when there were unmarried women in their own group ? That in tribes which practised female infanticide the men tried to make up the deficiency of women by capturing wives from other communities is conceivable enough, but it is hard to see why marriage with women of their own community should on this account have been prohibited, sometimes even on pain of death. Why should the men have refrained from marrying those women of their own community who were not killed ? Why should they have made these beings, whom they considered so useless, even more useless than they naturally were, by preventing them from becoming mothers of sons who would have increased the strength of the community ? And why should they have compelled those poor creatures either to remain unmarried or to fall a prey to a hostile captor ? Thus, even if McLennan's premises were right, which they are not, the conclusion which he draws from them would nevertheless be wrong. Like McLennan, Herbert Spencer assumes that primitive groups of men are habitually hostile, and on this postulate he bases his theory on the origin of exogamy. In all times and places victory is followed by pillage ; whatever portable things of worth the conquerors find they take. And of course they take women as they take other booty, because women are prized as wives, as concubines, or as drudges. But a captured woman, besides her intrinsic value, has an extrinsic value ; " like a native wife she serves as a slave, but unlike a native wife, she serves also as a trophy." Hence members of the tribe thus married to foreign women are held to be more honourably married than those married 1 Spencer, Principles of Sociology, i. 618. XX EXOGAMY 167 to native women. If the tribe, becoming successful in war, robs adjacent tribes of their women more frequently, there will then grow up the idea that the now considerable class having foreign wives form the honourable class, and non- possession of a foreign wife will come to be regarded as a proof of cowardice. An increasing ambition to get foreign wives will therefore arise ; and, as the number of those who are without them decreases, the brand of disgrace attaching to them will grow more decided ; until, in the most warlike tribes, it becomes an imperative requirement that a wife shall be obtained from another tribe-if not in open war, then by private abduction.1 This theory is open to objections very similar to those already brought against McLennan's theory. We find there the same belief in marriage by capture and the same assumption that the custom of robbing foreign tribes of their women leads to the prohibition of marrying native women. But whilst according to McLennan the custom of capture was due to a scarcity of women, it was according to Spencer due to the vanity of the men. The conclusion drawn from the assumed prevalence of this custom is equally unsatisfactory in either case. For even if every man had been able to procure a foreign wife, he might have married native women besides, without being disgraced by doing so ; on the conti ary, a plurality of wives is for savage men a source of wealth and reputation. Why should the women of a tribe which was constantly successful in war have been doomed to perpetual celibacy unless they were fortunate enough to fall into the hands of some hostile suitor ? And this would seldom happen if the adjacent tribes were habi- tually worsted in war. For in such tribes, according to Spencer, " marrying within the tribe will not only be habitual, but there will arise a prejudice, and eventually a law, against taking wives from other tribes."2 Spencer, like' McLennan, regards exogamy and endogamy as mutually excluding each other, and if his theory of the origin of exogamy were correct we should always expect to find exogamous and endogamous tribes as each other's neigh- 1 Ibid. i. 619 sqq. 8 Ibid. i. 627 s^. 168 THE HISTORY OF HUMAN MARRIAGE chap. bouts. But, as we have seen, clan exogamy prevails in large groups of tribes inhabiting the same area. How could all the exogamous clans within such tribes have been so constantly successful in war that the habit of marrying conquered women might have led to the prohibition of marrying native ones ? Where there is a vanquisher there must also be a vanquished. And if Spencer's theory also were to explain the origin of the prohibition of marriage between near relatives, as it has been supposed to do,1 we should have to infer that all peoples on earth have formerly been habitually successful in capturing foreign women. Lord Avebury, also, suggests that exogamy has originated in the practice of capturing women from foreign tribes, but he bases this suggestion on his theory of communal marriage. When all the men of. a tribe were the husbands of all the women, no one could appropriate one of the women to himself without impinging on the general rights of the tribe. But the women taken in war from a foreign tribe were in a different position; the tribe, as a tribe, had no right to these women, and they would become wives in our sense of the term. Marriage by capture would thus give rise both to individual marriage and to exogamy.2 This explanation of exogamy was accepted by Wilken,3 who also considered it certain that prohibitions of close intermarriage have everywhere originated in group exogamy.4 1 Mr. Huth, in the first edition of his work The Marriage of Near Kin (p. 157), suggests that marriage between parents and children is considered incestuous because marriage between old men and young women in general is considered so ; but in the second edition he seems to have given up this unfortunate hypothesis. For he says there (p. 18) that " the prohibition of marriage with those who were regarded as near of kin was derived from the same causes which made exogamy imperative," that is, the causes suggested by Spencer. * Avebury, Origin of Civilisation, p. 114. Idem, Marriage, Totemism, and Religion, p. 67 sqq. 3 Wilken, ' Over de primitieve vormen van het huwelijk en den oorsprong van het gezin,' in De Indische Gids, 1880, vol. ii. 612. A similar view has been expressed by Mr. Corin, Mating, Marriage, and the Status of Woman, p. 53. 4 Wilken, loc. cit. pp. 618, 619, 62^. XX EXOGAMY 169 Lord Avebury's theory is based upon two assumptions which I have already rejected as illegitimate, namely, that -man originally lived in a state where individual marriage was unknown, and that capture was once the general method of procuring a wife. A third assumption is that the captured woman became the wife of the captor, although he could claim none of the native women as his individual wife. But it may be asked: why should women taken in war have become the men's personal wives if the women of the tribe were not so ? McLennan made the remark that as war- captives are usually obtained by group-acts, or quasi group-acts, capture would be recognised as a regular mode of adding women to the group, subject to the customary rights of its male members, and that every man in the group would claim the communal right to women taken by others.1 As a matter of fact we find that, even where individual marriage exists, captured women may be regarded as common property. Parkinson states that the Moanus of the Admiralty Islands keep public women, who as a rule are women captured in war;2 and among the Brazilian Karaya unmarried men have recourse to captured Kayapo women, who at least among the Shambioa are found in every village and are regarded as common property.3 Again, among the Gournditch-mara, a small tribe in South-Western Victoria, according to the Rev. J. H. Stable, the man who captured a woman in war never kept her himself, but was compelled to give her to some one else.4 But even if Lord Avebury were right in all his assumptions, he would still have to explain how it happened that the men who obtained foreign wives lost their communal rights to the native women and sexual intercourse with them came to be prohibited, often on pain of death. Under the system of exogamy, as we have seen, not only marriage but also less regular connections between members of the same exogamous group are generally regarded as incestuous. 1 McLennan, Studies in Ancient History, p. 345. 2 Parkinson, Dreissig Jahre in der Siidsee, p. 396. 3 Ehrenreich, Beitrdge zur Volkerkunde Brasiliens, p. 28. ♦ Fison and Howitt, op. cit. p. 276. 170 THE HISTORY OF HUMAN MARRIAGE chap. Other writers have suggested that prohibitions of the marriage of near kin have arisen from observation of the injurious results of such unions. This was the opinion of L. H. Morgan,1 who also maintained that clan exogamy was instituted to prevent the marriage or cohabitation of blood- relations, especially brothers with sisters, which, according to him, had been common in early times.2 That unions between the nearest relatives are, as a rule, followed by injurious consequences is extremely probable, as we shall soon see ; but another question is whether the harmful consequences have been observed by the savages who have prohibited such unions. Sir Henry Maine thought that the men who discovered the use of fire and selected the wild forms of certain animals for domestication and of vegetables for cultivation might also have been able to find out that children of unsound constitution were born of nearly related parents.3 But other writers who have considered the subject are of a different opinion, especially because such knowledge could be gained only by lengthened observation beyond the capacity of the savage mind.4 Yet we are told of uncivilised peoples who ascribe evil results to close intermarriage. In a Greenland tale the father of Kakamak, finding that all his grandchildren have died before reaching the age of puberty, suggests to his son-in-law, " Perhaps we are too near akin."5 Concerning the Kenai of Alaska, Richardson states :-" It was the custom that the men of one stock should choose their wives from another, and the offspring belonged to the race of the mother. This custom has fallen into disuse, and marriages in the same tribe occur; but the old people say that mor- 1 Morgan, Ancient Society, p. 424. 2 Ibid. pp. 58, 425, 426, 502 sq. Morgan, Systems of Consan- guinity and Affinity of the Human Family, pp. 484, 485, 487 sqq. 3 Maine, Dissertations on Early Laiv and Custom, p. 228. 4 Darwin, Animals and Plants under Domestication, ii. 124. Peschel, Races of Man, p. 224. Lubbock (Lord Avebury), ' Customs of Marriage and Systems of Relationship among the Australians,' in Jour. Anthr. Inst. xiv. 300. Thomas, Kinship Organisations and Group Marriage in Australia, p. 70. Frazer, Totemism and Exogamy, iv. 154 sq. 6 Rink, Tales and Traditions of the Eskimo, p. 390 sq. XX EXOGAMY 171 tality among the Kenai has arisen from the neglect of the ancient usage."1 Mr. Dawson writes of some tribes of South-Western Victoria that " since the advent of the Europeans among them, the aborigines have occasionally disregarded their admirable marriage laws, and to this disregard they attribute the greater weakness and un- healthiness of their children."2 The Dieri of the Lake Eyre basin, according to Mr. Gason, have a tradition that after the creation fathers, mothers, sisters, brothers, and others of the closest kin intermarried promiscuously, until the bad effects of these marriages became manifest. A council of the chiefs was then assembled to consider in what way the evil might be averted, and the result of their deliberations was a petition to the Muramura, or Good Spirit. In answer to this he ordered that the tribe should be divided into branches, and distinguished one from the other by different names, after objects animate and inanimate, such as dogs, mice, emu, rain, and so forth, and that the members of any such branch should be forbidden to marry other members of the same branch.3 The Achewa, a Nyanja-speaking tribe in the Nyasaland Protectorate, have a very similar legend about the origin of the exogamy of their totemic clans. Formerly, they say, all their people bore the name of Piri, till their chieftainess Nyangu called them all together and, in order to prevent the evils of close intermarrying, gave each family a new name, which was to descend to the children and children's children.4 The Fanti of the Gold Coast, again, attribute the institution of their exogamous totem clans to a wise seer of old, and they are said to consider the practice of exogamy " of the greatest benefit for the improvement of the species " ;5 1 Richardson, Arctic Searching Expedition, i. 406. 2 Dawson, Australian Aborigines, p. 28. 3 Gason,' Manners and Customs of the Dieyerie Tribe of Australian Aborigines,' in Woods, Native Tribes of South Australia, p. 260 sq. See also Purcell, * Rites and Customs of Australian Aborigines,' in Verhandl. Berliner Gesellsch. Anthrop. 1893, p. 287. 4 Rattray, Some Folk-Lore Stories and Songs in Chinyanja, p. 175 sq. 6 Connolly, ' Social Life in Fanti-land,' in Jour. Anthr. Inst. xxvi. 132. THE HISTORY OF HUMAN MARRIAGE chap. 172 whereas the children of parents who belong to the same totem clan are believed not to live long.1 The Batamba in the Uganda Protectorate, who forbid marriage between all persons known to have been born of the same family ad infinitum, maintain that a transgression of this rule is followed by a disease called en diva de ya buko (the sickness of relationship), and that the children of such marriages pine away and die.2 The Washambaa in the former German East Africa think that incest leads to barrenness or early death of the offspring.3 Speaking of the custom prevalent among the mountain tribes in South Africa of a man marrying the daughter of his father's brother, Mr. McCall Theal observes that there is nothing else in their customs " that creates such disgust as this intermarriage does in the minds of the coast natives. They attribute to it the insanity and idiocy which are prevalent in the mountains, and they say the Basuto deserve to have idiots for children, as their marriages are like the marriages of dogs."4 Mr. Eyles wrote to me that the Zulus on the border of Pondoland regard sterility and deformity as consequences of consan- guineous unions ; and Mr. Cousins informed me that the Cis-Natalian Kafirs believe that their offspring would be of a more sickly nature if close intermarriage were allowed. The Herero, who favour marriage between children of a brother and a sister, but regard with horror marriage or sexual intercourse between children of two brothers or of two sisterS, assert that the offspring of unions of the latter kind are weak and die.5 Elsewhere, also, we find a distinction made between 1 van Hieu, in Harper and others, * Notes on the. Totemism of the Gold Coast,' in Jour. Anthr. Inst, xxxvi. 186. 2 Condon, ' Contribution to the Ethnography of the Basoga- Batamba, Uganda Protectorate,' in Anthropos, vi. 378. 3 Karasek-Eichhorn, ' Beitrage zur Kenntnis der Waschambaa,' in BaessJer-Archiv, i. 186. 4 Theal, History of the Boers in South Africa, p. 17. See also Idem, Yellow and Dark-shinned People of Africa south of the Zambesi, p. 219. 5 Bensen, quoted by Kohler, ' Das Recht der Herero,' in Zeitschr.f. ver "I. Rechtswiss. xiv. 300 sq. Dannert, Zum Rechte der Herero, P- 33 sq. xx EXOGAMY 173 different kinds of cousin marriages with regard to their effects on the offspring. " The Tibetans and Lepcha^," says Mr. Gait, " forbid cousins-german to marry, but the Bhotias confine the prohibition to cousins on the father's side, and more particularly to the children of the father's brother. The reason given is that the bone descends from the father's side and the flesh from the mother's, and should cousins on the father's side marry, the bone is pierced, resulting in course of time in various infirmities."1 The Rev. Dr. Alonzo Bunker kindly provided me with the following interesting information concerning the Karens of Burma. In some of their villages exogamy prevails, in others endo- gamy, but marriages between parents and children and between brothers and sisters are prohibited everywhere, and even first cousins very seldom intermarry, although there is no law against such connections. Dr. Bunker has found a striking difference with regard to stature, health, strength, and fecundity between the inhabitants of the exogamous and those of the endogamous villages, the latter being much inferior in all these respects. He has no doubt that this inferiority is due to the intermarriage of kinsfolk, and he asserts that even the natives themselves ascribe it to this cause, though they obstinately keep up the old custom, regarding marriage out of their own village as highly un- becoming. The Tikhup, one of the Old Kuki clans, are endogamous, though the union of first cousins either paternal or maternal is prohibited by them ; and we are told that the elders of the clan attributed the steady decline in their numbers to their custom of endogamy.2 The Maori, according to Mr. Best, hold the view that marriage between closely related persons is followed by a tipu-heke (degenera- tion or deterioration) in the offspring. They have an ancient saying, " Marry your tuahine (that is, sister or cousin) and the result will be puny offspring."3 In this connection may also be mentioned some statements 1 Gait, Census of India, 1911, vol. i. (India) Report, p. 252 sq. 2 Shakespear, Lushei Kuki Clans, p. 15,1. 8 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New Zealand Institute, xxxvi. 30 sq. 174 THE HISTORY OF HUMAN MARRIAGE chap. referring to peoples of a higher type. In ' Hadith,' the collection of Muhammadan traditions, it is said, " Marry among strangers ; thus you will not have feeble posterity." " This view," says Professor Goldziher, " coincides with the opinion of the ancient Arabs that the children of endogamous marriages are weakly and lean. To this class also belongs the proverb of Al-Meydani, ' . . . Marry the distant, marry not the near ' (in relationship)." A poet, praising a hero, says, " He is a hero, not borne by the cousin (of his father), he is not weakly; for the seed of relations brings forth feeble fruit."1 Two Muhammadan travellers of the ninth century tell us that the Hindus never married a relative because they thought that alliances between un- related persons improved the offspring.2 In High Albania Miss Durham was repeatedly told that if the strict rule for- bidding unions between all the descendants of a common male ancestor were transgressed, the offspring " would be blind, deaf, dumb, deformed-all kinds of misfortunes would befall. There would be a curse on such a union."3 It does not seem, however, that these facts give much support to the opinion that savages have by experience discovered the injurious effects of consanguineous marriages. Considering the detestation with which such unions are looked upon and considering further how readily all sorts of superstitions arise in connection with the sexual function, we may almost expect to find the belief that incest is attended with evil consequences, and the idea that these consequences will fall upon the incestuous brood is intelligible enough. In some cases it is quite obvious that the belief in the injurious effects on the offspring is a sheer superstition. Father Veniaminof tells us that among the early Aleut incest, which they considered the gravest crime, was believed to 1 Goldziher, * Endogamy and Polygamy among the Arabs,' in Academy, xviii. 26. Cf. Wilken, Das Matriarchal bei den alten Arabern, p. 61 ; Robertson Smith, Kinship and Marriage in Early Arabia, p. 60. 2 Reich, Geschichte, Natur- und Gesundheitslehre des ehelichen Lebens, p. 210 sq. 3 Miss Durham, ' High Albania and its Customs in 1908,' in Jour. Roy. Anthr. Inst. xl. 458. XX EXOGAMY 175 be always followed by the birth of monsters with walrus- tusks, beard, and other disfiguration p and among the Kafirs, according to Mr. Fynn, it is a general belief that the offspring of an incestuous union will be a monster-" a punishment inflicted by the ancestral spirit."2 Among the Batamba " the sickness of relationship," which causes the children of incestuous unions to pine away, will also be the result if a man and his wife sleep under the same roof as their parents, brothers, or sisters.3 The Akikuyu believe that if an incestuous marriage has been contracted knowingly, all children born in it will surely die ; whereas, if the sin of the parents has been committed unwittingly, the defilement which would otherwise prove fatal to the children can be removed by a sacrificial ceremony which is supposed to sever the bond of blood-relationship between the pair.4 In other cases, again, incest is supposed to be injurious to those who practise it or to the offspring as well. The Pasemahers of Southern Sumatra believe that if a man marries a woman of his own clan the gods will punish with destruction the guilty pair and their offspring.5 Among the Dinka incest is held to bring death to one or both of the culprits or their children, unless the misfortune be averted by means of a " sacrifice of atonement " ;6 or it is said to cause barrenness because it angers the jok, that is, the spirits of the old and mighty dead. " A girl guilty of this offence will have no children, even should she marry, until she has owned her sin, when the lover will be forced to provide a bullock to be sacrificed in atonement."7 The 1 Veniaminof, quoted by Petroll, Report on the Population, Indus- tries, and Resources of Alaska, p. 155. 2 Shooter, Kafirs of Natal and the Zulu Country, p. 45. 3 Condon, in Anthropos, vi. 377. 4 Hobley, ' Kikuyu Customs and Beliefs,' in Jour. Roy. Anthr. Inst. xl. 438. 5 Wilken, ' Over het huwelijks- en erfrecht bij de volken van Zuid-Sumatra,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-IndiS, xl. 155, 6 O'Sullivan, ' Dinka Laws and Customs,' in Jour. Roy. Anthr. Inst. xl. 187. 7 Seligman, ' Dinka,' in Hastings, Encyclopedia of Religion and Ethics, iv. 709. 176 THE HISTORY OF HUMAN MARRIAGE chap. Baganda thought that if a man had any communication with his father's sister's daughter or his mother's brother's daughter, they would both suffer from an illness " which would make their hands tremble and would unfit them for. any work."1 The Malagasy believed that the violation of a sexual tabu " entraine pour les coupables des sanctions redoutables, qui se manifestent sur cette terre par des mala- dies frappant soit les coupables eux-memes, soit leurs enfants, soit leurs conjoints, tout innocents que soient ceux-ci " ; and if children of a brother and of a sister married without undergoing the aspersion prescribed for removing the impediment from consanguinity, they were supposed to die young or to fall innocent victims to the poison ordeal whenever a false charge should be brought against them.2 The Tshi-speaking tribes of the Gold Coast maintain that if a man has unwittingly married a woman of his own clan and the act is persevered in, the children will die and bad luck will follow the parents.3 Some tribes in Sierra Leone believe that sickness will come upon any person committing incest; and when the illness comes, the whole of the family in the neighbourhood, which are supposed to become tainted with the offence, as well as the guilty parties, are publicly washed and whipped.4 Among the Fans, if a man marries a woman of his own village, any misfortune which befalls the parties is regarded as a just punishment of their misdeed.5 So also the Kafirs, according to Warner, believe that supernatural evil will befall persons who commit incest.6 The Akamba are of opinion that if a woman has sexual intercourse with her brother and conceives " she is sure to abort. The man has to purge his sin by bringing a big goat to the elders and the woman is ceremonially smeared with the 1 Roscoe, BUganda, p. 128 sq. 2 Grandidier, Ethnographic de Madagascar, ii. 149 sqq. 3 Harper and others, ' Notes on the Totemism of the Gold Coast, in Jour. Anthr. Inst, xxxvi. 182 sq. 1 Vergette, Certain Marriage Customs of some of the Tribes in the Protectorate of Sierra Leone, p. 5 sq. 6 Martron, ' Les " Eki " des Fang,' in Anthropos, i. 754. e Supra, ii. 138. XX EXOGAMY 177 contents of its stomach."1 In Nauru, one of the Marshall Islands, fainting fits, cramp, or unusually, severe agonies of death are taken as a certain sign that the patient has had connection with a near relative.2 In Mallicolo, of the New Hebrides, any violation of the rule which requires brothers and sisters to be very chaste in their intercourse is supposed to be punished by the man becoming afflicted with a sort of gravel, a dribbling of wind.3 In Mangaia, one of the Hervey Islands, should misfortune or disease overtake related couples, " the elders of the tribe would declare it to be the anger of the clan-god."4 The Koryak said that the relations of prohibited degrees would die soon if they should enter into cohabitation with one another.6 Sieroshevski knew one case of a Yakut taking a wife from his own clan, and when the woman shortly after marriage grew blind it was said that this was a punishment for breaking an old custom.6 The Caribs of Hayti, according to Oviedo, believed that a man would die if he married his mother, sister, or daughter.7 The Navaho say that if they married women of their own clan " their bones would dry up and they would die."8 The Californian Gualala, as noticed above, account it " poison " to marry a near relative.9 The Ainu of Japan, again, maintain that it is unlucky and displeasing to the gods if two sisters or two brothers marry into the same family. " One of the two sisters 1 Hobley, Ethnology of the. A-Kamba, p. 103. 2 Kohler, ' Das Recht der Marschallinsulaner,' in Zeitschr. f. vergl. Rechtswiss. xiv. 420. 3 Leggatt, ' Malekula, New Hebrides,' in Report of the Fourth Meeting of the Australasian Association for the Advancement of Science, held at Hobart, in 1892, p. 706. 4 Gill, ' Mangaia (Hervey Islands),' in Report of the Second Meeting oj the Australasian Association for the Advancement of Science, held at Melbourne, in 1890, p. 330. 5 Jochelson, Koryak, p. 737. 6 Sieroshevski, quoted by Miss Czapiicka, Aboriginal Siberia, p. 112. ' Oviedo, quoted by v. Martins, Beitrage zur Ethnographic Amerika's, i. x 16 n.*. 8 Bourke, Snake-Dance of the Moquis of Arizona, p. 279. 9 Supra, ii. 107. VOL. II N 178 THE HISTORY OF HUMAN MARRIAGE chap. will probably be punished, and die within a year after the marriage, should they marry two brothers ; or if indeed they do not die, there will be no issue."1 In many cases incest is supposed to involve the whole community in danger and disaster by causing epidemics, earthquakes, sterility of women, plants, or animals, or other calamities.2 It should be noticed that among the lower races the exoga- mous rules, especially those relating to the nearest relatives, are generally so strictly observed that no genuine knowledge could possibly be based on the few cases in which they are transgressed. And where the rules are becoming laxer and marriages which were formerly prohibited are becoming more numerous, it is only natural that the old people should look upon this disregard of ancestral custom as fraught with evil consequences. The case is different, however, when endogamous marriages are supposed to lead to injurious results in spite of being in accordance with native custom, as among the endogamous Karens and Tikhup and the ancient Arabs, who even considered that a man had a right to marry his father's brother's daughter.3 I think we have to admit that these cases may possibly represent accurate observations of facts, although, of course, we cannot be certain that they do so. But even if it could be proved that a few savages really have discovered the injurious effects of in-breeding, we should by no means be entitled to assume that such experi- ence ever has been acquired on so large a scale as is presup- posed by a theory which wants to explain rules of universal prevalence. And even if we had a right to make such an assumption, even if savage men everywhere had discerned that children born of marriage between closely related persons are not so sound and vigorous as others, we could certainly not be sure that they everywhere would have allowed this knowledge to check their passions. Con- sidering how seldom a civilised man who has any disease, or tendency to disease, which is likely to be transmitted to 1 Batchelor, Ainu and their Folk-Lore, p. 229. 2 Supra, ii. 123. Frazer, Psyche's Task, p. 45 sqq. 3 Supra, ii. 69. XX EXOGAMY 179 his descendants, hesitates to marry an equally unhealthy woman, it would hardly be reasonable to suppose that savages have greater forethought and self-command. In- deed, we just noticed instances of peoples who do not allow their belief in the injurious effects of in-breeding to prevent their practising it as a national custom. Finally, as for Morgan's suggestion that clan exogamy was instituted to prevent the marriage or cohabitation of blood-relations, especially brothers with sisters, I certainly agree that a very close connection exists between clan exogamy and the prohibitions of marriages of near kin, but I cannot believe that the extensive and cumbersome institution of clan exogamy should have been invented simply as a precaution against unions between the nearest relatives. It would have been just as easy to prevent such unions by prohibiting them only, as is done among savages who have no clan exogamy as well as among civilised peoples. Sir James G. Frazer believes that Morgan's view furnishes " the true key to the whole system of exogamy," and holds that it has been greatly strengthened by the additional knowledge which we have since acquired of the social organisation of the Australian aborigines. He maintains that the two-class system was instituted to prevent the marriage of brothers with sisters, the four-class system to prevent the marriage of parents with children, and the eight-class system.to prevent the marriage of the children of a brother and a sister respectively, the marriage of all other first Cousins having been already prevented by the institution of the two-class system. He makes, however, the important addition that the rules of class exogamy were not merely intended to prevent the marriage of a man with his sister, his mother, or his daughter in the physical sense in which we use these terms, but to prevent his marriage with his sister, his mother, and his daughter in the clas,si- ficatory sense of these terms.1 Of course, exogamous rules are intended to prevent such marriages as they prohibit. That their aim among the Australian natives is to prevent the union of persons who 1 Frazer, Totemism and Exogamy, i. 279 sqq. ; iv. 105 sqq. N 2 180 THE HISTORY OF HUMAN MARRIAGE chap. are considered to be too near to each other in blood, or, as they sometimes put it, are of " the same flesh," has been observed by several earlier writers.1 Thus Dr. Howitt thinks that there was originally an " Undivided Commune " which, owing to a reformatory movement arising within the council of elders (in which the tribal wizard holds no mean place), was for this purpose divided, by the ancestors of the Australian aborigines, into two classes, and that these classes were likewise intentionally divided into four and eight sub-classes.2 It may be questioned whether it is wise to speak of a " reformatory movement " in the past of a race of whose history we know nothing, especially when we have to do with rules of so universal a character as the pro- hibition of marriage between kindred; and even more conjectural is Frazer's suggestion that the system of class exogamy " was probably devised and introduced by the council of elders in some one local community, from which it may have gradually spread by peaceful transmission over the whole of Australia."3 Mr. Mathew assumes that the class relationship grew unconsciously, gradually, and naturally out of the actual relations of small family groups, or was the outcome of the contact of distinct races, or originated partly by the one cause, partly by the other- instead of assuming with Howitt that it was enacted as a remedy for incest. What he says about the contact of races is based on his theory that the two phratries or classes represent two ancient, distinct races, which amalgamated to form the Australian race, whilst the multiplication of classes from two to four and from four to eight was due to an amalgamation of tribes, each of which had two or 1 Lang, Aborigines of Australia, p. io. Curr, op. cit. i. 112. Hardman, ' Notes on some Habits and Customs of the Natives of the Kimberley District, Western Australia,' in Proceed. Roy. Irish Academy, ser. iii. vol. i. 71 sq. Stirling, "'Anthropology,' in Report on the Work of the Horn Scientific Expedition to Central Australia, iv. 49. 2 Howitt, ' Notes on the Australian Class Systems,' in Jour. Anthr. Inst. xii. 499 sqq. Idem, ' Australian Group-Relationships,' ibid, xxxvii. 287. Idem, Native Tribes of South-East Australia, p. 142 sqq. s Frazer, op. cit. i. 283. XX EXOGAMY 181 four classes.1 Whether there is any foundation for this theory 2 or not, I cannot see any reason to assume that the origin of the Australian class restrictions has been essentially different from the exogamous rules of other savage races. Sir James Frazer, also, suggests, although with the greatest caution, that exogamy may be due to a belief that the intercourse of near kin is injurious, not however to the progeny but to the whole community. He points out that among various peoples such intercourse is thought to render the women sterile and to endanger the common food-supply by preventing edible animals from multiplying and edible plants from growing. " The idea that sexual crime in general and incest in particular blights the crops is common among peoples of the Malayan stock in the Indian Archipelago and their kinsfolk in Indo-China; but it is also strongly held by some natives of West Africa, and there are grounds for thinking that similar notions as to the injurious influence of incest on women and cattle as well as on the com prevailed among the primitive Semites and the primitive Aryans, including the ancient Greeks, the ancient Latins, and the ancient Irish."3 An obvious objection to this theory is that the superstitions on which it is based have been found only among a few races, although the prohibition of marriage between kindred is universal. Moreover, as Sir James himself admits, all the peoples who are known to hold the beliefs in question appear to be agricultural, and what they especially dread is the sterilising effect of incest on their crops ; hence the known distribution 1 Mathew, Eaglehawk and Crow, p. 97. Idem, Two Representative Tribes of Queensland, pp. 138-140, 159. Idem, * Origin of the Austra- lian Phratries and Explanations of some of the Phratry Names,' in Jour. Roy. Anthr. Inst. xl. 166 sqq. Mr. Andrew Lang (Social Origins, p. 36) believes that the so-called " bi-section " of the Australian tribe into intermarrying classes " was really the result of an amalgamation of two separate and independent local totem groups " ; and something similar has been suggested by certain earlier writers (see ibid. p. 36). * For a discussion of it see Thomas, Kinship Organisations and Group Marriage in Australia, p. 53 sqq. 8 Frazer, Totemism and Exogamy, iv. 157 sq. The evidence is given in full in the same author's book Psyche's Task, p. 44 sqq. 182 THE HISTORY OF HUMAN MARRIAGE chap. of these beliefs suggests that they are a product of a culture somewhat more advanced than can be ascribed to the savages who started exogamy. It is particularly suspicious that no such superstition has been found among the Austra- lian aborigines, who treat incest with the greatest severity and whose beliefs have been more carefully investigated than those of most other savages ; and it is a poor argument to conjecture that a still more careful search might some day reveal the existence of the same superstition among them also. Mr. Curr says that he was unable to discover on what ground consanguineous marriages are held to be objec- tionable by those natives, their replies to questions on this head invariably being, " Our tribe always did as we do in this matter " j1 and later investigators have been unable to furnish us with any further information on the subject. But even if we knew that all races on earth believed in sterilising or injurious effects of incest upon women, animals, and plants, exogamy would not be explained thereby. The question how that belief arose would still call for an answer. I think the answer is quite obvious, but at the same time it is destructive to the whole theory. Incest is considered harmful because it is disapproved of, and it is not in the first place disapproved of because it is considered harmful. This appears from the fact that other forms of illicit love, such as adultery and fornication, are supposed to produce the same disastrous effects. Sir James is himself aware of this weakness in his theory. In 1909 and 1913 he actually wrote, in the two editions of ' Psyche's Task,' " Clearly the notion that such immorality interferes with the course of nature must have been secondary and deriva- tive : people must on independent grounds have concluded that certain relations between men and women were wrong and injurious before they extended the conclusion by false analogy to nature."8 It is therefore somewhat surprising that he in 1910, in his work on ' Totemism and Exogamy,' makes a suggestion which he has himself implicitly denounced 1 Curr, op. cit. i. 112. 2 Frazer, Psyche's Task, edition of 1909, p. 47, and edition of 1913, p. 102. xx EXOGAMY 183 both before and after.1 It is evidently an emergency measure to which Sir James has resorted so as not to be totally at a loss for an explanation of a world-wide institution which he has submitted to a most painstaking analysis. He writes, " At least if that is not the origin of exogamy I must confess to being completely baffled, for I have no other conjecture to offer on the subject."2 And yet he admits all the main premises on which I have based my theory of the prohibition of incest, but refuses to admit my conclusion. Whilst Frazer has offered no satisfactory solution of the problem of exogamy he has at any rate helped to destroy the theory that exogamy has originated in totemism.3 Professor Durkheim derives exogamy from a religious sentiment which is due to certain magical virtues attributed to blood, especially to the menstruous blood of women, and the religious awe for blood is traced by him to totemism.4 Nay, totemism is the ultimate source not only of clan exogamy but of all other prohibitions against incest as well; the rule of clan exogamy, he maintains, has been extended to near relatives belonging to different clans because they are in no less intimate contact with each other than are the members of the same clan.5 And when totemism and at the same time clan relationships disappeared, the rule of clan exogamy was entirely transformed into a prohibition of marriage between near relatives, which in the course of evolution narrowed down to a prohibition of marriage between ascendants and descendants and between brothers and sisters only.6 1 In his Totemism and Exogamy he only says (iv. 159), " It might be argued, as I have argued elsewhere, that all such notions as to the injurious natural consequences of incest are an effect rather than the cause of its prohibition." 2 Frazer, Totemism and Exogamy, iv. 160. 3 For a criticism of Professor Durkheim's theory see ibid. iv. 100 sqq. Father Schmidt (' Totemismus, viehziichterischer Nomadismus und Mutterrecht,' in Anthropos, x.-xi. 605) thinks that the majority of ethnologists share the opinion that exogamy has originated inde- pendently of totemism. 4 Durkheim, ' La prohibition de 1'inceste et ses origines,' in L'annde sociologique, i. 47 sqq. 5 Ibid. p. 19. 8 Ibid. p. 58. 184 THE HISTORY OF HUMAN MARRIAGE chap. Professor Durkheim thus tries to explain a phenomenon of universal prevalence through an institution which has been found among certain peoples only. How does he know that totem clans once prevailed among all peoples who now prohibit the intermarriage of near relatives ? Even exoga- mous clans are by no means always totemic. For example, the Masai of Africa and a number of tribes in Sumatra and other parts of the Malay Archipelago are divided into exogamous clans which are not, so far as appears, totemic ; and in India, especially, the institution of clan exogamy disjoined from the institution of totemism appears to be very widespread and is shared by the Aryan peoples, in- cluding the Brahmans, Rajputs, and other high castes. Indeed, Sir James Frazer thinks that exogamy unaccompanied by totemism is so common that it might furnish a theme for a separate treatise.1 And as clan exogamy may exist with- out totemism, so totemism may exist without clan exogamy. For example, among the Wahehe, Taveta, and Nandi in Africa, the Kacharis in Assam, the Kworafi tribe in British New Guinea, and the tribes in the heart of Australia, the totemic clans are not exogamous; and the traditions of the Central Australian tribes represent their ancestors as habitually marrying women of their own totem, thus pointing back to a time when totemism existed unaccom- panied by clan exogamy.2 But even if for the sake of argument we suppose that totemism was once universal and that exogamy was always combined with totemism, it still remains to be explained how totemism led to exogamy. Professor Durkheim maintains that totemism gave rise to that belief in blood and especially menstruous blood as a seat of magical virtues which he regards as the immediate cause of exogamy. " The totemic being," he says, " is immanent in the clan; he is incarnate in every individual, and it is in the blood that he resides. He is himself the blood. . . . When it flows, it is the god who is spilled. . . . The religious respect which it inspires forbids all idea of contact, and, since woman passes, so to say, a part of her life in blood, the same feeling extends to her, 1 Frazer, op. cit. iv. io sq. 2 Ibid. iv. 8 sq. XX EXOGAMY 185 stamps her with its impress, and isolates her."1 This notion of totemism, however, is entirely unsupported by known facts» as has been pointed out by Sir James Frazer, who has a more profound and comprehensive knowledge of totemism and the beliefs connected with it than anybody else, and who, moreover, must be recognised as a most impartial judge considering that it is just his original treatise on ' Totemism ' to which Professor Durkheim refers his readers for proof of the worship of the totem. Frazer writes :-" The conclusion to which the facts point is that the relation between a man and his totem is one of simple friendly equality and brotherhood, and by no means one of religious adoration of a deity mysteriously incarnate not only in the whole totemic species of animals or plants, but also in the flesh and above all in the blood of every man, woman, and child of the clan. A mystical religion of this abstract sort might be appropriate enough to sects like the Gnostics, the heirs of an ancient civilisation and of a long train of subtle philosophies ; it is wholly foreign and indeed incomprehensible to the simple, concrete modes of thought of a savage, and to attribute it to the extremely rude savages with whom the system of exogamy must unquestionably have originated is to commit the serious mistake of interpreting primitive thought in terms of advanced thought; it is to invert the order of development. A theory of exogamy which rests on such a basis is wholly untenable."2 That simple peoples have a dread of blood and particularly of menstruous blood is a well-known fact, but this dread is certainly not felt merely of the blood of members of one's own totem, and cannot therefore be the cause of totemic exogamy. M. Reinach approves of Professor Durkheim's idea that the prohibition of incest is a particular case of the blood tabu, but suggests that exogamy is intended to prevent the shedding of the blood of a woman of one's own clan at defloration, and that it consequently belongs to the same class of prohibitions as the commandment, " Thou shalt not kill."3 1 Durkheim, loc. cit. p. 52 sq. 2 Frazer, op. cit. iv. 101 sq. 3 Reinach, Cultes, mythes et religions, i. 165 sq. 186 THE HISTORY OF HUMAN MARRIAGE chap. But I fail to see that this suggestion brings us any nearer the solution of the problem. I am not aware that there is any general objection to harmless shedding of blood within the tribe ; on the contrary, the widespread practice of circumcision and various other mutilations clearly indicate that there is no such objection. Moreover, exogamy prevails among peoples whose girls are regularly deflowered before marriage by somebody else than the husband.1 Among some peoples it is said to be the custom for a father to deflower his daughter, and yet he is not allowed to marry her.2 And how could M. Reinach's theory account for the universal prohibition of marriage between mother and son ? Professor Durkheim's theory of the combination of exogamy with totemism has to some extent gained the support of Mr. Andrew Lang, who writes :-" As soon as the animal-named groups evolved the universally diffused beliefs about the wakan or mana, or mystically sacred quality of the blood as the life, they would also develop the various totem tabus, such as not to kill the totem animal, not to shed its blood, and the idea that, by virtue of this tabu, a man must not marry a maid who was of one blood with him in the totem." But Mr. Lang is of opinion that even without any blood tabu, the tabu on women of the same totem might arise, and quotes in support of this view the statement that " an Oraon clan, whose totem is the Kujzar-tree, will not sit in its shade." " So strong," he adds, " is the intertotemic avoidance." The belief grew to the pitch that a man must not " use " anything of his totem, and thus totemic exogamy, with the sanction of the sacred totem, was established 3 The instance of " inter- totemic avoidance," however, refers to the people's relations to their totem and not to their relations to each other. It would have been more to the point if Mr. Lang could have mentioned instances of " intertotemic avoidance " of the latter kind. But I doubt whether there is any such avoidance-except in the relations between the sexes. Mr. Lang, however, thinks that there was exogamy 1 See supra, i. 181 sqq. 2 Supra, i. 188 sq. 3 Lang, Secret op thb Totem, p. 125. XX EXOGAMY 187 before totemism. Following Mr. Atkinson,1 he suggests that man dwelt originally, as in Darwin's opinion, in small family groups, the sires in each case expelling the young males when they were arriving at puberty. This expulsion of the sons by the sire was his unspoken enforcement of exogamy, of marriage out of the brutal herd, out of the savage camp. And as progress advanced, " the sire was moved (by the tears, perhaps, of some female mate, in Mr. Atkinson's theory, and by a softening of his own heart, now becoming human) to let the son of his old age, his Benjamin, remain in the camp, so long as he did not interfere with any of the females, but found a mate outside the group. . . . Some early men must have begun the practice of permitting the young males to remain in the camp or fire- circle, but not to choose a mate within it. They were of milder mood ; the mothers, too, were growing more mater- nal ; had it not been so, we should all be more brutal than we are at this moment. Then came in Natural Selection. Groups which contained several fine young males would be ' the fittest,' would overcome in all encounters groups with only one male, perhaps a tottering old male ; and the fittest groups would survive. The reform would be imitated by other groups till ' the happy solution was repeated all through the species.' "2 One cannot help but admire the filial reverence of those exemplary sons, who never revolted against their father, even though he was a tottering old male. The young gorillas do not always seem to be equally reve- rential ; and yet it is on a statement made by Darwin with reference to the gorilla on which the whole theory is based. Darwin suggested that man originally either lived in small communities, in which each male had one or several wives for himself, or was not a social animal and yet lived with several wives, like the gorilla. "For all the natives ' agree that but one adult male is seen in a band ; when the young male grows up, a contest takes place for mastery, and the 1 Atkinson, Primal Law, p. 220 sqq. 2Lang, 'Theory of the Origin of Exogamy and Totemism,' in Folk-Lore, xxiv. 156 See also Idem, Social Origins, pp. 126, 238 sqq. ; Idem, Secret of the Totem, p. 113 sq. 188 THE HISTORY OF HUMAN MARRIAGE chap. strongest, by killing and driving out the others, establishes himself as the head of the community.' The younger males, being thus expelled and wandering about, would, when at last successful in rinding a partner, prevent too close inter- breeding within the limits of the same family."1 Darwin's authority was Dr. Savage, but from the statements of later writers it appears that the habits of the gorilla are not quite uniform. According to some accounts the old male some- times wanders companionless ; when he " becomes cross, or possibly, it may be, too infirm to travel with the company, he goes off by himself and spends the rest of his life without companionship." Sometimes, also, there seem to be two adult males in the same company, and " as the younger members grow up they take, or rather keep, their places in the company."2 But even if Mr. Lang's picture of the early family group were quite correct, it could not explain how it happened that the old male was prohibited from marrying his daughters. And any theory of the origin of the prohibition of incest which takes no account of the relation between father and daughter is obviously a failure. Messrs. Hose and McDougall have tried to make good this omission. They write :-" If we accept some such view of the constitution of primitive society as has been suggested by Messrs. Atkinson and Lang (' Primal Law '), namely, that the social group consisted of a single patriarch and a group of wives and daughters over all of whom he exercised unre- stricted power or rights ; we shall see that the first step towards the constitution of a higher form of society must have been the strict limitation of his rights -over certain of the women, in order that younger males might be incor- porated in the society and enjoy the undisputed possession of them. The patriarch, having accepted this limitation of his rights over his daughters for the sake of the greater security and strength of the band given by the inclusion of a certain number of young males, would enforce all the more strictly upon them his prohibition against any tampering with the females of the senior generation. Thus very strict 1 Darwin, Descent of Man, ii. 395. 2 Supra, i. 33 sq. XX EXOGAMY 189 prohibitions and severe penalties against the consorting of the patriarch with the younger generation of females, i.e., his daughters, and against intercourse between the young males admitted to membership of the group and the wives of the patriarch, would be the essential conditions of advance of social organisation. The enforcement of these penalties would engender a traditional sentiment against such unions, and these would be the unions primitively regarded as incestuous. The persistence of the tendency of the patri- arch's jealousy to drive his sons out of the family group as they attained puberty would render the extension of this sentiment to brother-and-sister union easy and almost inevitable. For the young male admitted to the group would be one who came with a price in his hand to offer in return for the bride he sought. Such a price could only be exacted by the patriarch on the condition that he maintained an absolute prohibition on sexual relations between his offspring so long as the young sons remained under his roof."1 Thus, brothers are prohibited from marrying their sisters because the old patriarch drove away his grown-up sons out of jealousy; but his jealousy was not strong enough to prevent other young males from joining the band. On the contrary, he allowed them to be incorporated in it, because they added to its strength; nay, he gave them his own daughters in marriage, and refrained henceforth himself from intercourse with these young women so rigorously that ever since a father has been prohibited from marrying his daughter. But the young men had to pay a price for their wives. It may be asked : why did not the old patriarch accept a price from his own sons or let them work for him, instead of mercilessly turning them out of their old home, although they would have been just as good, if not better, protectors of it as the strangers ? And why did he give the young men his daughters ? He might have kept the young women for himself and let the young men have the old ones. This is what is done by the old men in Australia, where the young girls are, as a rule, allotted to old men, and the boys, whenever they are allowed to marry, get old lubras as 1 Hose and McDougall, Pagan Tribes of Borneo, ii. 197 sq. n. 190 THE HISTORY OF HUMAN MARRIAGE chap. wives.1 Yet, in spite of this custom, there is no country where incest has been more strictly prohibited than in Australia. Finally, how can the theory set forth by Messrs. Hose and McDougall account for the prohibition of incest among low savages who have never been known to expect a bride price from a suitor ? That exogamy has partly originated in marriage by pur- chase has also been suggested by Professor Starcke. Accord- ing to him, unions between a father and daughter rarely occur because a father is unwilling to renounce the advantages of bestowing his daughter in marriage. Again, a marriage between a mother and son or between a brother and elder sister would altogether transform their relations, the husband being in all cases the lord of his wife ; and the breach in the respect due to their elders and the confusions and contra- dictions resulting from it would be quite enough to produce an aversion to such marriages. This aversion would be increased by the fact that it was generally impossible to contract unions of this kind, since the son possessed nothing which he could offer to the father as purchase money; and to enter the paternal house by force, in order to carry off the wife or daughter, would be an unheard-of crime among savages. If, then, " in this way the impression arises that there is something unusual and incompatible with other ideas in marriage between such persons, an occasional calamity which befalls any of them will be enough to excite the imaginative faculty in the highest degree; and if no prohibition previously existed, the absolute condemnation of such marriages would then be pronounced."2 According to this theory the prohibition of marriage between a father and his daughter was really due to the narrow intellect of savage men. The father did not see that, even though he did not get any purchase money by marrying his own daughter, he, on the other hand, had nothing to pay for his wife. He was like a farmer who does not eat his own corn because, if he did, he could not 1 See infra, on Group-marriage and other Gro up-relations. 2 Starcke, The Primitive Family, p. 229 sq. XX EXOGAMY 191 get money by selling it to somebody else. A son could not marry his mother or a brother his sister because Jie had nothing to offer to the father in return. How, then, could he get a wife at all, where all fathers expected a price for their daughters ? Again, if a man had to refrain from marrying his mother or elder sister on account of the respect he owed them, no such reason could have prevented him from marrying his younger sister. Yet this is also prohibited. I think it would be difficult to find another set of theories ever offered to explain a custom, law, or institution of world- wide occurrence so utterly unsatisfactory as those which have now been subject to our criticism. To each of them, as we have seen, the gravest objections may be raised ; and in addition there are other objections that may be raised to all of them. They all regard the exogamous rules as social survivals from very remote ages. They all suppose that these rules have originated in social conditions which no longer exist, or in ideas which have been found only among a few savages or which have never been found anywhere. Now, is it really possible to believe that a law like that of incest among ourselves could be traced to a pristine habit of female infanticide, or to the vain desire of savage men to have trophies in their wives, or to marriage by capture originating in the hypothetical period of primitive promiscuity, or to experience of the injurious influence of in-breeding made at an earlier stage of human development than that represented by any living savages but after- wards forgotten, or to a superstitious belief that incest endangers the crops, or to the furious jealousy of a gorilla- like ancestor ? Is it possible to believe that a restriction barring marriages which could be more conveniently arranged than any others, requiring neither capture nor purchase nor the consent of strangers, 'might have been preserved through ages without being relaxed though serving no useful purpose at all ? It should be noticed that the exogamous rules have not remained unaltered; on the contrary, they differ even among peoples of the same stock, 192 THE HISTORY OF HUMAN MARRIAGE chap. and we know that in Europe, in the course of a few cen- turies, they have been greatly changed in spite of the religious sanction given them by the Church. This proves that the exogamous rules are not dead fossils, but living parts of the social organism, subject to modifications accord- ing to the circumstances. Moreover, the theories in question imply that the home is kept free from incestuous intercourse by law, custom, or education.1 But even if social prohibitions might prevent unions between the nearest relatives, they could not prevent the desire for such unions. The sexual instinct can hardly be changed by prescriptions; I doubt whether all laws against homosexual intercourse, even the most draconic, have ever been able to extinguish the peculiar desire of anybody born with homosexual tendencies. Never- theless, our laws against incest are scarcely felt as a restraint upon individual feelings. And the simple reason for this is that in normal cases there is no desire for the acts which they forbid. Generally speaking, there is a remarkable absence of erotic feelings between persons living very closely together from childhood. Nay more, in this, as in many other cases, sexual indifference is combined with the positive feeling of aversion when the act is thought of. This I take to be the fundamental cause of the exogamous 1 This has also been the view of many other writers, such as Montesquieu (De I'esprit des lois, book xxvi. ch. 14 :-" Les pdres et les mdres ayent voulu conserver les moeurs de leurs enfans et leurs maisons pures "), Hume (' An Enquiry concerning the Prin- ciples of Morals,' sec. iv., in Philosophical Works, iv. 199 :-" Those who live in the same family have such frequent opportunities of licence of this kind, that nothing could preserve purity of manners, were marriage allowed, among the nearest relations, or any inter- course of love between them ratified by law and custom "), Tillier (Le Mariage, p. 111 :-•" Les rapprochements sexuels entre individus trop j eunes sont surtout faciles dans la famille & cause de la vie en commun et du contact continuel; les parents ont done dfl s'opposer a toute union sexuelle entre leurs enfants et exercer une surveillance attentive pour empficher ces rapprochements. . . . Des raisons analogues en principe ont dfl faire proscrire I'union sexuelle de la mdre et du fils et celle du pdre et de la fille "), Krauss (in Am Ur-Quell, iv. 151), Finck (Primitive Love, p. 49^.), and Freud (Three Contri- butions to the Theory of Sex, p. 84 sq.). XX EXOGAMY 193 prohibitions. Persons who have been living closely together from childhood are as a rule near relatives. Hence their aversion to sexual relations with one another displays itself in custom and law as a prohibition of intercourse between near kin. The existence of a feeling of the kind suggested, or at least of sexual indifference to housemates, has been recognised by various writers as a psychological fact proved by common experience.1 According to Bentham, " individuals accus- tomed to see each other and to know each other, from an age which is neither capable of conceiving the desire nor of inspiring it, will see each other with the same eyes to the end of life.''2 Dr. Havelock Ellis writes, " Between those who have been brought up together from childhood all the sensory stimuli of vision, hearing, and touch have been dulled by use, trained to the calm level of affection, and deprived of their potency to arouse erethistic excitement which produces sexual tumescence."3 Even between lads and girls who are educated together in the same school 1 Wagner, ' Die Kulturziichtung des Menschen gegeniiber der Naturziichtung im Tierreich,' in Kosmos, i. 29. v. Hellwald, Die menschliche Familie, p. 179 sq. Wake, Development of Marriage and Kinship, p. 55. Dalton, Descriptive Ethnology of Bengal, p. 248 note. My theory has been approved of by, A. R. Wallace (in his ' Introductory Note ' and in a letter to the author), Giddings {Principles of Sociology, p. 267), Howard {History of Matrimonial Institutions, i. 125 sqq.}, Miiller-Lyer {Die Familie, p. 37 sqq.}, and others. Tylor (in a review in Academy, xl. 289) thought that I was " well on the track." See also Crooke, Tribes and Castes of the North-Western Provinces and Oudh, i. pp. clxxix., clxxx., ccii. ; Russel!, Tribes and Castes of the Central Provinces of India, i. 155. Mr. William I. Thomas writes {Sex and Society, p. 194), " Familiarity with women within the group and unfamiliarity with women without the group is the explanation of exogamy on the side of interest, and the system of exogamy is a result of exchanging familiar women for others." He makes no mention of my theory, but says {ibid. p. 175) that the theories of exogamy are all unsatisfactory, and blames me in a general way for failing to perceive that " the lower races are intensely interested in sexual life " {ibid. p. 176). 2 Bentham, Theory of Legislation, p. 220. 3 Ellis, Studies in the Psychology of Sex, (vol. iv.) Sexual Selection in Man, p. 205 sq. VOL. II O 194 THE HISTORY OF HUMAN MARRIAGE chap. there is a conspicuous absence of erotic feelings, according to an interesting communication of a lady who has for many years been the head-mistress of such a school in Finland. One youth assured her that neither he nor any of his friends would ever think of marrying a girl who had been their schoolfellow ;1 and I heard of a lad who made a great distinction between girls of his own school and other, " real," girls, as he called them. According to other accounts boys may display erotic feelings towards younger girls in the school, but not towards girls of their own class. The normal want of inclination for sexual intercourse be- tween persons who have been living closely together from the childhood of one or both of them is no doubt a world- wide phenomenon. Plato observed that an unwritten law defends as sufficiently as possible parents from incestuous intercourse with their children and brothers from intercourse with their ' * 'ters, and that the thought of such a thing does not enter at all into the minds of most of them.2 When I asked my Berber teacher from the Great Atlas whether marriages between cousins were frequent in his tribe, his answer was, " How could you love a girl whom you have always seen ? " In reply to earlier writers who had accused the Hottentots of practising incest, Le Vaillant wrote, " The whole family inhabits one scanty hut-the father sleeps with his daughter-the brother with his sister- the mother with her son-but on the return of Aurora every one rises with a pure heart."3 Among the Maori of New Zealand, according to Colenso-who is considered a first- rate authority-adult brothers and sisters slept together, as they had always done from their birth, " not only without sin, but without thought of it."4 Speaking of the baisakh bihu5 of the Assamese, Colonel Dalton states that for many days before the actual festival the young people in the villages may be seen moving about in groups gaily dressed or forming circles, in the midst of which the prettiest girls dance with their long hair loose on their shoulders, but 1 Lucina Hagman, ' Fran samskolan,' in Humanitas, ii. 188 sq. 2 Plato, Leges, viii. 838. 3 Le Vaillant, Travels from the Cape of Good-Hope, ii. 136 sq. 4 Colenso, Maori Races, p. 47 sq. 5 See supra, i. 86. XX EXOGAMY 195 that on these occasions the girls " do not like to dance before the men of their own village."1 In other parts of India Mr. Russell2 and Mr. Sturrock3 have found facts supporting my view that exogamy has originated in a feeling against the marriage of persons who have lived closely together from early youth. In his account of Australian aborigines Mr. Mathew suggests that there may be " an auxiliary cause to exogamy among barbarians in what may be called an instinctive hankering after foreign women."4 Sir Richard Burton says, " As a general rule Somali women prefer amourettes with strangers, following the well-known Arab proverb, ' the new comer filleth the eye.' "5 Mr. Grubb explains the frequent unfaithfulness on the part of married men and women among the Lengua Indians of the Para- guayan Chaco by the fact that the people " mix in a very small circle, and meet with no partner with whom they could mate, except those whom they have played with from childhood. Little choice, therefore, is left them, and there is not that novelty of a fresh face and character which tends to become captivating, and eventually leads to love."6 Among the lower animals, also, there are indications that the pairing instinct fails to be stimulated by com- panions and seeks strangers for its gratification. The Marquis de Brisay, an authority on doves, says that " two birds from the same nest rarely couple. Birds coming from the same nest behave as though they regarded coupling as prohibited, or, rather, they know each other too well, and seem to be ignorant of their difference in sex, remaining unaffected in their relations by the changes which make them adults."7 The honey-bee never propagates in the 1 Dalton, op. cit. p. 81. 2 Russell, Tribes and Castes of the Central Provinces of India, i. 155. 3 Sturrock, Madras District Manuals : South Canara, i. 143. 4 Mathew, ' Australian Aborigines,' in four. & Proceed. Roy. Soc. New South Wales, xxiii. 403. 6 Burton, First Footsteps in East Africa, p. 119. 6 Grubb, An Unknown People in an Unknown Land, p. 215. 7 L'interm&diaire des biologist es, November 20th, quoted by Have- lock Ellis, op. cit. (vol. iv.) Sexual Selection in Man, p. 206 n. 3. 196 THE HISTORY OF HUMAN MARRIAGE chap. nest but flies out for this purpose.1 Among winged ants it is common for both males and females to have a " marriage flight." They leave their nest about the same time, their swarms mix with other swarms, even their hereditary enemies, ancient quarrels are forgotten, and there is nothing but joy and love-making.2 Among domesticated animals it has often been noticed that companionship has a dulling effect upon the sexual instinct and that preference is given to strangers. Montaigne wrote, " I was fain to turn out into the paddock an old stallion, as he was not to be governed when he smelt a mare : the facility presently sated him as towards his own, but towards strange mares, and the first that passed by the pale of his pasture, he would again fall to his importunate neighing and his furious heats as before."3 I myself have been told by a trustworthy person of a stallion that would not approach mares of the same stable. Mr. G. W. Harris wrote to me :-" It is well known, I believe, by dog breeders, that if you bring up from puppyhood dogs and bitches together, the bitches very frequently refuse to take the dogs to which they have been accustomed from puppyhood. There is no aversion, I believe, on the part of the male, but with regard to bitches, even from my own experience, I am able in a certain measure to corroborate this." Mr. Cuppies, quoted by Darwin, states that among dogs the male seems rather inclined towards strange females ;4 and Mr. Heape thinks that, in fact, all breeders will agree that animals when brought into r itact with strangers experience increased sexual stimulation.5 Professor Seligman has kindly provided me with the follow- ing communication :-■" I had about half a dozen fully plumaged drakes of the common mallard, born of domes- ticated stock, in a good-sized pen containing a small pool and natural herbage. They got along well together, and although constantly watched (I was studying colour 1 Hensen, Physiologic der Zeugung, p. 182. 2 Escherich, Die Ameise, p. 74 sqq. 3 Montaigne, Essays, book ii. ch. 15, vol. ii. 330. 4 Darwin, Descent of Man, ii. 294. See also Rohleder, Die Zeugung unter Blutsverwandten, p. 33 sq. 6 Heape, Sex Antagonism, p. 63. XX EXOGAMY 197 changes in plumage at the time) they were found to pay no particular attention to each other. Two strange drakes of the same class of bird and in the same plumage were intro- duced, and these were simply overwhelmed by the sexual attentions of the older inhabitants of the pen. I know this went on for two or three hours-it may have gone on for days; but later on they had all settled down to what seemed a normal life. The time of the year was early summer, probably early June." Sexual indifference, however, is not by itself sufficient to account for exogamous prohibitions. This seems to have escaped Dr. Havelock Ellis, and Mr. Crawley as well, who quotes his criticism of my theory with approval.1 On the one hand, Dr. Ellis says that " the explanation of the abhorrence of incest is really . . . exceedingly simple." On the other hand, he affirms that " the normal failure of the pairing instinct to manifest itself in the case of brothers and sisters, or of boys and girls brought up together from infancy, is a merely negative phenomenon due to the in- evitable absence under those circumstances of the conditions which evoke the pairing impulse " ; and he blames me for having assumed the existence of an innate tendency which "is as awkward and artificial an instinct as would be, let us say, an instinct to avoid eating the apples that grew in one's own orchard."2 But the " instinct " of which I spoke (I now prefer to avoid this term) was simply aversion to sexual intercourse with certain persons, and this is an extremely well known mental phenomenon, of which Dr. Ellis himself, in his excellent ' Studies in the Psychology of Sex,' ' has given us many instances, quite instinctive in character. Thus he observes that some of the male inverts described by him experience what is called horror feminae, that is to say, woman as an object of sexual desire is " disgusting " (not merely indifferent) to them ;3 and he also repeatedly speaks of the " abhorrence " of incest. As a matter of fact, sexual 1 Crawley, ' Exogamy and the Mating of Cousins,' in Anthropo- logical Essays presented to E. B. Tylor, p. 52 sq. 2 Ellis, op. cit. (vol. iv.) Sexual Selection in Man, p. 205. 3 Ibid. vol. ii. Sexual Inversion, p. 278 sq. 198 THE HISTORY OF HUMAN MARRIAGE chap. indifference is very generally combined with sexual aversion when the act is thought of ; indeed, I believe that this is normally the case whenever the idea of sexual intercourse occupies the mind with sufficient intensity and a desire fails to appear. And, as I have pointed out in another work,1 aversions which are generally felt readily lead to moral disapproval and prohibitory customs or laws. Apart from Dr. Ellis's criticism-which I consider to strengthen rather than weaken my position-various objec- tions have been raised to my theory. Thus it has been said that if close living together calls forth aversion to sexual intercourse, such aversion ought to display itself between husband and wife as well as between near relatives.2 But these cases are certainly not identical. What I have here spoken of is a lack of inclination for, and a feeling of aversion associated with the idea of, sexual intercourse between persons who have lived in a long-continued intimate relation- ship from a period of life when the action of sexual desire, in its acuter forms at least, is naturally out of the question. On the other hand, when a man marries a woman his feeling towards her is of a very different kind, and his love impulse may remain, nay increase, during the conjugal union. Yet even in this case long living together has undoubtedly a tendency to dull the sexual desire and sometimes even to lead to positive aversion. Dr. Bloch observes:-" The eternal uniformity of daily companionship puts love to sleep, damps its ardour, and even gives rise to a sense of latent or open hatred between a married pair. This hatred is observed most frequently in love-matches."3 1 Origin and Development of the Moral Ideas, i. 116, 117 ; ii. 740 ; and passim. 2 Durkheim, 'La prohibition de I'inceste et ses origines,' in L'annte sociologique, i. 64. Professor Durkheim refers in this connection to an article by Dr. Simmel, ' Die Verwandtenehe,' in Vossische Zeitung, June 3rd and 10th, 1894. But I cannot find that Dr. Simmel is really opposed to my view. He only says, " Das intime Beisammenleben wirkt keineswegs nur abstumpfend, sondern in vielen Fallen gerade anreizend, sonst wiirde die alte Erfahrung nicht gelten, dass die Liebe, wo sie beim Eingehen der Ehe fehlte, oft im Laufe derselben entsteht." 3 Bloch, Sexual Life of Our Time, p. 209. xx EXOGAMY 199 It has been argued that " the noisome list " of peoples practising adelphic incest is hostile, or even fatal, to my theory of sexual aversion among young camp-mates, whether brothers and sisters or not.1 But, as a matter of fact, the number of peoples reported to allow marriages between the nearest relatives is infinitesimal in comparison with the number of peoples who are known to prohibit such marriages. Moreover, as we have noticed above, some of the statements are obviously incorrect, others are of doubtful accuracy, and in many cases it is impossible to decide whether the statement refers to full or to half-brothers and sisters. The distinction between these two kinds of brothers and sisters is of importance in the present connection. We have seen that where marriage with a half-sister is allowed the brother and sister in nearly every case have the same father. This is explained by the fact that the children of different mothers are not brought into the same contact with one another as the children of the same mother. In polygynous families each wife and her children form a small group, very often living in a separate hut, and hatred and rivalry are of no rare occurrence among the members of the various sub-families. Nor does the father occupy the same place in each sub-family as he does in a monogamous family. Kubary says that in the Pelew Islands, where sexual relations between father and daughter occur although they are disapproved of, it very seldom happens that the several wives of the same family even see each other.2 With reference to the Athenian law which permitted a man to marry his half-sister by the father, Hume made the remark that " his step-mother and her children were as much shut up from him as the women of any other family."3 After speaking of the marriage of half-brother and half- sister allowed among the ancient Arabs, Robertson Smith observes, " Whatever is the origin of bars to marriage, they 1 Hartland, ' Totemism and Exogamy,' in Folk-Lore, xxii. 365 sq. Lang, ' Totemism and Exogamy,' ibid. xxii. 48S. 2 Kubary, Die socialen Einrichtungen der Pelauer, p. 62. 3 Hume, ' An Enquiry concerning the Principles of Morals, ' sec. iv., in Philosophical Works, iv. 199. 200 THE HISTORY OF HUMAN MARRIAGE chap. certainly are early associated with the feeling that it is indecent for housemates to intermarry."1 To the reported cases of marriages between parents and children and between brothers and sisters may no doubt be added occasional cases of non-matrimonial inces- tuous intercourse occurring both among savage and civilised peoples.2 In Europe they seem to have been more frequent in certain epochs, as in the period of the French Rococo.3 Mundt states that in the middle of the last century it was not very uncommon for French fathers to live in concubinage with their own daughters ; and he thinks that, generally, the French nature is not repelled to the same degree as the German by the idea of sexual union between persons nearly related by blood.4 "At the present time among civilised people," says Mr. Mortimer, " incest is not very rare, especially among the lowest classes in large cities. . . . And in the higher classes of our community intercourse between boys and girls of the same family is not so infrequent 1 Robertson Smith, Kinship and Marriage in Early Arabia, p. 170. 2 Latcham, ' Ethnology of the Araucanos,' in Jour. Roy. Anthr. Inst, xxxix. 354, 356, 361. Keating, Narrative of an Expedition to the Source of St. Peter's River, ii. 170 (Chippewa), v. Langsdorf, Voyages and Travels in Various Parts of the World, ii. 64 (Kaniagmiut). Bogoras, Chukchee, p. 576. Laufer, ' Preliminary Notes on Explora- tions among the Amoor Tribes,' in American Anthropologist, N.S. ii. 318 sq. Sumner, ' Yakuts, abridged from the Russian of Sieroshevski,' in Jour. Anthr. Inst. xxxi. 89 (where it is said that incest rather causes laughter than horror among the Yakut, and that it is met more frequently among them than among Russians ; whereas Dr. Jochelson states [Koryak, p. 734) that cohabitation of the nearest blood-relations are nowhere more frequent than in the hamlets of the Russians or Russianised natives in North-Eastern Siberia). MondUras, * Renseignements ethnographiques sur la Cochinchine,' in Bull. Soc. d'Anthr. Paris, ser. ii. vol. x. 118 (Annamese). Grandidier, Ethnographic de Madagascar, ii. 152 sqq. Czekanowski, ' Die anthropologisch-ethnographischen Arbeiten der Expedition S. H. des Herzogs Adolf Friedrich zu Mecklenburg,' in Zeitschr. f. Ethnol. xli. 597 (Avungura). Maurer, quoted by Bartels, ' Islandischer Brauch und Volksglaube in Bezug auf die Nach- kommenschaft,' ibid, xxxii. 55 (Icelandic tales). 3 Bloch, op. cit. p. 639. 4 Mundt, Pariser Kaiser-Skizzen, i. 141 sq. xx EXOGAMY 201 as most parents and guardians suppose ; indeed I have heard of several cases amongst brothers and sisters of twelve and a few years older, though after puberty the instances become much rarer. ... In by far the greater number 01 cases of incest in civilised countries, the intercourse is restricted to children before the age of puberty."1 Yet I do not understand how these facts could invalidate my theory. They are after all quite exceptional, and I have spoken of a general rule. I have no doubt that in the world generally, and in some countries particularly, homo- sexual practices are infinitely more frequent than incest; and nevertheless nobody would consider their frequency to be " fatal," or even " hostile," to the common view that there is normally a feeling of love between the sexes. Con- sidering the great variability to which the sexual instinct is subject, it is not astonishing that cases of what we consider incestuous intercourse do occur; it seems to me more remarkable that the exceptions to the rule should, compara- tively speaking, be so few. We must not forget that a lack of desire, and even a positive feeling of aversion, may in certain circumstances be overcome. The sexual instinct is so powerful that when it cannot be gratified in the normal manner it may seek for abnormal gratification. Thus homosexual practices are very frequently due to the absence of women,2 to say nothing of masturbation and bestiality. So also sexual intercourse with a near relative may be resorted to when another, more suitable, partner is out of reach. This may account for the practice of incest in small isolated communities, like those of some Brazilian Indians, the Chukchee, and the Bushmen, if certain reports of them are true. The brother-and-sister marriages among the Wateita are said to be caused by the difficulty experienced by young men in procuring the proper number of cows for the purchase of a wife.3 Mr. Thomas suggests that the marriage with a half-sister at Ososo in 1 Mortimer, Chapters on Human Love, p. 71. 2 Westermarck, Origin and Development of the Moral Ideas, ii. 466 sq, 3 Thomson, Through Masai Land, p. 51. THE HISTORY OF HUMAN MARRIAGE chap. 202 Nigeria may be put down to a scarcity of women.1 In Europe at the present day, says Dr. Bloch, " incest occurs almost exclusively as the result of chance associations-as, for example, in alcoholic intoxication, in consequence of close domestic intimacy in small dwellings, in the absence of other opportunity for sexual intercourse."2 As to the consanguineous marriages in royal families there can be little doubt that they are carried out with the aim of maintaining the purity of the royal blood.3 The suggestion has also been made that these marriages were introduced for the purpose, of giving the king's son the right of succession which, under a system of female kinship, was enjoyed either by the son of the king's sister or by the husband of the king's daughter.4 But this explanation has the disadvantage that it presumes the earlier existence of the system of female kinship among peoples who are not known to have had it; whereas the desire to ensure the purity of the royal blood is a well-known fact. The marriage with sisters in ancient Egypt has been explained as a method of keeping property, and especially landed property, together in the family.5 So also we are told that the extremely rare instances of brother-and-sister marriage among the Maori " generally arose from the desire to keep lands belong- ing to the woman in the same line as that of the man."8 As to the practice of next-of-kin marriage in ancient Persia, 1 Thomas, Anthropological Report on the Edo-speaking Peoples of Nigeria, i. 62. 2 Bloch, op. cit. p. 639. 3 See Malo, Hawaiian Antiquities, p. 80 ; Rivers, History of Melanesian Society, i. 382 ; Idem, ' Marriage (Introductory and Primitive),' in Hastings, op. cit. viii. 425 ; Shway Yoe, The Burman, p. 60 ; Alice C. Cook, ' Aborigines of the Canary Islands,' in American Anthropologist, N.S. ii. 478 ; Gilbert Murray, Rise of the Greek Epic, p. 144. 4 McLennan, The Patriarchal Theory, p. 95. Frazer, The Dying God, p. 193 sq. Griffith, ' Marriage (Egyptian),' in Hastings, op. cit. viii. 443. Miss White, ' Women in Ptolemaic Egypt,' in Journal of Hellenic Studies, xviii. 265. Miss Murray, ' Royal Marriages and Matrilineal Descent,' in Jour. Roy. Anthr. Inst. xlv. 308 sq. 5 Nietzold, Die Ehe in Agypten zur ptolemdisch-rdmischen Zeit, p. 13. Frazer, Adonis Attis Osiris, ii. 215 sq. fi Tregear, The Maori Race, p. 298. See also Shortland, Traditions and Superstitions of the New Zealanders, p. 119 sq. xx EXOGAMY 203 Mr, West observes that there were evidently two reasons for its establishment and continuance. One was the indis- pensable necessity of offspring, unfettered by duties towards any other family, for the purpose of maintaining the neces- sary periodical ceremonies for the souls of those passed away. The other was the wish of preventing any risk of religious perversion consequent upon marrying into a family of strangers or infidels. Both of these reasons, he adds, must have become intensified as the Mazda-worshippers diminished in numbers ; hence the increasing vehemence of priestly advocacy, until the foreign conquerors probably interfered and put a stop to the practice.1 Sir James G. Frazer has argued against me that if exogamy had resulted from a natural instinct there would have been no need to reinforce that instinct by legal pains and penalties ; the law only forbids men to do what their instincts incline them to do, and hence we may always safely assume that crimes forbidden by law are crimes which many men have a natural propensity to commit.2 This argument, which has been quoted with much appreciation by Dr. Freud,3 implies a curious misconception of the origin of legal pro- hibitions. Of course, where there is no transgression there is no law. But Sir James cannot be ignorant of the varia- bility of instincts and of the great variability of the sexual instinct, nor of the fact that there are circumstances in which a natural sentiment may be blunted or overcome. Would he maintain that there can be no general aversion to bestiality because bestiality is forbidden by law, and that the exceptional severity with which parricide is treated by many law-books proves that a large number of men have a natural propensity to kill their parents ? The law expresses 1 West, * The Meaning of Khvetuk-das or Khve'.hdad,' in Sacred Books of the East, xviii. 429. Cf. Geiger, Civilization of the Eastern Iranians, i. 66 sq. ; Bapp, ' Die Religion und Sitte der Perser und iibrigen Iranier nach den griechischen und romischen Quellen,' in Zeitschr. Deutsch. Morgenland. Gesellsch. xx. 113. 2 Frazer, Totemism and Exogamy, iv. 97. 3 Freud, Totem und Tabu, p. 114. A similar argument against my theory has been brought forward by Lord Avebury {Marriage, Totemism, and Religion, p. 52) 204 THE HISTORY OF HUMAN MARRIAGE chap. the general feelings of the community and punishes acts that shock them ; but it does not tell us whether an inclina- tion to commit the forbidden act is felt by many or by few. Frazer has also raised another, more important, objection to my theory, for which I owe him gratitude because it compels me to discuss in detail a point which I treated too lightly in the earlier editions of this book. In spite of his view on the origin of legal prohibitions just referred to, he admits that there seems to be some ground for believing in the existence of " a natural aversion to, or at least a want of inchnation for, sexual intercourse between persons who have been brought up closely together from early youth " ; but he finds it difficult to understand how this could have been changed into an aversion to sexual intercourse with persons near of kin, and maintains that, till I explain this satis- factorily, the chain of reasoning by which I support my theory breaks down entirely at the crucial point.1 And in this objection, also, he has the hearty support of Dr. Freud.2 My answer to this is that the transition which Frazer finds so difficult to understand is not only possible and natural but well-nigh proved by an exactly analogous case of equally world-wide occurrence and of still greater social importance, 1 Frazer, Totemism and Exogamy, iv. 96 sq. 2 Freud, Totem und Tabu, p. 114. Dr. Freud adds (p. 115) that " die Erfahrungen der Psychoanalyse die Annahme einer angeborenen Abneigung gegen den Inzestverkehr vollends unmoglich machen. Sie haben im Gegenteile gelehrt, dass die erstcn sexuellen Regungen des jugendlichen Menschen regelmassig inzestuoser Natur sind, und dass solche verdrangte Regungen als Triebkrafte der spateren Neurosen eine kaum zu tiberschatzende Rolle spielen." That the results of the so-called psycho-analysis are destructive to my theory is a supposition for which I must see some evidence before I can take it seriously. In Dr. Freud's terminology, " sexuelle Regungen " may imply mental states very different from the desire for sexual intercourse ; and the study of neurotic persons can hardly be regarded as a safe guide to the proper understanding of the normal manifestations of the sexual instinct. Dr. Jung, Freud's most distinguished disciple, says (Collected Papers on Analytical Psychology, p. 231 ; cf. Idem, Psychology of the Unconscious, p. 463), I am able to attribute as little strength to incestuous desires in childhood as in primitive humanity." xx EXOGAMY 205 namely, the process which has led to the association of all kinds of social rights and duties with kinship. As I have pointed out elsewhere,1 the maternal and paternal senti- ments, which largely are at the bottom of parental duties and rights, cannot in their simplest forms be based on a knowledge of blood-relationship, but respond to stimuli derived from other circumstances, notably the proximity of the helpless young, that is, the external relationship in which the offspring from the beginning stand to the parents. Nor is the so-called filial love in the first instance rooted in considerations of kinship ; it is essentially retributive, the agreeable feeling produced by benefits received making the individual look with pleasure and kindliness upon the giver. Here again the affection is ultimately due to close living together, and is further strengthened by it, as appears from the cooling effect of long separation of children from their parents. So also fraternal love and the duties and rights which have sprung from it depend in the first place on other circumstances than the idea of a common blood ; and the same may be said of the tie which binds together relatives more remotely allied. Its social force is ultimatelv derived from near relatives' habit of living together. " Men became gregarious by remaining in the circle where they were born ; if, instead of keeping together with their kindred, they had preferred to isolate themselves or to unite with strangers, there would certainly be no blood-bond at all. The mutual attachment and the social rights and duties which resulted from this gregarious condition were associated with the relation in which members of the group stood to one another -the relation of kinship as expressed by a common name,- and these associations might last even after the local tie was broken," being kept up by the common name.2 Here we have an immense group of facts which, though ultimately depending upon close living together, have been interpreted in terms of kinship. Why, then, may we not believe that the same has been the case with the aversion to incest and the prohibitory rules resulting from it ? 1 Origin and Development of the Moral Ideas, vol. ii. ch. xxxiv. 2 Ibid. ii. 203. 206 THE HISTORY OF HUMAN MARRIAGE chap. Frazer asks :-" If the root of the whole matter is a horror of marriage between persons who have always lived, with each other, how comes it that at the present day that horror has been weakened into a mere general preference for marriage with persons whose attractions have not been blunted by long familiarity ? . . . Why should the marriage of a brother with a sister, or of a mother with a son, excite the deepest detestation, . . . while the origin of it all, the marriage between housemates, should excite at most a mild surprise too slight probably to suggest even a subject for a farce, and should be as legitimate in the eye of the law among all civilised nations as any other marriage ? " 1 For my own part, I believe that marriage between a man and his foster-daughter or between a foster-brother and a foster- sister, in case the social relations between them have been exactly similar to those of blood-relatives of corresponding degrees, would cause more than a mild surprise, and appear unnatural and objectionable. Much, of course, depends upon the closeness of the social relationship and its nearness in time. Dr. Steinmetz's argument, that " the very sensual Frenchmen often seem to marry the lady friends of their earliest youth,"2 is certainly not to the point. Speaking of marriages between housemates among civilised peoples, Mr. Heape justly observes :-•" It must be recollected that, as civilisation progresses, the continuous living together from childhood upwards of blood strangers becomes more and more rare. It is also true that the Male's temporary absences from home become more and more frequent after puberty, and that housemates before puberty thus become strangers afterwards if considered from a sexual point of view. In fiction, which is a wonderfully accurate guide in such matters, the marriage of housemates is very rare, and when it is depicted it generally follows unsuccessful efforts of the Male to gain a stranger for wife."3 I certainly do not deny that unions between the nearest blood-relatives 1 Frazer, op. cit. iv. 96 sq. See also Steinmetz, ' Die neueren Forschungen zur Geschichte der menschlichen Familie,' in Zeitschr. f. Socialwiss ii. 818 sq. 2 Steinmetz, loc. cit. p. 818. 3 Heape, Sex Antagonism, p. 62 sq. XX EXOGAMY 207 inspire an aversion of their own ; but it is quite natural that they should do so, considering that from earliest times the aversion to sexual intercourse between persons living closely together has been expressed in prohibitions against unions between kindred, sanctioned by custom, law, and religion. Nor can it be a matter of surprise that the pro- hibitory rules so frequently refer to the marriage of kindred alone. Law only takes into account general and well- defined cases, and hence relationships of some kind or other between persons who are nearly always kindred are defined in terms of blood-relationship. This is true not only of the prohibitions of incest, but of many duties and rights inside the family circle. At the same time, as we have seen, adoption or fosterage also constitutes a very frequent bar to intermarriage. Among various peoples marriage is prohibited even between all persons belonging to the same village or other local group, whether they are related by blood or not. And innumerable facts show that the extent to which relatives are forbidden to intermarry is nearly connected with their close living together. Montesquieu observed long ago that marriage between cousins was prohibited by peoples among whom brothers and their children used to live in the same house. " Chez ces peuples," he says, " le manage entre cousins germains doit etre regarde comme contraire a la nature ; chez les autres, non."1 A comparison between the forbidden degrees of the Greeks and Romans is instructive. Among the former cousins and even half-brothers and sisters were allowed to intermarry, whereas among the latter marriage was prohibited also between more distantly related persons. This difference, as Rossbach points out, was due to the fact that the family feeling of the Greeks was much weaker than that of the Romans, among whom, in early times, a son used to remain in his father's house even after marriage, so that cousins on the father's side were brought up as brothers and sisters. Later on, the several families separated from the common household, and the prohibited degrees were 1 Montesquieu, op. cit. book xxvi. ch. 14. 208 THE HISTORY OF HUMAN MARRIAGE chap retrenched accordingly.1 Professor Kohler points out the con- nection between the extensive marriage prohibitions of the Hindus and their large households.2 The word gotra (clan) means a stall or cow-pen, and would thus originally signify those who lived together in one place like a herd of cattle.3 Many of the Southern Slavs to this day live in house communities each consisting of a body of from ten to sixty members or even more, who are blood-relations to the second or third degree on the male side, and who associate in a common dwelling or group of dwellings, having their land in common, following a common occupation, and being governed by a common chief.4 Sir Henry Maine remarks, " The common residence of so many persons of both sexes in the same household may be said to be only possible through their belief that any union of kinsmen and kins- women would be incestuous."5 In Wales, m former days, the exogamous joint-family consisted of kindred who dwelt together within one enclosure.6 " So long as the head of the family lived, all his descendants lived with him, apparently in the same homestead, unless new ones had already been built for them on the family land."7 The ancient Teutons, on the other hand, whose exogamous prohibitions seemed to have included only the nearest relatives, are not known to have had the joint-family institution, and, according to Tacitus, lived in scattered families at some distance from each other.8 1 Rossbach, Untersuchungen uber die rbmischeEhe, pp. 421-423, 429, 439. See also Fowler, ' Marriage (Roman),' in Hastings, op. cit.Nni. 464. 2 Kohler, ' Indisches Ehe- und Familienrecht,' in Zeitschr. f. vergl. Rechtswiss. iii. 362. 3 Russell, Tribes and Castes of the Central Provinces of India, i. 156. 4 Krauss, Sitte und Brauch der Sudslaven, pp. 75, 79 sqq. Maine, Dissertations on Early Law and Custom, p. 241 sqq. Utiesenovid, Die Hauskommunionen der Sudslaven, p. 20 sqq. Miler, ' Die Haus- kommunion der Sudslaven, ' in fahrbuch d. internal. Vereinigung f. vergl. Rechtswiss. iii. 199 sqq. 6 Maine, op. cit. p. 254 sq. 6 Lewis, Ancient Laws of Wales, pp. 56, 57, 196. 7 Seebohm, English Village Community, p. 193. Idem, Tribal System in Wales, p. 89 sqq. 8 Tacitus, Germania, ch. 16. See also Leist, Alt-arisches fus Civile, i. Anhang i. xx EXOGAMY 209 It is easy to explain, says Ewald, why among the Hebrews marriage between cousins was permitted though marriage between brothers and sisters was forbidden : the former " did not form one united household, and the more each house stood strictly by itself in the ancient fashion, the wider seemed the separation between cousins."1 I have already referred to Robertson Smith's statement relating to the ancient Arabs, that the intermarriage of housemates was held indecent. A Chinese woman, on marriage, alienates herself from her own family to be incorporated into that of her husband; hence, as Medhurst observes, children of brothers and sisters may marry at pleasure, while those of brothers cannot be united on pain of death.2 Among many of the lower races we notice a similar corre- lation between close living together and the prohibited degrees of relationship. Among peoples living in small family groups, like many of the Brazilian Indians and the Bushmen, the exogamous prohibitions are restricted to the nearest relatives only. Among the Senoi, one of the abori- ginal tribes of the Malay Peninsula, as we are told by Dr. Martin, a man " does not marry inside the local community, which mostly consists of a single Grossfamilie, but takes a wife from a neighbouring group, without asking whether they are blood-relatives or not."3 Speaking of some tribes of the North-West Amazons Mr. Whiffen says :- " Incest is unknown among them, and in that term I include promiscuous intercourse among any of the members of a household. The antipathy to this lies only between those living under the same roof, it does not extend to consan- guineous individuals who are members of different house- holds."4 Of the Pankas, a low weaving and watchman tribe in South Mirzapur, again, it is said that they " do not marry in their own family as long as the members are united and live together, no matter how distant relatives 1 Ewald, Antiquities of Israel, p. 197 sq. 2 Medhurst, ' Marriage, Affinity, and Inheritance in China,' in Trans. Roy. Asiatic Soc. China Branch, iv. 24 n. $ 3 Martin, Die Inlandstdmme der Malayischen Halbinsel, p. 872. 4 Whiffen, North-West Amazons, p. 262. VOL. II p 210 THE HISTORY OF HUMAN MARRIAGE chap. may reside under the same roof."1 " In Nanusa," Dr. Hickson remarks, " I understood that marriage was not permitted between members of the same household."2 Among the Rejangs of Sumatra, according to Marsden, a marriage must not take place between relatives within the third degree ; " but there are exceptions for the descendants of females who, passing into other families, become as strangers."3 Among the Mafulu mountain people of British New Guinea, whilst marriages within the clan are regarded as irregular and undesirable, marriages within the village are considered still more so ; each clan has its own villages, but two clans are never found represented in any one village.4 In the western islands of Torres Straits, as said above, the prohibition to marry a woman of the same totem clan did not extend to women of other islands.5 No Masai may marry a woman belonging to the same sub-division as himself " if both families live in the same district."6 An exogamous clan is very frequently a territorial group as well as a group of kindred. To take some examples. Among the Western Islanders of Torres Straits " the members of each clan lived in one locality, at any rate in Mabuiag."7 Among the Kiwai Papuans " each communal house is sup- posed to be inhabited by a single totemic clan, but some- times two or even three clans occupy separate parts of the same house."8 Among the Orang Mamaq on the eastern side of Sumatra, who are divided into a number of matrilineal exogamous clans called suku, the members of a suku live together and are very closely united to one another.9 Among 1 Crooke, Tribes and Castes of the North-Western Provinces and Oudh, iv. 114. 2 Hickson, A N aturalist in North Celebes, p. 197. 3 Marsden, History of Sumatra, p. 228. 4 Williamson, Mafulu Mountain People of British New Guinea, pp. t68, 83 sq. 6 Supra, ii. 124. 0 Hollis, Masai, p. 303. 7 Haddon and Rivers, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 159. 8 Landtman, Papuan Magic in the Building of Houses, (Acta Academiae Aboensis. Humaniora, i. no. 5), p. 9. 9 Graafland, ' De verbreiding van het Matriarchaat in het Land- schap Indragiri, in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, xxxix. 43. xx EXOGAMY 211 the Pasemahers of Southern Sumatra 1 and the aborigines of Burn,2 who trace descent through the father, each exo- gamous clan inhabits a separate district, so that local exogamy coincides with clan exogamy. Among the Kandhs of Orissa the clanspeople as a rule dwell together in the village or group of villages from which they take their name.3 The Toda clan system is territorial; each clan owns a number of villages, which are in general situated in the same part of the hills.4 So also among the Baganda " each clan had its family-estates, which were as a rule situated on some hill, with the gardens running down into the valley."5 Among the Fans, whose villages are exo- gamous, a village is essentially " un clan familiale."® In the tribes of the Upper Congo and its northern tributaries spoken of above,7 each exogamous clan occupies its own well-marked geographical district.8 Among the Pueblo Indians the members of each clan and phratry seem formerly to have dwelt together ; and to some extent this local grouping of kins may still be traced in the villages and towns of these Indians.9 " Related clans," we are told, " commonly built together, the newcomers seeking and usually obtaining permission to build with their kindred ; thus clusters of rooms were formed, each inhabited by a clan or a phratry."10 There are reasons for believing that most of the Navaho clans were originally local exo- gamous groups.11 In the towns of the Creeks " each group of 1 Wilken, Over de verwantschap en het huwelijks- en erfrecht bij de volken van het maleische ras, p. 13. 2 Wilier, Het eiland Boeroe, p. 100. 3 Risley, Tribes and Castes of Bengal, i. 400. 4 Rivers, Todas, p. 540 s^. 5 Roscoe, Baganda, p. 133. 0 Martrou, in Anthropos, i. 754. 7 Supra, ii. 148. 8 de Calonne Beaufaict, ' Zool&trie et totdmisme chez les peuplades septentrionales du Congo Beige,' in Revue des etudes ethnographiques el sociologiques, ii. 194. Czekanowski, in Zeitschr. f. Ethnol. xli. 596- 8 Frazer, Totemism and Exogamy, iii. 226. 10 Mindeleff, ' Localization of Tusayan Clans,' in Ann. Rep. Bur. Ethn. xix. 648. 11 Matthews, Navaho Legends, p. 31. Idem, ' Gentile System of the Navajo Indians,' in Jour. American Folk-Lore, iii. 104 sq. Bourke, ' Notes upon the Gentile Organization of the Apaches of Arizona, ibid. iii. 118, 120. 212 THE HISTORY OF HUMAN MARRIAGE chap. houses contained people of one gens only " ; and a similar distribution is observed in the villages and hunting and war camps of the Pani and southern Dakotan tribes and was very strictly enforced by them.1 The Kansa, whom Owen Dorsey found divided into sixteen clans grouped in seven phratries, camped in a circle, in which the clans had their fixed places. Eight clans camped in the semicircle on the left side of the line of march, and the other eight clans camped in the semicircle on the right side of the line of march. These two halves or sides of the tribe formed what may be called superior phratries, since no man was allowed to marry a wife from his side of the tribal circle.2 Commenting on this local segregation of the exogamous groups among the Kansa, and apparently among all the other Dakotan tribes of the Missouri who camped in circles, Sir James G. Frazer suggests that " it may help us to under- stand the method in which the somewhat complex relations between the groups were kept clear in the minds of those who had to observe them." It seems to him probable that when exogamy was first instituted the groups within which marriage was prohibited or the intermarrying classes into which the community dissolved itself were locally segregated from each other " in order to avoid confusion and mistakes."3 He thinks that this is true also of the exogamous moieties of the Australian tribes, and finds his conjecture supported by existing facts.4 In each of the Warramunga tribes in Central Australia the two exogamous moieties occupy separate territories, the Uluuru moiety inhabiting the southern territory, and the Kingilli moiety inhabiting the northern territory, with a more or less sharply marked boundary-line dividing them; and the totemic clans in like manner are locally divided between the two districts, all the clans that belong to the Uluuru moiety being found in the Uluuru district, and all the clans that belong to the 1 Gatschet, Migration Legend of the Creek Indians, i. 154. 2 Dorsey, ' Siouan Sociology,' in Ann. Rep. Bur. Ethn. xv. 230, 232. Cf. Frazer, Totemism and Exogamy, iii. 124. 3 Frazer, Totemism and Exogamy, iii. 124, 125, 358. 4 Ibid. i. 246 sqq. XX EXOGAMY 213 Kingilli moiety being found in the Kingilli district.1 Frazer, however, does not think it necessary to suppose that from the outset the whole country of the tribe was thus parcelled out into two great areas, of which one was assigned to one half of the tribe and the other to the other half : it would be enough that every local group should split into two sections, each of which should have its own camping and hunting grounds. And he observes that a trace of this practice seems to survive in the Arunta custom, according to which people of the same exogamous moiety always camp together and apart from the people of the other moiety, the two camps being regularly separated by some natural feature, such as a creek.2 It may be added that in the southern tribes of the Kulin nation the members of the two exogamous classes, Eagle-hawk and Crow, were also segregated from each other and dwelt in separate districts,3 and that in the Kariera and Mardudhunera tribes of Western Australia the members of one local group belong to only one couple of classes.4 Like Sir James Frazer, I hold it extremely probable that exogamous moieties or phratries and clans have originally been local groups with their own territories, but I cannot possibly believe that they were locally segregated from other phratries or clans for the purpose of more easily learning and more strictly observing the rule of exogamy. It cannot, after all, be so very difficult to learn that you must not marry women bearing the name of your own phratry or clan ; and I utterly fail to see how the Australian segregation could have taken place, as Frazer suggests,5 " in order to secure that the men and women who were forbidden to each other should not normally meet." I should have thought that the women whom a man normally meets are just those who belong to his own community; and where the exogamous moiety is a local group, the unmarried 1 Spencer and Gillen, Northern Tribes of Central Australia, p. 28 sqq. 2 lidem. Native Tribes of Central Australia, p. 31 sq. 3 Howitt, Native Tribes of South-East Australia, p. 127. 4 Brown, ' Three Tribes of Western Australia,' in Jour. Roy. Anthr. Inst, xliii. 43. 5 Frazer, op. cit. i. 248. 214 THE HISTORY OF HUMAN MARRIAGE chap. women in it are forbidden to him because they belong to his own moiety and the married ones who have been intro- duced into the group-with the exception of his own wife or wives-are forbidden to him because they are the wives of other men. But whilst exogamy cannot be the cause of the local separation of a phratry or clan, it is quite in agreement with my theory if the living together of its members is a cause of its exogamy. Very frequently, however, the members of an exogamous clan or phratry do not live in the same locality; indeed, as we have seen, the exogamous prohibition may even refer to the members of the same or an " equivalent " clan or phratry who ar- living in different tribes or nations.1 This is a natural consequence of the fact that the aversion to sexual intercourse between persons living closely together from childhood has been expressed in prohibitions against unions between kindred. The exogamous rules, though in the first place associated with kinship because near relatives normally live together, have come to include relatives who do not five together-just as social rights and duties connected with kinship, although ultimately depending upon close living together, have a strong tendency to last after the local tie is broken. Clan exogamy has its counterpart, for instance, in the blood feud as a duty incum- bent on the whole clan, whether the members of the clan live together or not. In this process the influence of a common name has undoubtedly been of great importance. As kinship is traced by means of a system of names, the name comes to stand for blood-relationship. This system is naturally one-sided. Though it will keep up the record of descent either on the male or on the female side, it cannot do both at once ;2 and the line which has not been kept up by such means of record, even when it is recognised as a line of relationship, is more or less neglected and soon forgotten. Hence the prohibited degrees, like the social rights and duties generally connected with clanship, extend much farther 1 Supra, ii. 108, 140. 2 Cf. Tylor, Early History of Mankind, p. 285 sq. XX EXOGAMY 215 on the one side than on the other side. It should also be remembered that, according to primitive ideas, the name itself constitutes a mystic link between those who have it in common. " In Greenland, as everywhere else," says Dr. Nansen, " the name is of great importance ; it is believed that there is a spiritual affinity between two people of the same name."1 The feeling against incest, however, may also be opposed to intercourse between persons one of whom has or has had sexual relations with the other's relative ; and, generally speaking, the feeling that two persons are intimately con- nected in some way or other may, through an association of ideas and feelings, give rise to the notion that marriage or sexual intercourse between them is improper or incestuous. Hence the prohibitions of marriage between relations by affiance, the prohibitions of the Roman and Greek Churches on the ground of " spiritual relationship," and certain other prohibitions mentioned above. But we must also take notice of the feeling of sexual shame, which is felt most acutely between relatives. Speaking of the prohibition of the Koryak against marriage with two sisters, Dr. Jochelson observes that " a Koryak who has two or more wives sleeps in one sleeping-tent with all of them, and shares his bed, now with one, and then with another. The sense of shame forbids a woman to be present at the acts of the intimate life of her sister."3 Among the Pima Indians of Arizona, we are told, cousins do not marry " out of respect of the parents for each other."3 Among the Khasis a man cannot marry his maternal uncle's daughter during the lifetime of the uncle, nor is marriage with a father's sister's daughter allowed during the lifetime of the father's sister; but after her death there is no ban, although such a union is looked upon with disfavour.4 For certain prohibitions there may also be quite particular reasons in particular cases. In Tikopia it is believed that if a man married his brother's widow " the ghost of the 1 Nansen, Eskimo Life, p. 230. 2 Jochelson, Koryak, p. 748 sq. 3 Russell, ' Pima Indians,' in Ann. Rep. Bur. Ethn. xxvi. 184. 1 Gurdon, Khasis, p. 78. THE HISTORY OF HUMAN MARRIAGE chap. 216 dead husband would injure the brother who has taken his place."1 And an old man of the Wotjobaluk tribe explained to Dr. Howitt his people's feeling against similar marriages by saying that " it was unpleasant to lie in the camp in the place of the deceased brother, and thus to be always reminded of him."2 Every hypothesis which pretends to give a fairly full explanation of the exogamous rules must inevitably assume the operation of the law of association. Hence it is some- what surprising to hear that my theory utterly fails to explain the origin of prohibitions referring to the marriage of kindred who live in different localities.3 Professor Durk- heim, while maintaining that my. theory could not apply to clan exogamy because the members of the same totem clan do not live together, is himself quite ready to resort to analogy to explain prohibitions extending outside the totem clan ; as when he says that the rule of clan exogamy has been extended to near relatives belonging to different clans because these are in no less close contact with each other than are the members of the same clan.4 Herr Cunow, again, assumes that the rule which forbids marriage between unrelated persons living in the same community has sprung from the prohibition of marriage between kindred because the local group very often coincides'with the kinship or totem group, whilst he rejects the idea that close living together could have given rise to the prohibition of marriage between kindred.6 That the extent of the prohibited degrees is closely con- nected with social intimacy, whether combined with actual living together or not, is also suggested by the regular co-existence of comprehensive exogamous prohibitions and the classificatory system of relationship. This was first pointed out by Tylor, who by his method of adhesions even was led to the conclusion that clan exogamy and classi- 1 Rivers, History of Melanesian Society, i. 346. 2 Howitt, Native Tribes of South-East Australia, p. 248. 3 Cunow, Die Verwandtschafts-Organisationen der Australneger, p. 185. Durkheim, in L'annee sociologique, i. 39 n. 2. Steinmetz, in Zeitschr. f. Socialwiss. ii. 819, 4 Durkheim, lac. cit. p. 19, & Cunow, op. cit. p. 187, XX EXOGAMY 217 ficatory relationship are in fact two sides of one institution. " In reckoning," he says, " from the present schedules the number of peoples who use relationship names more or less corresponding to the classificatory systems here considered, they are found to be fifty-three, and the estimated number of these which might coincide accidentally with exogamy, were there no close connection between them, would be about twelve. But in fact the number of peoples who have both exogamy and classification is thirty-three, this strong coincidence being the measure of the close causal connection subsisting between the two institutions. The adherence is even stronger as to cross-cousin marriage (i.e., that the children of two brothers may not marry, nor the children of two sisters, though the child of the brother may marry the child of the sister), of which twenty-one cases appear in the schedules, no less than fifteen of the peoples practising it being also known as exogamous." 1 Sir James G. Frazer observes, " So far as we can see at present, it may be laid down as a general rule, that every people who practise totemism and exogamy count their relationships according to the classificatory system."2 We have previously found reasons to believe that the classificatory terms are funda- mentally influenced by social relationships.3 And it seems to me highly probable that the same feelings of intimacy and kinship as have, for instance, led to the classification of cousins with brothers and sisters have also led to the rule which treats sexual intercourse between them as incest. I venture to believe that the' hypothesis here advocated is in agreement with the various facts given in the last chapter. It seems to me to explain how the aversion to incest may be independent of both experience and education ; why the exogamous rules refer not only to relations by blood, but very frequently to persons not at all so related ; why the prohibitions of consanguineous marriages vary so considerably with regard to the prohibited degrees, applying, 1 Tylor, ' On a Method of investigating the Development of Institutions,' in Jour. Anthr. Inst, xviii. 264. 2 Frazer, Totemism and Exogamy, ii. 330. 3 Supra, i. 250 sqq. 218 THE HISTORY OF HUMAN MARRIAGE chaf however, almost universally to persons living in the closest contact with each other; and why these prohibitions are so commonly extended much farther on one side, either the paternal or the maternal, than on the other. But an important question still calls for an answer :-How shall we explain the lack of inclination for and the positive aversion to sexual intercourse between persons who have been living closely together from childhood ?1 Before this question has been answered my theory of the origin of exogamy is incomplete. The sexual instinct is of such immense importance for the existence of the species that any satisfactory explanation of its normal characteristics must be sought for in their specific usefulness. That the psychical cause to which I have traced the origin of exogamy has a deep biological foundation is suggested by a multitude of facts relating both to animals and plants. These facts make it hardly possible to doubt that self-fertilisation in the case of plants and close in-breeding in the case of animals are, on the whole, injurious to the species. Darwin watched, from germination to maturity, more than a thousand individual plants, produced by crossing and self-fertilisation, belonging to fifty-seven species, fifty- two genera, and thirty large families, and including natives of the most various countries.2 This led him to the con- clusion that cross-fertilisation is generally beneficial and self-fertilisation injurious ; which is shown by the difference in height, weight, constitutional vigour, and fertility of the offspring from crossed and self-fertilised flowers, and in the number of seeds produced by the parent-plants.3 Hence, whenever plants which are the offspring of self- 1 I can by no means subscribe to Dr. Havelock Ellis's opinion (op. cit. Sexual Selection in Man, p. 206) that " the usual absence of sexual attraction between brothers and sisters requires no special explanation." 3 See Muller, Fertilisation of Flowers, p. 8. 3 Darwin, Effects of Cross, and Self Fertilisation in the Vegetable Kingdom, p. 436. XX EXOGAMY 219 fertilisation are opposed in the struggle for existence to the offspring of cross-fertilisation, the latter must have the advantage. That this is the case with the large majority of vegetable species constituted for cross-fertilisation is nowadays a generally accepted fact.1 As for the animal kingdom, Darwin remarks that almost all who have bred many kinds of animals and have written on the subject have expressed the strongest conviction on the evil effects of close interbreeding.2 Sir John Sebright writes :-" I have no doubt but that, by this practice being continued, animals would, in course of time, degenerate to such a degree as to become incapable of breeding at all. ... I have tried many experiments by breeding in-and-in upon dogs, fowls, and pigeons : the dogs became, from strong spaniels, weak and diminutive lap-dogs, the fowls became long in the legs, small in the body, and bad breeders."3 Low states that " dogs continually reproduced from the same litter exhibit, after a time, the aspect of feebleness and degeneracy. The hair becomes scanty, or falls off, the size diminishes, the limbs become slender, the eyes sunk, and all the characters of early age present themselves. Hogs have 'been made the subjects of similar experiments. After a few generations, the victims manifest the change induced in the system. They become of diminished size, the bristles are changed into hair, the limbs become feeble and short, the litters diminish in frequency and in the number of the young produced, the mother becomes unable to nourish them, and, if the experiment be carried as far as the case will allow, the feeble and frequently monstrous offspring will be incapable of being reared up, and thus the miserable race will utterly perish."4 The hog is said to be particularly susceptible to the injurious influence of in-breeding, and the sheep only a 1 See Baur, Einfuhrung in die experimentelle Vererbungslehre, p. 312 sqq. 2 Darwin, Animals and Plants under Domestication, ii. 116. See also Settegast, Die Thierzucht, p. 287 sqq. 3 Sebright, Art of Improving the Breeds of Domestic Animals, p. 12 sq. 4 Low, Domesticated Animals of the British Islands, p. Ixiii. sq. THE HISTORY OF HUMAN MARRIAGE chap. 220 little less so.1 Mr. Heape states that Dorset Horn sheep, when served by rams of their own breed, show a greater tendency to barrenness than when served by Hampshire Down rams ; but it is possible that what in this case appears to be barrenness is in reality very early abortion, the in-bred embryos tending to die at an early stage.2 Experiments made with four-horned goats have shown that much in- breeding has a weakening effect upon the bones and muscles, lowers the vital energy of the animal, and causes the loss of the sucking instinct.3 With regard to thoroughbred horses, which are notoriously in-bred, it is stated in one of the earlier reports of the Royal Commission on Horse-breeding that no less than forty per cent, of the thoroughbred mares in this country fail to have foals each year; but it seems probable that this large amount of sterility is not entirely due to the in-breeding.4 Professor Sheldon writes:- " ' In-breeding,' as it is termed-that is, the breeding for a time amongst near relations-generally results mischiev- ously on the systems and on the fertility of stock. Con- sanguinity, though valuable in the formation of pure-bred types, and up to a given point, is a great evil if carried too far. The effects of too close in-breeding, at all events in the bovine world, are most commonly infertility and general debility ; and what is only a slight and transient form of weakness in the first parents may, and often does, become a fixed and severe form in succeeding generations-indeed, what at first is only a weakened organic formation may be developed into a disease."5 So also according to Professor 1 Settegast, op. cit. p. 291. Bos, ' Untersuchungen uber die Folgen der Zucht in engster Blutsverwandtschaft,' in Biologisches Centralblatt, xiv. 81. Schultze, ' Zur Frage von den geschlechts- bildenden Ursachen,' in Archiv f. Mikroskopische Anatomic und Entwicklungsgeschichte, Ixiii. 212. 2 Heape, ' Abortion, Barrenness, and Fertility in Sheep,' in Jour. Roy. Agricultural Soc. England, ser. iii. vol. x. 248. 3 Muller, ' Inzuchtsversuch mit vierhornigen Ziegen,' in Zeitschr. f. indukt. Abstammungs- und Vererbungslehre, vol. vii., reviewed by Hirsch, in Archiv f. Rassen- und Gesellschafts-Biologie, ix. 511. 4 Marshall, Physiology of Reproduction, p. 208. 5 Sheldon, Live Stock in Health and Disease, p. 3. See also Idem, Dairy Farming, p. 4. xx EXOGAMY 221 Ewart, " nature only tolerates inbreeding up to a certain point, for while it may assist in perpetuating useful characters by inducing prepotency, it often does this at the expense of vitality-it may be of fertility as well. It is conceivable that inbreeding has played an important part in the extinc- tion of species ; it has undoubtedly been the means of deteriorating, if not actually destroying, many of the breeds and varieties artificially produced."1 Many writers maintain that if proper caution is observed in-breeding may be not only harmless but beneficial;2 and Mr. Huth quotes the evidence of numerous breeders whose choicest stocks have always been so bred.3 But in these cases, as Wallace remarks, " there has been rigid selection by which the weak or the infertile have been eliminated, and with such selection there is no doubt that the ill effects of close interbreeding can be prevented for a long time ; but this by no means proves that no ill effects are produced."4 Sir John Sebright observes that there may be families of domestic animals which even go through several generations without sustaining much injury from having been bred in-and-in ;5 and the offspring of self- fertilised plants do not always show any loss of vigour in the first generations. It seems to be an established fact that the breeding together of parent and progeny is almost always less injurious than that of brother and sister.6 But that in-breeding, in the long run at least,7 is generally productive of evil consequences of some kind or other can hardly be open to doubt. 1 Ewart, Penycuik Experiments, p. xliv. See also ibid. p. xlix. sqq. 2 Steinau, Pathological and Philosophical Essay on Hereditary Diseases, p. 38 sq. De Chapeaurouge, Einiges uber Inzucht und ihre Leistung auf verschiedenen Zuchtgebieten, reviewed by Roemer, in Archiv f. Rassen- und Gesellschafts-Biologie, vii. 233 sqq. Schmehl, Inzuchtstudien in einer deutschen Rambouillet-Stammschdferei, reviewed by Roemer, ibid. x. 194. Hesse, Inzucht- und Verer- bungsstudien bei Rindern, reviewed by Hilzheimer, ibid. xi. 519. 3 Huth, Marriage of Near Kin, ch. vi. 4 Wallace, Darwinism, p. 161. 5 Sebright, op. cit. p. 12. u Sheldon, Lite Stock in Health and Disease, p. 3. Idem, Dairy Farming, p. 5. Bos, loc. cit. p. 77. THE HISTORY OF HUMAN MARRIAGE chap. 222 The opinion on this point of eminent breeders is supported by many experiments made by biologists. According to Crampe's experiments with the brown rat {Mus decu- nianus'}, thirty-nine animals out of 153 born of related parents, that is, 25-5 per cent., died soon after birth, whereas of 299 animals of parents not related this was the case with twenty-eight only, that is, 8'4 per cent. The animals of incestuous broods were much smaller and lighter than others, and their fecundity was diminished.1 Ritzema Bos, who in the course of six years experimented with several thousands of rats, likewise found that continued close in-breeding reduced the fertility of the animals ; the number of infertile pairings increased and the average number of young per litter decreased. The vitality of the young and their mother's capacity of nourishing them were also diminished, whereas there was not so great reduction of size and weight, nor so much disease, as in the case of Crampe's rats-a difference which Bos explains by assuming that the animals with which Crampe started his experiments were weak and sickly.2 The result of von Guaita's experi- ments with mice was that the continued pairing together of brothers and sisters generally led to diminished fertility.3 Mr. Huth, who denies that in-breeding, however close, has proved to be in itself hurtful, nevertheless observed, when breeding rabbits in-and-in, that " after the fourth generation there was a diminution of fecundity analogous to the disgust that the stomach would feel at the same diet long continued." He noticed, however, no other evil effects; on the contrary, the in-bred offspring were some- what heavier than the non-related parent animals.4 Preyer made a similar observation with regard to guinea-pigs : 1 Crampe, ' Zuchtversuche mit zahmen-Wanderratten,' in Land- ivirthschaftliche Jahrbucher, xii., quoted by Diising, Die Regulierung des Geschlechtsverhdltnisses bei dev Vermehrung der Menschen, Tiere und Pflanzen, p. 246. 2 Bos, loo. dt. p. 78 sqq. 3 v. Guaita, ' Versuche mit Kreuzungen von verschiedenen Rassen der Hausmaus,' in Berichte der naturforschenden Gesellschaft zu Freiburg I.B. x. 331. 4 Huth, op. cit. p. 286 sq. XX EXOGAMY 223 in-breeding produced a considerable loss of fertility, but was accompanied with an increase of weight.1 Castle and his pupils, investigating the effects of in-breeding and cross-breeding upon the fertility of the pomace (Droso- phila ampclophila}, arrived at the conclusion that in-breeding probably tends to reduce the fertility to a slight extent, though the productiveness may be fully maintained under constant in-breeding if selection is made from the more productive families; and they found indications that cross-breeding increases the productiveness of closely in-bred families.2 Hensen has noticed albinoism as the result of in-breeding in guinea-pigs,3 and Hacker believes that the frequency of albinoism among certain birds may be due to a similar cause.4 It has been stated that in-bred canary birds do not learn by themselves to eat.5 In a certain moth, Lymantria japonica (a variety of Lymantria dispar), in-breeding has proved to cause hermaphroditism after some generations.6 Maupas has shown that among Protozoa the divided individuals die out if they do not pair with strange individuals.7 It has been argued that incest is constantly practised by wild animals a.nd therefore, if it were injurious, would lead to the destruction of the species.8 Our present know- ledge of their habits, however, by no means justifies this proposition.9 Among species that live in families the young leave the family when able to shift for themselves, or are said to refrain from pairing with one another. As 1 Preyer, Specielie Physiologic des Embryo, p. 8. 2 Castle, Carpenter, Clark, Mast, and Barrows, ' Effects of In- breeding, Cross-breeding, and Selection upon the Fertility and Variability of Drosophila,' in Proceed. American Academy of Arts and Sciences, xli. 784, 786. 3 Hensen, op. cit. p. 175. 4 Hacker, Der Gesang der Vogel, p. 90. Hensen, op. cit. p. 175. 6 Goldschmidt, ' Erblichkeitsstudien an Schmetterlingen,' in Zeitschr. f. induktive Abstammungs- und Vererbungslehre, vii. 1 sqq., and xi. 280 sqq. I am indebted to my friend Dr. Harry Federley for this reference. 7 Kraus and Dohrer, ' Blutsverwandtschaft in der Ehe und deren kolgen fur die Nachkommenschaft,' in v. Noorden and Kaminer, Krankheiten und Ehe, p. 62. 8 Huth, op. cit. p. 9. 9 Cf. Settegast, op. cit. p. 286. THE HISTORY OF HUMAN MARRIAGE chap. 224 regards gregarious animals it is quite possible that they prefer mating with strangers, as our domestic animals are known to do. The " marriage flight " of ants and bees obviously helps to prevent in-breeding.1 There are yet other arrangements in the animal world serving the same purpose.2 Thus among hermaphrodites self-fertilisation is very often prevented by the fact that the ova and the sperms in the same animal become mature at different times.3 It has been pointed out that the excited state of the males in the rutting season urges them to move about in search and pursuance of females, and the great strength of their passions has been accounted for as a means of securing the necessary mixture of blood.4 And in a previous chapter I have suggested that the secondary sexual characters consisting in colours, odours, and the production of sounds, are useful to the species not only because they facilitate reproduction by making it easier for the sexes to find each other, but also because they tend to prevent in-breeding by attracting individuals from a distance.5 So far as mankind is concerned the study of the effects of close in-breeding is prevented by the general absence of marriages between the nearest relatives. It has been said that there is no proof of the physical deterioration of peoples like the ancient Egyptians and Persians among whom such marriages did occur.6 But even if among these peoples close intermarrying was more frequent than we have a right to assume, it was certainly very far from being the exclusive kind of marriage ; and breeders of domestic animals inform us that the mixing-in even of a few drops of unrelated blood may be sufficient almost to neutralise the injurious effects of long continued in-breeding. Mr. Huth asserts that though the Ptolemies habitually married 1 See supra, ii. 195 sqq. 2 See Hensen, op. cit. p. 181 sqq. ; Hacker, op. cit. p. 90 ; Kraus and Dohrer, loc. cit. p. 62. 3 Hensen, op. cit. p. 182. Hacker, op. cit. p. 90. 1 Hacker, op. cit. p. 90 sq. Rohleder, op. cit. p. 33. 6 Supra, i. 488. 6 Adam, ' Consanguinity in Marriage,' in Fortnightly Review, iii. 81. XX EXOGAMY 225 their sisters, nieces, and cousins, they were neither sterile nor particularly shortlived.1 But Mr. Galton, on the contrary, sees in Ptolemaic experience a proof that close intermarriage is followed by diminished fertility ;2 and in an elaborate essay on the subject Ujfalvy has tried to show that the in-breeding of the Ptolemies led to physical and mental degeneration.3 The closest kind of intermarriage which we have oppor- tunities of studying is that between first cousins, and there is a considerable literature on the subject. But the opinions of the different writers often vary greatly. Some-for example, Perier,4 Voisin,5 and Huth-believe that such marriages are absolutely harmless unless the parents are afflicted with the same hereditary morbid tendencies ; whereas others-for example, Devay6 and Boudin7-express the most alarming opinions on the subject. I shall here give the results attained by a few investigators whose inquiries seem to be of some importance. Adopting a method different from that of his predecessors, G. H. Darwin endeavoured first to discover the proportion of consanguineous marriages in the whole population, and then to find out whether the offspring of those marriages exhibit a greater percentage of individuals, defective in one way or another, than the offspring of non-consanguineous marriages. His investigations tend decidedly to invalidate the exaggerated conclusions of many previous writers, but he thinks that " there are nevertheless grounds for asserting that various maladies take an easy hold of the offspring of 1 Huth, op. cit. p. 36. 2 Galton, Hereditary Genius, p. 152. 3 v. Ujfalvy, ' Die Ptolemaer,' in Archiv f. Anthropologic, N. S. ii. 73 sqq., particularly p. 119 sq. 4 Perier, ' Essai sur les croisements ethniques,' in Memoirs Soc. d'Anthr. Paris, i. 223. 5 Voisin, ' Contribution a 1'histoire des manages entre con- sanguins,' ibid. ii. 447. 6 Devay, Du danger des mariages consanguins au point de vue sanitaire, p. 24 sqq. 1 See various papers by Boudin mentioned by Huth, op. cit. p. 435 sqq. VOL. II Q 226 THE HISTORY OF HUMAN MARRIAGE chap. consanguineous marriages."1 He did not find evidence that the marriage of first cousins had any effect in the production of infertility, deaf-mutism, insanity, or idiocy, but he observed a slightly lowered vitality amongst the offspring of first cousins and a somewhat higher death-rate than amongst the families of non-consanguineous marriages.2 Moreover, the numbers of boating men belonging to the twenty boats at Oxford and thirty at Cambridge, in the first and second division, and those of selected athletes from some schools in England, justified, to some extent, the belief that offspring of first cousins are deficient physically, whilst at the same time they negatived the views of alarmist writers on the subject.3 It is curious that in spite of such unambiguous statements Darwin's paper has generally been quoted as evidence of the perfect harmlessness of first cousin marriages. Stieda has found that in the departments of France the number of bodily or mentally infirm people increases almost constantly in proportion to the number of consanguineous marriages, as will be seen from the following table :-• Group. Number of departments. Number of consanguineous marriages in each thousand marriages. Number of infirm people in each thousand inhabitants. I. . . . . IO . • ,5'4 • • • • 2-3 ir. . . . IO . . . • 8-3. . . . . 2-8 in. . . . . 14 . . . • 9'95 • • • ■ 3 IV. . . . IO . . . . 11*2 . . 24 v. . . . . 13 . . . • 12'5 • • . . 2-8 VI. . . . 8 . . . • 13'8 • . ■ • 3 VII. . . . . 14 . . . • 15-8 • • • • 3'5 VIII. . . . IO . . . . 19'2 • • 3'25 I.-IV. . . . 44 . . . • 9'2 . . • 2-65 V.-VIII. . . . 45 . . . . 14'8 • • 3'i 4 1 G. H. Darwin, ' Marriages between First Cousins in England,' in Fortnightly Review, N. S. xviii. 41. 2 Idem, 'Marriages between First Cousins in England,' in Jour. Statistical Soc. xxxviii. 181, 170, 182. 3 Idem, ' Note on the Marriages of First Cousins,' ibid, xxxviii. 344 sqq. 4 Stieda, ' Les mariages consanguins,' in Annales de demographic internationale, iii. 45 sq. XX EXOGAMY 227 A Danish physician, Dr. Mygge, published in 1879 a book on ' Marriage between Blood-Relations,' which has attracted less attention than it deserves1 owing to the trustworthiness of the method, the number of cases con- sidered, and the author's impartiality. He found, from the information he received from various parts of Denmark, that in that country, or at least in the parishes of it which came under his observation, there occurred among the children of related persons comparatively more idiots, lunatics, epileptics, and deaf-mutes than among others. He considers it probable, too, though not prov^i, that such children die in a higher ratio and are more liable to certain diseases. But, on the other hand, he did not notice any perceptible difference in fertility between consanguineous and crossed marriages.2 In these inquiries Dr. Mygge followed the method applied by the Norwegian physician Ludvig Dahl twenty years earlier. Through investigation of 246 marriages, eighty-five of which were between first cousins and four between still nearer relations, this inquirer was led to the conclusions that consanguineous marriages are somewhat less fertile than crossed marriages ; that they produce comparatively many more stillborn and sickly children ; and that insanity, idiocy, deaf-dumbness, and epilepsy occur about eleven times as often among the offspring of relations as among the offspring of unrelated parents. But he admitted that the numbers compared were too small to make his conclusions decisive.3 In this connection I may mention some facts which I gathered during my stay at Foula, the most isolated of the Shetland Islands. The inhabitants, altogether 200-250 individuals, are nearly all related by blood, and marriages between cousins are very frequent among them, although they believe that such marriages are far from harmless. The families are generally small; the children between seven and fourteen years of age formed only 14 per cent, of the whole population, whilst in Buna Isle, another island 1 It has escaped even Mr. Huth's keen observation. 2 Mygge, Om Aegteskaber mellem Blodbeslaegtede, pp. 162, 272. 3 Dahl, Bidrag til Kundskab om de Sindssyge i Norge, p. 99 sqq. 228 THE HISTORY OF HUMAN MARRIAGE chap. of the same group, where consanguineous marriages were said to be very rare, the children of the same age amounted to 22 per cent, of the population. The people of Foula seemed to be shorter than on other islands, and Mr. Morrison, who had spent ten years among them, was of opinion that their constitution was somewhat feeble, although diseases were not frequent. There were several cases of idiocy ; and I was told that in one family where husband and wife had been cousins no less than three children had been deaf-mute. According to Dr. Mayet's investigations, which are based on Prussian materials, the percentage of mentally defective persons is generally more than twice as large among the offspring of consanguineous marriages as among the offspring of non-consanguineous marriages, if there is an hereditary tendency for mental deficiency in the family.1 On the other hand, if no such tendency is known to exist, the percentage of such persons in general is even smaller among the children of related parents than among the children of non-related parents, whereas the percentage of idiots is very much larger. In Prussia the offspring of consanguineous marriages are at least 6-5 per cent', of the whole population, but among idiots who apparently have no hereditary tendency for mental deficiency in the family no less than 11-5 are offspring of consanguineous marriages. From this Dr. Mayet draws the conclusion that consanguineous marriages have a distinct tendency to produce idiocy.2 He further points out their tendencies to cause deaf-mutism and the eye disease called Retinitis pigmentosa,3 which have been noticed by various other writers. When pupils are admitted to American schools for the deaf, the question is usually asked whether their parents were related by blood ; and from the replies to this question statistics have been collated showing a considerable per- centage of deaf children born from consanguineous marriages 1 Mayet, ' Die Verwandtenehe und die Statistik,' in Jahrbuch dev inter nationalen Vereinigung fur vergleichende Rechtswissenschaft und Berlin, vi.-vii. 203. 06 sqq. 3 Ibid. p. 208 sq. xx EXOGAMY 229 in general.1 According to Virchow, there is in Berlin one deaf-mute among 3,179 Roman Catholics, 2,173 Protestants, and 673 Jews respectively; and. he connects these figures with the frequency of consanguineous marriages among the Jews as compared with their occurrence among the Christians.2 About a fourth or even a third part of all persons suffering from Retinitis pigmentosa are offspring of parents who are related by blood.3 The skin diseases Xeroderma pigmentosum and Ichthyosis congenita have been found about sixteen times more frequently among children of relatives than among other children.4 Dr. Schmolck thinks that in-breeding may be a cause of dwarfism. In Samnauntal, on the border between Tirol and Switzerland, he found among its 356 inhabitants no less than seven dwarfs, descended from two members of the same family who themselves were perfectly normal. He observes that from time immemorial consanguineous marriages have been the rule in the valley, and to this fact, combined with particular geographical conditions, he attributes the re- markable number of dwarfs among its population.5 Dr. Karutz states that the injurious consequences of marriages between blood-relations are generally known among the Basks, who on that account have prohibited such marriages.6 With reference to the effects of in-breeding in non- 1 Fay, Marriages of the Deaf in America, p. 105. On the greater percentage of deaf-mutes among the children of relatives see also, besides some statements previously quoted, Kraus and Dohrer, loc. cit. p. 65 sq. 2 Virchow, ' Ueber Erblichkeit,' in Deutsche fahrbucher fur 'Politik und Literatur, vi. 354. 3 Kraus and Dohrer, loc. cit. p. 64. Rudin, in Archiv f. Rassen- zind Gesellschafts-Biologie, iii. 598. 4 Adrian, ' Die Rolle der Consanguinitat der Eltern in der Aeti- ologie einiger Dermatosen der Nachkommen,' in Dermatologisches Zentral-Blatt, ix., quoted by Rudin, in Archiv f. Rassen- und Gesell- schafts-Biologie, iii. 598. 5 Schmolck, ' Mehrfacher Zwergwuchs in verwandten Familien eines Hochgebirgstales,' in Virchows Archiv fur pathologische Ana- tomie und Physiologic und fur klinische Medizin, clxxxvii. 105 sqq. 6 Karutz, ' Volksthiimliches aus den baskischen Provinzen,' in Verhandl. Berliner Gesellsch. f. Anthrop. 1899, P- 292. THE HISTORY OF HUMAN MARRIAGE CHAr. 230 European communities I have made a collection of facts which may be worth quoting in spite of their vagueness and more or less hypothetical character. They may at \east serve to show that the prevalence of narrow endogamous habits is no evidence of their harmlessness. According to von Martius, who is a great authority on Brazilian ethnography, it is a well-established fact observed everywhere that the smaller and more isolated of the Indian communities, scarcely any members of which marry members of other communities, are much more liable to every kind of deterioration than the larger groups.1 " It is probable," says Bates with special reference to the savage tribes on the Upper Amazon, " that the strange inflexibility of the Indian organisation, both bodily and mental, is owing to the isolation in which each small tribe has lived, and to the narrow round of life and thought, and close intetmarriages for countless generations, which are the necessary results. Their fecundity is of a low degree, for it is very rare to find an Indian family having so many as four children, and we have seen, how great is their liability to sickness and death on removal from place to place."2 The Isanna Indians, according to Wallace, are said not to be nearly so numerous, nor to increase so rapidly, as the Uaupes ; which may perhaps be owing to their marrying with relatives while the latter prefer strangers.3 Von Tschudi supposes that the low fecundity of the Botocudos is caused by their endogamous habits; for when their women marry out of their own horde, especially with whites or negroes, they are generally very fertile.4 Among the Calidonian Indians of the Isthmus of Darien, who are bound never to cross the breed with foreigners, inter- marriage is very constant and the race degenerates.5 The Pueblos in New Mexico are also said to deteriorate because of their constant intermarriage in the same village.6 1 v. Martius, Beitrage zur Ethnographic Amerika's, i. 334. 2 Bates, The Naturalist on the River Amazons, ii. 199 sq. 3 Wallace, Travels on the Amazon, p. 508. 4 v. Tschudi, Reisen durch Sudamerika, ii. 284. 5 Gisborne, Isthmus of Darien, p. 155. 0 Davis, El Gringo, p. 146. XX EXOGAMY 231 Concerning the Bakundu in North-West Kamerun, who marry within their own village, Bufe says that this in- breeding has in the course of time had very ill effects : " the Bakundu have become a degenerated race with a weakly bodily frame and slight power of resistance, and the mortality among them is terribly great."1 Speaking- of the present universal system of cousin marriage among the Awemba of Northern Rhodesia, Messrs. Gouldsbury and Sheane observe that " it is, undoubtedly, one of the main reasons which render the Wemba women less prolific than the wives of the Wiwa and other tribes where such close unions are prohibited."2 Barrow wrote of the Hottentots:- " The impolitic custom of hording together in families, and of not marrying out of their own kraals, has no doubt tended to enervate this race of men, and reduced them their present degenerated condition, which is that of a languid, listless, phlegmatic people, in whom the prolific powers of nature seem to be almost exhausted. To this may be added their extreme poverty, scantiness of food, and continual dejection of mind." Few of the women have more than two or three children, and many of them are barren. But this is not the case when a Hottentot woman is connected with a white man. " The fruit of such an alliance is not only in general numerous, but they are beings of a very different nature from the Hottentots."3 Mr. Sibrbe suggests that the sterility so frequent in Malagasy women may be due to the intermarriage of near relatives, as well as to early marriages and the licentiousness of the young.4 It has been said that one of the principal causes of the decrease of the Maori was their endogamy causing barrenness among the women.5 According to the Registrar-General of New Zealand " there is a much smaller birth-rate among 1 Bufe, ' Die Bakundu,' in Archiv f. Anthrop. N. S. xii. 231. 2 Gouldsbury and Sheane, Great Plateau of Northern Rhodesia, P- *73- 3 Barrow, Account of Travels into the Interior of Southern Africa, i. 144, 147. 4 Sibree, The Great African Island, p. 248. 6 Meade, Ride through the Disturbed Districts of New Zealand, p. 168. THE HISTORY OF HUMAN MARRIAGE chap. 232 the Maoris than among the rest of the population " ; and he maintains that their decrease is partly attributable " to con- stitutional feebleness caused by the practice of intermarriage between members of the same tribe or sub-tribe."1 The Lundu Sea Dyaks, according to Sir Spenser St. John, have decreased greatly in numbers-from a thousand families to ten. " They complain bitterly," he says, " that they have no families, that their women are not fertile ; indeed, there were but three or four children in the whole place. The men were fine-looking and the women well-favoured and healthy-remarkably clean and free from disease. We could only account for their decreasing numbers by their constant intermarriages."2 The rapid decrease of the Ainu of Japan has also been connected with their en- dogamous habits.3 I have previously quoted Dr. Bunker's statement that among the Karens of Burma there is a striking difference with regard to stature, health, strength, and fecundity between the inhabitants of the exogamous and those of the endogamous villages, and that the latter themselves ascribe their inferiority in these respects to the intermarriage of kinsfolk. But Dr. Bunker told me that in cases in which missionaries had been able to persuade young men to choose wives from another village the good effects of a cross appeared at once. Dr. Helfer also thinks that among the Karens of the Tenasserim Provinces close intermarrying is the reason why " they are a subdued, timid, effeminate, diminishing race."4 Colonel Dalton observes that among the Garos the chiefs have, in comparison with the lower classes, degenerated physically, and he is inclined to think that this degeneration is a result of in-breeding.5 Speaking of the cross-cousin marriage among the Dravidians of South India, Mr. Molony says that its ill effects in the transmission of infirmities have been emphasised by his 1 Gisborne, Colony of New Zealand, p. 325. 2 St. John, Life in the Forests of the Far East, i. 10. 3 Batchelor, Ainu of Japan, p. 290. 4 Helfer, ' Animal Productions of the Tenasserim Provinces,' in Jour. Asiatic Soc. Bengal, vii. 856. 6 Dalton, op. cit. p. 66. XX EXOGAMY 233 medical informants.1 Schotter believes that among the He-miao, one of the aboriginal tribes of China, the practice of cross-cousin marriage is the principal cause of the great number of idiots and deaf-mutes and the remarkable prevalence of illnesses which he has found among them.2 There are, however, also statements of a contrary character. Sir Basil Thomson found that in Fiji the offspring of first cousins were more numerous and showed greater vitality than the children of unrelated persons ; but he admits that his figures are-far too small to allow of any generalisation from them.3 Other writers have quoted instances of communities in which consanguineous marriages are said to have occurred constantly without any evil effects having appeared. One of these is Pitcairn Island, which was uninhabited till the year 1790, when it was peopled by nine white men, and six men and twelve women of Tahiti. In 1800 the population consisted of one man, five women, and nineteen children; and the descendants of these persons are stated by later travellers to be strong and healthy and without any traces of degeneration. But it should be noticed that since the colonisation of the island a few strangers have joined the little colony ; that it was once removed to Norfolk Island, and that of those who returned one was a Norfolk Islander who had married a Pitcairn girl; that the island has frequently been visited by ships with their crews ;4 and that, as Beechey expressly states, the same restrictions with regard to intermarriage of relations exist there as in England.5 There are several other isolated communities-in Java, Peru, Great Britain, France, Scandinavia, and so forth- where long continued in-breeding is said to have been accompanied with no discernible disadvantages. An often- quoted case is the community of Batz in Loire-Inf£rieur, 1 Molony, Census of India, 1911, vol. xii. (Madras) Report, p. 107. 2 Schotter, ' Notes ethnographiques sur les Tribus de Kouy- tcheou (Chine),' in Anthropos, vi. 320. 3 Thomson, Fijians, p. 195 sqq. 4 Huth, op. cit. p. 141 sqq. 5 Beechey, Narrative of a Voyage to the Pacific and Bccring's Strait, i. 86. - THE HISTORY OF HUMAN MARRIAGE chap. 234 situated on a peninsula. The inhabitants, amounting to 3,300 persons, have been in the habit of intermarrying among themselves from time immemorial; and yet they are almost all very well in health without any hereditary affection. But Dr. Voisin observes, " Les conditions climateriques de la commune de Batz, son voisinage de la mer, 1'hygiene et les habitudes de ses habitants, semblent s'accorder pour empecher la degenerescence de 1'espdce et paraissent expliquer 1'innocuite des mariages entre con- sanguins qui s'y pratiquent depuis plusieurs si&des."1 In other isolated communities the population is not so numerous and the sanitary conditions are perhaps not so favourable ; but in any case we may say that this local endogamy is generally something different from marriage with near relatives. Dr. Mitchell found that in almost all the isolated communities along the coasts of Scotland which had been given as instances of close in-breeding such marriages were comparatively rare.2 According to Dr. Mygge, the like is true of Lyp and Strynp in Denmark.3 And Dr. Andrew Wood states of the fisher-folk of Newhaven that though they keep themselves much segregated, they are very careful regarding intermarriage, and look upon the union of relatives as an infringement of the laws of morality.4 At the same time it seems that close intermarrying, even though continued for a considerable time, has in particular cases been harmless or almost so. In some parishes of Denmark Dr. Mygge found no evil effects of consanguineous marriages, whilst in others the effects were very conspicuous.5 And from the investigations of Darwin it appears that, notwithstanding the injury which most plants suffer from self-fertilisation, a few have almost certainly been propagated in a state of nature for thousands 1 Voisin, loc. cit. p. 447. 2 Mitchell, ' Blood-Relationship in Marriage considered in its Influence upon the Offspring,' in Memoirs Anthr. Soc. London, ii. 431 sq. 3 Mygge, op. cit. p. 126. 4 Edinburgh Medical Journal, vol. vii. pt. ii.'876. 5 Mygge, op. cit. p. 171. xx EXOGAMY 235 of generations without having been once intercrossed. It is impossible to understand, he says, why some individuals even of the same species are sterile, whilst others are quite fertile, with their own pollen.1 There is evidence that the bad consequences of self-fer- tilisation and clo-e interbreeding may almost fail to appear under favourable conditions of life. In-bred plants, when allowed enough space and good soil, frequently show little or no deterioration ; whereas, when placed in competition with another plant, they often perish or are much stunted.2 Crampe's experiments with brown rats proved that the breed- ing in-and-in was much less injurious if the offspring of the related parents were well fed and taken care of than it was otherwise.3 And this .is in striking accordance with Dr. Mitchell's observations as to consanguineous marriages in Scotland. The results there appear to be least grave, and are frequently almost nil, if the parents and children live in tolerable comfort, without anxiety or much thought for the morrow, and easily earning enough to procure good food and clothing-in short, when they work but do not struggle for existence. On the other hand, when they are " poor, pinched for food, scrimp of clothing, badly housed, and exposed to misery ; when they have to toil and struggle for the bare necessaries of life-never having enough for to-day and being always fearful of to-morrow,"-the evil may become very marked.4 If this is the case, we may perhaps suppose that consan- guineous marriages are more injurious in savage regions, where the struggle for existence is often very severe, than they have proved to be in civilised society, especially as it is among the well-to-do classes that such marriages occur most frequently.5 In England, according to G. H. Darwin, cousin-marriages among the aristocracy arc probably q| per cent. ; among the middle and upper middle class, or among 1 Darwin, Effects of Cross and Self Fertilisation in the Vegetable Kingdom, pp. 439, 458, 2 Ibid. p. 439. G. H. Darwin, in Jour. Statist. Soc. xxxviii. 175. 3 Crampe, quoted by Diising, op. cit. p. 249. 1 Mitchell, loc. cit. p. 447. 5 Cf. Devay, op. cit. p. 10. 236 THE HISTORY OF HUMAN MARRIAGE chap. the landed gentry, 3| per cent. ; but in London, comprising all classes, they are probably only i| per cent.1 He thinks that the slightness of the evils which he found to result from first cousin marriages perhaps depends upon the fact that a large majority of Englishmen live under what are on the whole very favourable circumstances.2 Taking, then, into consideration all the facts bearing on the subject which are known to me, I cannot but think that in-breeding generally is, in some way or other, more or less detrimental to the species. And here I find a quite sufficient explanation of the want of inclination for, and positive aversion to, sexual intercourse between persons who from childhood have lived together in that close intimacy which characterises the mutual relations of the nearest kindred. We may assume that in this, as in other cases, natural selection has operated, and by eliminating destructive tendencies and preserving useful variations has moulded the sexual instinct so as to meet the requirements of the species.3 It must not be argued that marriages between cousins have proved too slightly injurious to produce such a selection For if, as I maintain, the family consisting of parents and children prevailed as a social unit among our early human or ape-like progenitors, the peculiarity of the sexual instinct of which I am speaking 1 G. H. Darwin, in Jour. Statist. Soc. xxxviii. 163. 2 Ibid. p. 175 sq. 3 Professor Robertson Smith completely misunderstood my t heory when he argued (in a review of the first edition of the present work in Nature, xliv. 271) that it begins by presupposing the very custom which it professes to explain, the custom of exogamy- that " it postulates the existence of groups which through many generations (for the survival of the fittest implies this) avoided wiving within the group." What my theory postulates is not the existence of exogamous groups, but the spontaneous appearance of individual variations of the sexual instinct. Mr. Andrew Lang (Social Origins, p. 33) also regarded my whole argument as a " vicious circle," and thought that he could dispose of it simply by asking why any of our ancestors avoided marriage of near kin. He might just as well have called the theory of natural selection itself a vicious circle, since there never could be a selection of qualities that did not exist before. XX EXOGAMY 237 would have grown up as a consequence of the harmfulness of unions between the very nearest relatives, unless indeed it was an inheritance from a still earlier mammalian species. But once acquired, it would naturally show itself also in the case of more remote relatives or quite unrelated persons who lived in close intimacy from childhood, however harmless the unions between them might be. And through an association of ideas and feelings it might readily lead to the prohibition of sexual intercourse between individuals who did not live together at all. This explanation of the origin of exogamy is, so far as I can see, unaffected by the question why in-breeding is injurious to the species. Many writers on the subject believe that all the evils resulting from it are due to the combination and consequent increase of morbid tendencies common to both parents, and that therefore the presence or absence of such tendencies decides whether consan- guineous unions are harmful or not.1 By careful selection of hereditary good qualities, says Virchow, the best results may be obtained.2 Modern biologists, as Professor Kraus3 and Professor Feer,4 emphasise that no known facts compel us to suppose that unions between near relatives as such- that is, apart from the ordinary facts of heredity-are productive of evil consequences, because the related parents, although apparently perfectly healthy, may be bearers of similar baneful tendencies which are liable to be transmitted to the offspring and strengthened by combination. But at the same time the chances of their being so are considered to be sufficiently great to justify a warning against the marriage of relatives.5 Sir John Sebright, who likewise attributed all the ills of in-breeding to heredity alone, cautiously remarked that according to his belief there never did exist an animal without some defect, in constitution or form or some other essential quality, or that at least a 1 See Rudin, in Archiv f. Rassen- und Gesellschafts-Biologie, iii. 598. 2 Virchow, loo. cit. p. 355. 3 Kraus and Dohrer, loo. cit. p. 82. 4 Feer, Dey Einfluss der Blutsverwandtschaft der Eltern auf die Kinder, reviewed by Rudin, in Archiv f. Rassen- und Gesellschafts- Biologie, v. 272 sq. 5 Kraus and Dohrer, loc. cit. p. 82. 238 THE HISTORY OF HUMAN MARRIAGE chap. tendency to the same imperfection generally prevails in the same family.1 Darwin took another view of the question. He thought it highly probable that, although the injury has often partly resulted from the combination of morbid tendencies, the general cause is another. Considering the number of self-fertilised plants that were tried, he held it nothing less than absurd to suppose that in all these cases the mother- plants, though not appearing in any way diseased, were weakly or unhealthy in so peculiar a manner that their self-fertilised seedlings, many hundreds in number, were rendered inferior in height, weight, constitutional vigour, and fertility to their crossed offspring.2 Moreover, self- fertilisation and close interbreeding induce sterility, and this indicates something quite different from the augmentation of morbid tendencies common to both parents.3 Hence he concludes that, just as the sterility of distinct species when first crossed, and of their hybrid offspring, depends on their sexual elements having been differentiated in too great a degree, the evils of close interbreeding, or self- fertilisation in plants, result chiefly from their sexual elements not having been sufficiently differentiated.4 But we do not know why a certain amount of differentiation is necessary or favourable for the fertilisation or union of two organisms, any more than for the chemical affinity or union of two substances.5 A similar opinion has been expressed by Hensen ;c and Baur denies that all degeneration phenomena connected with self-fertilisation and in-breeding can be explained by a " Herausmcndeln of recessive types."7 Dr. Mayet maintains that the great prevalence of idiocy, deaf-mutism, and Retinitis 'pigmentosa among the children of related parents in families where these abnormalities 1 Sebright, op. cit. p. n sq. 2 Darwin, Cross and Self Fertilisation, p. 445. 3 Idem, Animals and Plants under Domestication, ii. 116. 4 Idem, Cross and Self Fertilisation, p. 443 sqq. 5 Ibid. p. 457. 6 Hensen, ' Wachstum und Zeugung,' in Schriften des Natur- ivissenschaftlichen Vere ins fur Schleswig-Holstein, xv. 266. r Baur, op. cit. p. 311. xx EXOGAMY 239 are not known to have existed before must be due to consan- guinity as such;1 and the attempts to trace it to latent hereditary tendencies alone2 certainly seem strained/ And how could the hermaphroditism of in-bred moths be accounted for as a result of heredity ? But, as already said, my theory of the origin of exogamy does not stand or fall with any particular explanation of the injuriousness of in-breeding. In support of my theory I wish still to emphasise certain facts. Its assumption that exogamy is rooted in a pecu- liarity of the sexual instinct is in full agreement with the general rule that exogamous prohibitions do not apply to marriage alone, but to sexual intercourse in general; the few exceptions to this rule have reference to remoter kindred only. It explains a world-wide institution by a mental characteristic which may be presumed to be common to all races of men. It traces the origin of this mental characteristic to the needs of the species. It co-ordinates three parallel groups of facts which seem intrinsically to belong together: the exogamous rules, the aversion to sexual intercourse between persons living together from childhood, and the injurious consequences of in-breeding. And it finds the same general law governing analogous phenomena in the two great kingdoms of the organic world : the cross-fertilisation of plants, the various arrangements to prevent in-breeding among animals, and the exogamy in mankind. 1 Mayct, loc. cit. p. 207 sqq. 2 Kraus and Ddhrer, loc. cit. p. 81. Feer, o/k cit., reviewed by Rudin, in Archiv f. Rassen- und Gesellschafts-Biologie, v. 272 sq. CHAPTER XXI MARRIAGE BY CAPTURE We shall now proceed to a discussion of the various modes of contracting a marriage. One is to take the woman by force both without her own consent and without the consent of her kindred, that is, marriage by capture. This method of obtaining a wife has been found in various parts of the world. In Tierra del Fuego marriage by capture sometimes occurs both among the Yahgans1 and the Onas. Of the latter it is said that when they make war on a neighbouring tribe they kill off the men and marry the women, whereas in times of peace, when they take wives from their own tribe, marriages are arranged by peaceful negotiations between the fathers of the parties.2 In many Brazilian tribes, also, women are captured from other tribes and married to the captors-3 among the Karaya wars seem to have been waged solely for this purpose ;4 but among the Bororo marriage by capture occurred even within the tribe.5 The same mode of marriage is found among the Indians of Ecuador. The 1 Hyades, ' Ethnographie des Fuegiens,' in Bull. Soc. d'Anthr. Paris, ser. iii. vol. x. 334. King and Fitzroy, Narrative of the Voyages of the Adventure and Beagle, ii. 224. 2 Cojazzi, Los indios del Archipelago Fueguino, p. 16 sq. Gallardo, Tierra del Fuego-Los Onas, p. 214. 3 v. Martins, Beitrage zur Ethnographie Amerika's, i. 107, 372, 620. von den Steinen, Unter den Naturvolkern Zentral-Brasiliens, p. 500. 4 Kranse, In den Wildnissen Brasiliens, p. 320. 5 von den Steinen, op. cit. p. 500. ch. xxi MARRIAGE BY CAPTURE 241 Macas Indians acquire wives by purchase if the woman belongs to the same tribe, but otherwise by force.1 The Jibaros frequently make war only in order to procure wives.2 The Zaparo Indians in many cases take forcible possession of women, the men running away with their neighbours' wives ; and though the women may sometimes object, they make no virtual difficulties, but, apparently, " rather like being run away with."3 The Caribs married female captives of war who pleased them.4 Among the Comanche marriage by capture occurred now and then.5 Among the Luiseno Indians on the coast of California one method of marriage is said to have been " for a man and several of his friends to carry off by force the woman he wished to marry, even from the house of her parents." 6 Among the Aht a man occasionally steals a wife from the women of his own tribe.7 Among the Thompson Indians of British Columbia it happened, although very seldom, that a man took a girl away by force.8 On the east coast of Greenland the men not infrequently abducted other men's wives, sometimes at the request of the woman's own family, in order that she should be better supported.9 The inhabitants of Unimak invaded the other Aleutian islands chiefly for the purpose of carrying off women.10 1 Buckley, quoted by Lubbock, ' Note on the Macas Indians,' in Jour. Anthr. Inst. iii. 30. 2 Rivet, ' Les Indiens Jibaros,' in L' Anthropologic, xviii. 605, 615. 3 Simson, Travels in the Wilds of Ecuador, p. 172 sqq. • 4 de Poincy, Histoire naturelie et morale des Iles Antilles de VAmerique, p. 545. Im Thum, Among the Indians of Guiana, p. 186. 5 ten Kate, Reizen en onderzoekingen in Noord-Amerika, p. 390. 8 Sparkman, ' Culture of the Luiseno Indians,' in University of California Publications in American Archeology and Ethnology, viii. 214. 7 Sproat, Scenes and Studies of Savage Life, p. 98. 8 Teit, ' Thompson Indians of British Columbia,' in Memoirs of the American Museum of Natural History, vol. ii. Anthropology, i. 324 sq. 9 Holm,' Konebaads-Expeditionen til Grqnlands 0stkyst 1883-85,' in Geografisk Tidskrift, viii. 92. 10 Coxe, Account of the Russian Discoveries between Asia and America, p. 257. VOL II R THE HISTORY OE HUMAN MARRIAGE chap. 242 Among the Chukchee Dr. Bogoras was told that in olden times " a company of young men would seize a young girl in the open, bind her hands and feet, and carry her to the house of one who wanted to have her for a wife. Not only the men of alien families, but even the relatives and the cousins, acted so, after having been refused by the father of the girl. The assault and the ravishing, however, were not considered as a reason for implacable hatred and feuds. The parents would come afterward and ask for ransom, which was paid, not in reindeer, but one woman for another. . . . Even at present a case of ravishment may happen now and then."1 Dr. Jochelson observes that " along with the other methods of obtaining wives, the Koryak, in former times, resorted to carrying away women, or taking away by force both girls'and married women." But, he adds, " if we take into consideration that among the Koryak marriage is rather endogamic than exogamic, and that in war the conquerors usually slew the children of the vanquished lest they should grow up to become avengers, and their women lest they should bear avengers, it seems plausible that the custom of capturing wives from foreign tribes or clans never prevailed among the Koryak to any extent."2 The great wars of the Gilyak, on the other hand, very often served the object of carrying off women ; and even now marriage by capture occurs among them, when a man lacks the means of buying a wife, although such an act is usually followed by blood- revenge.3 Among the Samoyed,4 Votyak,5 Ostyak,6 Vogul,7 Cheremiss,8 Mordvin,9 and Tartars of Crimea,10 1 Bogoras, Chukchee, p. 590. 2 Jochelson, Koryak, p. 743. See also ibid. p. 753. , 3 Sternberg, reviewed in L'Anthropologic, v. 343. Schrenck, quoted by Miss Czaplicka, Aboriginal Siberia, p. 101. 4 Castr^n, Nordiska resor och forskningar, ii. 168. 5 Muller, quoted by Hamalainen, Mordvalaisten, tSeremissien ja votjakkien kosinta- ja hddtavoista, p. 141. 0 Ibid. p. 186 sq. Patkanov, Die Irtysch-Ostjaken, p. 137. 7 Georgi, Beschreibung aller Nationen des russischen Reichs, p. 67. 8 Ibid. p. 56. Smirnov, Les populations finnoises des bassins de la Volga et de la Kama, p. 119 sq. 9 Lepehin, quoted by Hamalainen, op. cit. p. 23. 10 VAmbery, Das Turkenvolk, p. 541. XXI MARRIAGE BY CAPTURE 243 bride-stealing is, or has recently been, resorted to if the bridegroom cannot afford to pay the fixed purchase-sum. Marriage by capture formerly occurred among the Esthon- ians.1 According to the traditions of the Yakut, " if a man who was hunting in the forest with others saw a handsome woman, they watched to see where her husband went- to hunt. They then fell upon him, killed him, and took away his wife. If they could not take, her by force they took her by stratagem, enticing her out of her house by a call to help her husband bring home his game. Then they carried her off by force, in the same manner in which they brought home war captives. In their epic poetry, the stealing of women appears as a constant motive. The heroes help each other to find women outside the tribe, and they obtain them as payment for their heroic deeds, or for help given to others. In all the narratives about war, maidens fall to the victors as prize or ransom."2 Among the Kalmucks it sometimes happens that the suitor abducts the girl when neither she nor her parents are willing to yield to his wishes, but if the abducted girl has slept in his hut the parents are compelled to consent to his marrying her.3 In an article on the marriage customs of Tibet, Sarat Chandra Das states that in Purang a young man who wishes to marry a girl watches her movements and, when he finds a good opportunity, comes accompanied by one or two of his friends, and, finding her alone, carries her off by force. He then keeps her confined in a separate house, where he provides her with good food and nice clothes and remains near her to coax her and win her love. If the girl be un- willing to live with the captor, or if her parents do not permit her to marry him, the matter is settled by the village elders or the tribunal of the district chief ; and if they permit the union, an auspicious day is fixed for the marriage, when a 1 Willigerod, Geschichte Ehstlands, p. 9. v. Schroeder, Die Hochzeitsgebrauche der Esten, p. 19. For marriage by capture among Ugro-Finnic peoples see Hamalainen, op. cit. pp. 23-28, 100-103, 141-144, 186 sqq. 2 Sumner, ' Yakuts,' in Jour. Anthr. Inst. xxxi. 85 sq. 3 Bergmann, Nomadische Streifereien unter den Kalmiiken, ii. 147. THE HISTORY OF HUMAN MARRIAGE chap. 244 good deal of wine is consumed. Our informant adds that marriage by capture, as it now to some extent prevails in Purang and the country round Lake Manasarovara, existed in former times in Tibet and in the Cis-Himalayan countries.1 Among the Bhotias of the Darma pargana, Almora District, " it sometimes happens that a girl is carried away in reality by force from the rambang, but unless and until she eats dalong, datu and jan with her captor she is not considered to be married to him."2 The Sikhs were formerly accus- tomed to capture their wives in forays, " and hence perhaps it was considered of no consequence that the husband and wife should be of different faith."3 In Bastar girls are some- times regularly captured by their lovers, though the business is usually arranged by the couple beforehand ; and the same is the case among the Kolams of Wardha.4 Among the Malayalis, a Tamil-speaking hill tribe of North Arcot, the bride is now and then carried off by force, though this is viewed with much disfavour.5 Among the Mullaku- rumbers of Wynad, who are bowmen and hunters, if it so happens that the parents of the girl refuse consent to the match, " the man has the option to take her away by force and marry her after a fight."6 Among the Gonds of the Eastern Ghauts one mode of concluding a marriage is by compelling a girl or woman to become -one's wife. " She may not like the man and her parents may not countenance his match-making. But that does not deter the suitor. He waits for a convenient opportunity, for example, when the girl goes to fetch water from the river, and carries her off by force with the aid of his friends." Once caught, the girl is very closely confined ; and after a time she becomes reconciled to the man, and they are forthwith husband and 1 Das, ' Marriage Customs of Tibet,' in four. Asiatic Soc. Bengal, vol. Ixii. pt. iii. 8 sq. 2 Sherring, ' Notes on the Bhotias of Almora and British Garhwal, ' in Memoirs Asiatic Soc. Bengal, i. 108. 3 Russell, Tribes and Castes of the Central Provinces of India, i. 321. 4 Ibid. i. 147. 5 Cox, Madras District Manuals : North Arcot, i. 213. 6 Gopalan Nair, Wynad, p. 69. XXI MARRIAGE BY CAPTURE 245 wife.1 Marriage by capture is said to be commonly practised by some of the aboriginal tribes of the Chota Nagpur Plateau. " Among the Bhuiyas of the Orissa States, if a young man is in love with a girl, and either she or her parents will not consent to a marriage, he gathers together a band of his friends, and, when he gets an opportunity, carries her off, his companions guarding the flight. This method of obtaining a bride often leads to sanguinary conflicts, owing to the girl's friends attempting to prevent the abduc- tion or to rescue her."2 The Hos, also, " practise marriage by capture, the young man carrying off the girl from some dance or market in spite of any resistance, real or feigned, that she may make. In this case the bride-price is settled afterwards."3 Speaking of the Chittagong Hill tribes in general, Lewin states that " those who had few women went with arms in their hands, and took what they "wanted from a weaker community."4 Among the Morans of Assam the bridegroom occasionally carries off his bride by force, especially during the April Bihu, the union formed in this way being afterwards recognised by the girl's parents.5 Marriage by capture is found in various parts of the Malay Archipelago.6 In Bali it is almost the rule among the lower classes for a man to run away with the object of his desire, either with the girl's consent or against her will; in either case he has to pay a fine to her father, but the fine is in the latter case higher than in the former.7 On the Kei or Ewaabu Islands women were captured in war, and were 1 Hayavadana Rao, ' Gonds of the Eastern Ghauts,' in Anthropos, v- 795- 2 O'Malley, Census of India, 1911, vol. v. (Bengal, Bihar and Orissa and Sikkim) Report, p. 315. See also Macmillan, ' Bhuiyas,' in Calcutta Review, ciii. 177. 3 O'Malley, op. cit. p. 315. Cf. Dalton, Descriptive Ethnology of Bengal, p. 192. 4 Lewin, Wild Races of South-Eastern India, p. 92. Endle, Kacharis, p. 89. For the Bihu festival see supra, i. 86. r> Wilken, ' Plechtigheden en gebruiken bij verlovingen en huwelijken bij de volken van den Indischen Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, ser. v. vol. i. 181 sqq. " Jacobs, Eenigen Tijd onder de Baliers, p. 128. THE HISTORY OF HUMAN MARRIAGE chap. 246 regarded as the property of their captors, the children following the father.1 In the Babar Archipelago it is a glory to capture a woman from another village and bring her away, in which case, whether compensation be paid or not, the children follow the father.2 Among the Negritos of Negros, in the Philippines, marriages were, at least in the beginning of the seventeenth century, never concluded within the horde ; hence, it is said, the men had to capture women from neighbouring hordes, which led to endless wars.3 In the Mekeo tribes of British New Guinea, according to Professor Seligman, traditions of marriage by capture still linger in stories where a young man who surprises a girl while bathing and without her petticoat is said to exercise his right to make her his wife ; the marriage is consummated on the spot, after which the young man himself fastens the girl's petticoat.4 At Bogadjim, in New Guinea, when a man takes a wife from the mountains, he is said to " steal" her, and not long ago he was shot at with arrows.5 In the Bismarck Archipelago, and especially in the Admiralty Islands, women are often captured in wars.6 In the mountains of the Gazelle Peninsula of New Britain girls are forcibly ab- ducted from their parents, and sometimes a man carries off another man's wife ; if the men belong to different communities a feud ensues, whereas in other cases the matter is frequently settled by some influential man acting as arbiter.7 In Bougainville, where marriage by purchase is the rule, captured women are also made wives if their totem is no hindrance to it ;8 and in Buin, if a man for some reason or other finds it difficult to procure a wife in the ordinary peaceful way, it occurs that he carries off a girl from her relatives, who are afterwards appeased with shell money, 1 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, p. 236. 2 Ibid. p. 351. 3 Blumentritt, Ver such einev Ethnographic dev Philippinen, p. 7. 4 Seligman, Melanesians, p. 363 n. 1. 5 Hagen, Untev den Papua's, p. 243. 0 Schnee, Bilder aus der Sudsee, p. 38. " Burger, Die Kusten- und Bergvolker der Gazellehalbinsel, p. 54 sqq. 8 Parkinson, Dreissig Jahre in der Sudsee, p. 481. XXI MARRIAGE BY CAPTURE 247 though the amount of it is smaller than the sum other- wise paid for a wife.1 In Erromanga " marriage was frequently effected by capture," and " war between tribes generally arose from woman-stealing or disputes about land."2 In New Caledonia " les enlevements de femmes mariees ne sont . . . pas rares, et souvent ils attirent des guerres entre les tribus ou les families, a moins que celui qui a fait le rapt ne soit un chef ou relativement un grand personnage."3 On the larger islands of the Fiji Group was often found the custom of seizing upon a woman by apparent or actual force in order to make her a wife.4 Among the Maori female prisoners of war were generally appropriated by'members of the victorious party and taken as inferior or slave wives.5 Of the same people we are told that " an ancient and favourite way of marriage was to get up a war-party (or mimic war-party) and carry off the bride by force. There were so many relatives to be consulted, some of whom would be sure to feel aggrieved if their consent was not obtained, that abduction was easiest. There was often feigned abduc- tion and feigned defence, but it was at times very hard on the girl."6 The Tasmanians captured women for wives from alien and hostile tribes.7 In Australia marriage by capture has been found in- all parts of the continent.8 Mr. Mathew says 1 Thurnwald, Forschwngen auf den Salomo-Inseln und dem Bis- marck-Archipel, iii. 15, 19. 2 Robertson, Erromanga, pp. 396, 392. 3 Brainne, La Nouvelle-Calddonie, p. 251 sq. 4 Williams and Calvert, Fiji and the Fijians, p. 149. 6 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New Zealand Institute, xxxvi. 63. Idem, ' Notes on the Art of War, as conducted by the Maori of New Zealand,' in Jour. Polynesian Soc. xii. 163. Polack, Manners and Customs of the New Zealanders, i. 153. 6 Tregear, The Maori Race, p. 293. See also Taylor, Te Ika a Maui, p. 336. " Milligan, quoted by Nixon, Cruise of the Beacon, p. 29. 8 Hodgson, Reminiscences of Australia, p. 243. Turnbull, Voyage round the World, p. 98 sq. Collins, Account of the English Colony in New South Wales, i. 559. Sturt, Narrative of an Expedition into Central Australia, ii. 283. Taplin, Folklore, Manners, Customs, and Languages of the South Australian Aborigines, p. 93. Howitt THE HISTORY OF HUMAN MARRIAGE chap. 248 that " marriage by capture takes place between members of hostile communities. Sometimes a surprise party will be organised to attack a camp, slaughter the males and abduct and appropriate the females. This wholesale abduction is paralleled by individual cases of forcible abduc- tion, on which occasions the woman, if resisting, will be cruelly beaten."1 Speaking of some Central Australian natives, Willshire observes that if a man does not succeed in securing a wife in the formal and regular fashion, he is driven to the hazardous course of stealing one.2 Spencer and Gillen, however, say that capture is the " very rarest way in which a Central Australian secures a wife."3 Contrary to some statements made by early observers in New South Wales, it is usually reported to be merely an exceptional or occasional mode of contracting marriage in Australia.4 It leads to feuds or quarrels, hence the tribes very generally set themselves against the. practice.5 Among the Arunta and Loritja, according to Strehlow, a young man who steals a woman is speared to death if caught.6 Native Tribes of South-East Australia, pp. 195, 208, 216, 220, 225, 235, 236, 267, 280. Fison and Howitt, Kamilaroi and Kumai, p. 343. Oldfield, ' Aborigines of Australia,' in Trans. Ethn. Soc. N.S. iii. 250. Hardman, ' Notes on some Habits and Customs of the Natives of the Kimberley District, Western Australia,' in Proceed. Roy. Irish Academy, ser. iii. vol. i. 71. Withnell, Customs and Traditions of the Aboriginal Natives of North Western Australia, p. ±5 sq. See also following notes. 1 Mathew, Eaglehawk and Crow, p. 113 sq. 2 Willshire, Aborigines of Central Australia, p. 13. 3 Spencer and Gillen, Native Tribes of Central Australia, pp. 104, 554 sq. * Curr, The Australian Race, i. 108. Lang, Queensland, p. 337 sq. Idem, Cooksland in North-Eastern Australia, p. 394. Palmer, ' Notes on some Australian Tribes,' in Jour. Anthr. Inst. xiii. 301. Hodgkinson, Australia, from Port Macquarie to Moreton Bay, p. 230. Henderson, Excursions and Adventures in New South Wales, ii. ito. Taplin, ' Narrinyeri,' in Woods, Native Tribes of South Australia, p. 10. Malinowski, The Family among the Australian Aborigines, p. 53 sqqA 6 Curr, op. cit. i. 108. Beveridge, Aborigines of Victoria and Riverina, p. 23. 6 Strehlow, Die Aranda- und Loritja-Stdmme in Zentral-Austrahen, vol. iv. pt. i. 93, 103. XXI MARRIAGE BY CAPTURE 249 Of the Sakalava in Madagascar we are told that they every year steal many women from other tribes, and that some of their wives are Hova and Betsileo women from the interior, who are eagerly sought after by the cattle-lifters.1 Andersson states that among the Bushmen woman is only too often belli teterrima causa.2 Among the Auin, who belong to the Kalahari Bushmen, women and girls who are captured in war are adopted and married into the tribes of the captors.3 Concerning the Hottentots-whose marriage customs generally required that cattle should be given by the bridegroom to the nearest relatives of the bride-we read that " one of the principal objects in their wars with each other was to take females as prisoners, who were gener- ally regarded as mere concubines, but were sometimes raised to the dignity of wives. The difference . . . involved a right over the distribution of the milk, and upon it depended the inheritance of the children."4 So also it is the custom among the Fingu, a Kafir tribe, that a girl who has been captured in war can be legally married to the captor, if only he publicly declares that she is his wife ; although otherwise a woman whom a man takes to his hut without paying any- thing for her is regarded as a concubine or as a prostitute.5 In the different groups of the Ababua, Babati, and Baieu (Bantu-speaking) tribes along the Upper Wele andom Bo- kandiand Upper Rubi rivers,according to Grenfell, "marriage by capture is of common occurrence, but frequently results in war between the two villages ; properly the husband ought to buy his wife, her consent not being absolutely necessary." So also marriage by capture is said by the same writer to exist among the forest tribes in the north-eastern part of the Mongala basin :-" A man carries off a wife by force, betakes himself with her to the forest, lives there by 1 McMahon, ' Sakalava and their Customs,' in Antananarivo Annual and Madagascar Magazine, iv. 391 sq. 2 Andersson, Okavango River, p. 143. 3 Kaufmann, ' Die Auin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 154. 4 Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 85. 6 Kropf, Das Volk der Xosa-Kaffern im bstlichen Sudafrika, p. 138 sq. 250 THE HISTORY OF HUMAN MARRIAGE chap. hunting, and only returns to the village when the wife has a child and it is weaned. Returned home, he quits his temporary wife, and gives her half the proceeds of their hunting in exchange for the child."1 Of the Ngoni in British Central Africa we are told that " as the tribe existed by raiding other tribes, all who could bear arms might possess themselves of captive wives."2 In Nyasaland marriage by capture was formerly very prevalent but " has been practi- cally abolished within the Protectorate and adjoining terri- tories by the suppression of inter-tribal warfare."3 Constant raids bring to Uhehe large numbers of women and thus provide the men with many wives, who, together with cattle, constitute their most valuable property.4 In Bukoba, also, there is capture of women for wives, but it is an act of force rather than a genuine institution.5 Among the Bantu Kavirondo, " if the father is loth to hand over his daughter, the suitor sends a band of young men to waylay her at night and bring her to his village. If this step is taken by day the young men of the girl's village and her brothers turn out, and a fight with sticks takes place. The girl screams a great deal, pretends to make a great fuss, protests that she will run away, etc., but rarely carries out her threat. This abduction is usually only resorted to when the father of the girl is of an avaricious nature, and keeps postponing the marriage in order to force more goats, etc., out of the suitor."6 Among the Ewhe-speaking peoples of the Slave Coast " marriage by capture de facto still exists, for the great majority of the women captured in the annual raids are used as wives."7 Among the Ait Warain, a Berber tribe in the interior of Morocco, should a brother or uncle of the girl be left without the present which the young man generally 1 Johnston, George Grenfell and the Congo, ii. 676, 674. 2 Elmslie, Among the Wild Ngoni, p. 57. See also Miss Werner, Natives of British Central Africa, p. 133. 3 Dufi, Nyasaland under the Foreign Office, p. 315. 4 Fiilleborn, Das Deutsche Njassa- und Ruwuma-Gebiet, p. 22g. 5 Richter, ' Der Bezirk Bukoba,' in Mittheil. Deutsch. Schutzgeb. xii. 86. 0 Hobley, Eastern Uganda, p. 18. 7 Ellis, Ewe-speaking Peoples of the Slave Coast, p. 212. XXI MARRIAGE BY CAPTURE 251 gives to members of her family in order to induce them to help him in settling the ssdaq, or dowry, with the'girl's father, it may happen that he, on the occasion when she is going to be taken away from her home by the bridegroom's party, will step forward and claim the money which has not been given him. If it is refused, a fight ensues between the men of the bride's village and those of the bridegroom's party, first with stones and afterwards, if the refusal is persisted in, with guns. This is a fight in good earnest, and should anybody be killed there would be blood-revenge ; it has happened that the bride herself has been wounded by stones. If the men of the bride's village win, the bride- groom's people have to pay the demanded sum, whereas in the opposite case the latter carry away with them the bride by force.1 Capture of women for wives has prevailed among ancient Semites.2 In Arabia it was common before Muhammad.3 Among the Hebrews members of the military class were allowed to marry foreign women taken in war, contrary to the law which forbade intermarriage with the Gentiles.4 Marriage by capture has been found among the so-called Aryan peoples. According to the ' Laws of Manu,' one of the eight legal modes of concluding a marriage was the rakshasa mode, that is, " the forcible abduction of a maiden from her home, while she cries out and weeps, after her kinsmen have been slain or wounded, and their houses broken open." This mode was permitted to the Kshatriyas, or warrior caste, by the sacred tradition.5 According to Dionysius of Halicarnassus, marriage by capture at one time existed throughout Greece ;6 and according to Sakellarios it occurred there occasionally even in quite recent times.7 " Les 1 Westermarck, Marriage Ceremonies in Morocco, p. 80 sq. 2 Robertson Smith, Kinship and Marriage in Early Arabia, p. 72 sq. 3 Wellhausen, ' Die Ehe bei den Arabem,' in Nachrichten von der Konigl. Gesellsch. der Wissenschaften zu Gottingen, 1893, p. 435 sq. 4 Deuteronomy, xxi. 10 sqq. 5 Laws of Manu, ii. 33, 26. 6 Dionysius of Halicarnassus, Antiquitatum Romanarum quae supersunt, ii. 30. 5. 7 Sakellarios, Die Sitten und Gebrauche der Hochzeit bei den Neugriechen, p. 6 n. 1. 252 THE HISTORY OF HUMAN MARRIAGE chap. premiers Romains," says M. Ortolan, " d'apres leurs tradi- tions heroiques, ont ete obliges de recourir a la surprise et a la force pour enlever leurs premieres femmes."1 The ancient Teutons evidently captured women for wives.2 Of the Scandinavians Olaus Magnus says that they were constantly at war with one another " propter raptas virgines aut arripiendas."3 In the earliest Teutonic laws the " lapc- marriage " is no doubt a punishable offence, but still it is a marriage.4 In the Irish Nennius we read of a rape of wives by the Picts from the Gael.5 Marriage by capture also occurred among the Slavs in early times.6 The Cossacks of Little Russia and the Ukrainia practised it still in the seventeenth century,7 and many Southern Slavs in the beginning of the nineteenth or later.8 In High Albania forcible capture of a girl occasionally occurs even to this day.9 The same is the case among the Caucasian mountain tribes, who consider that a captured girl becomes the wife of her captor by being deflowered by him.10 This list of peoples among whom marriage by capture 1 Ortolan, Histoire de la legislation romaine, p. 81. 2 Dargun, Mutterrecht und Raubehe und ihre Reste im germanischen Recht und Leben, p. m sqq. Grimm, Deutsche Rechtsalterthiimer, p. 440. Nordstrom, Bidrag till den svenska samhdlls-forfattningens historia, ii. 12. Weinhold, Die deutschen Frauen in dem Mittelalter, i. 308 sqq. Heusler, Institutionen des Deutschen Privatrechts, ii. 277 sqq. 3 Olaus Magnus, Historia de gentibus septentrionalibus, p. 328. 4 Cf. Pollock and Maitland, History of the English Law before the Time of Edward I., ii. 364 ; Heusler, op. cit. ii. 277 sq. 5 Wood-Martin, Traces of the Elder Faiths of Ireland, ii. 35, 37. 6 Chronique dite de Nestor, p. 10. Macieiowski, Slavische Rechts- geschichte, ii. 189. Krek, Einleitung in die slavische Literatur- geschichte, pp. 56, 196 sqq. Piprek, Slawische Brautwerbungs- und Hochzeitsgebrauche, p. 155 sqq. Dargun, op. cit. p. 93 sqq. 7 Kovalewsky, Modern Customs and Ancient Laws of Russia, p. 24. 8 Krauss, Sitte und Brauch der Sudslaven, p. 245 sqq. 9 Durham, ' High Albania and its Customs in 1908,' in Jour. Roy. Anthr. Inst. xl. 459. 10 Darinsky, ' Die Familie bei den kaukasischen Vblkern,' in Zeitschr. f. vergl. Rechtswiss. xiv. 198, 199, 205. v. Seidlitz, * Die Abchasen,' in Globus, Ixvi. 40. XXI MARRIAGE BY CAPTURE 253 occurs or has occurred might easily be enlarged.1 But among no people is it known to have been the usual or normal mode of contracting a marriage. As appears from the statements quoted above, it is chiefly found either as an incident of war or as a method of procuring a wife when it is difficult or inconvenient to get one in the ordinary manner. We thus notice its occurrence among savages living in small family groups, like the Fuegians, various Brazilian tribes, and the Bushmen; in the Australian tribes, where, as we have seen, many young men for various reasons find it extremely hard to get married ; and among many peoples of a higher type as a substitute for marriage by purchase, by which a man tries to lower the price of the bride or to avoid payment altogether.2 Custom may require that the matter shall be settled afterwards with the parents of the captured girl, and in such cases the capture is rather a preliminary to marriage than a mode of contracting it. But Dr. Grosse undoubtedly goes too far when he maintains that marriage by capture has never been a form of marriage recognised by custom or law, though it has been an occasional and punishable act of violence.3 1 See McLennan's Studies in Ancient History, Tylor's Researches into the Early History of Mankind, Lord Avebury's The Origin of Civilisation, the works of Post and Dargun, Kulischer's ' Inter- communale Ehe durch Raub und Kauf,' in Zeitschr. f. Ethnol. vol. x., Kohler's ' Studien fiber Frauengemeinschaft, Frauenraub und Frauenkauf,'in Zeitschr. f. vergl. Rechtswiss. vol. v., Wilutzky's Vorgeschichte des Rechts, vol. i., The Material Culture and Social Institutions of the Simpler Peoples by Hobhouse, Wheeler, and Ginsberg. In 435 peoples of whose inodes of contracting a marriage the authors of the last-mentioned work obtained informa- tion, they found forty-one cases in all of capture as a reality, and many of these were only " occasional " or partial cases (see p. 154, and Hobhouse, Morals in Evolution, p. 153 n. 2). Tylor also found some forty cases of the genuine form of carrying off a bride from a strange tribe against her will and that of her relations (' On a Method of investigating the Development of Institutions ; applied to Laws of Marriage and Descent,' in four. Anthr. Inst, xviii. 259). 3 Cf. Abercromby, ' Marriage Customs of the Mordvins,' in Folk-Lore, i. 453 sq. ; Hildebrand, Recht und Sitte auf den verschic denen wirtschaftlichen Kultursiufen, pp. 10, 15. 3 Grosse, Die Formen der Familie, p. 105. 254 THE HISTORY OF HUMAN MARRIAGE chap. On the other hand, the prevalence of marriage by capture has undoubtedly been much exaggerated by some other writers. It has been represented as being at one time the normal mode of contracting a marriage among uncivilised peoples. But there is no evidence whatever that it was so. We have no reason to assume that a man in order to procure a wife was in ordinary circumstances compelled to take her by force from her relatives. Savages do not usually live at odds with all their neighbours ; among many of them wars are quite exceptional, and some of them are said to have no wars at all.1 It is impossible to believe that there ever was a time when friendly relations between families who could intermarry did not exist. The theory of an early stage of marriage by capture has been supported by reference to some very widespread customs which have been interpreted as survivals of capture in the past. But these customs do not prove what they are meant to prove, because they may be much more readily explained otherwise. First, there are a large number of cases in which sham fighting between the bridegroom or his party and the bride's family, or some other kind of resistance made by the latter, forms part of the wedding ritual. I shall quote a few instances of this class of supposed survivals of marriage by capture.2 1 See Westennarck, Origin and Development of the Moral Ideas, i. 334 ; Holsti, Relation of War to the Origin of the State, p. 16 sqq. ; Wheeler, The Tribe, and Intertribal Relations in Australia, pp. 149, 160 sq " It is not going beyond the evidence to say that peace, not war, is the normal condition of the Australian tribes." 2 For other instances see Munzinger, Bogos, p. 62 ; Weeks, Among the Primitive Bakongo, p. 144 sq. ; Rivers, History of Melanesian Society, i. 309 (Tikopians) ; Hagen, Unter den Papua's, p. 243 (Bogadjim) ; Landtman, Nya Guinea farden, p. 84 (Kiwai Papuans) ; Rowney, W7W Tribes of India, pp. 46, 47 (Kattis), 81 (Orions) ; Chatterji, ' Origin and Traditions of Kathis,' in Calcutta Review, cxxxi. 389 ; Dalton, Descriptive Ethnology of Bengal, pp. 86 (Meeh and Kacharis), 194 (Mundas), 319 (Kurmis) ; Crooke, Tribes and Castes of the North-Western Provinces and Oudh, i. p. cc. sq. ; Latimer, Census of India, 1911, vol. xiii. (North-West Frontier Province), p. T41 sq. ; Grenard, Tibet, p. 258 ; Ahmad Shah, Four Years in Tibet, p. 56 ; Clarke, Among the Tribes in South-West China, p. 130 sq. (Lolos) ; Loebel, Hochzeitsbrduche in der Turkei, p. 79 sq. (Kurds). XXI MARRIAGE BY CAPTURE 255 Among the Bushmen, according to Miss Lemue, " when the young people have settled it between themselves, they tell the parents, who fix a day for the marriage. The Bushmen then come from everywhere, and bring as much meat as they can get. The women smear themselves with red clay and put on their beads, when they all eat and are jolly. In the middle of this feast the young man catches hold of his bride ; her relations at once set on him with their ' kibis,' or digging sticks, and beat him on the head and everywhere ; all the Bushmen then begin to fight together, during which the young fellow must hold his bride fast and receive all the hammering they choose to give him, without letting his treasure escape ; if he can hold out they at length leave him, and he is a married man ; if not, and his charmer escapes from him, he will have to undergo a second ordeal some other time before he can again claim her."1 Among the Matabele, " when a young man has noticed a girl and wishes to marry her, he calls on the father and obtains his permission. . . When the father has given permission for the marriage to take place, the intended husband kills an ox or a sheep, according to his means, and takes part of it to the town where the girl's father lives. The young man stops outside, accompanied by his friends, and shouts, ' Here is meat for your child.' The young warriors of the town then rush out and drive the messengers away, and after a sham fight and pursuit they all come back and feast. A few days later the girl proceeds to the husband's town."2 Among the Mpfumo clan of the Thonga, when the bridegroom's relatives came to the bride's village, with forty or fifty hoes on their heads as bride price, the bride's brothers, provided with their sticks, tried to prevent the aggressors from entering, though in vain.3 Among the Akamba " in former days on the day of the marriage the bridegroom went with five or six brothers and friends and seized his prospective bride in the fields near the village, the girl would call out and her brothers 1 Miss Lemu6, quoted by Stow, Native Races of Soil th Africa, p. 96. 2 Decle, Three Years in Savage Africa, p. 158. 3 Junod, Life of a South African Tribe, i. 109. THE HISTORY OF HUMAN MARRIAGE chap. 256 would assemble and attack the bridegroom's party; they would fight with sticks and even swords and if the girl's brothers won they would take their sister back to their village. A palaver would then take place between the two families, and the girl's father would demand more dowry, the suitor would then pay up perhaps another ten goats and the second time the bridegroom would go alone to his father- in-law's village and would receive his bride without any trouble and take her to his father's village."1 Among the Banyankole, or Bahima, the bridegroom enters the kraal of the bride's family and is conducted to the hut in which the bride stands waiting. " He takes her right hand and leads her from the house and out of the kraal to the assembled guests. A strong rope is produced by one of the bride's relatives and tied to one of the bride's legs. Sides are then chosen by members of the bride's and bridegroom's clans and a tug of war takes place. The bride's clan struggle to retain their sister, and the bridegroom's clan strive to carry her off. During this contest the bride stands weeping because she is being taken from her old home and relatives; it is the correct thing to do. The bridegroom stands by her, still holding her hand, and when the final pull is given in his favour he slips the rope from her ankle and hurries her away a few yards to a group waiting near with a cow-hide spread on the ground. The bride sits upon this and the young men raise her up and rush off with her in triumph to the bridegroom's parents' house, chased by friends and relatives."2 In some parts of Morocco, when the bridegroom's party come to fetch the bride, stones are thrown at them ;3 or the bridegroom and two other men dressed like him are beaten by the men and women of the bride's village;4 or her brother or uncle, before he carries her into the nuptial chamber, has a sham fight with the bridegroom.5 In Pentecost Island, of the New Hebrides, at the wedding " a sort of sham-fight takes place ... in which sometimes 1 Hobley, op. cit. p. 62 sq. 2 Roscoe, Northern Bantu, p. 120. 3 Westermarck, Marriage Ceremonies in Morocco, pp. 180, 186. 4 Ibid. pp. 198, 199, 223. 5 Ibid. pp. 210, 211, 223. XXI MARRIAGE BY CAPTURE 257 men are hurt, the two sides being the kinsmen of the bride- groom and of the bride ; if one of the bridegroom's brethren is hurt, it is his business to make it up with him by a present."1 Among the Roro of British New Guinea on the wedding day a party of men belonging to the bride- groom's local group, but not including the bridegroom, surround the house of the girl's parents and carry it by mimic assault, with great fury and shouting. " The bride rushes out and runs away as fast as she can, and although she is soon overtaken and caught, she defends herself to the best of her ability, with hands, feet and teeth. Meanwhile a sham fight rages between the adherents of the bride and bridegroom. In the midst of the commotion is the bride's mother armed with a wooden club or digging stick, striking at every inanimate object within reach and shouting curses on the ravishers of her daughter. Finding this useless, she collapses, weeping for the loss of her child. The other women of the village join in the weeping. The girl's mother should keep up the appearance of extravagant grief for three days, and she alone of the girl's relations does not accompany the bride to her father-in-law's house." Subsequently " a mock-pillage of houses and gardens of the boy's local group also takes place, though it is clear that no expensive shell ornaments or other really valuable property, such as fishing nets, would be taken."2 Among the Tagbanuas of Culion and Busuanga, in the Philippines, when the necessary prepara- tions have been concluded, the family of the bridegroom go to the house of the bride's parents, where the messenger of the bridegroom and the messenger of the bride indulge in a mock combat at the door. The groom's messenger always wins.3 Among the Vonums of Formosa " the custom of pretending to seize a bride by force is in existence. . . . With a company of friends the bridegroom goes to the house of his intended, and in the face of pretended opposi- tion seizes his bride and carries her off to his own habitation. After a few days, a feast is given to the friends of the couple. 1 Codrington, Melanesians, p. 240 sq. 2 Seligman, Melanesians of British New Guinea, p. 268 sq. 3 Worcester, Philippine Islands, p. 493. VOL. II s 258 THE HISTORY OF HUMAN MARRIAGE chap. In some of the tribes, a sham fight occurs between the relatives of the bride and bridegroom, and the drawing of blood on the part of one of them is considered of good omen."1 Among the Kirghiz," after the women have sung the virtues of the bride, and the men have chanted those of the husband, telling of his great exploits, how many cattle he has stolen, and in how many marauding expeditions he has engaged, the young man must enter the kibitka (or tent) where the bride is seated and take her out, although both entrance and exit are forcibly opposed by all her friends. This," our informant adds, " is probably a remnant of the old primitive custom when marriage was an act of capture."2 Among the Mongols, when the bridegroom-conspicuous by the bow-and-arrow case he carries slung from his shoulder- and his party arrive at the bride's tent, her brother plants himself in front of the door and demands of the strangers what brought them there. "We want to enter your tent," they reply. " Then you'll have to fight for it 1 " is the answer. The strangers and the other party begin a scuffle, pulling each other about a good deal. After a sham fight lasting a few seconds, the defenders give in and invite the assailants to enter the tent.3 Mr. J. B. Lyall heard in Spitit that when the bridegroom's party go to bring the bride from her father's house, they are met by a party of the bride's friends and relations, who stop the path ; where- upon a sham fight of a very rough description ensues, in which the bridegroom and his friends, before they are allowed to pass, are well drubbed with good thick switches.4 Among the Kanjars, an aggregate of vagrant tribes probably of Dravidian origin in the North-Western Provinces of India, the youth goes with his father and as many other men as he can collect, all armed with their best weapons, and demands the girl in threatening tones, which imply that he is ready to seize her by force if she is refused. Yet the girl is always peacefully surrendered in virtue of the previous 1 Davidson, Istand of Formosa, p. 569. 2 Schuyler, Turkistan, i. 43. 3 Gilmour, Among the Mongols, p. 275 sq. 4 Lyall, quoted by Tupper, Punjab Customary Law, ii. 92. XXI MARRIAGE BY CAPTURE 259 compact, and the demonstration of force is a mere form.1 In most Hindu castes in the Central Provinces the bride- groom, on his arrival, performs some militant action, such as striking the marriage-shed or breaking one of its festoons.2 Among the Bagata, a class of fishermen in the Madras Presidency, " at a wedding the bridegroom is struck by his brother-in-law who is then presented with a pair of new cloths. This," we are told, " may be a relic of marriage by capture."3 In the Magh tribes of Chittagong, when the bridegroom approaches the steps leading to the bride's house, " the father or nearest male relative will oppose his advance, and make a feigned effort to prevent him ascending the steps; but the bridegroom courageously perseveres and is again met by the bride's brother or first cousin, who throws himself into the breach and with a small stick makes valiant pretence of striking the bridegroom seven times."4 A common feature of the marriage ceremonies of the Nagas is " the attendance on the bride of the young men of her clan who wrestle with the bachelors of her husband's clan, so as to determine the longevity of the parties to the match. Length of life goes to the partner whose champions win."5 In the country villages of Burma the custom is to tie a string across the road along which the bridegroom must pass to the house of his intended, when he comes in proces- sion with all his friends, carrying the greater portion of the belongings with which he intends to set up house. The people who have put up the string, usually young men intent on a jollification of their own, stop the happy man and threaten to break the string with a curse on the married couple unless some money is given them.0 1 Crooke, Tribes and Castes of the North-Western Provinces and Oudh, iii. 139 sq. 2 Russell, Tribes and Castes of the Central Provinces of India, i. 148. 3 Francis, Census of India, 1901, vol. xv. (Madras) Report, p. 143 sq. 4 Hutchinson, Account of the Chittagong Hill Tracts, p. 119. " Hodson, Ndga Tribes of Manipur, p. 145. See also Idem, ' Native Tribes of Manipur,' in Jour. Anthr. Inst. xxxi. 305. 0 Shway Yoe, The Burman, p. 58. See also Woodthorpe, ' Account of the Shans and Hill Tribes of the States on the Mekong,' in Jour. Anthr. Inst. xxvi. 22 ; Skeat, Malay Magic, p. 381. 260 THE HISTORY OF HUMAN MARRIAGE chap. Among the Teutonic, Slavonic, Romance, and other peoples of Europe it is a common marriage custom to barricade or stop the bridal procession on its way j1 and this custom has also been regarded by several writers as a survival of marriage by capture.2 It is known in Italy by the name of fare il serraglio or fare la barricata, and in Holland by the name of schutten or keeren. The barricading sometimes consists in throwing logs, or even weapons, before the bridal waggon, but more frequently only a rope or a string of flowers is spread across the way ; and the bridegroom has to pay a ransom in order that the waggon shall be allowed to pass. The barring of the wedding procession with a cord is also found in Gloucestershire3 and 1 Wintemitz, ' On a Comparative Study of Indo-European Customs, with Special Reference to the Marriage Customs,' in Trans. International Folk-Lore Congress, 1891, p. 268 sq. Samter, Geburt, Hochzeit und Tod, p. 162 sqq. v. Reinsberg-Diiringsfeld, Hichzeitsbuch, pp. 50, 51, 81, 104, 105, 169, 197, 260. Sartori, Sitte und Brauch, i. 85. Meyer, Deutsche Volkskunde, pp. 168, 169, 174. Idem, Badisches Volkslebenim neunzehnten Jahrhundert, pp. 276, 277, 321. Schonwerth, Aus der Oberpfalz, i. 90. Schell, ' Bergische Hochzeitsgebrauche,' in Zeitschr. d. Vereins f. Volkskunde, x. 165. Drechsler, Sitte, Brauch und Volksglaube in Schlesien, § 281, vol. i. 258. Tetzner, Die Slawen in Deutschland, pp. 258 sq. (Czechs), 278 (Moravians). Franzisci, Cultur-Studien uber Volksleben, Bitten und Brduche in Karnten, p. 68. Heyl, Sagen, Brauche und Meinungen aus Tirol, pp. 768, 770, 780. Piprek, op. cit. pp. 14, 16, 45, 82, 85, 86, 93, 95, 97, 104, 105, 112, 127, 128, 135 (Slavs), de Gubernatis, Storia comparata degli usi nuziali in Italia e presso gli allri popoli Indo-Europei, p. 182 sqq. Marchesi, ' In Valtellina,' in Archivio per lo studio delle tradizioni popolari, xvii. 419. Idem, ' In Capi- tanata,' ibid. xx. 25. Placucci, Usi e pregiudizj dei contadini della Romagna, p. 56. Estella Canziani, ' Courtship, Marriage and Folk- Belief in Vai d'Ossola (Piedmont),' in Folk-Lore, xxiii. 459. Sebillot, Le Folk-Lore de France, iii. 400 sq. (Bearn, valley of Aspe, Bigorre). a Winternitz, loc. cit. p. 268. Idem, ' Das altindische Hoch- zeitsrituell,' in Denkschriften der Kaiserl. Akademie der Wissenschaf- ten, Phil.-hist. Classe, xl. 68. Hillebrandt, Ritualliteraiur (Buhler, Grundriss der indo-arischen Philologies vol. iii. pt. ii ), p. 3. Dargun, Mutterrecht und Raubehe, p. 107. de Gubernatis, op. cit. p. 187. 3 Cowburn, * Roping the Wedding,' in Gloucestershire Notes and Queries, ii. 569. XXI MARRIAGE BY CAPTURE 261 in Wales.1 But in the eighteenth century and, in fact, until recent years a Welsh bridegroom met with more serious resistance. On the morning of the wedding day the groom with his friends demanded the bride. Her friends gave a positive refusal, upon which a mock scuffle ensued. The bride, mounted behind her nearest kinsman, is carried off and is pursued by the groom and his friends with loud shouts. When they have fatigued themselves and their horses, he is suffered to overtake his bride, and leads her away in triumph.2 The mock capture of a bride occurred a century ago in some parts of Scotland3 and Ireland4 as well. Some customs of this class may, no doubt, have been suggested by genuine capture of the bride. But this does not imply that capture was ever the usual mode of con- tracting a marriage. In a warlike tribe the capture of a woman for wife from an alien tribe may be admired as an act of bravery and therefore playfully imitated by ordinary people at their weddings. In some countries the bride- groom and bride are regarded as king and queen,5 but who would look upon this as a survival from a time when marriage was only contracted by royal persons ? The fight at a Bushman wedding, again, is called " an elaborate contest, when on a fixed wedding-day the endurance and sincerity 1 Hartland, in Trans. International Folk-Lore Congress, 1891, p. 289 sq. Dunnill, 'Welsh Fold-Lore Items,' in Folk-Lore, xxiv. 109 sq. (Newport). Notes and Queries, ser. ii. vol. v. 48 (Glamorgan- shire) . 2 Kames, Sketches of the History of Man, i. 450. See also Lewis, Ancient Laws of Wales, p. 197 ; Rhys, in Trans. International Folk- Lore Congress, 1891, p. 289; Marie Trevelyan, Folk-Lore and Folk- Stories of Wales, p. 271 sq. 3 Simpkins, County Folk-Lore, vol. v. Fife, p. 164 sq. 4 Wood-Martin, op. cit. ii. 32. 5 E.g., in Syria (Wetzstein, ' Die syrische Dreschtafel, ' in Zeitschrift f. Ethnol. v. 288) and in the Malay Peninsula (Skeat, Malay Magic, p. 388). In Morocco (see Westermarck, Marriage Ceremonies in Morocco, pp. 97, 102, 106, 113, 118, 274, 361) and Algeria (Gaudefroy- Demombynes, Les Ceremonies du mariage chez les indigenes de l'Algirie, p. 41 sq. ; Destaing, Etude sur le dialecte berbere des Beni- Snous, p. 289) the bridegroom is called " sultan." 262 THE HISTORY OF HUMAN MARRIAGE chap. of the bridegroom were tested to the uttermost."1 So, also, Dr. Jochelson considers the thrashing of the bridegroom by the bride's relatives among the Koryak " as the final act in testing his adroitness, bravery, and endurance, and not as a symbolic remnant of marriage by capture."2 In most cases, however, it seems that the ceremonial resistance of the girl's relatives is a symbolic expression of their un- willingness to give up the girl or of their feeling of sexual modesty, which is particularly felt with regard to the nearest relations. Of the sham fight at a wedding in Pentecost Island Dr. Codrington observes that it no doubt represents the feelings with which the bride's kinsmen regard the loss of her services; "it cannot be the loss of any rights of intercourse, since she was unapproachable by any of them."3 But the girl's parents and relatives may not only regret the loss of her services; they may also be sorry to part with her for purely sentimental reasons, and the mother's tears may be as genuine an expression of sorrow as they are among ourselves. The parents' reluctance, real or feigned, to give away their daughter may show itself in various ways. Dr. Jochelson says that among the Yukaghir the go-between puts his proposition to the girl's parents in a very disguised form, and that her father, wishing to display all the dignity of hiS house, usually answers at first with a refusal.4 The same is done in some Berber tribes of Morocco.5 At Lhasa, in Tibet, " the girl's mother will firmly and repeatedly insist upon the ugliness and uselessness of her debutante whenever a suggestion is made by the professional match- maker of the village."6 Among the Mikirs, when the bride- groom's party come to the bride's house, a colloquy ensues, in which the bride's father asks the bridegroom's party why they have come. The answer is that the wife of the speaker is becoming old and cannot work, hence he wants a wife for his son. The girl's father replies, " My daughter is unworthy, she does not know weaving and other household work."- 1 Stow, op. cit. p. 97. 2 Jochelson, Koryak, p. 743. 3 Codrington, op. cit. p. 241. 4 Jochelson, Yukaghir, p. 88. 5 Westermarck, Marriage Ceremonies in Morocco, pp. 42, 46. 6 Landon, Lhasa, p. 228. XXI MARRIAGE BY CAPTURE 263 " Never mind, we will teach her ourselves."1 Among the Ewhe-speaking peoples of the Slave Coast, on the day of the marriage ceremony the bridegroom sends a messenger with rum to the parents of the bride soon after daybreak and asks for his wife. The parents affect reluctance, and delay the messenger with various excuses till about noon, when the bridegroom despatches a second messenger on the same errand. This messenger also fails, and it is not until the arrival of a third one, who comes about sunset, that the parents overcome their hesitation.2 Of course, there may moreover be mercenary motives for refusing too early a settle- ment, as also for putting obstacles in the way of the bride- groom's party, which can only be overcome by the payment of a ransom. Dr. Samter, again, believes that the stopping of the bridal procession is meant as a measure against evil spirits.3 I have myself suggested that the sham fights at weddings in Morocco may partly have a purificatory signifi- cance, as is the case with similar fights on various other occasions.* Dr. Karsten suggests that the violence to which the bride is subject when forcibly abducted is intended to purify her and to rid her of invisible supernatural enemies, whilst the fast movement at the same time will help her to escape them.5 There is presumably some truth in these suggestions ; the great speed at which bride and bridegroom are driven to and from church at German peasant weddings is obviously meant as a safeguard against supernatural dangers.6 But Dr. Karsten undoubtedly goes too far when he asserts that " the so-called symbolical capture of the bride is probably in most cases nothing but a magical ceremony."7 Of the common practice of abusive language at weddings,8 which has also been supposed to be a relic of 1 Stack, Mihirs, p. 18. 2 Ellis, Ewe-speaking Peoples, p. 155 sq. 3 Samter, op. cit. p. 169. 4 Westermarck, Marriage Ceremonies in Morocco, pp. 128, 224, 261, 268, 325, 327. 6 Karsten, Studies in South American Anthropology, i. 198. 6 Sartori, Sitte und Brauch, i. 83, 89 sq. 7 Karsten, op. cit.i. 198. 8 See, e.g., Hagen, Unter den Papua's, p. 243 (Bogadjim) ; Dale, ' Customs and Habits of the Natives inhabiting the Bondei Country', in Jour. Anthr. Inst. xxv. 199; Junod, op. cit. pp. no, in, 120 (Thonga) ; Westermarck, Marriage Ceremonies in Morocco, p. 197. 264 THE HISTORY OF HUMAN MARRIAGE chap. marriage by capture, Mr. Crawley remarks that in many instances, as in European folk-custom, " the abuse is directed against the ' evil eye ' and possible external danger to the young couple."1 Very often resistance is made, or grief expressed, princi- pally or exclusively by the bride. And this has also been represented as a survival of earlier marriage by capture. Of the Araucanians, for example, we are told that their marriage ceremonies consist in nothing more than in carrying off the bride by pretended violence, which is considered an essential pre-requisite to the nuptials. " The husband, in concert with the father, conceals himself with some friends near the place where they know the bride is to pass. As soon as she arrives she is seized and put on horseback behind the bridegroom, notwithstanding her pretended resistance and her shrieks, which are far from being serious. In this manner she is conducted with much noise to the house of her husband, where her relations are assembled, and receive the presents agreed upon, after having partaken of the nuptial entertainment."2 In various tribes of the Amazons the bridegroom's party, in the middle of the wedding festivi- ties, suddenly kidnap the bride without any show of opposi- tion on the part of her friends and family, and among some of them " it is a point of honour for the bride to scream and protest while the groom carries her off with mock assistance from his friends."3 Among the Ipurina, on the Upper Rio Purus, when the girl's father has given his consent to her marriage she takes to flight, and if the young man succeeds in catching her they are without further ceremony regarded as husband and wife.4 On the east coast of Green- 1 Crawley, op. tit. p. 352 sq. 2 Molina, Geographical, Natural, and Civil History of Chili, ii. 115 sq. See also Smith, Araucanians, p. 215 ; Guevara, Historia de la Civilization de Araucania, i. 205 ; Latcham, ' Ethnology of the Araucanos,' in Jour. Roy. Anthr. Inst, xxxix. 359. For crying and resistance of the bride among the Mocobis of the Gran Chaco see Kohler, Pater Florian Baucke, p. 315. 3 Whiffen, North-West Amazons, p. 164. Cf. v. Martius, op. tit. i. 600; Wallace, Travels on the Amazon, p. 479 (Uaup£s). 4 Ehrenreich, Beitrage zur Volkerkunde Brasiliens, p. 65. XXI MARRIAGE BY CAPTURE 265 land, according to Dr. Nansen, the only method of con- tracting a marriage is still for the man to go to the girl's tent, catch her by the hair or anything else which offers a hold, and drag her off to his dwelling without further ado. Violent scenes are often the result, as single women always affect the utmost bashfulness and aversion to any proposal of marriage, lest they should lose their reputation for modesty. But " the woman's relations meanwhile stand quietly looking on, as the struggle is considered a purely private affair, and the natural desire of the Greenlander to stand on a good footing with his neighbour prevents him from attempting any interference with another's business."1 Among the Koryak, " when the bride's father has decided that it is time to end the probation service, he tells the bride- groom that he may seize the bride ; i.e., marry her. . . . The mother warns the bride that the bridegroom has obtained the right to take her. Custom requires that the bride shall not surrender without a struggle, even if she love her bride- groom. Should the bridegroom find his bride undressed in the separate sleeping-tent which she is given before marriage, he would not touch her, considering the accessi- bility as an offence to himself. The bride's resistance is a test of her chastity. Accordingly, with the aid of her friends; the bride ties up with thongs the sleeves and trousers of her combination-suit, so that it cannot be taken off without untying or cutting the thongs. On the day when the bride- groom obtains the right to seize the bride, the latter goes about thus tied up, and tries to run away when her bride- groom approaches her. The bridegroom seizes an oppor- tunity to attack her unawares, to tear or cut the garments with a knife, and touch her sexual organs with his hand. When he has succeeded in doing so, the bride ceases to resist, and submissively leads the bridegroom to her tent. . . . Being a symbol of copulation, the act of touching the bride's sexual organs makes her the man's wife. ... A similar sym- bolic act was performed also among the ancient Kamchadal." 2 1 Nansen, First Crossing of Greenland, ii. 316 sq. 2 Jochelson, Koryak, p. 741 sq. For the Kamchadal see Steller, Beschreibung von dem Lande Kamtschatka, p. 344 sq 266 THE HISTORY OF HUMAN MARRIAGE chap. Among the Turkomas, according to Vambery, " the young maiden, attired in bridal costume, mounts a high- bred courser, taking on her lap the carcase of a lamb or goat, and setting off at full gallop, is followed by the bride- groom and other young men of the party, also on horse- back ; but she is always to strive, by adroit turns, &c., to avoid her pursuers, that no one of them approach near enough to snatch from her the burden on her lap. This game, called Kokbiiri (green wolf), is in use amongst all the nomads of Central Asia."1 Among the Lisu tribes of the Burma- China frontier, on the marriage day the village elders proceed to the home of the maid, with the youths who will assist them in bearing away the bride. " At first she will make a show of resistance, kicking and biting her carriers, whilst her family cry to the ancestral ghost that their child is being borne away and that they are powerless to keep her. Arrived at the village boundary, however, the struggling maiden is released and she walks gaily to her future home with the wedding party."2 In many of the aboriginal tribes of India it is etiquette for the bride to assume reluctance, to make resistance or to cry.3 Thus among the Gonds she must weep formally for some hours or a day before the wedding, and she is sometimes taught to cry in the proper note.4 In the wild tribes of the Malay Peninsula the bridegroom has to pursue and catch the bride, if he is to become her husband.5 Among the Negritos of the Philippines, according 1 Vamb^ry, Travels in Central Asia, p. 323. For Tibet see Puini, Il Tibet secondo la relazione del viaggio del P. Ippolito Desideri, p. 130. 2 Rose and Brown, ' Lisu (Yawyin) Tribes of the Burma-China Frontier,' in Memoirs Asiatic Soc. Bengal, iii. 263. 3 Crooke, Tribes and Castes of the North-Western Provinces and Oudh, i. p. cci.; ii. 91, 416. Endle, Kachdris, p. 46. See also Leitner, Results of a Tour in Dardistan, &c., iii. 36 sq. 4 Russell, Tribes and Castes of the Central Provinces of India, i. 147. 5 Begbie, Malayan Peninsula, p. 13. Bourien, ' Wild Tribes of the Interior of the Malay Peninsula,' in Trans. Ethn. Soc. London, N.S. iii. 81. Logan, ' Orang Binua of Johore,' in four. Indian Archipelago, i. 270. Miklucho-Maclay, * Ethnological Excursions in the Malay Peninsula,' in Jour. Straits Branch Roy. Asiatic Soc. no. 2, p. 215. Newbold, Account of the British Settlements in the XXI MARRIAGE BY CAPTURE 267 to one account, " after a young man has shown his passion for a girl, and his advances have been well received by the parents of his intended, he catches her in his arms. She breaks loose, however, and runs. He follows hard behind. Again he catches her. She resists, and once more frees herself, running away from the eager arms of her ardent lover. After this play of struggle and chase have been kept up a little while, she finally yields, and he leads her triumphantly back to her home."1 In some parts of Fiji, " the betrothal gifts having been accepted some time before, the girl was waylaid and carried off. If she was unwilling she ran away to some one who could protect her ; if she was content the marriage feast was made oh the following morning."2 In Tahiti " la jeune epouse resiste a son mari, 1'egratigne, le frappe, engage en un mot un veritable combat au moment de se donner pour la premiere fois. Et cela dans des unions ou la femme est completement et librement consentante, heureuse meme de son choix." Dr. Tautain, our informant, says that this can only be explained as a survival of capture.3 Among the Auin the bridegroom seizes the bride's wrist and almost drags her to his hut; this has likewise been explained as a relic of earlier marriage by capture.4 A similar explanation has been given of the marriage ceremony of the Akikuyu, consisting in a seizure of the btide by two friends of the bridegroom, who carry her off to her new home after her father's leave to take her has been asked ; she struggles and makes much noise, but this is "as con- ventional as the tears of the early Victorian bride."5 Among the Nandi, when the bridegroom's emissaries have come to the house of the girl's parents, she has to be coaxed into the goats' department, where the emissaries are entertained, and Straits of Malacca, ii. 407. Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. 80 note. Martin, Die Inlandstdmme der Malayischen Halbinsel, p. 867. See also infra, ii. 295 sq. 1 Lala, Philippine Islands, p. 96. 2 Thomson, Fijians, p. 203. 3 Tautain, ' £tude sur le mariage chez les Polyn^siens des lies Marquises,' in L'Anthropologic, vi. 650 sq. 4 Kaufmann, in Mittheil. Deutsch. Schutzgeb. xxiii. 156. 6 Routledge, With a Prehistoric People, p. 130 sqq. 268 THE HISTORY OF HUMAN MARRIAGE chap. must be promised a cow by her father before she consents to enter and is handed over to them. On her arrival at the husband's house nothing will induce her to enter it until her father- and mother-in-law bribe her by promising her a cow and a goat respectively. When the bridal party have entered the house, all seat themselves except the bride, who remains standing and refuses to take off the things which have been tied on her back. And her father-in-law must promise her another cow before he can prevail upon her to lay aside her impediments.1 Among the Baganda, when the bride- groom with his friends came to the village of the bride, she began to weep and her father loaded him with reproaches. " There she is, take her," he said, adding that he was a churl, that he had not paid him enough, and similar compliments. The girl cried all the time, and said, " Oh, mother, you have sold me." One of the bridegroom's friends then took her up astride of his shoulders and carried her off, the father sending with her her younger sister as an attendant. During the first three days the wife did not speak to her husband, nor did she go abroad. But on the fourth day the husband gave her a goat, which she killed and cooked for him, and then she might speak.2 Among the pastoral clans of the Banyoro " the bride was expected to be in tears when going to her husband, because she was leaving her parents."3 In Morocco, too, the bride is expected to cry during the preparations for her departure.4 Among the Bedouins of Sinai, should the girl get an inkling of her betrothal, "it is considered etiquette for her to make a show of escaping to the mountains " ; and in one tribe she actually runs off to hide in the mountains for three days instead of remaining in a tent near her father.5 Among 1 Hollis, Nandi, p. 61 sq. 2 Decle, op. cit. p. 445 sqr Cf. Roscoe, Baganda, p. 90.- 3 Roscoe, Northern Bantu, p. 39. 4 Westermarck, Marriage Ceremonies in Morocco, p. 163 J see also Index, s.v. ' Crying of the bride.' 5 Palmer, Desert of the Exodus, p. 90 sq. Cf. Burckhardt, Notes on the Bedouins and Wahdbys, p. 153 ; Jaussen, Coutumes des Arabes au pays de Moab, p. 55 :-" Chez les Beni Saher, la femme s'enfuit . . . Ie premier soir du manage." XXI MARRIAGE BY CAPTURE 269 the Nestorians of Syria the bride remains in a comer of the apartment, or the church, during the greater part of the nuptial ceremony. " When they have reached a part where hands are to be joined, several women catch hold of the bride, veiled as she is, and pull her by main strength half across the room towards her intended husband, who is at first equally resolute in his modest reluctance, but finally yields and advances towards the bride. A smart struggle ensues in his effort to secure her hand; but he at length succeeds."1 ^Ingria The ceremonial reluctance or crying of the bride is found among all Indo-European peoples.2 In the Grihyasutras -" the Folk-Lore Journals of ancient India "-a certain prayer is enjoined for the bride's crying, which proves that this crying was an essential form at an ancient Hindu marriage, as it is still in modern India.3 The Roman bride fled to the lap of her mother, and was carried off by force by the bridegroom and his friends.4 The Spartan bride- groom carried off the bride with feigned violence.5 In modern Greece, when the bridal procession starts for church, the bride bursts into tears, and refuses to follow, and on the bridesman saying, " Leave her alone, as she weeps," she replies, " Take me away from here, but let me weep." In Germany it is a very general belief that the bride's crying is auspicious, that if she weeps during the marriage ceremony she will be happy in her married life. 1 Van-Lennep, Bible Lands, p. 552 sq. 2 Haas, ' Die Heirathsgebrauche der alten Inder,' in Weber's Indische Studien, v. 327 sq. Winternitz, in Denkschriften d. Kaiserl. Akademie d. Wissensch., Philos.-hist. Classe, xl. 42 sq. Idem, in Trans. Intern. Folk-Lore Congress, 1891, p. 275 sq. v. Schroeder, Die Hoch- zeitsgebrduche der Esten, p. 86 sqq., and ch. xxiii. ' Das Entlaufen Oder Entfliehen der Braut,' p. 141 sqq. Sartori, op. cit. i. 77. de Gubernatis, op. cit. p. 153 sqq. Placucci, op. cit. p. 53 sq. (Romagna). Flachs, Rumdnische Hochzeits- und Totengebrduche, p. 34. Piprek, op. cit. pp. 5, 6, 10, 17, 33, 45, 46, 55, 83, 94, 105, 106, 143, 146 sq. (Slavs). Hartknoch, Alt- und Neues Preussen, p. 179. Birlinger, Volksihumlich.es aus Schwaben, i. 479. 3 Winternitz, in Trans. Intern. Folk-Lore Congress, 1891, p. 277. 4 Rossbach, Untersuchungen uber die romische Ehe, p. 329. 6 Plutarch, Lycurgus, xv. 4. 270 THE HISTORY OF HUMAN MARRIAGE chap. So they say in the Upper Palatinate, " She who cries not before, must cry afterwards," or, " Laughing bride, weeping wife; weeping bride, happy wife." Among Slavonic peoples the crying of the bride is most essential. Thus in Russia much importance is attached to the bride's having " a good cry," and the more she cries, the more she gains the admiration of her friends.1 In Little Russia, when the bride's tresses have been unplaited and the cap is being put on her head, she is bound to resist with all her might, and even to fling her cap angrily on the ground. Then the groomsmen, at the cry of " Bayars, to your swords," pretend to seize their knives and make a dash at the bride, who is thereupon surrounded by her friends who come rushing to the rescue.2 In Serbia, when the elders had arranged a marriage between a young man and a young woman, it was formerly the custom for the bridegroom to seize the bride by force and carry her off; and it would have been considered highly indelicate for the girl to be a consenting party to the arrangement.3 Ceremonial resistance and crying on the part of the bride is also found among Ugro-Finnic peoples.4 Kulischer has tried to show that the official crying of the bride in modern Europe belongs to the survivals of marriage by capture;5 and this view is also favoured by De Gubernatis6 and Dr. Winternitz.7 Among some peoples it is the custom for the bride or newly-married wife to run away from her new home, or at least to make an attempt to do so. Thus among the Tsous of Formosa, after the bridegroom, with a sham exhibition of force, has carried the bride to his home, she on the following morning runs away and returns to the home of her mother. 1 Winternitz, in Trans. Intern. Folk-Lore Congress, 1891, p. 275 sq. Schonwerth, op. cit. p. 87 (Upper Palatinate). 8 Ralston, Songs of the Russian People, p. 285. 3 Forster and Daniell, Life and Letters of O.G. de Busbecq, i. 96. 4 Hamalainen, op. cit. p. 239 sq. 5 Kulischer, in Zeitschr. f. Ethnol. x. 208. 6 de Gubernatis, op. cit. p. 154. 7 Winternitz, in Denkschriften d. Kaiserl. Akademie d. Wissensch., Phil.-hist. Classe, xl. 43. Idem in Trans. Intern. Folk-Lore Congress, 1891, p. 276. See also Hillebrandt, op. cit. p. 3. XXI MARRIAGE BY CAPTURE 271 " Here she remains three days, when the friends of the bride- groom visit her house ; and again, with a show of force, she is carried to the home of her lover, where a social meeting is held and the marriage ceremony is considered at an end."1 Among the Hos, three days after the bride has been brought into the bridegroom's village she flies away, pretending that she does not love him and no longer wants to live with him ; but he pursues her and takes her back by force.2 The Zulu bride wanders about the bridegroom's kraal, followed by his female relatives. " She is ' crying for her father's house,' where she was well treated. Now she has come to a strange household where she may be ill used, and where she has only the certainty of hard work. She is supposed to be trying to run away, and the girls to be preventing her."3 Among the Herero, after the young wife has been brought to her husband's village with an exhibition of grief, she makes several attempts to run away to show her love of her parents and her old home.4 Among the Bakongo the bride is bound with bush-rope, carried to her husband's house, accompanied by her bridesmaids, and is then released by the husband, who is all amiability. " But a day or two afterwards she may run back home out of coquetry, outraged modesty, or for some real grievance with her new life. She is speedily recaptured, and perhaps receives a mock beating, or a real one, till at last she settles down into married life."5 The resistance and weeping of the bride can no more than the opposition made by her relatives be regarded as a survival of an earlier custom of marriage by capture. As her relatives are naturally reluctant or sorry to part with her, so she is sorry to be separated from them ; and in either case the feeling of sadness or grief is ceremonially expressed and 1 Davidson, op. cit. p. 571. For a similar custom among the Lolos of Western China see Henry, ' Lolos and other Tribes of Western China,' in Jour. Anthr. Inst, xxxiii. 106. 2 de Gryse, ' Les premiers habitants du Bengale,' in Missions Catholiques, 1897, p. 392. 3 Tyler, Forty Years among the Zulus, p. 203. 4 v. Francois, Nama und Damara Deutsch-Sild-West-Afrika, p. 196. 5 Johnston, George Grenfell and the Congo, ii. 680. 272 THE HISTORY OF HUMAN MARRIAGE chap. emphasised, at the wedding. But, as clearly appears from some of the statements quoted above, the behaviour of the bride is also largely due to coyness or sexual modesty, real or assumed. This was pointed out by Spencer as one origin of the ceremony of capture / and before him C. O. Muller had explained the Spartan ceremony as an ancient custom founded on the idea " that the young woman could not surrender her freedom and virgin purity, unless compelled by the violence of the stronger sex."2 As we have seen in an earlier chapter, sexual modesty shows itself in the pre- liminaries to, or the conclusion of, a marriage not only on the part of the woman but on the part of the man.3 The bridegroom, too, may have to be " captured." Among the Garos, on the day fixed for the celebration of marriage " some relatives of the bride go over to the bridegroom's machan (or hut) with a view to bring him in. On seeing them coming, or on being apprised of their coming, the latter takes to his heels and runs into a solitary room or enters a forest. The bride's people make diligent search, and on finding him out, try to bring him by force, at the same time holding out all sorts of temptations in order to induce him to consent to the marriage ; and on these means failing to have effect, they throw him down in a pool of water, and when on being ducked two or three times he at length expresses his consent, they take him out of the water and triumphantly lead him captive to the bride's house. The bride, too, on her part flees into a lonely room; but what makes her case somewhat different from that of the bridegroom is that she never goes into a forest."4 Among the Greeks of Southern Macedonia it is the bridegroom and not the bride who is " lifted." On the last day of the week-long celebration an envoy from the bride comes to gird the bridegroom, attempting at the same time to lift him from the ground, and he resists to the best of his ability.5 1 Spencer, Principles of Sociology, i. 623 sq. Idem, in Fortnightly Review, N.S. xxi. 897 sq. 2 Muller, History and Antiquities of the Doric Race, ii. 298 sq. 3 Supra, i. 424 sqq. 4 Chunder Dey, ' Account of the Garos,'in Calcutta Review, cxxviii. 159- Cf- Playfair, Garos, p. 67. 6 Garnett, Women of Turkey, i. 84. XXI MARRIAGE BY CAPTURE 273 There are yet other sham fights at weddings, which are obviously no survivals of marriage by capture, but accen- tuate the antagonism between different social groups and the solidarity of the members of each of them.1 The struggling bride may be assisted by her female friends, and in various cases the struggle is chiefly between them and the bridegroom.2 Among the Araucanians the attack made by the bridegroom and his male relations on the bride's dwelling is energetically resisted by all the womenfolk, the men of the household remaining passive spectators.3 Of the Mos- quito Indians we are told that after the wedding is all arranged and the presents have been paid, the bridegroom seizes his bride and carries her off, followed by her female relatives, who pretend to try to rescue her.4 Among the Kamchadal, after having served for his bride and at last obtained permission to seize her, the bridegroom " seeks every opportunity of finding her alone, or in the company of a few people ; for during this time all the women in the village are obliged to protect her. ... If the bridegroom happens to find her alone, or in company but with a few, he throws himself upon her, and begins to tear off her clothes, nets, and straps ; for to strip the bride naked constitutes the ceremony of marriage. This is not always an easy task ; for though she herself makes small resistance (and indeed she can make but little), yet, if there happen to be many women near, they all fall upon the bridegroom without any mercy, beating him, dragging him by the hair, scratching his face, and using every other method they can think of to prevent him from accomplishing his design." The victory is seldom obtained by him at once; indeed, the contest sometimes lasts a whole year, and after each attempt the 1 See infra, on Marriage Rites. 2 Crooke, 'Hill Tribes of the Central Indian Hills,' in Jour. Anthr. Inst, xxviii. 239. Ramsay, quoted by Das, in Jour. Asiatic Soc. Bengal, vol. Ixii. pt. iii. 3° (Ladakh). Skeat, Malay Magic, p. 381. Cummins, ' Sub-tribes of the Bahr-el-Ghazal Dinkas,' in Jour. Anthr. Inst, xxxiv. 150. Dundas, 'Wawanga and other Tribes of the Elgon District, British East Africa, ibid, xliii. 3^' 3 Latcham, in Jour. Roy. Anthr. Inst, xxxix. 359. 4 Bancroft, op. cit. i. 733. VOL. II 1 274 THE HISTORY OF HUMAN MARRIAGE chap. bridegroom is obliged to take some time to recover strength and to cure the wounds he has received.1 So also among the Lolos near Mount Wa in the interior of China the resist- ance offered the bridegroom's friends when they seize the bride is rather more than simulated ; for although her male friends only repel the attacking party with showers of flour and wood-ashes, the attendant virgins are armed with sticks, which they have the fullest liberty to wield.2 Among the Kandhs, when the bridegroom goes to the bride's place accompanied by all the men of his village, they are there lesisted by all the women of the bride's village, armed with stones and sticks; " elles sont disposees a ne pas laisser voler une de leurs filles sans se defendre."3 In his descrip- tion of the tribes of the Hindu-Kush Mr. Biddulph states that in Gilgit, when the bridegroom and his friends have fetched the bride, the women of her family follow them, assailing the bridegroom with abuse and pelting him with mud and filth, feigning anger. But after going a mile in this way, the bridegroom gives a present to the bride's mother, and is then allowed to depart in peace. " This," our informant adds, "is no doubt a relic of the practice of marriage by capture."4 In Hazara,in the Punjab, the women who are the guests and bystanders at the ceremony in the bride's place " find an immoral delight in pelting the bride- groom's procession with such abuse as gives us an appalling view of the standard of social morality common among the generality of the population " ; 5 and among the Magh tribes of Chittagong " the unfortunate bridegroom is the butt of jests from the female friends of the bride."® Among some South Arabian Bedouins the bridegroom and his bachelor friends have to search for the bride when they come to fetch her ; and when they find the cavern into which 1 Krasheninnikoff, History of Kamschatka, p. 212 sq. 2 Baber, ' Travels and Researches in the Interior of China,' in Roy. Geograph. Soo. Supplementary Papers, i. 69. 3 Rossillon, ' Moeurs et Coutumes du peuple Kui,' in Anthropos, vii. 102. 4 Biddulph, Tribes of the Hindoo Koosh, p. 80. 6 Wace, quoted by Tupper, op. cit. ii. 92. 0 Hutchinson, op. cit. p. 120. XXI MARRIAGE BY CAPTURE 275 she has fled, they are received by her girl companions with a shower of stones.1 Mr. Crawley says that in such cases the bride is " captured" and taken away from her sex, " who, by psychological necessity, take her part."2 To this M. van Gennep makes the objection that she is not removed from her sex, which she cannot change ; " mais elle quitte une certaine societe sexuelle restreinte, tant familiale que locale, pour etre agregee a une autre soci£t£ sexuelle restreinte tant familiale que locale." And he says that he knows of no case " ou la solidarity sexuelle soit generale, c'est-a-dire ou les filles et femmes de la famille du clan et de la tribu du jeune homme s'opposent de leur cote a 1'entree de la fiancee."3 But in Morocco the bridegroom is sometimes attacked by all the women assembled outside his house,4 or they curse both his and the bride's father, as if the marriage were an offence against their sex ;6 and the sex antagonism is also conspicu- ous in the fights which take place between the bachelors and the unmarried women6 or the women in general,7 in the young men's attempts to take something from the bride, who is defended by the other women and her bridesmen,8 and in the robberies which the men of the bridal procession commit on the bridegroom's mother and sisters as well as on the bride.9 Among the Arabs of Moab, again, the women of the bridegroom's tribe attack the bride when the pro- cession approaches a camp in the tribe ; they do not want to receive the stranger, whose arrival they regard as an insult to their own beauty and attractions.10 Marriage by capture is supposed to have left behind traces in the language spoken by the women. To such an extent, we are told, did the Caribs capture women from neighbouring tribes that the men and women spoke different languages or, at any rate, had two more or less distinct vocabularies, 1 v. Wrede, Reise in Hadhramaut, p. 262 sq. 2 Crawley, op. cit. pp. 352, 361, 370. 8 van Gennep, Les rites de passage, p. 179 sq. 4 Westermarck, Marriage Ceremonies in Morocco, pp. 211, 223. 6 Ibid. pp. 197, 223, 346. 6 Ibid. pp. 237, 238, 269, 346. 7 Ibid. pp. 245, 247, 261, 268, 346. 8 Ibid. pp. 204, 223, 346. 9 Ibid. pp. 204, 223, 346. 10 Jaussen, op. cit. p. 53. 276 THE HISTORY OF HUMAN MARRIAGE chap. one of which was used by the men and by the women when speaking to the men and the other only by the women between themselves or by the men when repeating in oratio obliqua some saying of the women.1 This theory, however-• which may be traced back to a suggestion by Breton in his ' Dictionnaire Caraibe-Fran^ais ' afterwards accepted by several writers2-has been justly criticised. Differences between the speech of men and women are found all over America and in many other parts of the world,3 but nowhere -not even among the Caribs4-are these differences so great as to constitute two different languages in the proper sense of the word. They may be due to various causes. Sapper5 and Lasch® have emphasised the social-economic factor of differentiation of occupation and labour. Peculi- arities of speech are always apt to arise among people who are closely associated with each other, as the inhabitants of the same district or the members of the same class of society; and the segregation of the sexes naturally leads to a similar result. As Mr. Crawley points out, " in modern Europe sexual separation to some extent still influences popular language, women and men respectively using certain terms peculiar to each sex."7 There is the specifically sexual taboo : a special vocabulary for the generative organs and functions is very widespread.8 The so-called " lan- guage " of the Kafir women, known as the uku hlonipa, is said to arise from the repugnance which the women have to mention the names of their fathers or fathers-in-law and any word which resembles such names.9 As regards the 1 McLennan, Studies in Ancient History, p. 34. Avebury, Origin of Civilisation, p. 90. Im Thurn, op. cit. p. 186. Rat, ' Carib Lan- guage as now spoken in Dominica,' in Jour. Anthr. Inst, xxvii. 311. 2 Lasch, ' Uber Sondersprachen und ihre Entstehung,' in Mitteil. Anthrop. Gesellsch. Wien, xxxvii. 95 sqq. 3 Ibid. p. 98 sqq. Crawley, op. cit. p. 46 sq. * Lasch, loc. cit. p. 97. Rat, loc. cit. p. 311. 8 Sapper, ' Mittelamericanische Caraiben, ' in Internationales Archiv f. Ethnographic, x. 57. fi Lasch, loc. cit. p. 97. 7 Crawley, op. cit. p. 47. 8 Ellis, Studies in the Psychology of Sex, i. Evolution of Modesty, p. 67. 8 Oppert, * On the Classification of Languages in conformity with Ethnology,' in Jour. Anthr. Inst. xiii. 41. XXI MARRIAGE BY CAPTURE 277 women's dialect among the Karaya on the Araguaya, Ehrenreich1 and Krause2 suggest that this dialect, with its fuller tone, represents an older form of the tribal speech, which has been retained by the women. " Of a distinct language," says the latter authority, " which might have arisen through the reception of female captives from foreign tribes, there is no question here. On the one hand, the taking over of foreign women (Tapirape, Kayapo, Savaje) is too limited to have enabled their language to have any influence, and, on the other hand, the deviations from the men's language are too small to allow us to look for a foreign derivation of the women's language."3 The comparative isolation of the women from the outside world undoubtedly accounts for the fact, noticed by myself among the Berbers of the Great Atlas, that the women use the old Berber num- erals in cases where the men invariably use Arabic loan- words. Finally it should be remembered that, even if certain peculiarities of the women's speech are possibly due to exogamous habits, there may be exogamy without marriage by capture. Many other practices have been regarded by imaginative writers as survivals of marriage by capture. This is the case with the lifting of the bride over the doorstep,4 the veiling of the bride,5 the use of rings at weddings,6 the throwing of a slipper after the departing bride and bride- groom,7 the avoidance of parents-in-law,8 and even our honeymoon " during which the bridegroom keeps his bride away from her relatives and friends."9 Some of these customs have been mentioned above, in the chapter on sexual modesty, and the others will be dealt with in our discussion of marriage rites. None of them has the faintest claim to be in any way associated with marriage by capture. 1 Ehrenreich, ' Materialien zur Sprachenkunde Brasiliens, ' in Zeitschr. f. Eihnol. xxvi. 23. 2 Krause, In den Wildnissen Brasiliens, p. 344. 3 Ibid. p. 344. 4 See infra, ii. 537 n. 1. 5 v. Schroeder, Die Hochzeitsgebrauche der Esten, p. 77. 6 Kulischer, in Zeitschr. f. Ethnol. x. 208 sq. 7 McLennan, op. cit. p. 14 sq. n. 3. Avebury, op. cit. p. 103. 8 Avebury, op. cit. p. 103. 9 Ibid. p. 102 sq. CHAPTER XXII CONSENT AS A CONDITION OF MARRIAGE Unlike the unions between the sexes in the lower animal world, human marriage is a social institution, which concerns not only the contracting parties but other individuals as well. Hence the conclusion of a marriage may require the consent of the latter or even be arranged by them; and, on the other hand, the consent of the bride or the bridegroom or both may be dispensed with. We shall begin with the lower races. We have seen that infant- or child-betrothals are common among many of them,1 and in such cases the consent of the parties is out of the question. But often enough the betrothal is not con- sidered binding on either party,2 or is regarded as binding on the female only,3 although we may assume that as a rule 1 Supra, ch. x. 2 Appun, ' Die Indianer von Britisch-Guayana,' in Das Ausland, xliv. 446. Boas, ' Central Eskimo,' in Ann. Rep. Bur. Ethnol. vi. 578 ; Hodge, Handbook of American Indians, i. 809 (Eskimo). Petroff, ' Report on the Population, &c. of Alaska,' in Tenth Census of the United States, p. 158 (Atkha Aleut). Batchelor, Ainu and their Folk-Lore, pp. 223, 226 sq. Rivers, Todas, p. 504. Hahl, ' Ueber die Rechtsanschauungen der Eingeborenen der Blanchebucht und des Innern der Gazelle Halbinsel,' in Nachrichten uber Kaiser Wil- helms-Land und den Bismarck-Archipel, 1897, p. 78 sq. de Rochas, La Nouvelle Caledonie, p. 231. Tautain, ' Rtude sur le mariage chez les Polynesiens des lies Marquises,' in L'Anthropologic, vi. 645. Routledge, With a Prehistoric People, p. 124 sq. (Akikuyu). Decle, Three Years in Savage Africa, p. 490 (Akamba). Torday, Camp and Tramp in African Wilds, p. 94 (Bambala). 3 Ellis, Eiie-speaking Peoples of the Slave Coast, p. 201. Dennett, At the Back of the Black Man's Mind, p. 198 sq. Sarbah, Fanti Customary Laws, p. 45 sq. Guillemard, Cruise of the Marchesa, p. 389 (Nufoor Papuans). ch. xxn CONSENT AS A CONDITION OF MARRIAGE 279 the marriage is consummated.1 Among other savage peoples marriage contracts are concluded by the parents of the parties even when these are grown-up.2 This was the case among the Algonkin, of whom Charlevoix wrote :- " Treaties of marriage are entirely carried on by the parents : the parties interested do not appear at all, and give them- selves up entirely to the will of those on whom they depend. . . . However, the parents do not conclude any thing without their consent; but this is only a formality."3 Among the Natchez no match was ever concluded without the agree- ment of the heads of both families, who were usually great- grandfathers ; but the young man first obtained the girl's own consent.4 Among the natives of the Gazelle Peninsula of New Britain it is the rule that a wife is chosen not by the future husband himself but by his maternal uncle or other relatives; but if the parties object to the choice, the matter drops.5 In Tikopia, one of the Santa Cruz Islands, and in the Banks Islands, again, when a man is old enough to marry, a wife is chosen for him by his father's sister, or 1 Among the Maori the breaking of a betrothal made in infancy was always regarded as an insult and had to be avenged (Tregear, Maori Race, p. 285). 2 Hill Tout, 'Report on the Ethnology of the South-Eastern Tribes of Vancouver Island,' in Jour. Roy. Anthr. Inst, xxxvii. 310 (Salish tribes). Sproat, Scenes and Studies of Savage Life, p. 97 (Aht). La Flesche, ' Osage Marriage Customs,' in American Anthropologist, N.S. xiv. 127. Marston, ' Letter to Jedidiah Morse,' in Blair, Indian Tribes of the Upper Mississippi Valley, ii. 165 (Sauk and Foxes; sometimes). Lumholtz, Unknown Mexico, ii. 93 (Huichol; formerly). Geiseler, Die Oster-Insel, p. 41 (Easter Islanders). Shand, ' Moriori People of the Chatham Islands,' in Jour. Polynesian Soc. vi. 145. Thomson, Fijians, p. 175. Kubary, ' Die Palau-Inseln in der Siidsee,' in Jour. Museum Godeffroy, iv. 55. Finley and Churchill, Subanu of Mindanao, p. 39. Davidson, Island of Formosa, p. 573 (Tsalisens). 3 Charlevoix, Voyage to North-America, ii. 38. See also Hodge, op. cit. i. 809 ; Speck, Family, &c. of various Algonkian Bands of the Ottawa Valley, p. 23 (Timagami band of Ojibwa), 4 Le Page du Pratz, History of Louisiana, p. 344. 5 Hahl, loc. cit. p. 78. Parkinson, Dreissig Jahre in der Siidsee, pp. 62, 65. Cf. Pfeil, Studien und Beobachtungen aus der Siidsee, p. 28 (Bismarck Archipelago). 280 THE HISTORY OF HUMAN MARRIAGE chap. if he himself selects one, the choice is ratified by her ; "a man would never marry against the will of his father's sister."1 Among the Bagobo of Mindanao the parents of the boy as a rule selected the girl and negotiated the match ; and it was not customary for the young people to refuse to carry out the wishes of their elders.2 In many of the un- civilised tribes of India marriages are generally planned and arranged by the parents of the young people con- cerned.3 Among the Mundas, for example, the selection is ordinarily made for the boy as well as for the girl by the father or guardian ; but the boy's approval of the selection is frequently sought for and generally obtained.4 Among the Oraons " the boy and girl have absolutely nothing to say in the matter. Everything is settled by the parents."5 And among the Kisans " there is no instance on record of a youth or maiden objecting to the arrangement made for them."6 In various African tribes, also, the parents, and especially the father, arrange for the son's marriage.7 1 Rivers, History of Melanesian Society, i. 39, 49, 309. 2 Cole, ' Wild Tribes of Davao District, Mindanao,' in Field Museum of Natural History, Anthropological Series, xii. 101 sq. 3 Stoll, ' Notes on the Yoon-tha-lin Karens,' in Madras Journal of Literature and Science, N.S. vi. 61 sq. Endle, Kacharis, pp. 44 (Kacharis), 95 (Deoris). Rowlatt, ' Expedition into the Mishmee Hills,' in Jour. Asiatic Soc. Bengal, vol. xiv. pt. ii. 488. Hutchinson, Account of the Chittagong Hill Tracts, p. 96 (Chukmas). Dalton, Descriptive Ethnology of Bengal, pp. 194, 215, 279 (Muhdas, Santals, Marias). Hunter, Rural Bengal, iii. 83 (Kandhs). Malcolm, ' Essay on the Bhills,' in Trans. Roy. Asiatic Soc. i. 83. Thurston, ' Anthropology of the Todas and Kotas of the Nilgiri Hills,' in the Madras Government Museum's Bulletin, i. 195 (Kotas). Sherring, ' Notes on the Bhotias of Almora and British Garhwal,' in Memoirs Asiatic Soc. Bengal, i. 98. 4 Dalton, op. cit. p. 194. Sarat Chandra Roy, Mundas, p. 437. 5 Dehon, ' Religion and Customs of the Uraons,' in Memoirs Asiatic Soc. Bengal, i. 161. 6 Dalton, op. cit. p. 132. 7 Grandidier, Ethnographic de Madagascar, ii. 166 (certain tribes). Johnston, Uganda Protectorate, p. 878 (Nandi). Baumann, Durch Massailand zur Nilquelle, p. 221 (Warundi). Felkin, ' Notes on the Madi or Moru Tribe of Central Africa,' in Proceed. Roy. Soc. Edin- burgh, xii. 320. Roscoe, Northern Bantu, pp. 38, 39 (Banyoro), 172 sq. (Bagesu). xxii CONSENT AS A CONDITION OF MARRIAGE 281 Among the Batoro, south of Lake Albert Nyanza, the father of the young man goes to the father of the girl, and the matter is arranged ; " the young people are told about it, but they have no voice in the matter,"1 Among the Xosa Kafirs it is the rule that the young man's father chooses the first, and sometimes even the second, wife for his sons.2 When a Basuto youth, through his behaviour, indicates that he wants to marry, no questions are asked as to whom he wishes for a wife. " His wishes are very secondary considerations, and not to be weighed for a moment against those of his father. If the wife selected by his father does not happen to be the ' lady of his heart,' he is at liberty to choose a second wife for himself, as soon as he can pay the dowry or persuade his father to pay it for him."3 Among the Herero the young people have as a rule no liberty of choice, their, fathers or parents, or other elderly relatives as well, arranging the matter without consulting their wishes.4 Among some peoples a marriage is considered to require the consent not only of some near relatives but of the communities to which the parties belong. Among the Maori, according to Mr. Best, " in the arranging of a marriage it is not only the families of the young couple who take part in such, but also the family group, or the hafiu (that is, clan), or perhaps even the whole tribe-i.e., in a marriage of important persons. Indeed the parents often have little to say in regard to the marriages of their children, the leading part in the arrangements being taken by the brothers and sisters of the parents." The girl's consent is also asked, before all the assembled people, but even now it is a common thing, when a marriage occurs, to hear the remark made, " The tribe married them."5 In the kingdom of Acheh, on 1 Cunningham, Uganda, p. 52. 2 Kropf, Das Volk der Xosa-Kaffern im ostlichen Sudafrika, p. 133 sq. See also Kidd, The Essential Kafir, p. 212. 3 Minnie Martin, Basutoland, p. 81. See also Casalis, Basutos, p. 186. 4 Biittner, ' Sozialpolitisches aus dem Leben der Herero in Damaraland,' in Das Ausland, Iv. 853. Kohler, ' Das Recht der Herero,' in Zeitschr. f. vergl. Rechtswiss. xiv. 301. 5 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New Zealand Institute, xxxvi. 34 sqq. See also Colenso, Maori Races of New Zealand, p. 25. 282 THE HISTORY OF HUMAN MARRIAGE chap. the east coast of Sumatra, the parents of the young man have before taking the first step asked the permission of the headman of their own gampong, or village community, for the proposed marriage, and the parents of the girl on their part do the same with their local authorities. The headmen have power to prevent a proposed marriage from taking place, although the headman of the girl's gampong will seldom refuse his consent. Thus marriage is here by no means a mere family matter, but at least as much an affair of the whole village community.1 Among the Australian aborigines marriages are often arranged by the camp council or the leading men of the community. In some Queensland tribes, according to Dr. Roth, " each male can at least have two wives-an official one supplied him, as a member of the community, by the camp in general council assembled, and an unofficial one of his own choice, whose love, such as it is, he finds reciprocated. ' '2 Among the Narrinyeri " marriages are generally, but not always, arranged by the clans. The marriage ceremony consists in the father, or eldest brother, or nearest male relative of the woman, formally giving her to her future husband in the presence of the assembled clans or relatives. She signifies her acceptance of the giving by making a fire for her husband."3 Among the Dieri wives are given " by the elders and chief of the tribes, after consultation with the near relatives."4 Among some peoples a young man who wants to marry must first obtain the permission of his chief ;5 and the Dahoman theory is that the King gives wives to the people, parents having no sort of property in their children.6 We do not know, however, the actual extent of interference in such cases. It may be said that 1 Snouck Hurgronje, Achehnese, i. 299. 2 Roth, Ethnological Studies among the North-West-Central Queensland Aborigines, p. 180. See also Lang, Queensland, p. 337 ; Idem, Cooksland in North-Eastern Australia, p. 394. 3 Taplin, Folklore, &c. of the South Australian Aborigines, p. 35. 4 Gason, ' Dieyerie, &c.,' in Jour. Anthr. Inst. xxiv. 16g. 5 Johnston, Uganda Protectorate, p. 632 (Bahima). Isabella Bird, Unbeaten Tracks in Japan, ii. 99 (Ainu). 6 Norris, quoted by Ling Roth, Great Benin, p. 40 n. 2. xxii CONSENT AS A CONDITION OF MARRIAGE 283 among every people the selection of mates is controlled by its customs or laws. And sometimes-as in the case of obligatory levirate, sororate, and cross-cousin marriage-the choice of partners may be entirely independent of all indi- vidual wishes. Marriage brings together strange families or larger groups of kindred, or constitutes a new tie between friendly ones. It is therefore not surprising that the fathers or parents or other relatives of the parties want to have a voice in the matter. Speaking of the careful supervision of marriages among the Maori, Mr. Best observes, "To a great extent it was caused by tribal anxiety to avoid a mesalliance, to prevent a person of good birth from marrying into a family of ware, or low bom people, to keep unmixed the blood of the rangatira class, to uphold the rank, fame, and dignity of first-bom lines of descent, and hence to prevent all ti-puheke, or degeneration, of blue-blooded lines."1 The power of interference depends, of course, upon the authority which the families or their heads possess over the individual members of the family. Among the Xosa Kafirs, for example, the father rules as long as he lives over the whole of his family, including the married sons.2 Among the Kandhs, in each family the absolute authority rests with the house- father : the sons have no property during his lifetime, and all the male children, with their wives and descendants, continue to share the father's meal, prepared by the common mother.3 The father chooses a full-grown woman as a wife for his young son. " In the superior age of the bride," says Colonel Macpherson, " is seen but a proof of the supremacy of the paternal authority amongst this singular people. The parents obtain in the wives of their sons during the years of their boyhood very valuable domestic servants, and their selections are avowedly made with a view to utility in this character."4 In this and other uncivilised tribes of India where not even the grown-up son is allowed to choose his mate, Hindu influence may have made itself felt. It is said 1 Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 50. 2 Kropf, op. cit. p. 134. 3 Hunter, Rural Bengal, iii. 72 4 Macpherson, Memorials of Service in India, p. 70 sq. 284 THE HISTORY OF HUMAN MARRIAGE chap. that among the Mundas young folk of both sexes in earlier times had a freer hand than now in the choice of their part- ners ;x and it seems as if the same was the case among the Oraons.2 Among most savage peoples a father has apparently little if any authority over his full-grown son,8 and the latter marries independently of his will. Sometimes parental arrangements may be a necessary consequence of the separation betw'een the sexes. Among the Osage, according to Mr. La Flesche, " young people, unless near relatives, are not allowed to mingle or even to speak to one another. They are strictly guarded, so that no couple can arrange their own marriage affairs, and open courtship or love-making becomes impossible." A suitable maiden to be found is one who in the opinion of the young man's family would make him a good wife.4 Much more frequently than the young man is the young girl dependent upon somebody else's will in the choice of a partner. Although the subject of family authority among the lower races requires much further investigation, I think we may safely say that among most of them a girl is in her father's power till she marries, whilst in some instances his authority over her continues even after her marriage.5 Yet among various peoples the consent of a mother,6 1 Sarat Chandra Roy, Mundas, p. 436 sq. 2 Cf. Dehon, in Memoirs Asiatic Soc. Bengal, i. 161, and Dalton, op. cit. p. 252. 3 See Westermarck, Origin and Development of the Moral Ideas, i. 601 sq. 4 La Flesche, in American Anthropologist, xiv. N.S. 127. 5 See Westermarck, Origin and Development of the Moral Ideas, i. 601. 6 Bovallius, Resa i Central-Amerika, ii. 301 (Simoo). Keating, Narrative of an Expedition to the Source of St. Peter's River, ii. 157 (Chippewa). Morgan, League of the Iroquois, p. 320. Hodge, op. cit. i. 809 (Hurons ; " proposals made to the girl's mother were sub- mitted by her to the women's council, whose decision was final among the Hurons "). Hardisty, ' Loucheux Indians,' in Smithsonian Report, 1866, p. 322. Meinicke, Die Inseln des Stillen Oceans, ii. 407 (Ladrone Islanders). Kubary, ' Die Bewohner der Mortlock Inseln,' in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, p. 260. Hagen, Unter den Papua's, p. 241 (Bogadjim). Howitt, Native Tribes of South-East Australia, pp. 177, 198 (Dieri), 217 (Wollaroi), xxn CONSENT AS A CONDITION OF MARRIAGE 285 brother,1 or maternal uncle2 is regarded as particularly essen- tial to a girl's marriage, and in such cases the father may even 222 (Wakelbura). Weule, Native Life in East Africa, p. 307 (Makua and Makonde). Draper, quoted by v. Rosen, Trdskfolket, p. 420 (Baleng: of North-Western Rhodesia). Kaufmann, ' Die Auin. Ein Beitrag zur Buschmannforschung,' in Mittheil. Dexitsch. Schutz- geb. xxiii. 156. Torday, quoted by Johnston, George Grenfell and the Congo, ii. 683 (Bahuana). Weeks, Among the Primitive Bakongo, P- *43- 1 v. Martius, Beitrdge zur Ethnographic Amerika's, i. 217 (Guarayos). Rivet, ' Les Indiens Jibaros,' in L'Anthropologic, xviii. 606. Thom- son, Story of New Zealand, i. 177 ; Shortland, Traditions and Super- stitions of the New Zealanders, p. 118 ; Tregear, op. cit. p. 293 (Maori). Percy Smith, ' Niue Island,' in Jour. Polynesian Soc. xi. 206. Marcuse, Die Hawaiischen Inseln, p. 108 (higher class Hawaiians). Howitt, op. cit. pp. 217 (Wollaroi), 260 (Narrinyeri) ; Holden, in Taplin, Folklore, Manners, &c. of the South Australian Aborigines, p. 17 (Maroura tribe) ; Newland, ' Parkengees on the Darling River,' in Proceed. Roy. Geograph. Soc. Australasia : South Australian Branch, ii. 21 ; Lang, Aborigines of Australia, p. 10 sq. (natives of Moreton Bay ; for brothers exchanging their sisters among the Australian natives, see also infra, ii. 354 sq.). Carey and Tuck, Chin Hills, i. 190. Dalton, op. cit. p. 201 sq. (Hos). Roscoe, Northern Bantu, p. 149 sq. (Bakene). 2 Nicholas, ' Aborigines of the Province of Santa Marta, Colombia,' in American Anthropologist, N.S. iii. 647 (Goajircs). Bartram, ' Creek and Cherokee Indians,' in Trans. American Ethn. Soc. vol. iii. pt. i. 65 ; Schoolcraft, Archives of Aboriginal Knowledge, y. 268 (Creeks; the consent of the uncles, aunts, and brothers of the girls asked for). La Flesche, in American Anthropologist, N.S. xiv. 128 (Osage ; " if the parents of the girl are favourably inclined they communicate with the maternal uncle and consult him. If he gives his consent, all the other relatives agree to the proposed marriage "). Kubary, in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, p. 260 (Mortlock Islanders). Howitt, op. cit. p. 195 (Australian tribes extending up the Murray River). Sarat Chandra Das, ' Marriage Customs of Tibet,' in Jour. Asiatic Soc. Bengal, vol. Ixii. pt. iii. 15 sq. (in Sikkim, &c. it is necessary to arrange the marriage business not only with the parents of the girl but also with the maternal uncle). Cain, ' Bhadrachellam and Rekapalli Taluqas,' in Indian Antiquary, viii. 34 (Kois). Thurston, Castes and Tribes of Southern India, iii. 79 (Kalians of the Madura district). Munzinger, Ostafrikanische Studien, p. 528 (Barea and Bazes). Delafosse, ' Le peuple Sidna ou S^noufo,' in Revue des Etudes ethnographiques et sociologiques, i. 484. ' Negersitten,' in Das Ausland, liv. 1026 (Negroes of Bondo). Weeks, Among the Primitive Bakongo, pp. 107, 142. 286 THE HISTORY OF HUMAN MARRIAGE chap. have little or nothing to say in the matter. The necessity of the father's or somebody else's consent, however, by no means implies that the girl is, or can be, given in marriage against her own will. And even where her wishes are not consulted, she may know how to make her influence felt. I think it is a mistake to suppose that among the lower races women are, as a rule, married without having any voice of their own in the matter. In many cases, as will be seen, their liberty of selection is, on the contrary, very considerable, apart from those already noticed in which they are actually the courters. Among the Yahgans of Tierra del Fuego, according to Bove, the father decides whom his daughter shall marry, and the daughter makes no resistance j1 but the same traveller also speaks of the eagerness with which the women seek for young husbands and of the surprising fact that they nearly always attain their aims.2 Thomas Bridges writes of the same people, " It frequently happens that there is inseparable aversion on the girl's part to her husband, and she leaves him, and if she persists in hating him she is then given to one she likes."3 Cojazzi states that among the Onas there are marriages concluded from genuine affection. The young man goes to the hut of his beloved one and, without saying a word, gives her his bow, after which he retires to some place close by to wait for an answer. The girl consults her mother before she makes up her mind. If she decides to refuse the proposal she sends back the bow by a boy, whereas if she accepts it she goes herself to the suitor with the bow in her hand. They are now considered to be married to one another without further ceremony, and .the young woman accompanies her husband to his hut.4 A very similar statement is made by Gallardo, but according to him the suitor first receives the consent of the girl's father and then waits for the moment when he can present his bow 1 Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 132. * Ibid. p. 138. 3 Bridges, ' Manners and Customs of the Firelanders,' in A Voice jar South America, xiii. 184. Cf. King and Fitzroy, Voyages of the Adventure and Beagle, ii. 182. ' Cojazzi, Los indios del Archipelago Fueguino, p. 17. xxn CONSENT AS A CONDITION OF MARRIAGE 287 to the girl.1 These statements do not justify the opinion that the Fuegian girls are given in marriage without having themselves any voice in the matter.. Among the Tehuelches of Patagonia girls are never forced to marry against their will.2 The Lengua Indians of the Paraguayan Chaco, both men and women, are free to make their own choice of a partner, although in doing so they are influenced by their parents and friends.3 Of the Guanas it is said, " Aucune femme ne consent a se marier, sans avoir fait ses stipulations preliminaires tr^s-detaillees avec,son pretendu, et avec son pere et ses parents, A 1'egard de leur genre de vie reciproque."4 In the eighteenth century Father Sanchez Labrador wrote of the Guaycurus that if a man wanted to marry a certain girl he only had to get her parents' and her own consent, and then took her with him to his hut.5 Among the Matacos, living on the shores of the Rio Bermejo and the right shore of the Pilcomayo, the consent of the parties themselves is all that is needed for the conclusion of a marriage ; even the parents have nothing to say in the matter.6 Of the Chiriguanos of Bolivia, north-east of Tarija, we are told by Father Chome that if one of them wanted to marry he tried to gain the favour of his beloved one by presenting her with fruit and game. He then placed a load of fuel outside her hut; if she took it in herself they were at once regarded as husband and wife, whereas if she left it where it was the man had to start afresh to find another wife.7 A somewhat similar ceremony is found among the Zaparo Indians of Ecuador. Sometimes, it is said, the matri- monially-inclined swain goes into the woods and hunts. On his return his sport is thrown at the feet of his elect, and immediately afterwards sufficient firewood for cooking 1 Gallardo, Tierra del Fuego-Los Onas, p. 215. 2 Musters, At Home with the Patagonians, p. 186. 3 Grubb, An Unknown People, p. 214. 4 Azara, Voyages dans V Amerique meridionale, ii. 92. 5 Sanchez Labrador, El Paraguay Catolico, ii. 26. 6 Cardiis, Las Misiones Fransiscanas entre los infieles de Bolivia, P- 254- 7 Chom6, ' Dritter Brief an Rev. Patrein Vanthiennen,' in Stoeck- lein, Der Neue Welt-Bott, vol. iv. pt. xxix. 72. 288 THE HISTORY OF HUMAN MARRIAGE chap. it. Should the maiden arise and employ herself in cooking his game, he may consider himself an accepted suitor.1 Among the Arawaks the young man first makes sure that he will not be refused by the object of his love, and then pays a visit to her parents. If, after his talk with them, the girl puts before him some food, she thereby shows that his proposal has been accepted, and by his eating it the marriage is concluded.2 Among the Macusis young men and women who have not been betrothed as children choose themselves their partners.3 Among the Uaup£s a young girl is always free to choose her husband, and the consent of both parties makes them a married couple. The fair sex enjoys the greatest liberties among them :-" Les dames prennent, laissent et reprennent leurs maris ; les demoiselles prennent, laissent et reprennent leurs amants."4 Among the Central Brazilian Bororo the consent of parents is not asked ; if they are opposed to the match, a fight ensues and the defeated party leaves the village.5 Among the Karaya, on the River Araguaya, a suitor addresses himself to the mother or parents of the girl, but it rests with the girl herself to decide whether the proposal shall be accepted or not.6 Speaking of the Witoto and Boro of the North- West Amazons, Mr. Whiffen says :-" In every marriage the contracting parties are allowed complete freedom of choice. This is absolute on the part of the man, and, with the rare exception of young girls adopted into a family with a view to marriage, equally so on the part of the woman.".7 Among the Tarahumare of Mexico " the young women enjoy absolute liberty " ; 8 and among the Huichol they 1 Simson, Travels in the Wilds of Ecuador, p. 173. 2 Appun, in Das Ausland, xliv. 124. 3 Ibid. p. 446 sq. 4 Coudreau, La France equinoxiale, ii. 174 sq. 5 von den Steinen, Unter den Naturvblkern Zentral-Brasiliens, p. 501. 6 Ehrenreich, Beitrage zw Volkerkunde Brasiliens, p. 29. Krause, * Bericht uber seine ethnographische Forschungsreise in Zentral- brasilien,' in Zeitschr. f. Ethnol. xli. 499. Idem, In den Wildnissen Brasiliens, p. 325. 7 Whiffen, North-West Amazons, p. 164. 8 Lumholtz, Unknown Mexico, i. 266. XXII CONSENT AS A CONDITION OF MARRIAGE 289 are, as a rule, able to decide their own fates. Among the latter it often happens that neither the boy nor the girl asks the parents' consent, although according to the ancient custom, still in vogue among some of them, only the old people can arrange marriages properly.1 Among the Moqui the young are generally allowed to follow their own inclina- tions, and most marriages are said to be love-matches;2 and of the Pueblos generally it is said that " no girl is forced to marry against her will, however eligible her parents may consider the match." 3 Very similar statements are made with reference to a large number of other North American tribes-such as the Pawnee in former times,4 the Natchez,5 the Alibamu and Illinois,6 the Shawnee,7 Omaha,8 Kiowa,9 and others.10 Among the Hidatsa Indians of Dakota " girls are left much to their own choice in selecting husbands for themselves. Parents sometimes, by persuasion, but rarely by any harsh coercion, endeavour to influence a 1 Ibid. p. 92 sq. 2 Solberg, ' Gebrauche der Mittelmesa-Hopi (Moqui) bei Namenge- bung, Heirat und Tod,' in Zeitschr. f. Ethnol. xxxvii. 629 sq. Voth, * Oraibi Marriage Customs,' in American Anthropologist, N.S. ii. 238 sq. 3 Bancroft, Native Races of the Pacific States of North America, i. 549 n. 206. 4 Grinnell, Story of the Indian, p. 41. Idem, ' Marriage among the Pawnees,' in American Anthropologist, iv. 275. 5 Le Page du Pratz, op. cit. p. 344. 0 Bossu, Travels through Louisiana, i. 128, 232. 7 Ashe, Travels in America, p. 249 . 8 Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethnol. iii. 259. 9 Hodge, op. cit. i. 809. 10 Schoolcraft, op. cit. v. 269 (Creeks in former times). In his book on The Indian in his Wigwam (p. 72) Schoolcraft asserts that marriages among North American Indians are brought about " sometimes with, and sometimes against, the wishes of the graver and more prudent relatives of the parties," the marital rite consisting chiefly in the consent of the parties. See also Catlin, Illustrations of the Manners, &c. of the North American Indians, i. 120 ; Adair, History of the American Indians, p. 141. Heckewelder quotes instances of Indians who committed suicide because they had been disappointed in love, the girls on whom they had fixed their choice, and to whom they were engaged, having changed their minds, and married other lovers (Buchanan, Sketches of the History, &c. of the North American Indians, p. 184). VOL. II U THE HISTORY OF HUMAN MARRIAGE chap. 290 daughter in the reception or rejection of an offer."1 Among the Apache " the girls are wholly free in their choice of husbands. Parents never attempt to impose suitors upon their acceptance, and the natural coquetry of the sought- for bride is allowed full scope."2 Among the Chippewa the mothers generally settle the preliminaries to marriage without consulting the children, but the parties are not considered husband and wife till they have given their consent.3 Morgan states that among the Iroquois the mother, when she considered her son of a suitable age for marriage, looked about for a maiden whom she thought likely to accord with him in disposition and temperament, and that remonstrance or objection on the part of the children was never attempted ;4 but according to Loskiel, who wrote in the eighteenth century, the marriages of the Iroquois solely depended on the free will of the parties them- selves, as there was never any compulsion in the matter.5 Perrot said of the Algonkin that the youth tried to gain the consent of the girl, and that only then the alliance was proposed by his parents to the girl's family.6 Among the Lillooet Indians, in the south-western interior of British Columbia, at a so-called " touching " dance, any man who wished a certain girl ran up to her and seized her belt or the loose end of her sash. If the girl did not want him she pushed him off or snatched the end of her sash from his hands, and then he had to desist; whereas, if she favoured him, he danced with her, holding her by her belt or sash. When the dance was finished the chief called out the couples, who each in their turn, stepped out in front of the people ; and mentioning their names he said in a loud voice, " So and so holds so and so." If the girl did not then shake 1 Matthews, Ethnography and Philology of the Hidatsa Indians, P- 52. 2 Gremony, Life among the Apaches, p. 246. 3 Keating, op. cit. ii. 157 sq. 1 Morgan, League of the Iroquois, pp. 321, 323. 5 Loskiel, History of the Mission of the United Brethren among the Indians in North America, i. 56. 6 Perrot, ' Memoir on the Manners, &c. of the Savages of North America,' in Blair, op. cit. i. 67 sq. xxii CONSENT AS A CONDITION OF MARRIAGE 291 the man off, they were 'considered husband and wife.1 Among the Thompson Indians, when a proposal was made to a girl's parents, a meeting of her nearest kin was called by them, and the subject of the proposed marriage discussed. If all agreed in thinking the young man a suitable person, the girl was asked whether she liked him ; and we are told that she generally assented, not caring to go against the wishes of her relatives.2 Among the Nootka,3 Tlingit,4 and Kaniagmiut,5 the suitor has to consult the wishes of the young lady. Of the Indians and Eskimo of the Ungava district of Labrador Mr. Turher says that if the parents of the girl have agreed and she herself refuses, she is forcibly abducted by her lover.6 But concerning the Point Barrow Eskimo Mr. Murdoch writes, " As well as we could judge, the marriage bond was regarded simply as a contract entered into by the agreement of the contracting parties."7 Among the Kamchadal,8 the Yurak of the Yenisei,9 and other uncivilised peoples of Northern as well as Central Asia10 the inclinations of a daughter are nearly always con- sulted when she is given in marriage. Among the Ainu young people need not marry unless they choose ; even though they have been betrothed in childhood by their parents, both the young man and his fiancee have a final word in 1 Teit, ' Lillooet Indians,' in Publications of the Jesup North Pacific Expedition, ii. 268. 2 Idem, ' Thompson Indians of British Columbia, in Memoirs of the American Museum of Natural History, vol. ii. Anthropology, i. 322. 3 Macfie, Vancouver Island and British Columbia, p. 447. 4 Holmberg, ' Ethnographische Skizzen uber die Volker des russischen Amerika,' in Acta Soc. Scient. Fennica:, iv. 314. 5 Sauer, Expedition to the Northern Parts of Russia performed by Billings, p. 177. 6 Turner, * Ethnology of the Ungava District,' in Ann. Rep. Bur. Ethnol. xi. 270, 188. 7 Murdoch, ' Ethnological Results of the Point Barrow Expedi- tion,' in Ann. Rep. Bur. Ethnol. ix. 411. 8 Steller, Beschreibung von dem Lande Kamtschatka, p. 345. 9 Miss Czaplicka, My Siberian Year, p. 104. 10 Clarke, Among the Tribes in South-West China, p. 78; Gray China, ii. 393 (aboriginal tribes). VamMry, Das Turkenvolk, p. 369 (Usbeg). 292 THE HISTORY OF HUMAN MARRIAGE chap. the matter. If the parents raise objections to the marriage of two lovers, the latter take the law into their own hands and become husband and wife without ceremony; " and that relationship is sacred, and stands good in Ainu society."1 Among the Yukaghir it is only after the affair has been quite settled between the young people that the lover sends a go-between to the girl's father to ask for his consent ; and although a girl seldom acts in such matters against her father's will, there is really no absolute submission to paternal authority.2 Among the Koryak, " though nominally the father can marry off his daughter on his own authority, he nevertheless not only consults his wife and eldest son, but often takes into consideration even the likes or dislikes of his daughter. Cases occur where the daughter does not submit to the father's or elder relative's authority in the choice of a bridegroom. Their will is not forced upon her., This attitude is also mirrored in the myths."3 According to Patkanov, a Tungus girl is free to choose her husband; and if her father does not approve of her choice, she elopes with the man she has chosen.4 In Dardistan " no objection to lawful love terminating in matrimony is ever made unless the girl or the boy is of a lower caste."5 Among the Bhotias of Almora the young ladies, when courted, " are allowed full liberty in exercising a preference."6 In Nepaul, says Waddell, " marriage is . . . almost always an affair of the heart. Young men and maidens become acquainted with each other, and courtships and real love-matches are the rule. Whereas, with the Indian plains-people everything is arranged by the friends, and it is seldom that the bride and bridegroom see each other before marriage."7 We have noticed that 1 Batchelor, Ainu and their Folk-Lore, p. 223 sq. See also v. Sie- bold, Ethnologische Studien uber die Aino auf der Insel Yesso, p. 30. 2 Jochelson, Yukaghir, p. 89 sq. 3 Idem, Koryak, p. 744. 4 Miss Czaplicka, Aboriginal Siberia, p. 105. 5 Leitner, Results of a Tour in ' Dardistan,' &c., iii. 36. 8 Sherring, ' Notes on the Bhotias of Almora and British Garhwal,' in Memoirs Asiatic Soc. Bengal, i. 106. 7 Waddell, Among the Himalayas, p. 311. XXII CONSENT AS A CONDITION OF MARRIAGE 293 in various uncivilised tribes of India marriages are arranged by the parents of the parties ; but generally there seems to be great liberty of choice among them. The young couple often settle the affair entirely between themselves, even though it is ostensibly arranged for them ;1 or the parents, before they give their children in marriage, consult them and, as a rule, follow their likings.2 Among the Yanddis of the Nellore district in the Madras Presidency, " the parties feel free in every respect to select their future partners, whose consent is formally endorsed by the elders."3 Among the Ulladans of Travancore " a small round building is made of leaves, and inside this the bride is ensconced. All the eligible young men of the village then assemble and form a ring round the hut. At a short distance sits the girl's father or nearest male relative with tom tom in his hands, and a few more musical instruments complete the scene. . . . The young men, each armed with a bamboo, begin dancing round the hut into which each of them thrusts his stick. This continues for an hour, when the owner of whichever bamboo she seizes becomes the fortunate husband of the concealed bride. A feast then follows, and the ceremony is complete."4 Among the Paliyans, a nomadic tribe of the Palni Hills in South India, " the girl is free to choose a husband for herself."5 Among the Badagas6 and Kotas7 of 1 Rowney, Wild Tribes of India, pp. 67, 76, 159 (Kols, Santals, Abors ; as for the Santals, cf. Dalton, op. cit. p. 215 ; Man, Sonthalia, p. 102 ; Hunter, Rural Bengal, i. 205 sq.; Ymer, v. p. xxiv.). Shortt, ' Account of the Hill Tribes of the Neilgherries,' in Trans. Ethn. Soc. N.S. vii. 242 (Todas ; cf. Marshall, A Phrenologist amongst the Todas, p. 212 ; Rivers, Todas, p. 504). 2 Dalton, op. cit. pp. 29, 57, 91, 125 (Miris, Khasis, Koch, Mudsis); Watt, ' Aboriginal Tribes of Manipur,' in Jour. Anthr. Inst. xvi. 358 sq. (Kolyas). Cunningham, ' Notes on Moorcroft's Travels in Ladakh,' in Jour. Asiatic Soc. Bengal, vol. xiii. pt. i. 204 (Ladakhis). 3 Ranga Rao, ' Y An Adis of the Nellore District,' in the Madras Government Museum's Bulletin, iv. 98. 4 Anantha Krishna Iyer, Cochin Tribes and Castes, i. 61 sq. 6 Dahmen, ' Paliyans,' in Anthropos, iii. 27. 6 Hough, Letters on the Climate, Inhabitants, &c. of the Neilgherries, p. 91. 7 Ward, in Grigg, Manual of the Nilagiri District, Appendix, p. Ixxvii. 294 THE HISTORY OF HUMAN MARRIAGE chap. the Nilgiris marriage is said to be concluded by the consent of both parties. The Bhuiyas, an Orissa hill tribe, consider it essential that the parties should have attained manhood and womanhood, and " that there should be no coercion used by parents or others, though parents and friends some- times give advice."1 So also the Saorias of the Rajmahal Hills require that the parties shall be "of an age to please themselves," hence there are no infant-marriages among them.2 Among the Nagas " many marriages are arranged by the young people themselves on the basis of affection " ;3 and among the Angami Nagas " the choice lies mainly with the woman."4 Of the Mikirs, a Tibeto-Burman people in the Province of Assam, we are told that when the bride- groom's party come to the bride's house, her father asks his wife to inquire of the girl if she will take the lad ; if the girl is reported to consent, the beer and spirits brought by the bridegroom's party are drunk by the two fathers, whereas without her consent they cannot be accepted.5 Concerning the Chittagong Hill tribes, Captain Lewin says that the women's " power of selecting their own husband is to the full as free as that enjoyed by our own English maidens."6 MacMahon states that among the Indo-Burmese border tribes in the good old times chivalry demanded that the wishes of the weaker sex should be taken into account.7 Among the Shans mutual consent is required to constitute a valid union ;8 and of the Lisu tribes of the Burma-China frontier we are told that their young people " evidently 1 Macmillan, ' Bhuiyas,' in Calcutta Review, ciii. 175. 2 Bainbridge, ' Saorias of the Rajmahal Hills,' in Memoirs Asiatic Soc. Bengal, ii. 56. 3 Hodson, Ndga Tribes of Manipur, p. 87. 4 Prain, ' Angami Nagas,' in Revue coloniale inter nationale, v. 492. Cf. Butler, Travels and Adventures in the Province of Assam, p. 148. 6 Stack, Mikirs, p. 18. 8 Lewin, Wild Races of South-Eastern India, p. 347. Cf. ibid. PP- x45> *79. 285. 7 MacMahon, Far Cathay and Farther India, p. 275. 8 Anderson.. Mandalay to Momien, p. 301. See also Woodthorpo, ' Account of the Shans and Hill Tribes of the States on the Mekong,' in Jour. Anthr. Inst. xxvi. 22. xxii CONSENT AS A CONDITION OF MARRIAGE 295 have a personal vote in the management of their love affairs.'1 " In Tavay," says Low, " the Karean lover pays his addresses in form, and if the girl approves of his suit they are soon constituted a married pair by the giving of a feast to their friends."2 Among the aboriginal tribes of the Malay Peninsula the girl's consent to her marriage seems to be essential. Thus among the Sakai, according to Low, " a young man pays his addresses in person. If the girl approves, he gives a present to her family of spears, knives and household utensils."3 Of the same people Maxwell was told that when a man wanted to marry he first of all spoke to the girl and then, if she agreed, went to her father and mother, taking some jungle produce as a gift ;4 and de Morgan states that the consent of the girl is required, together with that of her father or, if he is dead, the eldest surviving member of the family.5 Among the Mantra, according to Mr. Knocker, " marriage is merely a mutual compact entered into by the two parties concerned, and cohabitation is sufficient to acknowledge a man and woman as husband and wife."6 " The bride and bridegroom are led by one of the old men of the tribe, towards a circle more or less great, according to the presumed strength of the intended pair ; the girl runs round first, and the young man pursues a short distance behind ; if he succeed in reaching her and retaining her, she becomes his wife ; if not, he lose*s all claim to her. At other times, a larger field is appointed for the trial, and they pursue one another in the forest. The race, according to the words of the chronicle, ' is not to the swift nor the battle to the strong,' 1 Rose and Brown, ' Lisu (Yawyin) Tribes of the Burma-China Frontier,' in Memoirs Asiatic Soc. Bengal, iii. 263. 2 Low, ' Karean Tribes or Aborigines of Martaban and Tavai,' in Jour. Indian Archipelago, iv. 418. 3 Ibid. iv. 430 sq. 4 [Maxwell, in] Jour. Boy. Asiatic Soc. Straits Branch, no. 1, 1878, p. 112. 5 de Morgan, ' Moeurs, &c. des Negritos de 1'intdrieur de la presqu' ile Malaise,' in Society normande de Geographic, Bulletin de I'annfe 1885, vol. vii. 421., 6 Knocker, ' Aborigines of Sungei Ujong,' in Jour. Roy. Anthr. Inst, xxxvii. 293. 296 THE HISTORY OF HUMAN MARRIAGE chap. but to the young man who has had the good fortune to please the intended bride."1 The same may probably be said of the following custom, which according to Malay accounts prevails among the Benua. When a marriage has been agreed upon, " the relatives of both parties assemble at the house of the bride, who is placed in a canoe by herself, supplied with a paddle, and sent down the stream. When she has got a start of one or two reaches the bridegroom enters a canoe and gives chase. Should he succeed in over- taking the fair one, she is his wife. If he fail, the match is broken off." Logan, who reports this account, thinks that it is inaccurate, although it may have some foundation in the practice of former days;2 but Mr. Skeat believes that, in view of all the evidence, it may be taken as substantially correct.3 In the Malay Archipelago, according to Wilken, most marriages are contracted by the mutual consent of the parties.4 Moszkowski states that among the Sakai of Sumatra, when a young man desires to marry a certain girl, he mentions it to her mother, who then asks her daughter if the suitor pleases her ; and if the answer is in the negative the matter drops.5 Among the Orang Akit, another non- Malayan tribe in the eastern part of Sumatra, if two young persons like each other they simply settle down together without much ceremony.6 Among the Orang Mamaq " love-sick men have only to make sure that the girls they desire consent."7 Of the Rindan Kubu we are told that as 1 Bourien, ' Wild Tribes of the Interior of the Malay Peninsula,' in Trans. Ethnol. Soc. London, N.S. iii. 8x. 2 Logan, ' Orang Binua of Johore,' in Jour. Indian Archipelago, i. 270. 8 Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. 80. Cf. supra, ii. 267. 4 Wilken, ' Plechtigheden en gebruiken bij verlovingen en huwe- lijken bij de volken vail den Indischen Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, ser. v. vol. i. 159. 5 Moszkowski, Auf neuen Wegen dur ch Sumatra, p. 105. e Ibid. p. 40. 7 Schneider, quoted by. Speiser, ' Beitrage zur Ethnographic der Orang Mamma auf Sumatra,' in Archiv f. Anthrop. N.S. ix. 86. xxn CONSENT AS A CONDITION OF MARRIAGE 297 soon as a young man and a girl have made up their minds to become husband and wife, they inform their respective parents about it and are then formally married to one another; and it never happens that any obstacles are laid in their way.1 The Battas regard the consent of the bride as by no means unessential even though she is sold; for she is seldom compelled to marry a suitor whom she does not accept.2 Among the Hill Dyaks of Borneo a girl is always allowed to " choose her partner without let or hindrance."3 The girls of the Sea Dyaks accept or refuse their lovers as they like ; " as a rule the parents of the girl approve of her choice, or they would not have allowed her to receive visits from the man."4 Among the Punans, a Borneo jungle tribe, " there is little or no formal arrange- ment of marriages by the elders on behalf of the young people."5 Among the Minahassers of Celebes courtship or love-making " is always strictly an affair of the heart and not in any way dependent upon the consent or even wish of the parents."6 Similar statements are made by Riedel with reference to some of the smaller islands.7 Among the Agtas (Negritos) of the Philippines, " when a young man has made his choice, his friends or his parents make a demand for the young girl; a refusal is never given. A day is chosen ; and on the morning of that day the young girl is sent into the forest, where she hides herself or not, just as she pleases, and according as she wishes to be married to the young man who has asked her. An hour after her departure, the young man is sent to find out his bride. If he has the good luck to find her, and to bring her back to 1 van Dongen, quoted by Hagen, Orang Kubu auf Sumatra, p. 130. 2 Junghuhn, Die Battaldnder auf Sumatra, ii. 132. 3 Hornaday, Two Years in the Jungle, p. 455. 4 Gomes, Seventeen Years among the Sea Dyaks of Borneo, p. 120 sqq. For the liberty of choice among the Dyaks see also Boyle, Adventures among the Dyaks of Borneo, p. 236 ; Brooke, Ten Years in Sarawak, i. 69. 5 Hose and McDougall, Pagan Tribes of Borneo, ii. 183. 8 Hickson, A Naturalist in North Celebes, p. 272. 7 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, p. 302 (Tenimber and Timo riant) 298 THE HISTORY OF HUMAN MARRIAGE chap. her parents before sunset, the marriage is concluded, and she becomes his wife without fail; but if, on the contrary, he returns to the camp without her, he is not allowed to renew his addresses."1 Among the Li of Hainan parents have nothing else to do with the betrothals of their children than to ratify them ; it is the inclinations of the parties themselves that decide the matter.2 In most of the savage tribes of Formosa there is great freedom of marriage for both sexes.3 Among the Tsous the young brave presents to the lady of his choice an ornamental hair-pin made of deer-horn, and the acceptance of the gift signifies consent.4 Among the Paiwans he goes to her house with fuel and water, which he places before the door, and if the damsel puts them to the use for which they are intended it means that he is accepted.5 In the Mortlock Islands most marriages depend on the free choice of the parties.6 In the Oleai Islands, situated between the Pelew Islands and the Marshall Group, even many a well-to-do suitor fails to gain the girl he desires for the simple reason that she refuses his proposal.7 Of some of the Marshall Islanders it is said that " marriages depend on a free convention."8 Love-matches are frequent in many parts of New Guinea.9 In the Massim area court- ship consists of more or less prolonged cohabitation ; the proposal, as we have seen, comes from the girl; and it is 1 Proust de la Girondiere, Twenty Years in the Philippines, p. 271 sq. Piehlcr, ' Die Ajitas (Aetas) der Philippinen,' in Globus, xcvi. 201. Concerning the Negritos of Zambales, however, see infra, ii. 312. 2 Strzoda, ' Die Li auf Hainan,' in Zeitschr. f. Ethnol. xliii. 203. 3 Davidson, op. cit. pp. 569, 57L 575. 577. 581 sq. 4 Ibid. p. 571. 6 Ibid. p. 575. 6 Kubary, in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, p. 261. 7 Born, ' Einige Betrachtungen ethnographischer Natur fiber die Oleai-Inseln,' in Mittheil. Deutsch. Schutzgeb. xvii. 189. 8 v. Kotzebue, Voyage of Discovery into the South Sea, iii. 172. 9 Landtman, Nya Guinea farden, p. 84 (Kiwai Papuans). Vetter, ' Bericht fiber papuanische Rechtsverhaltnisse,' in Nachrichten uber Kaiser Wilhelms-Land und den Bismarck-Archipel, 1897, p. 90 (Jabim). xxu CONSENT AS A CONDITION OF MARRIAGE 299 evidently very unusual for parents to make any difficulties.1 It is also obvious that the consent of the girl is of essential importance among the Mafulu mountain people2 and at Bogadjim.3 The same is the case in New Britain and neighbouring smaller islands ;4 Romilly states that after the man has worked for years to pay for his wife and is finally in a position to take her to his house, she may refuse to go, and he cannot claim back from the parents the large sums he has paid them in yams, cocoa-nuts, and sugar-canes.5 In New Caledonia " lorsqu'un jeune homme se presente pour £pouser une jeune fille, il n'est pas rare qu'il eprouve un refus, s'il n'est pas chef ou d'un sang noble, et il arrive quelquefois que, par suite de refus successifs, il est reduit a rester c61ibataire. Si la jeune fille agree la demande, le mariage se conclut immediatement, sans autres ceremonies. Les parents ne sont pas toujours consultes ; il arrive souvent aussi que des parents font le mariage de leurs enfants sans les consulter."6 In Samoa, where marriages were frequently arranged by the parents without any reference to the woman's feelings, the latter was at any rate " sometimes able to follow her own inclination or affection " ; and among the higher ranks it occurred that the messengers of three or four rival suitors were for some time struggling for the prize amidst much coquetting on the part of the fair one, until at length she formally announced her choice.7 In Arorae, of the Gilbert Group, " in choosing a husband the 1 Seligman, Melanesians of British New Guinea, pp. 499, 502, 503, 708 sq. 2 Williamson, Mafulu Mountain People of British New Guinea, p. 172. 3 Hagen, Unter den Papua's, p. 241 sq. 4 Parkinson, op. cit. pp. 65, 160. Hahl, loc. cit. p. 78. Schnee, Bilder aus der Sudsee, p. 99. Vogel, Eine Forschungsreise im Bis- marck-Archipel, p. 223. Cf. Burger, Die Kiisten- und Bergvolker der Gazellehalbinsel, p. 55. 5 Romilly, ' Islands of Now Britain Group,' in Proceed. Roy. Geo- graph. Soc. N.S. ix. 10. 0 Brainne, La Nouvelle-Caledonie, p. 251. Cf. Moncelon, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. ix. 368. ' Stair, Old Samoa, p. 171 sqq. Kramer says {Die Samoa-Inseln, i. 39) that the girl must submit to the will of her relatives. 300 THE HISTORY OF HUMAN MARRIAGE chap. lady sat in the lower room of the house, and over her head were let down through the chinks of the floor of the upper room two or three cocoa-nut leaflets, the ends of which were held by her lovers. She pulled at one, and asked whose it was. If the reply was not in the voice of the young man she wished to have, she left it and pulled at another leaf, and another, until she found him, and then pulled it right down. The happy man whose leaf she pulled down sat still, while the others slunk away."1 Mariner supposed that in Tonga perhaps two-thirds of the girls had married with their own free consent.2 In the Society Islands the women of the middle and lower ranks had the power to choose husbands according to their own wishes ; and that the women of the highest classes sometimes as- serted the same right appears from the addresses a chief of Eimeo had to pay to the object of his attachment before she could be induced to accept his offer.3 In Nukahiva " la femme est libre; elle choisit en quelque sorte de son plein gre celui qui sera son epoux, pendant un temps plus ou moins limite."4 Among the Maori, according to Mr. Tregear, " there were three kinds of courtship, the annual or parliamentary form, arrangement between rela- tives, and free choice." The first took place at a solemn gathering in " the House of Amusement," but the young men and girls selected each other there also. Yet, he adds, " freedom of choice could hardly be allowed in cases where tribal alliances might be cemented, or powerful families conciliated. Where freedom of choice was exerted among young men and girls of high birth the result was generally a romance."5 On the other hand, as we have seen, even when the families or clans of the parties took an active part in the matter, the consent of the girl was required.6 Thom- son wrote of the Maori:-" Girls not betrothed in childhood were allowed, on growing up, to bestow their favours on 1 Turner, Samoa, p. 295 sq. 2 Mariner, Account of the Natives of the Tonga Islands, ii. 167. 3 Ellis, Polynesian Researches, i. 270, 267 sq. Cf. Waitz-Gerland, Anthropologic der Naturvolker, vi. 99 sq. 4 Vincendon-Dumoulin and Desgraz, Iles Marquises ou Nouka- Hiva, p. 287. 5 Tregear, op. cit. p. 286 sqq. 6 Supra, ii. 281. xxn CONSENT AS A CONDITION OF MARRIAGE 301 whom they pleased. The more suitors they had, the more valuable were they reckoned. Should a girl have a pre- ference for one of her suitors, and particularly on becoming enceinte, she went home with him, and they lived as man and wife."1 Mr. Macmillan Brown speaks of " the romance that hung round youthful love in all their legends, arguing great independence in the daughter of the household."2 And in a Maori proverb it is said, " As a kahawai (a fish which is very particular in selecting the hook that most resembles its food) selects the hook which pleases it best out of a great number, so also a woman chooses one man out of many."3 Among the Moriori of the Chatham Islands a girl who had been betrothed by her parents without regard for her own feelings, if determined and obstinate, gener- ally had her own way.4 In most Australian tribes a girl seems to be given in marriage by her father or some other relative5 or the camp council or leading men without having herself any voice in the matter.6 But we also hear of attempts to gain the 1 Thomson, Story of New Zealand, i. 177. 2 Brown, Maori and Polynesian, p. 67. 3 Taylor, Te Ika a Maui, p. 299. 4 Shand, in Jour. Polynesian Soc. vi. 148. 5 The other relative may be her mother, brother, or maternal uncle (see supra, ii. 284 sq. nn. 6, 1 sq.). Among the Yerkla-mining " it is the father who gives his daughter, but he may be overruled by his elder brother, especially if the latter has the support of the principal, that is, the oldest medicine-man of the local group " (Howitt, op. cit. p. 257 sq.), and in the Mukjarawaint tribe living in the northern parts of the Grampian mountains the paternal grand- parents had a voice in the disposal of their granddaughter (ibid. P- 243)- 6 This is implied in many statements and expressly said, e.g., by Curr, The Australian Race, i. 108; Idem, Recollections of Squatting in Victoria, p. 248 ; Beveridge, Aborigines of Victoria and Riverina, p. 22 ; Holden, in Taplin, Folklore, Manners, &c. of the South Australian Aborigines, p. 17 (Maroura tribe, Lower Darling); Gason, ' Dieyerie, &c.,' in Jour. Anthr. Inst. xxiv. 170 ; Schulze, 'Aborigines of the Upper and Middle Finke River,' in Trans, and Proceed. Roy. Soc. South Australia, xiv. 236; Wilhelmi, 'Manners and Customs of the Australian Natives, in particular of the Port Lincoln District,' in Trans. Roy. Soc. Victoria, v. 179 ; Petrie, Reminiscences of Early Queensland, p. 59. THE HISTORY OF HUMAN MARRIAGE chap. 302 girl's consent by courting j1 and " the use of love letters is perfectly understood by the Kabi natives of Queensland."2 Concerning the Narrinyeri Mr. Taplin says that although the consent of a female is not considered a matter of the first importance, yet it is always regarded as desirable.3 We shall see that the women in Australian tribes practi- cally enjoy a greater liberty of choice than might be supposed, owing to the frequency of marriage by elopement, which is quite a recognised institution among them. Of the Tasmanians it is said that they courted with flowers, and that " the female occasionally possessed a negative."4 In Madagascar there are tribes where young people have perfect liberty to choose their partners, although in others girls are given in marriage without being consulted, or matches are arranged by the parents of both parties accord- ing to their own will.5 Thus " all matrimonial connections among the Sakalava are contracted by the. individuals themselves, and not by the ordering of the parents, as is the custom among the Hova and Betsileo."6 Among the Bushman tribes we find pairing by mutual consent ;7 " in general," says Mr. McCall Theal, " there was no marriage ceremony, the mere consent of both parties being all that was needed."8 According to earlier writers on the Bushmen9 and Hottentots,10 the lover of a girl who has grown 1 Howitt, op. cit. p. 232 sq. (Kulin of Victoria). Eylmann, Die Eingeborenen der Kolonie Sudaustralien, p. 129. 2 Mathew, Eaglehawk and Crow, p. 114. 3 Taplin, ' Narrinyeri,' in Woods, Native Tribes of South Australia, p. 10. 4 West, History of Tasmania, ii. 78 sq. 5 Grandidier, op. cit. ii. 166. 6 Walen, in Antananarivo Annual, no. viii. (1884) p. 54. 7 Stow, Native Races of South Africa, p. 96 sq. 8 Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 47. Cf. Kaufmann, ' Die Anin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 156. 9 Burchell, Travels in the Interior of Southern Africa, ii. 59. Chap- man, Travels in the Interior of South Africa, i. 258. Fritsch, Die Eingeborenen Sud-Afrika's, p. 444. 10 Thunberg, ' Account of the Cape of Good Hope,' in Pinkerton, Collection of Voyages and Travels, xvi. 141. Le Vaillant, Travels from the Cape of Good-Hope, into the Interior Parts of Africa, ii. 67. CONSENT AS A CONDITION Ob' MARRIAGE 303 up to womanhood without being previously betrothed must gain her approbation as well as that of the parents. Among the Hottentots of the Korana clans, " when a young man is attached to a female, and wishes to marry her, he and his companions take an ox and place it before her house. If she allow the ox to be killed, it is considered as giving her consent to the union, and the parties are im- mediately regarded as married persons. It occasionally happens that the ox is brought three or four times before her approval is obtained."1 Von Francois contrasts the right of the young Hottentot girl to accept or refuse the proposal of a suitor with the power of a Herero father to dispose of his daughter's hand even against her inclination, although he adds that among the Herero also the will of the girl often decides the matter.2 Among the Kafirs a woman is frequently forced to marry against her wish ; " yet it is undoubtedly true that most good-tempered Kafir girls can, as a rule, get out of a hateful marriage," whereas girls who are self-assertive and strong-willed may find difficulty, since men like to tame such characters and teach them lessons.3 Among the Xosa Kafirs, although the father selects a husband for his daughter, the latter can by strongly opposing his choice make the marriage contract invalid.4 In his book on the Zulus and Amatongas Mr. Leslie remarks that it is " a mistake to imagine that a girl is sold by her father in the same manner, and with the same authority, with which he would dispose of a cow."5 Among the Barotse " free women who have not been given away or sold as slaves are allowed to choose what husbands they please."6 Among the Mpfumo clan of the Thonga " a father very seldom obliges his daughter to accept a husband 1 Campbell, Travels in South Africa, Second Journey, ii. 347. 2 v. Francois, Nama und Damara Deutsch-Sud-West-Afrika, pp. 213, 195. 3 Kidd, The Essential Kafir, p. 220 sq. According to Maclean {Compendium of Kafir Laws and Customs, p. 69), the Kafir girl is seldom or never consulted about the matter. 4 Kropf, op. cit. p. 141. 5 Leslie, Among the Zulus and Amatongas, p. 194. 0 Holub, Seven Years in South Africa, ii. 293, 298. Cf. ibid. ii. 206. 304 THE HISTORY OF HUMAN MARRIAGE chap. whom she dislikes, except in the case of debts."1 In the Ruwuma district of " German East Africa " the decision rests with the girl.2 Among the Wakonde she " is at perfect liberty to choose her husband."3 Among the Sango the consent of the girl is necessary ; if her parents want to give her in marriage against her will she appeals to the chief, who as a rule decides the case in her favour.4 When a Shambaa wishes to marry a certain girl he first tries to gain her favour, and, if successful, is then introduced by her to her father, whose consent is likewise necessary ; a grown-up girl is never compelled by her father to marry a man she does not like.5 Among the Wayao " girls are not in reality so passive in the matter as we are apt to assume, but most certainly expect to have their wishes consulted ; and many a carefully-planned match has come to nothing merely because the girl loved another man. In this respect," Dr. Weule adds, " there is not the slightest difference between white and black." 6 In British Central Africa, " when a man takes a fancy to a girl whom he finds to be disengaged, he first of all comes to an understanding with her, then goes to her village and tells her family-all the relations, in- cluding the grandparents, the elder brother, and above all, the maternal uncle, are consulted, as well as the parents-and, if they have no objection, he then goes to his own people."7 Among the East African Wagiriama and Wasania the girl's consent is required.8 Among the Akamba, " if a young 1 Junod, Life of a South African Tribe, i. 103. See also Idem, Les Ba-Ronga, p. 31. 2 Fiilleborn, Das Deutsche Njassa- und Ruwuma-Gebiet, p. 60. 3 v. Behr, ' Die Volker zwischen Rufiyi und Rovuma,' in Mittheil. Deutsch. Schutzgeb. vi. 74. See also Fiilleborn, op. cit. p. 346. 4 Heese, ' Sitte und Brauch der Sango,' in Archiv f. Anthrop. N.S. xii. 135 sq. 5 Dahlgriin, ' Heiratsgebrauche der Schambaa,' in Mittheil. Deutsch. Schutzgeb. xvi. 219, 224. 8 Weule, Native Life in East Africa, p. 306. 7 Miss Werner, Natives of British Central Africa, p. 131. See also Stannus, ' Notes on some Tribes of British Central Africa,' in Jour. Roy. Anthr. Inst. xl. 308. 8 Barrett, ' Notes on the Customs and Beliefs of the Wa-Giriama, etc., British East Africa,* in Jour. Roy. Anthr. Inst. xli. 20, 21, 30. Xxn CONSENT AS A CONDITION OF MARRIAGE 305 man sees a girl he fancies he will talk to her now and again, and eventually tell her he wants to marry her. If he acquiesces his father goes to her father and asks him if he agrees and if he does then the youth's father returns with a present of two goats ; this ratifies the bargain."1 In many of their families the betrothal of infants is practised, but " when a girl comes of age she is at liberty to refuse a proffered husband if she does not like him," even though the dowry has all been paid up in advance.2 The same is the case among the Akikdyu. Among them " a girl's betrothal is entirely her own affair," and if-as occasionally happens-bespoken as a child by some older and wealthy man, she is not obliged to marry him on coming to years of discretion. When Mrs. Routledge asked the leading wife of one of their chiefs what she should tell the English wemen about the women of Kiktiyu, the answer was :-" Tell them two things. One is, that we never marry any one we do not want to ; and the other is, that we like our husbands to have as many wives as possible."8 If a Masai girl strongly objects to marrying a man who wants her, her parents do not force her to become his wife.4 Of the Batamba in the Uganda Protectorate we are told that a girl up to her fifteenth year has no say whatever in the matter of marriage, but that beyond this age she may make her choice.5 Among the Baganda a man who desired to marry a special girl ad- dressed himself to her brother, who in his turn informed his paternal uncle and consulted with him ; but the uncle asked the girl whether she wished to marry the man or not, and she had now an opportunity for either accepting or rejecting him.6 So also among the Bakene-a small Bantu tribe dwelling in floating huts on the Mpologoma River and some neighbouring lakes where the Nile has its sources- " the girl has the right to accept or reject a man's offer " ; 1 Hobley, Ethnology of A-Kamba and other East A frican Tribes, p. 62. 2 Decle, Three Years in Savage Africa, p. 490. 3 Routledge, With a Prehistoric People, pp. 124, 125, 127. 4 Hinde, The Last of the Masai, p. 71. 5 Condon, ' Contribution to the Ethnology cf the Basoga-Batamba,' in Anthropos, vi. 370 sq. 11 Roscoe, Baganda, p. 87 sq. VOL. II - X 306 THE HISTORY OF HUMAN MARRIAGE chap. and among the Basoga, in the Eastern Province of the Uganda Protectorate, she has an opportunity to decide the question of marriage for herself.1 Among the Lendu of Uganda " the custom is for the young man to propose to the young woman, and if she accepts him, he then applies to her father."2 In the Sese Archipelago, in the Victoria Nyanza, " marriages are easily arranged. The girl takes her lover to her father. If she is old enough to marry, she proceeds at once to her new home " ; and afterwards cows are paid. This, at any rate, was the old custom.3 Among the Bavuma -inhabiting the large island of Buvuma off the north coast of the Victoria Nyanza-" the unmarried woman is a very independent being, and submits to no control, especially in her love affairs. This is the only people," says Mr. Cunningham, " I have met where the father has nothing whatever to do with arranging a marriage for his daughter. She it is who is asked directly by her lover whether she will marry him, and, if she consents, such consent constitutes the whole of the marriage ceremony, and she proceeds straightway to her new home with her husband. If the consent is given whilst she is even temporarily away from home, she does not take the trouble of informing her parents that she has accepted a husband."4 The Madi girls, says Emin Pasha, enjoy great freedom, and are able to choose companions to their liking.5 Among the Shulis, according to Dr. Felkin, the women have a voice in the selection of their husbands.6 No Dinka girl is obliged to marry a man against her will.7 Among the Batua, a Pygmy people of the Belgian Congo, a father gives his daughters perfect liberty to choose their husbands.8 Among the Warega " il est rare que Ton de- mande une jeune fillc en mariagc sans s'etre entendu d'abord 1 Idem, Northern Bantu, pp. 149, 211 sq. 2 Cunningham, Uganda, p. 328. 3 Ibid. pp. 92, 94. 4 Ibid. p. 129. 5 Emin Pasha in Central Africa, p. 103. 6 Wilson and Felkin, Uganda and the Egyptian Soudan, ii. 61. ' O'Sullivan, ' Dinka Laws and Customs,' in Jour. Roy. Anthr. Inst. xl. x8o. 8 Hutereau, Notes sur la Vie famiiiale el juridique de quelques populations du Congo Beige, p. 3. xxn CONSENT AS A CONDITION OF MARRIAGE 307 avec elle."1 Among the Boloki, on the Upper Congo, if a girl has not been given away in her early childhood, she has evidently her say in the matter.2 Among the Bangongo the most usual method of concluding a marriage is that the couple meet in the bush, and the young man makes his proposal, the acceptance of which is immediately followed by consummation ; afterwards negotiations are made with the parents of the girl, and the bride price is paid. The woman's consent is regarded as necessary.3 Grenfell states that among the Bakongo marriage usually takes place by mutual consent, " so that there are marriages of affection."4 As to the West African Negroes, Winwood Reade informed Darwin that " the women, at least among the more intelli- gent Pagan tribes, have no difficulty in getting the husbands whom they may desire, although it is considered unwomanly to ask a man to marry them."5 Among some of the Ibo- speaking people of Nigeria, " as soon as the girl is six or eight she will be consulted and if she declares that she does not like her suitor her father will wait until she makes her choice, and will repay the first suitor from the monies handed over by the second suitor."6 Among the Yoruba- speaking peoples of the Slave Coast " parents cannot force a daughter to marry a suitor who is distasteful to her, but they can prevent a girl from marrying a man of whom they do not approve. ... If, however, she runs off with him, they usually take no further trouble."7 So also among the 1 Delhaise, Les Warega, p. 170. 2 Weeks, Among Congo Cannibals, p. 123. 3 Torday and Joyce, Les Bushongo, p. 115. See also ibid. pp. 113, 272 ; Torday, Camp and Tramp in African Wilds, p. 134 (Bayaka). 4 Johnston, George Grenfell and the Congo, ii. 678. 6 Darwin, Descent of Man, ii. 408. See also Reade, Savage Africa, pp. 260, 390, 453, 554; Beecham, Ashantee and the Gold Coast, p. 125 ; Bosman, ' New Description of the Coast of Guinea,' in Pinkerton, Collection of Voyages and Travels, xvi. 419 (Negroes of the Gold Coast) ; Merolla da Sorrento, ' Voyage to Congo and several other Countries,' ibid. xvi. 236 (Negroes of Sogno) ; Soyaux, Aus West-Afrika, pp. 152, 161 (Negroes of Loango). • Thomas, Anthropological Report on Ibo-speaking Peoples of Nigeria, iv. 63. See also ibid. i. 62. Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 185. 308 THE HISTORY OF HUMAN MARRIAGE chap. Ewhe-speaking peoples a girl " cannot be forced into a union that is absolutely repugnant to her," although marriage is as a rule arranged without any reference to her wishes.1 Among the Bambara of the Ivory Coast a woman cannot be married without her consent, and of the Baoul6 we are told that if a girl consents to a proposal her parents also invariably agree unless there are particular reasons for not doing so.2 Among the Baya " il y a des unions pro- jetees entre les parents des le plus jeune age de leurs enfants. . . . D'autres fois, au contraire, les enfants se sont spontanement choisis. Les parents n'auront plus, a un certain age, qu'a ajouter leur consentement a cette union d'inclination, a moins de voir naitre des desordres."3 Among the Bah tribes of Kamerun the conclusion of a marriage depends either upon the free choice of the parties, or upon the agreement of the parents (not infrequently before the birth of the children), or upon pure purchase.4 Among the ancient inhabitants of the Canary Islands the lover asked his bride of her father, who referred the decision to her.5 These statements are sufficient to show that in the savage world the consent of the woman to her marriage is very frequently, not only asked as a matter of fact, but even required by custom. There are no doubt many instances to the contrary as well; but from my collection of data I have come to the conclusion that these are less numerous than those in which the woman's wishes are, or must be, consulted. A similar inference may be drawn from the list complied by Messrs. Hobhouse, Wheeler, and Ginsberg, 1 Idem, Ewe-speaking Peoples of the Slave Coast, p. 199. Bowdich says (Mission from Cape Coast Castle io Ashantee, p. 303), " No Ashantee forces his daughter to become the wife of the man he wishes, but he instantly disclaims her support and protection on her refusal." 2 Clozel and Villamur, Les coutumes indigenes de la C6te d'Ivoire, pp. 318, 101. 3 Poupon, ' £tude ethnographique des Baya de la circonscription du M'Bimou,' in L'Anthropologic, xxvi. 126. 4 Hutter, Wanderungen wnd Forschungen im Nord-Hinterland von Kamerun, p. 379. 5 Alice C. Cook, ' Aborigines of the Canary Islands,' in American Anthropologist, N.S. ii. 479. xxii CONSENT AS A CONDITION OF MARRIAGE 309 recording 103 cases in which the woman's consent appears to be generally required and 811 in which it is not required.1 As to the circumstances influencing the matter, we may with- out the slightest hesitation make the negative statement that her liberty of choice among the lower races has not increased in proportion to their advancement in culture. Professor Hobhouse and his collaborators give the following figures representing the percentage of cases in which the woman's consent is not required among simpler peoples at different stages of economic culture : among lower hunters and food-collectors 0*71, among higher hunters o • 44, among incipient agriculturists O'22, among lower pastoral peoples 0-65, among pure agriculturists 0-38, among higher pastoral peoples 0'86, and among cattle-keeping agriculturists 0-30. The first of these figures, referring to lower hunters, how- ever, is of little value. It is based on 8| cases of consent required and 21 cases of consent not required, but 3 of the former and 18 of the latter refer to Australian tribes.2 So far as my knowledge goes, there is not a single Australian tribe of which it can be said that the woman's consent to her marriage is required, although it may be held desirable, especially as a safeguard against elopement. But, as has been pointed out before, the various Australian tribes can certainly not be treated as independent units in a statistical investigation following the lines adopted by Professor Hobhouse and his collaborators.3 The other three negative eases in their list refer to the Shoshoni, Fuegians, and Bush- 1 Hobhouse, Wheeler, and Ginsberg, Material Culture and Social Institutions of the Simpler Peoples, p. 157. The scorn with which Mr. Finck (Primitive Love and Love-Stories, p. 338 sqq.) treats my statement that " it would be a mistake to suppose that among the lower races women are as a rule married without having any voice of their own in the matter " seems thus to be rather undeserved. 2 Hobhouse, Wheeler, and Ginsberg, op. cit. p. 176 sqq. 3 Our authors speak of North West Australians, Victoria tribes, and the Bangerang as requiring the woman's consent. For the Bangerang reference is made to Curr, who, however, says the very reverse (Recollections of Squatting in Victoria, p. 248 sq.), namely, that " the bride had no voice in the matter, but was simply required to go to the hut of the man to whom her father, brother, or uncle, as the case might be, had given her." 310 THE HISTORY OF HUMAN MARRIAGE chap. men. As to the Shoshoni, it is doubtful whether they should be reckoned among the lower hunters ; and statements quoted above hardly justify the assumption that the woman's consent is not required among the Fuegians and Bushmen. Some other tribes at about the same stage of economic culture, like the Sakai of the Malay Peninsula and the Negritos of the Philippines, whom we have also mentioned before, have not, in this connection at least, been noticed by our authors. Concerning the Pygmy tribes generally, Father Schmidt, in his monograph on them, observes that both sexes enjoy a remarkable degree of liberty in the selection of partners.1 This may in fact be said of all or nearly all the lower hunters, with the exception of the Australian tribes, as also of the great bulk of incipient agriculturists, many of whom can hardly be distinguished from the lowest hunting and food-collecting tribes. If the Australian natives are excepted, the woman's liberty of choice may thus be said to be decidedly greater among the lowest savages than among the more advanced savages. It is greater among the lower hunters than among the higher hunters, greater among the incipient agriculturists than among the pure and the cattle-keeping agriculturists, greater among the lower pastoral tribes than among the higher pastoral tribes, and greater among the hunters and agriculturists as a whole than among the pastoral peoples. This is borne out both by my own collection of facts and by that of Professor Hobhouse and his collaborators, although their general conclusions are somewhat different owing to their inclusion of the Australian tribes, which in my opinion must be dealt with separately. How shall we explain these facts ? There can be little doubt that among the lower races marriage by purchase is the principal cause of the giving away of a girl in marriage without any regard for her own wishes. Among the lower hunters the liberty of choice goes hand in hand with a general absence of marriage by purchase ; small gifts may be presented to the father or 1 Schmidt, Die Stellung der Pygmdenvolker in dev Entwicklungs- geschichte des Menschen, p. 169 sqq. xxn CONSENT AS A CONDITION OF MARRIAGE 311 the family of the girl, but they cannot be regarded as a bride price in the proper sense of the word. So far as savage peoples at other stages of economic culture are concerned, I find in the tables given by Professor Hobhouse and his collaborators that among 69 cases of marriage by purchase there are 42 in which the woman's consent is not required and only 27 in which it is required, whilst out of all cases in which something is said about her consent it is required in 94 cases and not required in 59. Thus marriage by purchase prevails in 80 per cent, of the cases in which the woman's consent is not required and only in 30 per cent, of the cases in which it is required. My own collection of facts shows a still more frequent coincidence between the absence of marriage by purchase and freedom of choice. Out of the cases mentioned above there are at least 130 in which the woman's liberty of choice seems to be very great, even though it is not positively said in all of them that her consent is indispensable ; and only in 21 of these cases- that is, 16 per cent.-I find clear evidence of genuine marriage by purchase. In these calculations, however, the lower hunters are also taken into account, and this makes the percentage somewhat smaller than it would otherwise be. The majority of cases in which liberty of choice is combined with actual purchase come from Africa (14 out of 21), whilst in other African cases the prevalence of marriage by purchase is expressly denied although the man does not get his bride for nothing.1 That there is a close connection between marriage by purchase and disregard for the woman's own wishes is also directly indicated by many observers. In East Mallicolo, one of the New Hebrides, the father offers his daughter to the highest bidder, and receives for her ten pigs at least.2 Among the Battas of Sumatra it sometimes happens that an exceptionally high offer induces the parents to give their 1 Fiilleborn, op. cit. p. 61 (natives of the Ruwuma district). Weule, op. cit. p. 306 (Wayao). Delhaise, op. cit. p. 173 (Warega). Henry, L'Ame d'un peuple africain, p. 203 (Bambara). 2 Serbelov, ' Social Position of Men and Women among the Natives of East Malekula, New Hebrides,' in American Anthro- pologist, N.S. xv. 278. 312 THE HISTORY OF HUMAN MARRIAGE chap. daughter in marriage against her will.1 Among the Negritos of Zambales, in Luzon, " a daughter is regarded as an asset of so much value, not to be parted with until that price is paid, and, while she is allowed some freedom in the choice of a husband, parental pressure usually forces her to the highest bidder."2 In the Chin Hills, on the southern borders of Assam and Manipur, parents practically sell their daughters to be wives, and love is not taken into consideration at all.3 Among the Bedouins of Mount Sinai, who have marriage by purchase, no father thinks it necessary to consult his daughter before selling her, whereas among the Arabs of the eastern plain, the Aeneze, &c., according to Burckhardt, " the father never receives the price of the girl, and therefore some regard is paid'to her inclinations."4 In Ruanda, in East Africa, a girl may express her reluctance against a suitor, but the decision rests in the last instance with her father, and the scale of his favour is almost without exception turned by the highest offer.5 Among the Thonga of South-East Africa, a girl to be married is in principle entirely at the mercy of her family as regards the choice of her husband. It is true that her consent is generally asked before any decision takes place, but this is done only as long as the male masters of the woman have no special, interest in the choice. " Should they have reason to impose a certain husband upon her, they will not hesitate to force her to accept him ; a young girl will be given up to a dirty old man, for whom she has no sympathy, on account of a lobola (that is, bride price) debt of twenty years standing."6 Among the Angoni Zulus, who formerly allowed a girl to choose her husband, the father does not nowadays as a rule consult her wishes, but sells her to the man who offers him the largest number of oxen.7 In most parts of the territory of the Edo-speaking people in Nigeria it appears to be rare 1 Junghuhn, op. cit. ii. 132. 2 Reed, Negritos of Zambales, p. 56. 3 Carey and Tuck, Chin Hills, i. 189. 4 Burckhardt, Notes on the Bedouins and Wahdbys, p. 149 sq. 6 Schumacher, ' Das Eherecht in Ruanda,' in Anthropos, vii. 6. 0 Junod, Life of a South African Tribe, i. 264. 7 Wiese, ' Beitrage zur Geschichte der Zulu im Norden des Zambesi, namentlich der Angoni,' in Zeitschr. f. Ethnol. xxxii. 191. xxii CONSENT AS A CONDITION OF MARRIAGE 313 to put pressure upon a girl to marry a suitor to whom she objects ; but in the Sobo country, says Mr. Thomas, " owing possibly to the considerably higher money payments as part of the bride price, and to the consequent greater difficulty of refunding the bride price, it appears to be the custom to compel a girl to accept a suitor."1 Of the Dakota Indians it is said that a lover sometimes " buys his wife without her consent."2 Concerning the Pawnee Mr. Grinnell writes :- " In the olden times, before they had horses, when their dogs, their simple arms, and their clothing constituted all their possessions, the Pawnees married for love. The affection which existed between two young people was then the only motive which brought about a union, and this affection was seldom interfered with unless there was a very great difference between the social standing of the family of the boy and of the girl. . . . After the Pawnees obtained horses and began to accumulate property-as the people acquired wealth, and their circumstances became easier-the practice arose of giving presents to the immediate relatives of the girl whose hand was sought in marriage. These presents were given in order to conciliate those relations who controlled the girl. Originating merely in this desire to gain the good will of her family, the custom gradually became more and more firmly established until it had come to be a matter of course to give presents, and finally a matter of necessity if the young man hoped to gain the consent of the girl's family to his proposal of marriage."3 Mr. Matthews says that travellers who speak of the " marriage by purchase " among the Indians as a mere sale of the woman to the highest bidder misrepresent the custom, " unless where their remarks may apply to some modern irregularities among the least reputable persons."4 We are told that among the Creeks, if a man desires to make a woman his wife " conformably to the more ancient and 1 Thomas, Anthropological Report on the Edo-speaking Peoples op Nigeria, i. 58 sq. 2 Mary Eastman, Dahcotah, p. 106. 3 Grinnell, Story of the Indian, p. 41. 4 Matthews, Ethnography and Philology of the Hidatsa Indians, P- 52, 314 THE HISTORY OF HUMAN MARRIAGE chap. serious custom of the country," he endeavours to gain her own consent by regular courtship.1 Now we are in a position to explain why progress in economic culture has among the lower races been un- favourable to the woman's liberty of choosing her husband. It has led to marriage by purchase, and this has tended to restrict her liberty. Whilst marriage by purchase hardly occurs among the lower hunters and is rare among incipient agriculturists, it is more frequently found among higher hunters and especially among pure and cattle-keeping agriculturists ; it is more prevalent among higher pastoral tribes than among lower pastoral tribes, and, though very frequent among the higher agriculturists, it is more preva- lent among the pastoral tribes as a whole than among the agricultural tribes. The development of marriage by purchase thus follows the same lines as the restriction of the woman's liberty of choice to an extent which makes it impossible to doubt that there has been a causal connection between them. But it is worth noticing that among the agricultural peoples there are a comparatively larger number of cases in which marriage by purchase is combined with freedom of choice than among hunters and pastoral tribes. Out of 49 agricultural peoples mentioned in Professor Hob- house's tables who have marriage by purchase, 24, that is 47 per cent., consider the woman's consent to her marriage necessary and 25 do not consider it necessary, whereas out of 20 higher hunting and pastoral peoples who have marriage by purchase only 3, that is 15 per cent., require.her consent and 17 do not do so ; and out of 27 cases in which marriage by purchase is combined with obligatory consent, 24, or 89 per cent., refer to agricultural peoples and only 3 to higher hunters and pastoral tribes. The result derived from my own collection of facts is very similar : out of 21 cases of combined marriage by purchase and liberty of choice 18, or 87 per cent., refer to agriculturists and 3 to higher hunters and pastors; but as in one of the latter cases the woman's liberty of choice is somewhat doubtful, the percentage of the casesreferring to agriculturists should here also perhaps be 89. 1 Schoolcraft, Archives of Aboriginal Knowledge, v. 269. xxn CONSENT AS A CONDITION OF MARRIAGE 315 We may assume, however, that advancement in economic culture has exercised an unfavourable influence upon the woman's liberty of choice not only by giving rise to marriage by purchase, but for another reason as well. By leading to accumulation of property and the distinction between richer and poorer people it naturally increases the interest which the family takes in the marriage of its members, and its head becomes less willing to allow individual inclinations to have their free play. Speaking of the Hottentots, among whom the first youth who could gain a girl's affections readily obtained the consent of her parents, Le Vaillant observes that in a country where there is no difference in birth or rank and all the schemes for heaping fortune on fortune in the same coffer are utterly unknown, " parents have not a single motive which can tempt them to oppose those sentiments of predilection, which their children may form for one object in preference to another."1 So also we are told that among the Moqui, or Mittelmesa-Hopi, most marriages are love-matches because the differences in wealth are so insignificant that from the family's point of view it is a matter of little moment whom its members marry.2 But the case may be different among tribes where there are rich and poor and inheritance is an important source of increasing wealth. In our discussion of the circumstances which have inter- fered with savage women's liberty to choose their husbands we have so far taken no notice of the Australian aborigines, who, contrary to the lower hunters in other parts of the. world, generally seem to allow their women no voice at all in the matter. Marriage by ordinary purchase does not exist among them, although the man may have to make small presents of game or weapons to the kindred of his bride. Schulze speaks of betrothals made from mercenary motives, a father promising his little daughter to a man who is many years older than the fiancee since "it is seemly that the son-in-law prove himself grateful by frequently handing over to his father-in-law and his bride part of his hunting 1 Le Vaillant, op. cit. ii. 67 sq. 2 Solberg, in Zeitschr. f. Ethnol. xxxvii. 629. 316 THE HISTORY OF HUMAN MARRIAGE chap. booty, and helping thus to provide for him in his old age."1 But the chief reasons for the compulsion to which the Australian women are subject are undoubtedly the excep- tionally great prevalence of infant-betrothals; the habit of the old men of appropriating to themselves the comeliest women ; and the custom of procuring a wife by the exchange of a sister or other female relative, which gives the woman no voice in the matter whether the barter takes place in her infancy or at a later age. These customs are connected with the extraordinary tyranny exercised by the old men and the great difficulties the younger men experience in getting wives, or young wives, in any other peaceful way than by infant-betrothals or the exchange of related females. Hence the utter disregard of the woman's wishes among the Australian natives is due to particular circumstances of an essentially local character and can by no means be regarded as a survival from a more primitive stage than that represented by the other lower hunters, whose social organisation is certainly of a less developed type than that of the best known Australian tribes. Professor Hobhouse and his collaborators say that it is only with some hesitation that they have included the whole of the Australians in the class of lower hunters,2 and I fully share their hesitation. In view of all these facts it is impossible to agree with Letourneau in his sweeping statement that during a very long period woman was married without her wishes being at all consulted.3 There is, on the contrary, every reason to believe that in primitive ages she was much freer in her choice of partner than she is now in the Australian tribes and among many other savages on a higher level of culture. It should be added that even where custom, strictly speaking, gives to the woman no voice in the matter she may nevertheless possess means of preventing a marriage which is distasteful to her or of breaking one which has been forced upon her against her will. She may enlist the 1 Schulze, ' Aborigines of the Upper and Middle Finke River,' in Trans, and Proceed. Roy. Soc. South Australia, xiv. 236. 2 Hobhouse, Wheeler, and Ginsberg, op. cit. p. 18. 3 Letourneau, Sociology, p. 378. xxii CONSENT AS A CONDITION OF MARRIAGE 317 sympathy of her mother or other female relatives, whose intervention would prevail upon the father to change his mind.1 Among the Xosa Kafirs she can on the day fixed for the wedding render the marriage contract null and void by smearing herself with human excrement, which induces her father to return the cattle paid for her.2 The meaning of this custom is not stated, but I presume that there is at the bottom of it some superstition similar to that which Berber girls in Morocco sometimes employ as a weapon against an objectionable proposal. Among the Aith Sadden in the neighbourhood of Fez, when the young man's mother, accompanied by some other women, comes to the house or tent of the girl's parents to let them know that her husband intends to sue for their daughter on behalf of his son, it lies in the girl's power to influence the proceedings although she is not at all consulted in the matter. If she is fond of the young man she dresses herself in fine clothes and sits down with the women, trying to be as attractive as possible ; on the other hand, if she dislikes him she makes use of bad and ominous words which should not be mentioned on this occasion,3 or she behaves like a woman at a funeral, scratch- ing her face and dirtying herself with cowdung. The result of this may be that no further steps are taken, for fear lest a marriage arranged in such circumstances should be unlucky. Moreover, it has sometimes happened in the same tribe that the girl has prevented the proposed marriage by running away on this occasion, or on the day of the intended wedding. Of some West African Negroes, again, we are told that although a father can force his daughter to marry against her will, " such marriages are troublesome, and generally end in the man putting the woman away.' 4 1 See Turner, ' Ethnology of the Ungava District, Hudson Bay District,' in Ann. Rep. Bur. Ethnol. xi. 188 (Labrador Eskimo) ; Zache, ' Sitten und Gebrauche der Suaheli,' in Zeitschr. f. Ethnol. xxxi. 78. 2 Kropf, op. cit. p. 141. 3 In Morocco, as elsewhere, certain words must be avoided as unlucky on certain occasions. See Westermarck, Marriage Cere- monies in Morocco, p. 37. 4 Nassau, Fetichism in West Africa, p. 8. 318 THE HISTORY OF HUMAN MARRIAGE chap. Among the Chimariko Indians of California, if a girl is sent by her parents as a wife to a man whom she dislikes, she will bite his hands and scratch him until he sends her back to her home.1 Of various peoples we learn that it is a common custom for women to run away from men who have been forced upon them by their parents.2 In Pentecost, of the New Hebrides, when a woman has left her husband several times and repeated beatings have not changed her mind, " her parents pay back the money and the husband gives up his wife."3 Among the Bogos of North-Eastern Africa a woman who has been compelled to marry a man against her will may dissolve the union by three times running away from him.4 Among the Boloki on the Upper Congo, " when a free woman does not want to marry the man who is trying to arrange for her, she will tell him frankly that if he persists in marrying her, she will run away from him. But if, in spite of this threat, he completes the arrangements, then a few days after the marriage she will escape to a neighbouring town and put herself under the protection of the chief by tearing his cloth. The chief then gives the husband notice of what has happened, and before he can claim his wife he has to pay the chief. 600 brass rods = 39s. as compensation for his torn cloth. If the husband does not then permit her to marry the man she wants, she runs away again and again, and every time she runs it will cost her husband 600 brass rods. A sensible man will take warning by the first threat, and will not complete the marriage."5 It would be interesting to know whether the tearing of the chief's cloth, which induces him to interfere, is supposed to contain a conditional curse, as is the case with the Moorish 'dr, which is also, in some places, resorted to by a married woman as a means of ridding 1 Dixon, Chimariko Indians and Language, p. 301. 2 King and Fitzroy, op. cit. ii. 153 (Patagonians). Dobrizhoffer, Account of the Abipones, ii. 207. Brett, Indian Tribes of Guiana, p. 354 (Caribs). Fries, Grbnland, p. 111 (Greenlanders). 3 Speiser, Two Years with the Natives in the Western Pacific, p. 234 sq. 4 Munzinger, Ueber die Sitten und das Recht der Bogos, p. 64. 5 Weeks, Among Congo Cannibals, p. 126. xxii CONSENT AS A CONDITION OF MARRIAGE 319 her of her husband and at the same time of compelling another man to marry her. Among the Berbers inhabiting the mountain regions of Central Morocco I found the following custom. A woman who does not like to remain with her husband may fly to another man's house or tent, and embrace the pole supporting the roof or one of the vertical tent-poles, or, if there is no such pole, take hold of the handmill and turn it round as if she were grinding. Then the owner of the house or tent is obliged to marry her, whether he be a bachelor or a married man and whatever be the number of his wives ; and in addition he has to pay to the abandoned husband a compensation, fixed by custom, which varies considerably in different tribes and even in different divisions of the same tribe-among the Aith Sadden it amounts to five hundred dollars. This singular custom is based on the idea that some grave misfortune would befall the man if he did not wed the woman who in the said manner took refuge with him. For by taking hold of the pole of his dwelling or turning round his handmill, she makes 'ar on him, that is, transfers to him a conditional curse.1 A very common method by which a woman can obtain the husband she desires without the consent or against the will of her parents is to elope with him.2 Whilst this practice 1 See Westermarck, Marriage Ceremonies tn Morocco, p. 60 sqq. 2 James, Indians of the Painted Desert Region, p. 228 (Havasupai of Arizona). Dodge, Our Wild Indians, p. 199 (Plains Indians). Dunbar, ' Pawnee Indians,' in Magazine of American History, iv. 265 ; Grinnell., Story of the Indian, p. 46 (Pawnee). Hodge, op. cit. i. 809 (Siksika). Schoolcraft, Indian Tribes of the United States (Archives of Aboriginal Knowledge'), iii. 238 (Dakota); v. 683 (Comanche). Alary Eastman, Dahcotah, p. 103 sq. Matthews, Hidatsa Indians, p. 53. Dorsey, in Ann. Rep. Bur. Ethnol iii. 260 (Omaha). Harmon, Journal oj Voyages and Travels in the Interior of North America, p. 341 (Blackfeet, Chipewyan, Cree, &c.). Teit, ' Thompson Indians of British Columbia,' in Memoirs American Museum Natur. Hist. vol. ii. Anthropology, i. 324 sq.- Bogoras, Chukchee, p. 590 sq. Jochelson, Koryak, p. 744. Sternberg, reviewed in L'Anthropologic, v. 343 (Gilyak). Mendiarov, reviewed ibid. vi. 205 sq. (Cheremiss of Oufa). Yavorski, reviewed ibid. viii. 356 (Turkomans). Pat- kanov, quoted by Miss Czaplicka, Aboriginal Siberia, p. 105 (Tungus). Patkanov, Die Irtysch-Ostjaken und ihre Volkspoesie, i. 140. Kova- lewsky, Coutume contemporaine et loi ancienne, p. 167 (Ossetes). 320 THE HISTORY OF HUMAN MARRIAGE chap. indicates that the mutual consent of the parties is not always sufficient for the conclusion of a marriage, it also provides a remedy for the insufficiency. It is resorted to if the young man is too poor to pay the price asked for the girl or other- wise, from no fault of hers, is unable to marry her in the St. Elie, ' La Femme du desert autrefois et aujourd'hui,' in An- thropos, iii. 189 (Bedouins). Elphinstone, Account of the Kingdom oj Kaubul, i. 239 sq. (Afghans). Sarat Chandra Das, ' Marriage Customs of Tibet,' in Jour. Asiatic. Soc. Bengal, vol. Ixii. pt. iii. 9 sq. (Purang). Sherring, ' Notes on the Bhotias of Almora and British Garhwal,' in Memoirs Asiatic Soc. Bengal, i. 108. Thurston, Castes and Tribes of Southern India, v. 94 (Muduvars). Hayavadana Rao, ' Gonds of the Eastern Ghauts,' in Anthropos, v. 795. Dalton, Descriptive Ethnology of Bengal, pp. 192, 299 sq. (Hos, Boad, Kandhs). Hutchinson, Account of the Chittagong Hill Tracts, pp. 97, 167 (Chukmas, Mriis). Watt, in Jour. Anthr. Inst. xvi. 355 (Kaupuis of Manipur). Endle, Kachdris, p. 43 sq. Gilhcdes, ' Mariage et condition de la femme chez les Katchins (Birmanie),' in Anthropos, viii. 371. Rose and Brown, ' Lisu Tribes of the Burma-China Frontier,' in Memoirs Asiatic Soc. Bengal, iii. 262 (Lihsaws). Shway Yoe, The Burman, p. 54. Young, Kingdom of the Yellow Robe, p. 85 (Siamese). Kanakasabhai Pillai, Tamils Eighteen hundred Years ago, p. 120. Schmidt, Liebe und Ehe im alten und modernen Indien, p. 345 sqq. Marsden, History of Sumatra, p. 235 ; Crawfurd, History of the Indian Archipelago, iii. 129 sq. (Rejangs of Sumatra). Moszkowski, Auf neuen Wegen durch Sumatra, p. 240 sq. ; Junghuhn, op. cit. ii. 132 (Battas). Brooke Low, quoted by Ling Roth, Natives of Sarawak and British North Borneo, i. 118 (Dyaks). Jacobs, Eenigen Tijd onder de Baliers, p. 128. Anna Forbes, Insulinde, p. 171 (Tenimberese).- Kubary, ' Die Verbrechen und das Strafverfahren auf den Pelau-Inseln,' in Original-Mittheilungen aus der ethnologischen Abtheilung der kbnigl. Museen zu Berlin, i. 79 (Pelew Islanders). Seligman, Melanesians of British New Guinea, pp. 271, 363 n. 1 (Roro, Mekeo). Williamson, Mufulu Mountain People of British New Guinea, p. 173. Hagen, Unter den Papua's, p. 241 (Bogadjim). Vetter, in Nach- richten uber Kaiser Wilhelms-Land, 1897, p. 89 (Jabim). Thurnwald, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 16 (natives of Songa in Vellalavella). Codrington, Melanesians, p. 240 sqq. (Banks and Lepers' Islanders). Thomson, Fijians, p. 203. Wilkes, Narrative of the United States Exploring Expedition, v. 101 (Gilbert Islanders). Turner, Samoa, p. 95 sq. ; Schultz, ' Principles of Samoan Family Law,' in Jour. Polynesian Soc. xx. 49. Polack, Manners and Customs of the New Zealanders, i. 139 ; Colenso, Maori Races, p. 25; Tregear, The Maori Race, p. 294. - Shooter, xxii CONSENT AS A CONDITION OF MARRIAGE 321 ordinary way.1 Among many peoples elopement is a veritable institution, recognised by custom as a method of concluding a marriage or at least as a preliminary to it. Frequently the lover may afterwards have to pay for his bride or to conciliate her people with some gift.2 Among the Pawnee and Siksika, in case of elopement, the subsequent presentation of gifts legitimised the marriage and removed the disgrace which would otherwise attach to the girl and her family.3 Among the Kacharis, who look upon elope- ment as a blameworthy and irregular but not invalid pro- cedure, " the bride's parents claim an immediate payment of Rs. 5/- from the bridegroom, and also exact the bride- price at a higher rate than usual " ;4 but these conditions are quite exceptional for their severity. Among the Pangwe in West Africa the father of the girl is appeased by a part, Kafirs of Natal and the Zulu Country, pp. 57, 60. Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 215 (Bantu). Junod, Life of a South African Tribe, i. 120 sq. (Thonga). Heese, * Sitte und Brauch der Sango,' in Archiv f. Anthrop. N.S. xii. 136. Roscoe, Northern Bantu, p. 209 (Basoga). Cunningham, Uganda, pp. 112-114, 368 sq. (Basoga, Bavuma, Latuka). Hollis, Nandi, p. 61 n. 2. Merker, Die Masai, p. 46 sq. Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 185. Tessmann, Die Pangwe, ii. 260. Poupon, in L' Anthropologic, xxvi. 126 (Baya). Westermarck, Marriage Customs in Morocco, p. 36 (Berbers of the Aith Sadden in Central Morocco). 1 Sometimes sham elopements are arranged with the consent of the bride's parents in order to save the expenses of a regular wedding (Blau, ' Nachrichten uber kurdische Stamme,' in Zeitschr. Deutsch. Morgenland. Gesellsch. xvi. 624 [Dusik-Kurds in the Dersim Mountains south of Erzingan]; Ahmad Shah, Four Years in Tibet, p. 57; Schotter, 'Notes ethnographiques sur les tribus de Kouy- tcheou (Chine),' in Anthropos, vi. 320 [He-miao] ), or, when the bridegroom is too poor to pay the required price, " to sa.ve the credit of the bride's family in not having received a sufficient price for her " (Biddulph, Tribes of the Hindoo Koosh, p. 81 [Yassin] ). 2 Plains Indians, Pawnee (according to Dunbar), Siksika, Chere- miss of Oufa, Ostyak, people of Purang, Gonds of the Eastern Ghauts, Kacharis, Lihsaws, Battas (according to Moszkowski), natives of Bali and Tenimber, Roro and Mekeo tribes and Mafulu of British New Guinea, Jabim, people of Songa in Vellalavella, Banks Islanders, Thonga, Basoga, Latuka, Nandi, Pangwe, Baya. 8 Hodge, op. cit. i. 809. 4 Endle, op. cit. p. 43. VOL. II Y THE HISTORY OF HUMAN MARRIAGE chap. 322 perhaps one half, of the ordinary purchase-sum and generally grants delay of the payment of the rest of it.1 Of the Thonga we are told that if the lover does not succeed in bringing the regular payment for the girl, the first girl bom from their marriage will take its place-■" the child will pay for her mother " ;2 and a similar custom is found among the Latuka, a tribe in the British sphere, bordering on the Sudan.3 Among the Dyaks of Borneo, " when a young woman is in love with a man who is not acceptable to her parents, there is an old custom called nunghup bui, which permits him to carry her off to his own village. She will meet him by arrangement at the water-side, and step into his boat with a paddle in her hand, and both will pull away as fast as they can." When his pursuers arrive at his house he gives them food to eat and toddy to drink, and sends them home satisfied; and in the meanwhile he is left in possession of his wife.4 Among various peoples the elopement is by itself sufficient to make the run-away couple husband and wife. Among the Havasupai of Arizona, if parents refuse to sell their daughter to a suitor and the couple elope, " this ends the matter. The ethics of the tribe are such that cohabitation once entered upon, the parents have no authority to declare the marriage void. And, as a further penalty for his obdurate obstinacy, the father loses the ten dollars or its equivalent he might have had by being kind and complaisant to the desires of the young couple."5 Among the Thompson Indians, even if the father of a girl who has eloped with her lover brought her back, " he could only deliver her up to the young man, as custom declared them already married."6 The Dakota, according to Mary Eastman, have two kinds of marriages, buying a wife and run-away matches; for " it is an understood thing that, when the young people run 1 Tessmann, op. cit. ii. 260. 2 Junod, Life of a South African Tribe, i. 121. 3 Cunningham, Uganda, p. 369. 4 Brooke Low, quoted by Ling Roth, .Natives of Sarawak, i. 118. 5 James, Indians of the Painted Desert Region, p. 228 sq. 6 Teit, in Memoirs American Museum Nat. Hist. vol. ii. Anthro- pology, i. 324. xxn CONSENT AS A CONDITION OF MARRIAGE 323 away, they are to be forgiven at any time they choose to return, if it should be the next day, or six months after wards."1 Among the Hidatsa Indians of Dakota, if the eloped pair remain out on the prairie for a week or so and then return to the village, this usually ends the trouble and they are considered married; yet " such marriages are looked upon as undignified, and different terms are applied to a marriage by elopement and a marriage by parental consent."2 The Omaha thought that a father, brother, or uncle had no reason to feel vexed if his daughter, sister, or niece eloped with a lover, and ridiculed him if he showed anger on that account; for a man could run away with a woman who consented to it, whether she belonged to his own or to another tribe.3 Should a Koryak girl run away to her lover's house against the will of her kindred, " her parents would not demand her return, as she went of her own accord."4 Among the Bhotias of the Almora district, when the eloped girl formally eats the ceremonial food- dalong, datu, and jam-with her lover, her relations are in process of time " compelled to accept what cannot be mended."5 Among the Muduvars of Southern India, if a man and woman do not obtain the consent of their relatives to marrying each other, they run away into the jungle or a cave, visiting the village frequently and getting grain and other food from sympathisers ; but when the anger aroused by their behaviour has subsided, " they quietly return to the village, and live as man and wife."6 The Chukmas of the Chittagong Hills have marriage by elopement; the parents of the girl can demand her restitution on three separate occasions, but " if the ardent lover can successfully bring off a fourth elopement, he has secured the prize and won his wife."7 Among the Masai it not infrequently happens that a girl immediately on becoming marriageable leaves the hut where she is living with her mother and flies 1 Eastman, op. cit. p. 103. 2 Matthews, op. cit. pp. 53, 167, 208. 3 Dorsey, in Ann. Rep. Bur. Ethnol. iii. 260. 4 Jochelson, Koryak, p. 744. 6 Sherring, in Memoirs Asiatic Soc. Bengal, i. 108. " Thurston, Castes and Tribes of Southern India, v. 94. 7 Hutchinson, op. cit. p. 97. 324 THE HISTORY OF HUMAN MARRIAGE chap. to her lover. If her father-as is often the case-then sends a troop of boys carrying sticks to fetch her back, the lover runs away with her to the woods threatening forcible resistance if anybody is trying to separate them. " As soon as the couple have eaten a head of cattle in the woods, they return to the village and are without further ceremonies recognised as man and wife."1 Nowhere is marriage by elopement more frequent, and indeed more needed, than in many Australian tribes.2 Among the Kumai, according to Howitt, a man, with rare exceptions, " could acquire a wife in one way only, namely, by running off with her secretly and with her own consent. Marriage, therefore, was by elopement." It was the busi- ness of the bunjil-yenjin, a medicine-man, to aid the elope- ment of young couples, and this gave sanction to the prac- tice, which arose from the difficulty in finding a wife.3 If the parties are prohibited from intermarrying on account of the relationship in which they stand to each other, the elopement is punished with great severity,4 whereas other- wise the lover may, at least in certain circumstances, retain the woman with whom he eloped. In many cases 1 Merker, op. cit. p. 46 sq. 8 See Malinowski, The Family among the Australian Aborigines, pp. 36-47, 55 sqq., and the various authorities quoted by him ; Browne, ' Die Eingebomen Australiens, ihre Sitten und Gebrauche,' in Mittheilungen aus Justus Perthes' Geographischer Anstalt, 1856, p. 450 (natives of King George's Sound) ; Wilhelmi, ' Manners and Customs of the Australian Natives, in particular of the Port Lincoln District,' in Trans. Roy. Soc. Victoria, v. 179; McKiernan, ' Some Notes on the Aborigines of the Lower Hunter River, New South Wales,' in Anthropos, vi. 887. According to Dr. Howitt (' On the Organisation of Australian Tribes,' in Trans. Roy. Soc. Victoria, vol. i. pt. ii. 118), marriage by elopement may be safely said to obtain " in all tribes in which infant betrothal occurs, and where the young men, or some of them, find more or less difficulty through this, practice, or by there being no female relative available for exchange, or indeed wherever a couple fall in love with each other and cannot obtain consent to their marriage." Dr. Malinowski observes (op. cit. p. 47) that " there seems to have been not a single tribe in which elopement was completely absent." 8 Howitt, Native Tribes of South-East Australia, pp. 273, 274, 279. 4 Ibid. pp. 222, 246. xxn CONSENT AS A CONDITION OF MARRIAGE 325 he has to fight the man to whom she has been promised or, if she is already married, her husband,1 or a more general struggle ensues in which kindred are involved;2 and in either case the issue of the fight decides whether he will be allowed to keep the woman or not. Among the Wiimbaio " the girl was pursued by her father and brothers, and the man she had eloped with would have to allow them to strike him on the head with a club, after which in some cases he would retain her. But in other cases there was a fight between her kindred, male and female, and those of the man she went off with."3 In the Wollaroi tribe, in cases of elopement with the wife of another man, it was the cus- tom for the abductor " to stand out before a number of the woman's kindred, who were armed with spears, he having merely a spear for his protection, to turn them aside " ; and if he passed through the ordeal safely he was allowed to keep the woman.4 In certain tribes within fifty miles of Maryborough in Queensland the young man has to fight all her male relatives who choose to take the matter up, and the girl is at the same time severely beaten by her kindred ; but he may also placate both her relatives and her promised husband by sending presents to them before he returns.5 In some tribes the man, in order to be allowed to retain the woman with whom he has run away, has to give a female relative in exchange for her, but even then he may have to fight or be subject to the usual ordeal.6 Among the Yuin in New South Wales, if the loving couple could escape and remain away until a child was born, nothing would be done to them, especially if the man could find a sister to exchange for his bride ;7 and in some other tribes, also, the man would be allowed to retain the woman with 1 Ibid. pp. 216, 222, 258. Roth, Ethnological Studies among the North-West-Central Queensland Aborigines, p. 181. Brown, * Three Tribes of Western Australia,' in Jour. Roy. Anthr. Inst, xliii. 158 (Kariera tribe). 2 Howitt, op. cit. pp. 194, 236 sq. Petrie, Reminiscences of Early Queensland, p. 60. - 3 Howitt, op. cit. p. 194. * Ibid. p. 217. 5 Ibid. p. 233. 6 Ibid. pp. 246, 255. Lang, Aborigines of Australia, p. 12. 7 Howitt, op. cit. p. 263. 326 THE HISTORY OF HUMAN MARRIAGE chap. whom he eloped if they stayed away till a child was born 1 or the girl was with child.2 Among the Goulburn tribe of Victoria, if " the happy couple straightway elope, and remain together in the bush for two nights and one day, in order to elude the pretended search of the tribe to whom the female belonged," the man may return with her as his wife to his own tribe.3 In other cases, again, repeated elopements make the marriage lawful.4 When passing from the savage and barbarous races of men to those next above them in civilisation, we find paternal, or parental, authority and filial reverence at their height.6 In ancient Mexico necessitous parents were allowed to dispose of any one of their children in order to relieve their poverty, although a master could not sell a well-behaved slave without his consent.6 Clavigero says that children were bred to stand so much in awe of their parents that, even when grown up and married, they hardly durst speak before them.7 The following is an exhortation which an Aztec gave to his son :-" Honour all persons, particularly thy parents, to whom thou owest obedience, respect, and service. Guard against imitating the example of those wicked sons, who, like brutes that are deprived of reason, neither reverence their parents, listen to their instruction, nor submit to their correction ; because whoever follows their steps will have an unhappy end, will die in a desperate or sudden manner, or will be killed and devoured by wild beasts."8 A youth was seldom allowed to choose a wife for himself but was expected to abide by the selection of his parents. Hence it rarely happened that a marriage 1 Ibid. p. 259 (Narrang-ga tribe of Yorke Peninsula). 2 Fison and Howitt, Kamilaroi and Kumai, p. 202 (Kurnai). 3 Blandowski, quoted by Wilhelmi, in Trans. Roy. Soc. Victoria, v. 179 sq. 4 Dawson, Australian Aborigines, p. 34 (South-West Victorian tribes). Fraser, Aborigines of New South Wales, p. 26 sq. 5 Cf. Westermarck, Origin and Development of the Moral Ideas, i. 607 sqq. 6 Clavigero, History of Mexico, i. 360. 7 Ibid. i. 331. a fbid j 232. xxn CONSENT AS A CONDITION OF MARRIAGE 327 took place without the sanction of parents or other kinsfolk, and he who presumed to choose his own wife or married without such consent had to undergo penance and was looked upon as ungrateful, ill-bred, and apostate.1 The belief was, according to Torquemada, that an act of that kind would be punished with some misfortune.2 Concerning the Guatemalans Mr, Bancroft says :-" It seems incredible that the young men should have quietly submitted to having their wives picked out for them without being allowed any voice or choice in the matter. Yet we are told that so great was their obedience and submission to their parents, that there never was any scandal in these things." 3 In the greater part of Nicaragua matches were arranged by the parents ; though there were certain independent towns in which the girls chose their husbands from among the young men while the latter sat at a feast.4 In ancient Peru, Inca Pachacutec confirmed the law that sons should obey and serve their fathers until they reached the age of twenty- five, and that none should marry without the consent of his own parents and of those of the girl, a marriage without such consent being invalid and the children illegitimate.5 But we are told that the preference of the parties was also to be consulted.6 Similar ideas formerly prevailed, or to some extent still prevail, among the civilised nations of the Old World. Confucius taught:-Of all the actions of man there is none greater than filial piety. In filial piety there is nothing greater than the reverential awe of one's father. In the reverential awe shown to one's father there is nothing greater than the making him the correlate of Heaven."7 1 IbidA. 319. Bancroft, op. cit. ii. 251. 2 Torquemada, Monarchia Indiana, ii. 415. 3 Bancroft, op. cit. ii. 666. 4 Ibid. ii. 667. Squier, ' Observations on the Archaeology and Ethnology of Nicaragua,' in Trans. American Ethn. Soc. vol. iii. pt. i. 127. 5 Garcilasso de la Vega, First Part of the Royal Commentaries of the Yncas, ii. 207. 8 Prescott, History of the Conquest of Peru, p. 54. 7 Hsido King, 9 (Sacred Books of the East, iii. 476). Cf. ibid. 1, 7 (Sacred Books of the East, iii. 466, 473). 328 THE HISTORY OF HUMAN MARRIAGE chap. But the idea that filial piety is the fundamental duty of man was not originated by Confucius ; it had obtained a firm hold of the national mind long before his time.1 And it has up to our days dominated the Chinese legislation relating to the family. The house-father reigns almost supreme in his household, and not even marriage withdraws the son from his power.2 The old Chinese code, it is true, prohibits him from killing 3 and selling4 his children ; but it is only in supreme cases that the State has interfered between the head of a household and his family belongings, and the sale of children has been practically allowed.5 No person, of whatever age, can act for himself in matri- monial matters during the lifetime or in the neighbourhood of his parents or near senior kinsfolk. The power of these guardians is so great that they may contract a marriage for a junior who is absent from home, and he is bound to abide by such engagement even though already affianced elsewhere without their privity or consent.6 The consequence of this system is that in many cases the betrothed couple scarcely know each other before marriage, the wedding being the first occasion on which the man catches a glimpse of his wife's face.7 In Corea it is the rule that men never see their brides until the wedding day, all preliminaries being arranged by the fathers of the young couple, whose inclinations are not consulted at all.8 In Japan the authority of a house-father was in former times as strong as in China.9 It was the established principle of Japanese customary law that a member of a house must 1 Douglas, Confucianism and Taouism, p. 118. 2 de Groot, Religious System of China, (vol. ii. book) i. 507. s Ta Tsing Leu Lee, sec. cccxix. p. 347. 4 Ibid. sec. cclxxv. p. 292. 5 Douglas, Society in China, p. 78. Staunton, in his translation of Ta Tsing Leu Lee, p. 292 n.* Doolittle, Social Life of the Chinese, ii. 209. 6 Medhurst, ' Marriage, Affinity, and Inheritance in China,' in Trans. Roy. Asiatic Soc. China Branch, iv. 11. 7 Gray, China, i. 205. 8 Saunderson, ' Notes on Corea and its People,' in Jour. Anthr. Inst. xxiv. 305. 9 Rein, Japan, i. 583. Griffis, Religions of Japan, p. 122 sq. xxii CONSENT AS A CONDITION OF MARRIAGE 329 obtain the consent of the head of the family for his or her marriage,1 and often the consent of the parties themselves was not required.2 The House-law of the Taiho Code also required the consent of grandparents, parents, and other relatives before a marriage could be celebrated.3 According to the regulations of the new Civil Code of Japan, the free con- sent of the persons who marry is necessary in addition to that of the parents " who are in the same house " ; and when a man has completed his thirtieth year and a woman her twenty-fifth, the consent of the parents is no longer required.4 But as a matter of fact, marriages are still generally arranged by the parents of the parties.5 In ancient Chaldaea the father likewise had great power over his children.6 Hommel believes that the mother's authority was as great as the father's ;7 whereas Meissner concludes that it was less, from the fact that her children are not seldom found to be at law with her in matters of succession.8 According to the latter, a daughter was given away in marriage by her father without being able to raise any objection to his choice.® As Dr. Koschaker observes, a Babylonian woman was said not to marry, but to be married.10 We are also told that a son could not conclude a valid marriage without the consent of his father, although he could take a concubine without such consent j11 but it is not known whether the father's right to interfere with the 1 Hozumi, Ancestor-Worship and Japanese Law, p. 131 sq. 2 Nakajima, ' Marriage (Japanese and Korean),' in Hastings, Encyclopedia of Religion and Ethics, viii. 459. 3 Hozumi, op. cit. p. 132. * Civil Code of Japan, § 772, p. 203. 6 Crasselt, ' Die Stellung der Ehefrau in Japan,' in Anthropos, iii. 546. • Oppert, review of Haupt, ' Die sumerischen Familiengesetze,' in Gbttingische gelehrte Anzeigen, 1879, p. 1604 sqq. Hommel, Die semitischen Vbiker und Sprachen, i. 416. Meissner, Beitrage zum altbabylonischen Privatrecht, p. 14 sq. 7 Hommel, op. cit. i. 416. 8 Meissner, op. cit. p. 15. 9 Ibid. p. 13. 10 Koschaker, Rechtsvergleichende Studien zur Gesetzgebung Hammu- rapis, p. 119. Cf. ibid. p. 123 sq. 11 Kohler and Peiser, Aus dem Babylonischen Rechtsleben, ii. 7 sqq. Peiser, Skizze der Babylonischen Gesellschaft, p. 10 (Mitteilungen der Vorderasiatischen Gesellschaft, 1896, p. 154). 330 THE HISTORY OF HUMAN MARRIAGE chap. matrimonial affairs of his son lasted till his death. That his power over his sons was great appears from the facts that a disobedient son might be sold as a slave,1 and that, according to the Laws of Qammurabi, a man might give not only his daughter but his son as a hostage for debts,2 although he could not disown his children at discretion;3 The importance which the ancient Hebrews attached to the duties of a child to its parents may be at once learned from the placing of the law on the subject among the Ten Commandments, and from its position there in the im- mediate proximity to the commands relating to the duties of man towards God.4 A father might sell his child to relieve his own distress, or offer it to a creditor as a pledge.6 He had not only unlimited power to marry his daughters,6 but even to sell them as maids into concubinage, though not to a foreign people.7 He also chose wives for his sons,8 the selection, however, sometimes being made by the mother and there is no indication that the subjection of sons ceased after a certain age.10 According to the later Jewish law, the consent of parents is no legal requirement when the parties to the marriage are of age.11 But Mielziner states that, in consequence of the high respect and veneration in which father and mother have ever been held among Israelites, " the- cases of contracting marriage without the parents' consent fortunately belong to the rarest exceptions."12 1 Hommel, op. cit. i. 416. Meissner, op. cit. p. 1. Cf. Koschaker, op. cit. p. 126 sqq. 2 Laws of Qammurabi, § 117 (Johns' trans, p. 21 sq. ; Winckler's trans, p. 33 sqq.). 3 Ibid. § 168 (Johns' trans, p. 34; Winckler's trans, p. 49). 4 Cf. Ewald, Antiquities of Israel, p. 188 ; Gans, Das Erbrecht in weltgeschichtlicher Entwickelung, i. 134. * Ewald, op. cit. p. 190. Wellhausen, Prolegomena to the History of Israel, p. 465. 8 Benzinger, Hebraische Archaologie, p. 107. 7 Exodus, xxi. 7 sq. 8 Genesis, xxiv. 4 ; xxviii. 1 sq. ; xxxviii. 6. Exodus, xxxiv. 16. Deuteronomy, vii. 3. 9 Genesis, xxi. 21. 10 Cf. Michaelis, Commentaries on the Laws of Moses, i. 444. 11 Maimonides, Ishuth, quoted by Mielziner, Jewish Law of Marriage and Divorce, p. 69. Lichtschein, Die Ehe nach mosaisch-talmudischer Auffassung, p. 41. 12 Mielziner, op. cit. p. 69 sq. xxii CONSENT AS. A CONDITION OF MARRIAGE 331 Of the ancient Arabs Wellhausen says that a woman was betrothed by her wait, or guardian, that is, her father, brother, or cousin ; but " of course, the daughter is often asked, by affectionate parents, whether she wants to have the suitor."1 According to Muhammadan law, the woman's consent is not required if she is still in her father's power. Among the Hanafis and the Shia'hs the father's right to marry his daughter without her consent comes to an end when she arrives at puberty,2 but this is not the case among the Maliki school of Muhammadans. Among them she ceases to be in his power only by his death, or by her being expressly emancipated by him during his lifetime,3 or by her marriage (unless she has been married before puberty and the marriage has then been dissolved, or the marriage has been dissolved, without having been consummated, before she has lived a year in her husband's dwelling), or, according to some jurists, when she has reached the age of at least thirty.4 On the other hand, when marriage is con- tracted on behalf of a woman who is no longer in her father's power, it is necessary that she should give her consent to it, either in express terms, or, if she be a virgin, at least by implication ; and in the latter case her silence or laugh is construed to imply consent.6 In Morocco, Algeria,6 Tunis,7 and many parts of Palestine8 it is the general rule that the 1 Wellhausen, ' Die Ehe bei den Arabern,' in Nachrichten von der Kdnigl. Gesellschaft der Wissenschaften zu Gbttingen, 1893, p. 431 sq. 2 Ameer Ali, Mahommedan Law, ii. 335. Milliot, La Femme musulmane au Maghreb, p. 81. 8 Emancipation, however, may be only for the purpose of enabling her to select a husband for herself, while her business affairs remain subject to the paternal control (Russell and Abdullah al-Ma'mun Suhrawardy, ' A Manual of the Law of Marriage,' from the Mukbtasar of Sidi Khalil, p. 7 n. 2). * Sidi Halil, Muhtasar, § 24 sq. (Russell and Abdullah al-Ma'mun Suhrawardy, op. cit. p. 6 57). Milliot, op. cit. pp. 79, 80, 87 sq. 8 Ameer Ali, op. cit. ii. 334, 335, 343 sq. Milliot, op. cit. p. 101 sq. Sidi JIalil, op. cit. § 40 (p. 12). e Villot, Mceurs, coutumes et institutions des indigenes de l'Algdrie, p. 76. 7 Sellami, ' La Femme musulmane,' in Revue Tunisienne, iii. 435. 8 Klein, ' Mittheilungen uber Leben, Sitten und GebrS,uche der Fellachen in Palastina,' in Zeitschr. d. Deutschen Palaestina-Vereins, vi. 88 sq. 332 THE HISTORY OF HUMAN MARRIAGE chap. parents of a girl marry her without asking for her consent. But the Bedawin of the desert " differ entirely from all Mohammedan natives of Palestine in allowing their girls to accept or reject a proposal";1 among the Aeneze of Northern Arabia the girl's wishes are consulted, and it is never supposed that she should be compelled to marry against her inclination ; 2 and in Mecca a virgin is seldom forced into marriage, although it is considered proper that she should behave as though she submitted to her father's will only because she has to obey him.3 According to all the Muhammadan schools, a son is at liberty to contract a marriage without his father's consent when he has attained his puberty, which is mostly presumed on the completion of the fifteenth year.4 But although the father certainly has the right to impose the status of marriage on his children during their minority, sons and daughters alike, the law takes care that this right shall not be exercised to the preju- dice of the infant; any act of the father which is likely to injure the interests of the minor is considered illegal, and entitles the judge to interfere in order to prevent the com- pletion of such act or, if complete, to annul it.5 As a matter of fact, however, in Morocco and elsewhere6 parents not infrequently arrange the marriage of their son even though he be grown-up, according to their own taste ; and custom may require that he should comply with their wishes. Where the separation between the sexes is so strict as it often is in the Muhammadan world, the inter ference of parents in the matrimonial affairs of their son can hardly be felt as a burden by the young man, especially as he can readily divorce a wife whom he does not like. And for a girl it would be no easy mattet to choose between suitors whom she does not know. Among the ancient Romans, in relation to the house- father, " all in the household were destitute of legal rights- 1 Robinson Lee, Witness of the Wilderness, p. 120. 2 Burckhardt, Notes on the Bedouins and Wahdbys, p. 61. 3 Snouck Hurgronje, Mekka, ii. 157. 4 Ameer Ali, op. cit. ii. 273 sq. 8 Ibid. ii. 273, 275 sq. * Jaussen, Coutumes des Arabes au pays de Moab, p. 43. Snouck Hurgronje, op. cit. ii. 157 (Mecca). xxii CONSENT AS A CONDITION OF MARRIAGE 333 the wife and the child no less than the bullock or the slave."1 The father not only had judicial authority over his child/en -implying the right of inflicting capital punishment on them, though not without a just cause,8-but he could sell them at discretion.3 Even the grown-up son and his children were subject to the house-father's authority,4 and in marriage without conventio in manum a daughter remained in the power of her father or tutor even after marriage. The consent of the paterfamilias was indispensable to the marriage of children, sons and daughters alike ;6 and so strict was this rule that down to the reign of Marcus Aurelius the children of a mente capitis could not contract a legal marriage while in the power of their father, the latter being incapable of giving his consent.6 It has been suggested by Sir Henry Maine and others that the fatria potestas of the Romans was a survival of the paternal authority which existed among the primitive Aryans.7 But no clear evidence of the general prevalence of such unlimited authority among other so-called Aryan peoples has been adduced.8 The ancient jurist observed, " The power which we have over our children is peculiar to Roman citizens; for there are no other nations possessing the same power over their children as we have over ours."8 That among the Greeks and Teutons the father had the right to expose his children in their infancy, to sell them in case of urgency as long as they remained in his power,10 1 Mommsen, History of Rome, i. 74. 2 Westermarck, Origin and Development of the Moral Ideas, i. 393, 611. 8 Dionysius of Halicarnassus, Antiquitates Romanae, ii. 27. 4 Institutiones, i. 9. 3. 8 Ibid. i. 10. Rossbach, Untersuchungen uber die romische Ehe, p. 393. Mackenzie, Studies in Roman Law, p. 104. 6 Mackenzie, op. cit. p. 104 n. 4. 7 Maine, Ancient Law, p. 138. Fustel de Coulanges, La cite antique, p. 106 sq. Hearn, Aryan Household, p. 92. 8 For Greece cf. v. Wilamowitz-Moellendorff and Niese, Staat und Gesellschaft der Griechen und Rbmer, p. 35. 9 Institutiones, i. 9. 2. 10 Leist, Graeco-italische Rechtsgeschichte, p. 60 sq. Grimm, Deutsche Rechtsalterthilmer, p. 461 sq. Brunner, Deutsche Rechts- geschichte, i. 76. 334 THE HISTORY OF HUMAN MARRIAGE chap. and to give away his daughters in marriage without consulting their wishes,1 does not imply the possession of a sovereignty like that which the Roman house-father exercised over his descendants of all ages. It was very frequently the lot of a Greek woman to be given in marriage to a man whom she did not know.2 In Greece 3 and among all the Teutonic nations4 the father's authority over his son came to an end when the latter grew up and left his home ; and a grown-up son might choose his own wife.6 But it seems that among the Teutons a man, in doing so, might have been required by custom to take counsel with his kinsfolk.6 Nor is there any evidence that the ft atria potestas of the Roman type ever prevailed in full in India, great though the father's or the parents' authority has been, and still is, among the Hindus. In Vedic times the father seems to have been the head of the family only as long as he was able to be its protector and maintainer,7 decrepit parents being even allowed to die of starvation.8 Macdonell and Keith maintain that considerable freedom was probably left both to man and woman in selecting a wife or a husband, and deny the existence of any clear evidence in favour of Zimmer's assertion that the consent of the father was needed/ although 1 Cauvet, ' De Torganisation de la famille k Athenes,' in Revue de legislation, xxiv. 147. Muller, Doric Race, ii. 298. Hruza, Beitrdge zur Geschichte des griechischen und romischen Familienrechtes, i. 9. Beauchet, Histoire du droit privi de la Republique Athenienne, i. 113, 120, 121, 132-134, 157. Schroeder, Lehrbuch der deutschen Rechtsgeschichte, p. 299. Sohm, Das Recht der Eheschliessung aus dem deutschen und canonischen Recht geschichtlich entwickelt, p. 50. Wilda, Das Strafrecht der Germanen, p. 802. Weinhold, Die deutschen Frauen in dem Mittelalter, i. 303. Olivecrona, Om makars giftorcitt i bo, p. 48. Roeder, Die Familie bei den Angelsachsen, p. 24. 2 Becker-Goll, Charikles, iii. 347 sq. Hermann-Blumner, Lehrbuch der griechischen Privatalterthiimer, p. 261. 8 Cauvet, loc. cit. p. 138. Leist, op. cit. p. 62 sq. 4 Grimm, op . cit. p. 462. Koenigswarter, Histoire de 1'organisation de la famille en France, p. 140. Brunner, op. cit. i. 75 sq. 6 Cauvet, loc. cit. p. 138. Beauchet, op. cit. i. 134, 157. • Tacitus, Germania, ch. 18. Olivecrona, op. cit. p. 143. Winroth, Aktenskapshindren, p. 49 sq. 7 Rig-Veda, i. 70. 5. 8 Zimmer Altindisches Leben, p. 328. 9 Ibid. p. 309, xxn CONSENT AS A CONDITION OF MARRIAGE 335 no doubt parents often arranged suitable marriages for their children.1 According to some sacred books from a later age, the father and the mother have power to give, to sell, and to abandon their son, because " man formed of uterine blood and virile seed proceeds from his mother and his father as an effect from its cause " ; but an only son may not be given or received in adoption, nor is a woman allowed to give or receive a son except with her husband's permission.2 In other books it is said that " the gift or acceptance of a child and the right to sell or buy a child are not recognised," 8 and that he who casts off his son-unless the son be guilty of a crime causing loss, of caste-shall be fined by the king six hundred panasA According to the ' Laws of Manu ' a daughter might choose her husband in accordance with her own wish, but the legislator disapproves of such a " voluntary union of a maiden and her lover . . . which springs from desire and has sexual intercourse for its pur- pose."6 The four marriages-brahma, daiva, drsha, and prajapatya-in which the father gives away his daughter are blessed marriages, and from them spring sons radiant with knowledge of the Veda, honoured by good men, and destined to live a hundred years. But the remaining four marriages-those effected by purchase, voluntary union, forcible abduction, or stealth-are blamable marriages, from which spring sons who are cruel and untruthful, who hate the Veda and the sacred law.6 Narada says that a maiden shall be given in marriage by her father, or by her brother with the father's authority, or by her paternal grandfather, or by her maternal uncle, or by agnates or cognates, or in default of all these by her mother in case she is competent to act as guardian, or by the distant connections if the mother is not competent. If no such persons be in 1 Macdonell and Keith, Vedic Index of Names and Subjects, i. 482. See also Delbriick, ' Die indogermanischen Verwandt- schaftsnamen,' in Abhandlungen der philologisch-historischen Classe der Konigl. Sachsischen Gesellsch. f. Wissensch. xi. 574. 2 Vdsishtha, xv. i. sqq. Baudhdyana Parisishta, vii. 5. 2 sqq. 3 Apaslamba, ii. 6. 13. II. 1 Laws of Manu, viii. 389. Cf. ibid. xi. 60. 5 Ibid. iii. 32. ti Ibid. iii. 39 sqq. 336 THE HISTORY OF HUMAN MARRIAGE chap. existence, the maiden shall have recourse to the king, and with his permission betake herself to a bridegroom of her own choice.1 Whatever be the strict legal rights of a parent, filial piety is a most stringent duty in a child.* A man has three Atigurus, or specially venerable superiors : his father, mother, and spiritual teacher. To them he must always pay obedience. He must do what is agreeable and service- able to them. He must never do anything without their leave.3 " By honouring these three all that ought to be done by man is accomplished ; that is clearly the highest duty, every other act is a subordinate duty." 4 Similar feelings prevail among the modern Hindus.6 Sir W. H. Sleeman observes, " There is no part of the world, I believe, where parents are so much reverenced by their sons as they are in India in all classes of society " ; whilst the duty of daughters is from the day of their marriage transferred entirely to their husbands and their husbands' parents.6 According to the existing customs of the Hindus, " the con- sent of the parents on the girl's side is essential, and on the boy's it is considered necessary on a first marriage, and is always advisable." 7 A native writer states that " the Hindu youth has to maintain an attitude of utter indifference about every proposal regarding his marriage, and when any arrangement in that respect is made by his parents, grand- parents, uncles or elder brothers, he has to go through the ceremony out of his sense of duty to obey or oblige them."8 According to ancient Russian laws, fathers had great power over their children ;9 but it is not probable that a son could be sold as a slave.10 Baron von Haxthausen, who wrote before the Emancipation in 1861, says that 1 NArada, xii. 20 sqq. 2 Apastamba, i. 4. 14. 6. Laws of Manu, ii. 225 sqq. ; iv. 162 ; &c. 3 Institutes of Vishnu, ch. 31. 4 Laws of Manu, ii. 237. 6 Nelson, View of the Hindu Law, p. 56 sq. Ghani, ' Social Life and Morality in India,' in International Journal of Ethics, vii. 312. 4 Sleeman, Rambles and Recollections of an Indian Official, i. 330 sqq. 7 Steele, Law and Custom of Hindoo Castes, p. 162. 8 Jogendra Nath Bhattacharya, Hindu Castes and Sects, p. 12. 9 Accurse, quoted by de Lauriere, in Loysel, Institutes coutumi^res, i. 82. 10 Macieiowski, Slavische Rechtsgeschichte, iv. 404. xxn CONSENT AS A CONDITION OF MARRIAGE 337 " the patriarchal government, feelings, and organisation are in full activity in the life, manners, and customs of the Great Russians. The same unlimited authority which the father exercises over all his children is possessed by the mother over her daughters."1 Even the adult son was subject to his father's authority until he had himself children arrived at years of discretion, or had in his turn become the head of the family.2 As has been noticed above, it was a common custom for a father to marry his young sons to grown-up women ;3 and we are told that in Poland also, in ancient times, a father used to select a bride for his son.4 According to Professor Bogisic, the power of the father is not so great among the Southern Slavs as among the Russians.6 But a son is not permitted to make a proposal of marriage to a girl against the will of his parents; and among the Croatians and Serbians it is quite exceptional for the young man himself to look about for his future wife.6 A. daughter, of course, enjoys still less freedom of disposing of her own hand.7 Caesar states that in Gaul the house-father had the power of life and death over his wife and children.8 M. d'Arbois de Jubainville maintains that according to early Celtic law, as in Rome, the authority of the father lasted till his death ; but this is not quite borne out by the facts he quotes.® In ancient Ireland the son was under the father's control till formally emancipated ; but it is not known at what age the emancipation took place.10 The Welsh laws refer to the giving of a daughter in marriage by her kindred as well as by her father. She does not seem to have been 1 v. Haxthausen, Russian Empire, ii. 229 sq. 2 Leroy-Beaulieu, Empire of the Tsars and the Russians, i. 488. 3 Supra, i. 386. 4 Macieiowski, op. cit. ii. 189. 5 BogiSid, quoted by Maine, Dissertations on Early Law and Custom, p. 244 note. 6 Krauss, Sitte und Brauch der Sudslaven, pp. 314, 313. Wesnitsch, ' Die Blutrache bei den Sudslaven,' in Zeitschr. f. vergl. Rechtswiss. ix. 50. 7 Krauss, op. cit. p. 320. Wesnitsch, loc. cit. p. 50. 8 Caesar, Comment ar ii de bello Gallico, vi. 19. 3. 9 d'Arbois de Jubainville, Cours de literature celtique, vii. 244 sqq- 10 Joyce, Social History of Ancient Ireland, ii. 13. VOL. II Z 338 THE HISTORY OF HUMAN MARRIAGE chap. entirely at the disposal of the latter, nor to have been, in theory, entirely free.1 In Europe the paternal authority of the archaic type has gradually yielded to a system under which the father has been divested of the most essential rights he formerly possessed over his children-a system the inmost drift of which is expressed in the words of the French Encyclopedist, " Le pouvoir patemel est plutot un devoir qu'un pouvoir."2 Already in pagan times the Roman patria Jotestas was subject to important restrictions. The life of the child was practically as sacred as that of the parent long before Christianity became the religion of Rome.3 Alexander Severus limited the father's right of punishing his children to simple correction,4 and Diocletian and Maximilian took away the power of selling freeborn children as slaves.5 Under the jurisprudence of Justinian a father could not force his son or daughter in marriage.6 But at the same time the right of a voice in his children's marriages was stoutly maintained : the consent of the head of the family remained essential to the validity of the marriage of anyone under his power, irrespective of age.7 Canon Law adopted the principle that no marriage can be concluded without the consent of the persons who marry ; but, unlike Justinian's law, as a consequence of its doctrine that marriage is a sacrament, it ruled that, however young the bridegroom and bride may be, the consent of their parents or guardians is not necessary to make the marriage 1 Rhys and Brynmor-Jones, The Welsh People, p. 209. 2 Encyclopedic methodique, Jurisprudence, vii. 77, art. Puissance paternelle. 8 Milman, History of Latin Christianity, ii. 25. Sohm, Institutes, p. 483. 4 Codex Justinianus, viii. 46. (47.) 3. 8 Ibid. iv. 43. 1. Constantine, however, made an exception of newly-born infants when the parents were in extreme misery (ibid. iv. 43. 2). 6 Digesta, xxiii. 2. 21. Codex Justinianus, v. 4. 14. Cf. Hunter, Exposition of Roman Laiv, p. 680 ; Winroth, op. cit. p. 51. 7 Karlowa, Rbmische Rechtsgeschichte, ii. 174. Hunter, op. cit. p. 680. Winroth, op. cit. p. 51. XXII CONSENT AS A CONDITION OF MARRIAGE 339 valid.1 The Church disapproved of marriages contracted without such consent: the lack of it was a " prohibitory impediment " (impedimentum impediens) rendering the marriage illicit, but not a " diriment impediment " (impedi- mentum dirimens) rendering it null and void.2 The stipula- tions of Canon Law influenced secular legislation. An edict of Clothaire I. in 560 prohibited the forcing of women to marry against their will.3 According to the Laws of Cnut, no woman or girl could be compelled to marry a man whom she disliked.4 In an Anglo-Saxon betrothal formula from the tenth century the girl's consent is unconditionally re- quired.5 And various early Teutonic law-books in con- tinental countries likewise prohibited the forcing of a woman into marriage against her will.6 As to the canonical prescrip- tion that a marriage is valid without the consent of parents or guardians, it seems that the English temporal law more or less acquiesced in it, although it regarded " wardship and marriage " as a valuable piece of property ;7 and in the later Middle Ages German women were able to marry without parental consent, though at the risk of being disinherited.8 1 Gratian, Decretum, ii. 27. 2. 2 (Migne, Patrologies cursus, clxxxvii. 1392) :-" Sufficiat secundum leges solus eorum consensus, de quorum conjunctionibus agitur." 2 Friedberg, Lehrbuch des katholischen und evangelischen Kirchen- rechts, p. 422. Winroth, op. cit. p. 52. 3 Pardessus, Loi Salique, p. 666. 4 Laws of Cnut, ii. 75. 5 Roeder, op. cit. p. 24 sq. c Nordstrom, Bidrag till den svenska samhalls-fbrfattningens his- toria, ii. 15 sq. Wilda, op. cit. p. 803. Weinhold, Die deutschen Frauen in dem Mittelalter, i. 304. Winroth, op. cit. 55 sq. Ludlow, Consent to Marriage,' in Smith and Cheetham, Dictionary of Christian Antiquities, i. 434 sq. According to Saxo Grammaticus (Historia Danica, book v. vol. i. 186), a woman was allowed to dispose of her own hand before the days of King Frotho. 7 Pollock and Maitland, History of the English Law before the Time of Edward L, ii. 389. Cf., however, Roeder, op. cit. p. 25. 8 Kraut, Die Vormundschaft nach den Grundsdtzen des deutschen Rechts, i. 326. Schroeder, op. cit. p. 733. Sohm, Das Recht der Eheschliessung aus dem deutschen und canonischen Recht geschichtlich entwickelt, p. 51 sq. Friedberg, Das Recht der Eheschliessung in seiner geschichtlichen Entwicklung, p. 104 sq. 340 THE HISTORY OF HUMAN MARRIAGE chap. The 4 Schwabenspiegel,' which is a faithful echo of canonical ideas, says that when a young man has completed his fourteenth year he can take a wife without the consent of his father, that a maiden is marriageable at twelve years, and that her marriage subsists even if contracted in spite of her father or other relatives.1 But the feelings of the people seemed to have been opposed to such a marriage, and required the consent of parents. Ulrich von Lichtenstein says in his 4 Frauenbuch,' 44 A girl who has no parents should follow the advice of her kinsfolk ; if she gives herself to a man of her own accord, she may live with shame."2 Attempts were made to induce the Church to change its law on the subject, but in vain : the matter was definitely settled at the Council of Trent, after a lively discussion.3 Luther and other reformers were of a different opinion : they maintained that a marriage contracted without the consent of parents should be regarded as invalid, unless the consent was given afterwards.4 This principle was gradually accepted by most legislators in Protestant countries, but with the modification that parental consent could be refused for good reasons only and, in case of need, the consent of the authorities could take its place.5 In Roman Catholic countries, also, the canonical doctrine met with opposition ; legislators declared parental consent to be necessary for the validity of a marriage, and no appeal could be made in the case of refusal.6 Henry II. of France decreed, in 1556, that a marriage contracted by a minor without the consent of ascendants was null and void; and the later legislation went further in the same direction. If a marriage was contracted without such consent by a person who was below 1 Der Schwabenspiegel, Landrecht, § 55. 2 Weinhold, op. cit. i. 305. 8 Friedberg, Das Recht der Eheschliessung, p. 122 sq. Winroth, op. cit. p. 52 sq. 4 Colberg, Ueber das Ehehinderniss der Entfuhrung, p. 114 sqq. Friedberg, Das Recht der Eheschliessung, p. 105 sq. Idem, Lehrbuch des katholischen und evangelischen Kirchenrechts, p. 422 sq. • Colberg, op cit. p. 121. Friedberg, Das Recht der Eheschliessung, p. 106. Sohm, Das Recht der Eheschliessung, p. 206 sq. n. 16. 6 Cf. Winroth, op. cit. p. 53. xxn CONSENT AS A CONDITION OF MARRIAGE 341 the age of twenty-five, it was annulled; if contracted by a person between twenty-five and thirty, it was valid, but disinheritance might be the consequence ; and if contracted by a person above the age of thirty, it had still to be previously notified to the ascendants by " three respectful acts."1 Indeed, according to the French ' Code Civil,' a son under twenty-five and a daughter under twenty-one could not, until 1907, marry without parental consent.2 Generally speaking, in France and other Latin countries the Roman notions of paternal rights and filial duties have to some extent survived among the people throughout the Middle Ages and long after. In the literature of the eleventh century, says M. Bernard, the paternal character " is every- where honoured, and filial piety everywhere praised and rewarded. In the romances of chivalry fathers are never ridiculous; nor sons insolent a.nd mocking. . . . Above the majesty of the feudal baron, that of the paternal power was held still more sacred and inviolable. However powerful the son might be, he would not have dared to outrage his father, whose authority was in bis eyes always confounded with the sovereignty of command."3 Bodin wrote, in the latter part of the sixteenth century, that although the monarch commands his subjects, the master his disciples, the captain his soldiers, there is none to whom nature has given any command except the father, " who is the true image of the great sovereign God, universal father of all things."4 Du Vair remarked that we ought to regard our fathers as gods on earth.6 In the Duke of Sully's Memoirs we read that in his days in France children were not per- mitted to sit in the presence of their parents without being commanded to do so.6 Speaking of the women among the nobility and upper classes in France during the eighteenth century, Messrs, de Goncourt remark :-" Generalement le 1 Gu6tat, Histoire 6l6mentaire du droit franfais, p. 364 sq. 2 Code civil, art. 148. 3 Bernard, quoted in Spencer's Descriptive Sociology, p. 38. 4 Bodin, De republica, i. 4, p. 31. 6 Du Vair, quoted by de Ribbe, Les families et la sociM en France avant la Revolution, p. 51. 8 de Sully, Memoirs, v. 180. 342 THE HISTORY OF HUMAN MARRIAGE chap. mariage de la jeune fille se faisait presque immediatement au sortir du couvent, avec un mari accepte et agree par la famille. Car le mariage etait avant tout une affaire de famille, un arrangement au gre des parents, qui decidaient des considerations de position et d'argent, des convenances de rang et de fortune. Le choix 6tait fait d'avance pour la jeune personne, qui n'etait pas consultee."1 According to the present law of France, a son and daughter under the age of twenty-one cannot marry without the consent of the father and mother, or of the father only if they disagree, or of the survivor if one be dead. If both father and mother are dead, or in a condition which renders them unable to consent, the grandparents take their place. Between the ages of twenty-one and thirty the parties must still obtain parental consent, but if this be refused it can be regulated by means of an act before a notary, and if the consent is not given within thirty days the marriage can take place without it.2 In the Netherlands a minor, that is, a person under the age of twenty-three, cannot marry without the consent of parents, or at least of the father ; and here also a formal request is prescribed for individuals under thirty years of age, although, in the case of refusal, the marriage can be contracted after three months.3 In Italy the consent of parents, or of the father, or of the survivor if one of the parents is dead, is required for a son who has not completed his twenty-fifth year and for a daughter who has not completed her twenty-first; but in case of refusal of consent provision is made for an appeal to a court.4 In Austria minors, that is, individuals under the age of twenty-four, are incapable of contracting a valid marriage without the consent of their father.5 In Germany a legitimate child requires before the completion of his twenty-first year of age the approval of the father for 1 de Goncourt, La Femme au dix-huiti&me siicle, p. 20. 2 Code civil, art. 148-151, 154. 3 ' Code civil,' §§ 92, 99 sqq., in Les codes Nderlandais, p. 69 sq. Roguin, Traite de droit civil comparL Le mariage, p. 53. 4 Codice civile del regno d'Italia, §§ 63, 67. 5 Das allgemeine biirgerliche Gesetzbuch fur das Kaisertum Oester- reich, §§ 21, 49. xxn CONSENT AS A CONDITION OF MARRIAGE 343 concluding a marriage, whilst an illegitimate child requires before the same age the approval of the mother ; if the father is dead, the mother takes his place.1 In Sweden parental consent is required for persons under the age of twenty-one,2 in Switzerland for persons under the age of twenty.3 In England, by the common law, the marriages of minors who had attained the age of consent-fixed at fourteen years for males and twelve years for females'-were valid without the consent of parents until the year 1753, when Lord Hardwicke's Marriage Act (26 Geo. 2, c. 33, § 11) declared such marriages void.4 According to the present law of England, " where a person, not being a widower nor widow, is under the age of twenty-one years, the father, if living, or, if he is dead, the guardian or guardians, or one of them, or if there is no guardian lawfully appointed, then the mother, if she has not remarried, has authority to consent to his or her marriage ; and such consent is required except where there is no person having authority to give it."5 Yet the. marriage of a minor without the requisite consent is not invalid, whether it is by banns or licence or superintendent registrar's certificate ; but there may be forfeiture of all the rights and interest in any property accruing to the offending party by force of the marriage.6 In Scotland, on the other hand, no consent of parents or guardian is required even for minors who have attained the age of puberty ;7 and by the common law of the United States, which was not affected by Lord Hardwicke's Mar- riage Act, the marriage of minors without the parental consent is likewise good. There are " statutes which forbid the celebration of the nuptials of minors without permission from the parent or guardian ; but, in the absence of a clause 1 Biirgerliches Gesetzbuch, § 1305. 2 Lag om dktenskaps ingdende och upplbsntng av den 12 nov ember 1915, ii. 2. 3 Schweizerisches Zivilgesetzbuch vom 10. Dezember 1907, art. 14, 98. 4 Blackstone-Kerr, Commentaries on the Laws of England, i. 408 sq. 5 Earl of Halsbury, Laws of England, xvi. 296. 8 Ibid. xvi. 297 sq. ' Erskine-Rankine, Principles of the Law of Scotland, p. 61. 344 THE HISTORY OF HUMAN MARRIAGE chap. of nullity, which most of them do not contain, a marriage in disobedience is valid, while yet the participators in it may be subject to a penalty or punishment."1 In many European countries a marriage entered into without the prescribed parental consent is annulled if the parent whose consent is required demands it, whereas in other countries such a marriage is not invalid but may entail disinheritance.2 Before leaving our present subject, we must still consider the origin of that authority which has given fathers or parents a right to interfere with the marriage of their children even in cases when such interference cannot be regarded merely as a safeguard of the children's own interests. In the first place, the parental authority obviously depends upon the natural superiority of parents over their children when young, and on the helplessness of the latter ; and for similar reasons the daughter, though grown-up, still remains in her father's power. Parents are, moreover, considered to possess in some measure proprietary rights over their offspring, being their originators and maintainers ;3 and in various cases, it seems, the father is also regarded as their owner because he is the owner of their mother. Filial duties and parental rights to some extent spring from the children's natural feeling of affection for their parents, particularly for their mother,4 and from the debt of gratitude which they are considered to owe to those who have brought them into existence and taken care of them while young.6 The authority of parents is much enhanced and extended by the sentiment of filial reverence, as distinct from mere affection. From their infancy children are used to look up to their 1 Bishop, New Commentaries on Marriage, Divorce, and Separation, i 239 sq. a Winroth, Aktenskapshindren, p. 89 sq. Roguin, op. cit. p. 54 sq. 8 Cf. Vdsishtha, xv. 1 sq. ; BaudhAyana Parisishta, vii. 5. 2 sq. * See Westermarck, Origin and Development of the Moral Ideas, i. 618 n. 6 sq. It is said in the Talmud that the child loves its mother more than its father, whilst it fears its father more than its mother (Deutsch, Literary Remains, p. 55). 8 HsiAo King, 9 (Sacred- Books of the East, iii. 479). Laws of Manu, ii. 227. Plato, Leges, iv. 717. xxn CONSENT AS A CONDITION OF MARRIAGE 345 parents, especially the father, as to beings superior to them- selves ; and this feeling, which by itself has a tendency to persist, is all the more likely to last even when the parents grow old, as it is based not only on superior strength and bodily skill, but on superior knowledge and wisdom, which remain though the physical power be on the wane. " In the Russian people," says Leroy-B.eaulieu, " paternal power is supported by religious feeling and reverence for age. . . . ' Where white hairs are, there is good sense, there is right '■- such is, with variations, the burden of many popular proverbs."1 " Long life and wisdom," say the Iroquois, " are always connected together " ;2 and throughout West Africa the aged are " the knowing ones."8 Among peoples who possess no literature the old men are the sole authorities on religion as well as on custom. In Australia the great deference shown to them is partly due to the superstitious awe of certain mysterious rites which are known to them alone, and to the knowledge of which young persons are only very gradually admitted? Dr. Rivers takes the magical power of the old men to be the original source of their dominance in Melanesia.5 Among the East African Embe "it is only by means of the rankest superstition that the old men are able to maintain their supremacy over the hot-blooded youths " ; they convince the warriors, by presenting them with some magic emblem, that in the hands of the sages alone rest the fate and fortune of those who fight in a battle. And old women, also, are often believed to possess supernatural power, in which case their influence, in spite of the subservient position of their sex in general, is almost as great as that of a medicine-man.6 Old age itself inspires a feeling of mysterious awe. The 1 Leroy-Beaulieu, op. cit. i. 489. 2 Loskiel, History of the Mission of the United Brethren among the Indians in North America, i. 15. 8 Miss Kingsley, West African Studies, p. 142. 4 Schuermann, ' Aboriginal Tribes of Port Lincoln,' in Woods, Native Tribes of South Australia, p. 226. Cf. Nelson, ' Eskimo about Bering Strait,' in Ann. Rep. Bur. Ethn. xviii. 304. 6 Rivers, History of Melanesian Society, ii. 108. 6 Chanler, Through Jungle and Desert, pp. 247, 252. 346 THE HISTORY OF HUMAN MARRIAGE chap. Moors say that, when becoming old, a man becomes a saint, and a woman a jinnta, or evil spirit-there is something supernatural in both. The beliefs held regarding the dead also influence the treatment of the aged whose lives are drawing to an end. Certain Central African tribes are reported to treat their old people with every kindness in order to secure their good will after death.1 A missionary in East Africa heard a negro say with reference to an old man, " We will do what he says, because he is soon going to die." 2 The Niase is an egqist even in his respect for the old, because he hopes that they will protect and assist him when they are dead.3 In China the doctrine that ghosts may interfere at any moment with human business and fate, either favourably or unfavourably, " enforces respect for human life and a charitable treatment of the infirm, the aged, and the sick, especially if they stand on the brink of the grave." 4 The regard for the aged and the worship of the dead are often mentioned together in a way which suggests that there exists an intrinsic connection between them.5 In such cases, however, it is impossible accurately to distinguish between cause and effect. Whilst the worship of the dead is, in the first place, due to the mystery of death, it is evident that the regard in which a person is held during his lifetime also influences the venera- tion which is bestowed on his disembodied soul. Among the peoples of archaic culture, in particular, there is a close connection between filial submissiveness and religious beliefs. In China and Japan the reverence for parents almost forms a part of the worship of ancestors. As to the Israelites, Philo Judaeus remarks that the com- mandment enjoining obedience to parents occupies its position immediately after those prescribing the duties of man towards God because parents are something between divine and human nature, partaking of both-of human nature inasmuch as it is plain that they have been born 1 Arnot, Garenganze, p. 78 note. 2 Lippert, Kulturgeschichte dev Menschheit, i. 229. 3 Modigliani, Viaggio a Nias, p. 467. 4 de Groot, op. cit. (vol. iv. book) ii. 450. 5 See Westermarck, op. cit. i. 62 t. xxn CONSENT AS A CONDITION OF MARRIAGE 347 and that they will die, and of divine nature because they have engendered other beings, and have brought into existence what did not exist before. What God is to the world, that parents are to their children ; they are " the visible gods."1 The religious character of filial duties is very conspicuous both in Muhammadanism and Hinduism. Disobedience to parents is considered by Moslems as one of the greatest of sins, and is put, in point of heinousness, on a par with idolatry, murder, and desertion in an expedition against infidels.2 According to ancient Hindu ideas, a father, mother, and spiritual teacher are equal to the three Vedas, equal to the three gods, Brahman, Vishnu, and Siva.3 A man who shows no regard for them derives no benefit from any religious observance ; whereas, " by honouring his mother, he gains the present world ; by honouring his father, the world of gods ; and by paying strict obedience to his spiritual teacher, the world of Brahman." 4 In the Greek writings there are numerous passages which put filial piety on a par with the duties towards the gods.5 To the ancient Romans the parents were hardly less sacred beings than the gods.8 In Russia the father, like the Tsar, " was thought to hold from Heaven a sort of right divine, to rebel against which would have been sacrilege." 7 Accord- ing to a Slavonic maxim, " a father is like an earthly god to his son." 8 Among the ancient nations of culture the father was invested with sacerdotal functions. In primitive antiquity, says Fustel de Coulanges, " the father is not only the strong man, the protector who has power to command obedience ; he is the priest, he is heir to the hearth, the continuator of the 1 Philo Judaeus, Opera, i. 759 sqq. 2 Lane, Manners and Customs of the Modern Egyptians, p. 70. 3 Institutes of Vishnu, xxxi. 7. Laws of Manu, ii. 230. 4 Institutes of Vishnu, xxxi. 9 sq. Cf. Laws of Manu, ii. 233 sq. 5 Schmidt, Die Ethik der alten Griechen, ii. 141 sq. Infra, ii. 349 sq. 8 Valerius Maximus, i. 1. 13 : " Pari vindicta parentum ac deorum violatio expianda est." Servius, In Virgilii Georgicon, ii. 473 : " Sacra deorum sancta apud illos sunt, sancti etiam parentes." 7 Leroy-Beaulieu, op. cit. i. 488. 8 Maine, Dissertations on Early Law and Custom, p. 243. 348 THE HISTORY OF HUMAN MARRIAGE chap. ancestors, the parent stock of the descendants, the deposi- tary of the mysterious rites of worship, and of the sacred formulas of prayer. The whole religion resides in him."1 Another very important reason for the connection between filial submissiveness and religious beliefs was no doubt the extreme importance attached to the curses and blessings of parents. The Israelites believed that parents, and especially a father, could by their blessings or curses deter- mine the fate of their children ; 2 indeed, we have reason to assume that the reward which in the fifth commandment is held out to respectful children was originally a result of parental blessings. We still meet with the ancient idea in Ecclesiasticus, or 'The Book of Sirach,' where it is said :- " Honour thy father in word and deed, that a blessing may come upon thee from him. For the blessing of the father establisheth the houses of children ; but the curse of the mother rooteth out the foundations."8 The Moors have a proverb that " if the saints curse you the parents will cure you, but if the parents curse you the saints will not cure you " ; in other words, the curse of a parent is even stronger than that of a saint. The notion that the parents' blessings beget prosperity and that their curses bring ruin prevailed in ancient Greece. Plato says in his ' Laws ' :-Neither God, nor a man who has understanding, will ever advise any one to neglect his 1 Fustel de Coulanges, La cite antique, p. 106. Idem, Ancient City, trans, by Small, p. 116. 2 Genesis, ix. 25 sqq. ; xxvii. 4, 19, 23, 25, 27 sqq. ; xlviii. 9, 14 sqq.', xlix. 4, 7 sqq. Judges, xvii. 2. Cf. Cheyne, ' Blessings and Curses,' in Encyclopedia Biblica, i. 592; Nowack, ' Blessing and Cursing,' in Jewish Encyclopedia, iii. 244. 3 Ecclesiasticus, iii. 8 sq. Cf. ibid. iii. 16. It seems to have been a notion of the ancient Egyptians that a son who accepts the word of his father will attain old age on that account {Precepts of Ptah- Hotep, 39). In the Dhammapada (109) it is said that to him who always greets and constantly reveres the aged four things will increase, namely, life, beauty, happiness, and power. We have previously noticed the old Aztec idea that irreverence to parents will be followed by an unhappy end {supra, ii. 326). It is reasonable to suppose that in these cases, also, the consequences of reverence or irreverence were originally regarded as the result of blessings or curses. Cf. Westermarck, op. cit. i. 624 sq, xxii CONSENT AS A CONDITION OF MARRIAGE 349 parents. ... If a man has a father or mother, or their fathers or mothers treasured up in his house strickerf in years, let him consider that no statue can be more potent to grant his requests than they are, who are sitting at his hearth, if only he knows how to show true service to them. . . . Oedipus, as tradition says, when dishonoured by his sons, invoked on them curses which every one declares to have been heard and ratified by the gods, and Amyntor in his wrath invoked curses on his son Phoenix, and Theseus upon Hippolytus, and innumerable others have also called down wrath upon their children, whence it is clear that the gods listen to the imprecations of parents ; for the curses of parents are, as they ought to. be, mighty against their children as no others are. And shall we suppose that the prayers of a father or mother who is specially dishonoured by his or her children, are heard by the gods in accordance with nature ; and that if a parent is honoured by them, and in the gladness of his heart earnestly entreats the gods in his prayers to do them good, he is not equally heard, and that they do not minister to his request ? . . . Therefore, if a man makes a right use of his father and grandfather and other aged relations, he will have images which above all others will win him the favour of the gods."1 We may assume that originally the efficacy of parents' curses and blessings were ascribed to a magic power immanent in the spoken word itself, and that their Erinyes, like those of guests and suppliants2 and beggars,3 were only personi- fications of curses pronounced in case of ill treatment or neglect.4 But in this, as in other similar instances,5 the 1 Plato, Leges, xi. 930 sq. Cf. ibid. iv. 717. 2 Aeschylus, Eumenides, 545 sqq. Plato, Epistolae, viii. 357. Apollonius Rhodius, Argonautica, iv. 1042 sq. 3 Odyssea, xvii. 475. . * See Ilias, xxi. 412 sq. ; Sophocles, Oedipus Coloneus, 1299, 1434 ; v. Lasaulx, Der Fluch bei Griechen und Rbmern, p. 8 ; Muller, Dissertations on the Eumenides, p. 155 sqq ; Rohde, ' Paralipomena, ' in Rheinisches Museum fur Philologie, 1895, pp. 7, 16 sq. Aeschylus (Eumenides/ 416 sq.) expressly designates the Erinyes by the title of " curses " (dpai), and Pausanias (Descriptio Graeciae, viii. 25. 6) derives the name Erinys from an Arcadian word signifying a fit of anger. s See Westermarck, op. cit. i. 563-565, 585 ; ii. 67 sq. 350 THE HISTORY OF HUMAN MARRIAGE chap. fulfilment of the' curse or the blessing came afterwards to be looked upon as an act of divine justice. According to Plato, " Nemesis, the messenger of justice," watches over unbecoming words uttered to a parent; 1 and Hesiod says that if anybody reproaches an aged father or mother, " Zeus himself is wroth, and at last, in requital for wrong deeds, lays on him a bitter penalty."2 It also seems to be beyond all doubt that the divi parentum of the Romans, like their dii hosfiitales,3 were nothing but personified curses. For it is said, " If a son beat his parent and he cry out, the son shall be devoted to the parental gods for destruction."4 In aristocratic families in Russia children used to stand in mortal fear of their father's curses ;5 and the country people still believe that a marriage without the parents' approval will call down the wrath of Heaven on the heads of the young couple.6 Some of the Southern Slavs maintain that if a son does not fulfil the last will of his father, the soul of the father will curse him from the grave.7 The Serbians say, " Without reverence for old men, there is no salvation."8 Various uncivilised peoples, also, ascribe great efficacy to the curses or blessings of parents. Among the natives of the Lower Congo children are terribly afraid of their father's curses.9 Among the Mpongwe of Western Africa " there is nothing which a young person so much deprecates as the curse of an aged person, and especially that of a revered father."10 Among the Nandi, " if a son refuses to obey his father in any serious matter, the father solemnly strikes the son with his fur mantle. This is equivalent to a most serious curse, and is supposed to be fatal to the son 1 Plato, Leges, iv. 717. 2 Hesiod, Opera et dies, 331 sqq. (329 sqqj, 3 Westermarck, op. cit. i. 580, 585. 4 Servius Tullius, in Bruns, Fontes Juris Romani antiqui, p. 14. and Festus, De verborum significations, ver. Plorare : " Si parentem puer verberit, ast olle plorassit, puer divis parentum sacer esto.'' Cf. Leist, Alt-arisches Jus Civile, i. 184. x 5 I am indebted to Prince Kropotkin for this statement. 8 Kovalewsky, Modern Customs and Ancient Laws of Russia, p. 37 7 Krauss, op. cit. p. 119. 8 Maine, Dissertations on Early Law and Custom, p. 243. 8 Weeks, Among the Primitive Bakongo, p. 155. 10 Wilson, Western Africa, p. 393. xxii CONSENT AS A CONDITION OF MARRIAGE 351 unless he obtains forgiveness, which he can only do zby sacrificing a goat before his father."1 The Barea and Kunama are convinced that any undertaking which has not the blessing of the old people will fail, that every curse uttered by them must be destructive.2 Among the Bogos nobody takes an employment or gives it up, nobody engages in a business or contracts a marriage, before he has received the blessing of his father or his master.3 Among the Herero, " when a chief feels his dissolution approaching, he calls his sons to the bedside, and gives them his benediction."4 Why are the blessings and curses of parents supposed to possess such an extraordinary power ? One reason is no doubt the mystery of old age and the nearness of death. Not parents only, but to some extent old people generally, are held capable of giving due effect to their good and evil wishes, and this capacity is believed to increase when life is drawing to its close. The Herero, according to Biittner, know really no blessing save that conferred by the father on his death-bed.5 According to Teutonic ideas, the curse of a dying person was the strongest of all curses.6 A similar notion prevailed among the ancient Arabs;7 and among the Israelites the father's mystic privilege of determining the weal or woe of his children was particularly obvious when his days were manifestly numbered.8 But, at the same time, parental benedictions and imprecations possess a potency of their own owing to the parents' superior position in the family and the respect in which they are naturally held. The influence which such a superiority has upon 1 Johnston, Uganda Protectorate, p. 879. Among the Masai, again, a maternal uncle " exercises great influence over his nephews, as it is believed that if he were to curse them they would die " (Hollis, ' Note on the Masai System of Relationship,' in Jour. Roy. Anthr. Inst. xl. 478). 2 Munzinger, Ostafrikanische Studien, p. 475. 3 Idem, Ueber die Sitten und das Recht der Bogos, p. 90 sq. 4 Andersson, Lake Ngami, p. 228. 5 Biittner,' Sozialpolitisches aus dem Leben der Herero in Damara- land,' in Ausland, Iv. 852. 6 Grimm, Teutonic Mythology, iv. 1690. 7 Wellhausen, Reste arabischen Heidentums, pp. 139, 191. 8 Cheyne, in Encyclopedia Biblica, i. 592. 352 THE HISTORY OF HUMAN MARRIAGE chap. the efficacy of curses is well brought out by certain facts. According to the Greek notion, the Erinyes avenged wrongs done by younger members of a family to elder ones, even brothers and sisters, but not vice versd.1 The Arabs in Morocco say that the curse of a husband is as potent as that of a father. The Tonga Islanders believe that curses have no effect " if the party who curses is considerably lower in rank than the party cursed/'2 Moreover, where the father was the priest of the family, his blessings and curses would for that reason also be efficacious in an exceptional degree.3 However, the facts which we have hitherto considered are hardly sufficient to account for the extraordinary development of the paternal authority in the archaic State. Great though it be, the influence which magical and religious beliefs exercise upon the paternal authority is largely of a reactive character. A father's blessings would not be so eagerly sought for, nor would his curses be so greatly feared, if he were a less important personage in the family. So, too, as Sir Henry Maine aptly remarks, the father's power is older than the practice of worshipping him. " Why should the dead father be worshipped more than any other member of the household unless he was the most prominent- it may be said, the most awful-figure in it during his life ? "4 We must assume that there exists some connection between the organisation of the family and the political constitution of the society. At the lower stages of civilisation-though hardly at the very lowest-we frequently find that the clan has attained such an overwhelming importance that only a very limited amount of authority could be claimed by the head of each separate family. But, as I have pointed out elsewhere,5 this was changed when clans and tribes were united into a State. The new State tended to weaken and destroy the clan system, whereas at the same time the family 1 Ilias, xv. 204 : " Thou knowest how the Erinyes do always follow to aid the elder-born." Cf. Muller, Dissertations on the Eumenides, p. 155 sq. 2 Mariner, Account of the Natives of the Tonga Islands, ii. 238. 3 Cf. Nowack, in Jewish Encyclopedia, iii. 234 sq. 4 Maine, Dissertations on Early Law and Custom, p. 76. 5 Origin and Development of the Moral Ideas, ii. 222 sq. xxn CONSENT AS A CONDITION OF MARRIAGE 353 tie grew in strength. In early society there seems to be an antagonism between the family and the clan. Where the clan bond is very strong it encroaches upon the family feeling, and where it is loosened the family gains. Hence Dr. Grosse may be right in his assumption that the father became a patriarch, in the true sense of the word, only as the inheritor of the authority which formerly belonged to the clan.1 But whilst in its earlier days the State strengthened the family by weakening the clan, its later development had a different tendency. When national life grew more intense, when members of separate families drew nearer to one another in pursuit of a common goal, when along with industrial progress the younger members of a family became economically more independent of their parents-the family again lost in importance. Other factors also, inherent in progressive civilisation, contributed to the downfall of the paternal power-the extinction of ancestor-worship, the decay of certain superstitious beliefs, the declining influence of religion, and, last but not least, the spread of a keener mutual sympathy throughout the State, which could not tolerate that the liberty of children should be sacrificed to the despotic rule of their fathers. 1 Grosse, Die Formen der Familie, p. 219. VOL II CHAPTER XXIII MARRIAGE BY CONSIDERATION AND BY THE EXCHANGE OF PRESENTS. GIFTS TO THE BRIDE AND TO THE BRIDEGROOM Among the lower races consent to the contraction of a marriage is not generally given for nothing. In most cases some consideration has to be offered to the father or other relatives of the bride, either in the form of the exchange of bride for bride, or of service, or of the giving of property of some kind or other. "The Australian male,'' says Mr. Curr, " almost invariably obtains his wife or wives, either as the survivor of a married brother, or in exchange for his sisters, or later on in life for his daughters.''1 Dr. Howitt makes a similar statement:- " It may be safely laid down as a broad and general proposi- tion that among these savages a wife was obtained by the exchange of a female relative, with the alternative possibility of obtaining one by inheritance (Levirate), by elopement, or by capture. ... It seems to me that the most common practice is the exchange of girls by their respective parents as wives for each other's sons, or in some tribes the exchange of sisters, or of some female relatives by the young men themselves. It must be always borne in mind that in such cases it is not merely the own sisters, but also the tribal sisters, who are thus available.''2 The prevalence of the exchange of bride for bride among the Australian aborigines has been attested by many other writers with regard to 1 Curr, The Australian Race, i. 107. See also Idem, Recollections of Squatting in Victoria, pp. 128, 248. * Howitt, ' On the Organisation of Australian Tribes,' in Trans. Roy. Soc. Victoria, vol. i. pt. ii. 115 sq. ch. xxm MARRIAGE BY CONSIDERATION, ETC. 355 different tribes,1 although the custom does not seem to occur in all parts of Australia.2 Mr. Taplin says that among the Narrinyeri in South Australia "it is considered disgraceful for a woman to take a husband who has given no other woman for her," and that the right to give a woman away is often purchased from her nearest male relative by those who have no sisters.3 In the western islands of Torres Straits the exchange of " sisters," in the classificatory sense of the word, was the usual method of obtaining a wife.4 So also the practice of exchanging women as wives is the rule among the Kiwai Papuans,5 and seems to be widespread in New Guinea.6 In 1 Idem, Native Tribes of South-East Australia, pp. 177, 178, 211, 217, 222, 237, 242-244, 249, 252, 253, 260, 262 sq. Fison and Howitt, Kamilaroi and Kumai, pp. 276, 285, 343. Jung, ' Aus dem Seelenleben der Australier,' in Mittheil. d. Vereins f. Erdkunde zu Leipzig, 1877, p. 31 (Narrinyeri). Holden, in Taplin, Folklore, Manners, Customs, and Languages of the South Australian Aborigines, p. 17 (Maroura tribe of the Lower Darling). Newland, ' Parkengees, or Aboriginal Tribes on the Darling River,' in Proceed. Roy. Geograph. Soc. Australasia : South Australian Branch, ii. 21. Beveridge, Aborigines of Victoria and Riverina, p. 24 n. * Brough Smyth, Aborigines of Victoria, i. 79, 84. Oberlander, ' Die Eingeborenen der australischen Kolonie Victoria,' in Globus, iv. 279. Mathew, Eaglehawk and Crow, p. 113. Lang, Aborigines of Australia, p. 10 sq. (natives of Moreton Bay). Petrie, Reminiscences of Early Queensland, p. 59. Roth, Ethnological Studies among the North-West-Central Queensland Aborigines, p. 181. Lumholtz, Among Cannibals, p. 164. Brown, ' Three Tribes of Western Australia,' in Jour. Roy. Anthr. Inst, xliii. 156. Malinowski, The Family among the Australian Aborigines, p. 36 sqq. Frazer, Folk-Lore in the Old Testament, ii. 195 sqq. 1 See Malinowski, op. cit. p. 50 sq. 3 Taplin, op. cit. p. 35. Cf. Jung, loc. cit. p. 31. 4 Haddon, in Reports of the Cambridge A nthropological Expedition to Torres Straits, v. 231. See also Rivers, ibid. v. 241. 5 Hely, quoted ibid. v. 189. Landtman, Nya Guinea far den, p. 84. 6 Beardmore, ' Natives of Mowat, Daudai, New Guinea,' in Jour. Anthr. Inst. xix. 460 sq. Vetter, ' Papuanische Rechtsver- haltnisse bei den Jabim,' in Nachrichten uber Kaiser Wilhelms-Land und den Bismarck-Archipel, 1897, p. 90. Thurnwald, ' B&naro Society,' in Memoirs of the American Anthropological Association, iii. 258 sqq. van Hasselt, ' De Huwelijksregeling voor de Papoesche Christenen, op Noord-Nieuw-Guinea,' in Mededeelingen van ivege het Nederlandsch Zendelinggenootschap, Iviii. 215. A A 2 356 THE HISTORY OF HUMAN MARRIAGE chap. Erromanga of the New Hebrides " girls were often exchanged, though both parties paid as well; and when a girl was given by one tribe, the people who got her were expected to give in return to her village their first marriageable girl."1 In the Buin district of Bougainville, in the Solomon Group, the exchange of women is considered the orthodox form of marriage; but here also payment is needed in addition, although in case the brother of the bride takes the sister of the bridegroom " the payment of an equal amount of money and wares is carefully executed, so that the price for the brides is evenly exchanged. ''2 Among the Sumatrans, according to Marsden, in lieu of paying the jujur, or bride price, a virgin is sometimes exchanged for another. A man who has a son and daughter gives the latter in exchange for a wife to the former, and the person who receives her disposes of her as his own child or marries her himself. A brother will give his sister in exchange for a wife, or, in default of such, will procure a cdusin for the purpose ; and it is not unusual to borrow a girl from a friend or relation in order to exchange her for a wife, the borrower binding himself to replace her, or to pay her jujur when required.3 Among the Buryat the paying of the ordinary bride price, which is very high, is avoided by the custom called adlayi, according to which the parents in two families who have sons and daughters exchange brides for their sons ;4 and a similar custom exists among some of the Tungus, although among them the father still receives a small kalym, or bride price, from the groom.6 Among the wild tribes of the Afghan frontier the chief exception from purchase is marriage by exchange ; if in each of two families there is an unmarried son and an unmarried daughter, 1 Robertson, Erromanga, p. 396. 2 Thurnwald, in Memoirs of the American Anthropological Associa- tion, iii. 274, 285 sq. Idem, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 18. 8 Marsden, History of Sumatra, p. 259. 4 Melnikow, ' Die BurjSten (Burjaten) der Irkutskischen Gou- vernements,' in Verhandl. Berliner Gesellsch. Anthrop. 189g, p. 441. 6 Samokvasoff, quoted by Miss Czaplicka, Aboriginal Siberia, p. 105. XXIII MARRIAGE BY CONSIDERATION, ETC. 357 they frequently arrange a mutual double marriage without any payments.1 The practice of exchanging daughters in marriage is much in vogue among the tribes of Baluchistan 2 and in Jammu Province of Kashmir, especially in the higher hills, where a man having no daughter to give in exchange finds it very difficult to marry his son.3 Among the Bhotias of Almora and British Garhwal marriage may likewise be by exchange, or adala badala, as it is called, a man giving his own daughter and taking for his son or his brother the other man's daughter.4 In the North-Western Provinces of India this custom is in a great measure confined to the lower castes;5 and Mr. Ibbetson remarks that in the east of the Punjab " exchange of betrothal is thought disgraceful, and, if desired, is effected by a triangular exchange,-A betrothing with B, B with C, and C with A." But he adds that " in the West, on the contrary, among all classes, in the Hills and Submontane Districts, apparently among all but the highest classes, and among the J&ts, almost every- where, except in the Jumna District, the betrothal by exchange is the commonest form."6 Among the Madigas of Mysore exchange o£ daughters in marriage " is most commonly in use, the reason being the saving of the bride price by both parties " ; 7 whilst among the Idigas, another Dravidian caste of the same province, " when two families exchange daughters, the tera or bride price is not, as a rule, paid by either party."8 Among the Santals, a primitive tribe of Bengal, a man who has a son and a daughter of marriageable age, and who is not in a position to pay the 1 Pennell, Among the Wild Tribes of the Afghan Frontier, p. 193. 1 Bray, Census of India, 1911, vol. iv.. (Baluchistan) Report, p. 101. 3 Matin-uz-Zaman Khan, Census of India, 1911, vol. xx. (Kashmir) Report, p. 140. * Sherring, ' Notes on the Bhotias of Almora and British Garhwal,' in Memoirs Asiatic Soc. Bengal, i. 98. 3 Crooke, Tribes and Castes of the North-Western Provinces and Oudh, i. p. cciv. • Ibbetson, quoted ibid. i. cciv. 7 Nanjundayya, Ethnographical Survey of Mysore, xvii. 11, quoted by Frazer, op. cit. ii. 210. 8 Ibid, xviii. 6, quoted by Frazer, op. cit. ii. 212. 358 THE HISTORY OF HUMAN MARRIAGE chap. price for a wife for his son, commissions a go-between to look out for a family in a like position, so that they may exchange daughters for wives to their Sons; but in such cases the sister must be younger than her brother.1 In ancient Arabia two fathers or other guardians might exchange their daughters or the girls under their guardian- ship instead of paying the mahr, or bride price.2 Among the peasantry of Palestine, also, the exchange of sisters as wives is not infrequently resorted to by poor people in order to escape the paying of a dowry.3 Among the Senoufo of the French Sudan the price of purchase is often replaced by a woman, the bride's brother receiving, instead of a dowry, a wife who is generally the sister of the bridegroom.4 So also the Mossi of the French Sudan practise the exchange of daughters, although the usual way of obtaining a bride among them is to give presents to her parents.5 In Southern Guinea, according to Leighton Wilson, " intermarriages among the sea-coast tribes are effected by interchanging sisters or daughters, and not by purchase. The party who make the first overture must take some kind of offering to the parents of the girl, but this is not regarded in the light of purchase money by either party." 6 In a previous chapter we have noticed the practice of married men exchanging their wives either for a time or for ever.7 From these statements it appears that the practice of exchanging bride for bride often occurs side by side with marriage by ordinary purchase as an economic measure intended to save the bride price, whilst in other cases it is the regular method of concluding a marriage. The latter is particularly the case among many of the Australian aborigines. Sir James G. Frazer observes that on account 1 Campbell, ' Santal Marriage Customs,' in Jour. Bihar and Orissa Research Society, ii. 306, 331, quoted by Frazer, op. cit. ii. 218. 2 Wellhausen, ' Die Ehe bei den Arabern,' in Nachrichten von der Konigl. Gesellschaft der Wissenschaften zu Gottingen, 1893, p. 433. 3 Wilson, Peasant Life in the Holy Land, p. no. 4 Delafosse, Haut-Sendgal-Niger. Le Pays, les Peuples, les Langues, I'Histoire, les Civilisations, iii. 70 n. 1. 5 Tauxier, Le Noir du Soudan, p. 544 sq. 8 Wilson, Western Africa, p. 266 sq. 7 Supra, i. 230 sqq. XXIH MARRIAGE BY CONSIDERATION, ETC. 359 of their extreme poverty a wife is among them a man's most valuable possession, and that, having no equivalent in property to give for a wife, an Australian aboriginal native is generally obliged to get her in exchange for a female relative.1 But there are many peoples who are equally poor or poorer and yet have never been known to give daughters or sisters in exchange for wives. It seems that an important reason for the remarkable prevalence of this practice in Australia is the unusual difficulty in getting a wife, which must to some extent be due to the rigorous class and clan rules greatly restricting the circle within which a man is allowed to marry.2 This difficulty does not exist for a man who has a sister to be given in exchange for a wife, because he can thereby get a woman from a class and clan into which he is permitted to marry. If a Kumbo man can marry a Muri woman, the brother of the latter, who is a Muri man, can also marry the sister of the former, who is a Kumbo woman. With special reference to the exchange system in the Solomon Group, Dr. Thumwald expresses the opinion that " mutual exchange of women probably originated as a pledge of good will in the establish- ment of friendly relations between two communities " ; and he believes that this form of marriage was the original one from which the buying of women with objects of value has been derived. " In cases where mutual exchange became impossible, return was made by objects of value." 3 Speaking of the same practice in Baluchistan, Mr. Bray observes that " such a marriage system is in keeping with the whole spirit of a country where most affairs are con- ducted on a brotherly basis of mutual co-operation."4 In this connection may be mentioned the custom of giving a woman in marriage, not in exchange for a bride, but, in the case of an act of homicide, as compensation for the inflicted injury or as a means of reconciling the family of 1 Frazer, op. cit. ii. 195. 2 See infra, on Group-marriage and other Group-relations. 3 Thumwald, in Memoirs of the American Anthropological Associa- tion, iii. 258. See also Idem, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 18. * Bray, op. cit. p. 101. 360 THE HISTORY OF HUMAN MARRIAGE chap. the manslayer with that of his victim. This custom is found among the Beni-Amer 1 and Bogos 2 of North-Eastern Africa, some Ewhe-speaking tribes on the Slave Coast,3 and the Afghans.4 Among the Arabic-speaking moun- taineers of Northern Morocco, also, a homicide sometimes induces the avenger to abstain from his persecution by giving him his sister or daughter in marriage. Among the Avesta people in ancient Iran women and maidens were in the most serious cases offered as blood price, and Geiger believes that they were married to their new possessors.5 More widespread than marriage by exchange is the custom of obtaining a wife by services rendered to her father. This practice is found among many North and South American Indians and some Eskimo6; among various 1 Munzinger, Ostafrikanische Studien, p. 322. 2 Idem, Die Sitten und das Recht der Bogos, p. 84 sq. 3 Spieth, Die Ewe-Stamme, pp. 194, 742. ♦ Kohler, Nachwort zu Shakespeare vor dem Forum der Juris- prudenz, p. 15 sq. & Geiger, Civilization of the Eastern Iranians in A ncient T imes, ii. 34. 6 Bridges, * Manners and Customs of the Firelanders,' in A Voice for South America, xiii. 201 ; King and Fitzroy, Narrative of the Voyages of the Adventure and Beagle, ii. 182 ; Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 132 (Yahgans of Tierra del Fuego). Nordenskiold, Indianliv i- El Gran Chaco (Sydamerika), p. 195 (Chiriguanos, Chan6). v. Martius, Beitrdge zur Ethnographic Amerika's zumal Brasiliens, i. 107, 108, 645. Krause, In den Wildnissen Brasiliens, p. 325 (KarayA). von den Steinen, Enter den Natur- vdlkern Zentral-Brasiliens, p. 332 (Bakairi). Simson, Travels in the Wilds of Ecuador, p. 104 (Canelos Indians). Juan and Ulloa, ' Voyage to South America,' in Pinkerton, Collection of Voyages and Travels, xiv. 521 (Quito Indians). Gumilla, El Orinoco ilustrado, ii. 325 sq. Brett, Indian Tribes of Guiana, p. 101 (Arawaks). Im Thum, Among the Indians of Guiana, p. 221 sq. Schmidt, * Ueber das Recht der tropischen Naturvblker Sudamerikas,' in Zeitschr. f. vergl. Rechtswiss. xiii. 306 sqq. de Herrera, General History of the Vast Continent and Islands of America, commonly called the West Indies, iv. 172 (Indians of Yucatan). Fewkes, ' Aborigines of Porto Rico and Neighbouring Islands,' in Ann. Rep. Bur. Ethn. xxv. 48. Perrot, ' Memoir on the Manners, Customs, and Religion of the Savages of North America,' in Emma Helen Blair's Indian Tribes of the Upper Mississippi Valley and Region of the Great Lakes, i. 69 (some Algonkin). Marston, * Letter to Rev. Dr. Jedidiah XXIII . MARRIAGE BY CONSIDERATION, ETC. 361 Siberian peoples1 and the Ainu of Japan2; in a large number of aboriginal tribes in China,3 Indo-China,4 and India5; Morse,' ibid. ii. 166 sq. (Sauk and Foxes, Chippewa, Ottawa, Pota- watomi). Forsyth, * Account of the Manners and Customs of the Sauk and Fox Nations of Indians,' ibid. ii. 214 (Sauk and Foxes). Carver, Travels through the Interior Parts of North America, p. 373 (Naudowessies). Goddard, Life and Culture of the Hupa, p. 56. Dixon, ' Notes on the Achomawi and Atsugewi Indians of Northern California,' in American Anthropologist, N.S. x. 217. Swanton, Haida, p. 50. Richardson, Arctic Searching Expedition, i. 406 sq. (Kenai). Dall, Alaska, p. 402 (Aleut). Lisiansky, Voyage round the World, p. 198 (Kaniagmiut). 1 Steller, Beschreibung von dem Lande Kamtschatka, p. 343 ; Krasheninnikofi, History of Kamtschatka, p. 212 ; Dobell, Travels in Kamtschatka, i. 82. Bogoras, Chukchee, pp. 579, 586, 587, 609- Jochelson, Yukaghir, p. 87 sqq. Idem, Koryak, p. 740 ; Dittmar ' Uber die Koraken und die ihnen sehr nahe verwandten Tschuk- tschen,' in Melanges russes tiris du bulletin historico-philologique de VAcademic Impdriale des Sciences de St-Ptersbourg, iii. 24 ; Kennan, Tent Life in Siberia, p. 135 (Koryak). Dall, op. cit. p. 519 (Tungus). Georgi, Beschreibung alter Nationen des Russischen Reichs, p. 195 (Barabinze). Hamaliinen, Mordvalaisten, tseremissien ja votjakkien kosinta- ja hdatavoista, p. 194 (Siberian Tartars). See also ibid. p. 189 (Laplanders). 2 Batchelor, Ainu and their Folk-Lore, p. 230. Dall, op. cit. p. 524. Miss Czaplicka, Aboriginal Siberia, p. 102. 3 Gray, China, ii. 304. ♦ Scott and Hardiman, Gazetteer of Upper Burma and the Shan States, i. 522 (some forest tribes on the borders of Burma). Mason, ' Physical Character of the Karens,' in Jour. Asiatic Soc. Bengal, vol. xxxv. pt. ii. 18. Aymonier, quoted by Hartland, Primitive Paternity, ii. 47 sq. (Cambodians). 8 Hodson, Naga Tribes of Manipur, p. 91. Gait, Census of India, '1891, Assam, Report, pp. 251, 284 (Kuki-Lushais, Turungs). Dalton, Descriptive Ethnology of Bengal, pp. 41, 47, 104, no (Nagas of Upper Assam, Kukis, Limbus and Kirintis, Tipperahs). Stack. Mikirs, p. 18 sq. Butler, Travels and Adventures in the Province of Assam, pp. 83, 138 (Kukis, Mikirs). Soppitt, Short Account of the Kuki-Lushai Tribes, p. 14 sq. Hutchinson, Account of the Chittagong Hill Tracts, pp. 135, 155, 167 (Kuki tribes of Chittagong, Tipperahs, Mrijs). Endle, Kachdris, pp. 44, 45, 85, 86, 95 (Kachiris, Rdbhis, Hajongs, and Deoris of Assam). Lewin, Wild Races of South- Eastern India, pp. 202, 234 (Tipperahs, Mriis). O'Malley, Census of India, 1911, vol. v. (Bengal, Bihar and Orissa and Sikkim) Report, p. 3T5 (Santals, Mundas, Orions). Hooker, Himalayan Journals, i. 125 (Lepchas). Hodgson, Miscellaneous Essays relating to Indian 362 THE HISTORY OF HUMAN MARRIAGE chap. in many islands of the Indian Archipelago1; in Ponape of the Caroline Islands,2 and in New Britain3; and among several African peoples.4 The man generally has to go and live with the family of his future spouse for a certain time. Subjects, i. 402 (Kirantis of the Central Himalayas). Kealy, Census of India, 1911, vol. xxii. (Rajputana and Ajmer-Merwara) Report, p. 165. Campbell, ' Note on the Limboos,' in Jour. Asiatic Soc. Bengal, vol. ix. pt. i. 603. Hodgson, ' On the Origin, &c. of the K6cch, Bodo, and Dhimal People,' ibid. vol. xviii. pt. ii. 735 (B6do and Dhimals). Hay, ' Tiiran Mall Hill,' ibid. xx. 507 (Bhils). Forsyth, Highlands of Central India, p. 148 sq. (Gonds and Korktis). Grant, Gazetteer of the Central Provinces of India, p. 277 (Gonds), Hayavadana Rao, ' Gonds of the Eastern Ghauts,' in Anthropos, v. 794 sq. Russell, Tribes and Castes of the Central Provinces of India, iii. 80, 395, 467, 558 sq. ; iv. 133, 134, 166 (Gonds, Kawars, Kandhs, Korkiis, Mahars, Marars of Balaghat and Bhandara). Cox, Madras District Manuals North Arcot, i. 213 (Malay Alis). Thurston, Castes and Tribes of Southern India, iv. 445 (Mandadan Chettis). Dubois, Mceurs, institutions et cdr&monies des peuples de I'Inde, i. 295 sq. 1 Marsden, op. cit. p. 262 sq. (Sumatrans). Wilken, ' Over het huwelijks- en erfrecht bij de volken van Zuid-Sumatra,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indii, xl. 162 sq. Kreemer, ' De Loeboes in Mandailing,' ibid. Ixvi. 321. Junghuhn, Die Battalander auf Sumatra, ii. 132 sqq. (Batta, natives of Passumah and Palembang). Snouck Hurgronje, Het Gajoland en zijne Bewoners, p. 270 sq. (Gayos of Achin in the north of Sumatra). Adriani and Kruijt, De Bare'e-sprekende Tor adj a's van Midden- Celebes, ii. 27 sq. Treffers, ' Het landschap Laiwoei in Z. O. Celebes,' in Tijdschrift van het Koninkl. Nederlandsch Aardrijkskundig Genoot- schap, ser. ii. vol. xxxi. 209 sq. Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, pp. 67, 131, 171, 205 sq. (Amboyna, Ceram, Ceramlaut, Watubela Islands). Mallat, Les Philippines, i. 58 ; ii. 45. Bowring, Visit to the Philippine Islands, p. 144. Lala, Philippine Islands, p. 90 sq. Blumentritt, Ver such einer Ethnographic der Philippinen, p. 14 (Tagalese, Bisayans). Jagor, Reisen in den Philippinen, p. 235 (Bisayans). Cole, * Wild Tribes of Davao District, Mindanao,' in Field Museum of Natural History, Anthropological Series, xii. 101-103, 144, 157, 192. Finley and Churchill, Subanu. Studies of a Sub-Visayan Mountain Folk of Mindanao, p. 40. 2 Christian, Caroline Islands, p. 74. 3 Romilly, ' Islands of New Britain Group,' in Proceed. Roy. Geograph. Soc. N.S. ix. 8. 4 Chapman, Travels in the Interior of South Africa, i. 259 (Bushmen). Hahn, Tsuni-Goam, p. 18 ; Francois, Nama und Damara Deutsch- XXIII MARRIAGE BY CONSIDERATION, ETC. 363 during which he works as a servant. The period of service varies greatly among different peoples : it seldom lasts less than a year, and may even last as many as ten or twelve or fifteen years.1 During this time he may2 or may not3 have access to the girl. He may have to serve after his marriage as well, until a child is born or longer,4 and he may have to remain with his wife's family for ever. But, as has been justly pointed out, a form of marriage under Sild-West-Afrika, p. 214 (Hottentots). Casalis, Basutos, p. 183. Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 220 (Makaranga). Livingstone, Missionary Travels and Researches in South Africa, p. 622 (Banyai of the Zambesi River). Fromm, ' Ufipa,' in Mittheil. Deutsch. Schutzgeb. xxv. 96. Heese, ' Sitte und Branch der Sango,' in Archiv f. Anthrop. N.S. xii. 135. Fiillebom, Das Deutsche Njassa- und Ruwuma-Gebiet, p. 447 (natives of Ukinga). Fraser, Winning a Primitive People, pp. 153, 155 (Tumbuka of British Central Africa). Cunningham, Uganda, pp. 302, 304 (Basukuma). Merker, Die Masai, p. 232 (Wandorobbo, belonging to the Masai). Tauxier, op. cit. p. 366 sq. (Zangas of the French Sudan). Thomas, Anthropological Report on Ibo-speaking Peoples of Nigeria, i. 63 (people of the Edo country) ; iv. 61 (Asaba Ibo). Idem, Anthropological Report on Sierra Leone, i. 100 (Susu). Ellis, Tshi-speaking Peoples of the Gold Coast, p. 281. Allen and Thomson, Narrative of the Expedition sent to the River Niger, in 1841, ii. 203 (Edeeyahs of Fernando .Po). Lamouroux,' La region du Toubouri,' in L'Anthropologic, xxiv. 682. Poupon, ' litude ethnographique des Baya,' ibid. xxvi. 124.-For the custom of serving for a wife see also Post, Grundriss der ethnologischen Jurisprudenz, i. 318 sqq., and especially Frazer, Folk-Lore in the Old Testament, ii. 342 sqq. 1 Dobell, op. cit. i. 82 (Kamchadal). Kealy, op. cit. p. 165 (Rajputana and Ajmer-Merwara). Endle, op. cit. p. 45 (Kach&ris). Thomas, Ibo-speaking Peoples of Nigeria, iv. 61. 2 v. Martins, op. cit. i. 107 sq. (some Brazilian tribes). Gait, op. cit. p. 251 (Rangkhol of the Kuki-Lushais). Lewin, op. cit. p. 202 (Tipperahs). Riedel, op. cit. pp. 67, 131 (Amboyna, Ceram). 3 v. Martius, op. cit. i. 108 (some Brazilian tribes). Jochelson, Koryak, p. 735. Kealy, op. cit. p. 165 (Rajputana and Ajmer- Merwara). Russell, op. cit. iii. 80; Grant, op. cit. p. 277 (Gonds). 4 Marston, loc. cit. p. 166 (Sauk and Foxes). Stack, op. cit. p. 18 sq. (Mikirs). Soppitt, op. cit. p. 15 (Kuki-Lushais). Hutchinson, op. cit. p. 155 (Tipperahs). Hayavadana Rao, in Anthropos, v. 795 (Gonds of the Eastern Ghauts). Cole, loc. cit. pp. 102, 144, 192 (Bagobo, Bila-an, Mandaya of Mindanao). Kreemer, loc. cit. p. 321 (Lubus of Sumatra). Troffers, loc. cit. p. 210 (natives of South- Eastern Celebes). 364 THE HISTORY OF HUMAN MARRIAGE chap. which a man is permanently transferred to his wife's faifiily is to be distinguished from the form of marriage under which he serves his father-in-law for a limited time for the wife whom he will afterwards take away with him to his own home ; 1 and I can find no sufficient ground for Mak- simoff's suggestion that the custom of serving for a wife is the relic of a former custom by which a man went to live permanently in his father-in-law's house.2 Of several peoples we are told that the first wife only is obtained by service.3 In various tribes service is the regular, if not exclusive, method of acquiring a wife. Of the Ainu Pilsudski says that if purchase of a wife occurs either in real life or in the myth, this is usually in places where the Ainu have come into contact with the Gilyak and are influenced by them ;4 but it also occurs that an Ainu girl who falls in love with a young man enslaves herself to his parents as a price for their son.6 Among the Kamchadal, Chukchee, Koryak, and Yukaghir serving for a wife is likewise the customary form of marriage. The Chukchee call it by a term which means " serving as a herdsman in payment for the bride " ; and this term is so firmly established that it is used even by the Maritime Chukchee, though they have no herds and the bridegroom simply lives in the house of the girl's father and works for him during a certain period.6 Yet when a rich Chukchee wants to marry a girl of a poor family, the time of service may dwindle down to nothing; he gives to the girl's father a few reindeer-which, however, are not called pay for the bride, but a " joyful gift "-or, more frequently, he invites the poor family of his new wife to come to his camp and to live there on his own herd.7 Dr. 1 Frazer, op. cit. ii. 357 sq. Starcke, The Primitive Family, p. 39. ' Maksimoff, quoted by Miss Czaplicka, op. cit. p. 83. 3 v. Martius, op. cit. i. 108 (Brazilian tribes). Carver, op. cit. p. 373 (Naudowessies). Bogoras, op. cit. p. 586 (Chukchee), Cole, loc. cit. pp. 103, 144 (Bagobo, Bila-an of Mindanao). Allen and Thomson, op. cit. ii. 203 (Edeeyahs of Fernando Po). 4 Pilsudski, quoted by Miss Czaplicka, op. cit. p. 102. 5 Batchelor, op. cit. p. 230. 9 Bogoras, op. cit. p. 579. 1 Ibid. p. 586. xxiii MARRIAGE BY CONSIDERATION, ETC. 365 Jochelson says that on the tundra or steppe " the Yukaghir have borrowed from the Tungus the custom of purchasing the bride and taking her to the bridegroom's house " ; but it is only after serving from one to three years that the bridegroom can carry off his wife to his own home.1 The same authority remarks that among the Reindeer Koryak a wealthy reindeer-breeder undoubtedly would prefer paying for his wife with reindeer to serving for her, but that this does not occur; yet if the suit is pressed by an elderly or wealthy man, the service is reduced to a minimum and is performed in a formal manner only.2 Among the Kuki-Lushais of Assam, according to Mr. Soppitt, the preliminaries to an ordinary marriage are as follows :-" A man having taken a fancy to a girl offers a present of liquor to the parents and talks the matter over. Should they be willing to accept him as a son-in-law, he takes up his abode with them for three years, working in the jhums, and practically becoming a bond servant. At the end of this period he is allowed to marry the girl, but even then is not free, as he has to remain on another two seasons, working in the same manner as he did before. At the completion of the five years he is free to build a separate house and start life on his own account. Two rupees is the sum ordinarily paid the parents of the girl, a sum paid evidently more for the purpose of proving a contract than for anything else, the long period of servitude being the real price paid."3 In another account of the Kuki-Lushai tribes we are told that the Rangkhol tribe prefers marriage by service, the bridegroom residing from three to seven years in his future father-in-law's house; whereas the Thadoi tribe prefers marriage by purchase.4 Among the Tipperahs of the Hill Tracts of Chittagong, when a match is made with the consent of the parents, the young man must serve three, or according to another account two,5 years in his father-in-law's house before he obtains his wife or is formally married, though during his time of servitude 1 Jochelson, Yukaghir, pp. 87, 93. 2 Idem, Koryak, p. 740. 3 Soppitt, op. cit. p. 14 sq. 1 Gait, op. cit. p. 251. 6 Hutchinson, op. cit. p. 155. 366 THE HISTORY OF HUMAN MARRIAGE chap. or probation the girl is practically his wife.1 So also among the Mr us, living in the same region, a suitor has to serve three years for his wife ; but if he be wealthy he can dispense with this service by paying 200 or 300 rupees down2- which suggests that among them marriage by purchase is only a substitute for marriage by service. Among the Malay4Hs of North Arcot a man has to serve at least for a year in the house of the bride in order to receive the consent of her parents to marry her.3 In Cambodia, among families in easy circumstances the period of probation is sometimes no longer than fifteen or twenty days.4 Among the Bisayans of the Samar and Leyte islands in the Philippines " the suitor has to serve in the house of the bride's parents two, three, and even five years before he takes his bride home ; and money cannot purchase exemption from this onerous restriction."5 In various other tribes in the same group and other islands of the Indian Archipelago service is evidently a regular or necessary preliminary to marriage. Thus it is said of the Lubus of Mandailing in Sumatra that a man is obliged to serve his prospective parents-in-law for two years before marriage, during which he has to perform all kinds of drudgery for them ; and even after his marriage custom imposes on him many obligations as to field labour for the benefit of his wife's father and mother.6 Among the Hottentots the son-in-law was obliged to spend the first year or years of married life in the service of his father-in-law.7 Among the Banyai of the Zambesi River, when a young man takes a liking to a girl of another village and her parents have no objection to the match, he is likewise obliged to come and live at their village, where he has to perform certain services for the mother-in-law, such as keeping her well supplied with firewood ; but if he becomes tired of living in this state of vassalage, he may return to his own village on condition that he leaves all 1 Lewin, op. cit. p. 202. 2 Ibid. p. 234. 3 Cox, op. cit. i. 213. 1 Aymonier, quoted by Hartland, op. cit. ii. 48. 6 Jagor, op. cit. p. 235. 6 Kreemer, loc. cit. p. 321. 7 Hahn, op. cit. p. 18. v. Francois, op. cit. p. 214. xxni MARRIAGE BY CONSIDERATION, ETC. 367 his children behind.1 Among the Tumbuka of British Central Africa, when a young man's engagement was fixed, he had to go and build a house in the village of his future father-in-law, and, when the rain fell, help him to hoe his garden ; and when all these preliminary arrangements were completed, the marriage took place and the husband became a member of his wife's village. After some years, however, he might return to his own people if he desired to do so, on condition of presenting a slave or a cow to his parents-in- law to redeem himself, though he could never redeem his children.2 Among the Edeeyahs of Fernando Po the betrothal must continue at least for two years, " during which time the aspirant, to Edeeyah beauty is obliged to perform such labour as would otherwise fall to the lot of his intended wife ; carrying the palm-oil to the market, water for household purposes, planting yams, etc." ; but among this, as among various other peoples, it is only for his first wife that a man has to serve.3 Among the Ekoi of Southern Nigeria a man who desires to marry a certain woman must serve her family for some considerable time, usually from two to three years, during which time he is also expected to make presents to her relations.4 Among the Zangas of the French Sudan a suitor does not pay for his wife, but he works instead once a year for three years on the fields of his father- in-law, or rather of the head of the family group to which his father-in-law belongs.5 It seems that in most South American tribes which have marriage by service, a man can in no other way obtain a father's consent to marry his daughter. Among the Brazilian Indians, according to von Martius, it is especially in the larger, settled hordes and tribes that a wife is acquired by work in the house of the prospective father-in-law; and a young man may have to work there even for several years with indefatigable diligence, although he lives with his own relations.6 Among the Canelos Indians of Ecuador all the 1 Livingstone, op. cit. p. 622. a Fraser, Winning a Primitive People, p. 153 sqq. 3 Allen and Thomson, op. cit. ii. 203. 4 Talbot, op. cit. p. 105. 5 Tauxier, op. cit. p. 366 sq. 8 v. Martius, op. cit. i. 107. 368 THE HISTORY OF HUMAN MARRIAGE chap bridegroom has to do " is to clear the ground necessary for the chacra, or plantain and yuca plantation."1 Of the Arawaks of British Guiana Mr. Brett says that " the wife's father expects the bridegroom to work for him in clearing the forest, and in other things, and the young couple often remain with him until an increasing family renders a separate establishment necessary."2 But according to Sir Everard F. Im Thurn's description of the Indians of British Guiana, marriage is certainly sometimes by purchase, although it occurs that a girl is given by her parents to a man in recom- pense for some service done. But " the marriage once arranged, the husband immediately transports his possessions to the house of his father-in-law, and there he lives and works. The head of his family, for whom he is bound to work, and whom he obeys, is not his own father, but his wife's."^ Among the Indians of Yucatan, after the conclusion of a marriage, the husband used to serve his father-in-law four or five years ; but if he failed to complete his term of service, he was turned adrift and the woman given to another man.4 Among the Naudowessies in the region of the Great Lakes it was customary for a young man to obtain his first wife by residing for a year as a menial servant in the tent of the Indian whose daughter he wished to marry ; during that time he hunted and brought all the game he killed to the family of his future wife, and when the year expired the marriage was celebrated.5 Marston states that among the Sauk and Foxes the young man has to serve the parents of the girl according to custom, which is until she has a child, hunting in the most industrious manner ;6 whilst Forsyth says of the same Indians that " if the parents of the girl will not agree to receive property but insist on servitude, the young Indian must come to hunt for his wife's parents for some one, two, or three years as may be agreed on before the parents will relinquish their right to their daughter."7 Of the Chippewa, Ottawa, and Potawatomi we are told 1 Simson, op. cit. p. 104. 2 Brett, op. cit. p. 101. 3 Im Thurn, op. cit. p. 221 sq. 1 de Herrera, op. cit. iv. 172. 8 Carver, op. cit. p. 373.. 8 Marston, loc. cit. p. 166. 7 Forsyth, in Emma Helen Blair's Indian Tribes of the Upper Mississippi Valley, ii. 214. xxin MARRIAGE BY CONSIDERATION, ETC. 369 that a wife is sometimes purchased by the parents of the young man, when she becomes at once his own property, but that the most common mode of procuring a wife in all these tribes is by servitude ; and it frequently happens that when an Indian's servitude for one wife has expired, he will take another-his wife's sister perhaps-and again serve her parents according to custom.1 Perrot wrote of some Algonkin that after the consummation of the marriage the youth was obliged to live with his mother-in-law and work for her during two years.2 Among the Haida at Skidegate, from the time a youth became engaged " he went to live with the girl's family, if not already there, and worked for them until his marriage."8 Among the Kenai, an Eskimo people of Alaska, a man must perform a year's service for his bride. He goes to the house of his prospective father-in-law, and, without a word of explanation, begins to heat the bath-room, to bring in water, and to prepare food; if his suit is not rejected he remains as a servant in the house a whole year, after which he receives a reward for his services from the father, and takes his wife home with him.4 In many cases marriage by service occurs, not as the regular form of marriage, but as a substitute for marriage by purchase, when the suitor is too poor to pay the ordinary bride price. The Abbe Dubois, in his book on the ' People of India,' which refers chiefly to the Madras Presidency, says that " as the marriage expenses are considerable, we find in all castes a number of young men destitute of the means of defraying them who, in order to procure a wife, have recourse to the same expedient as Jacob employed with Laban " ; and he adds that the number of years of service required in these cases is the same as it was in ancient Israel, namely seven.5 Marriage by service as a substitute for marriage by purchase is practised in various parts of India, although the period of service varies,6 and in other countries as 1 Marston, loc. cit. p. 167. 2 Perrot, loc. cit. i. 69. 3 Swanton, op. cit. p. 50. 4 Richardson, op. cit. i. 406 sq. 6 Dubois, op. cit. i. 295 sq. 0 Hayavadana Rao, in Anthropos, v. 794 (Gonds of the Eastern Ghauts in the Madras Presidency). Grant, op. cit p. 277 (Gonds of the Central Provinces). Russed op- cit. iii. 80, 467 (Gonds of the VOL. II B B 370 THE HISTORY OF HUMAN MARRIAGE chap. well.1 Thus in Palembang, in the south of Sumatra, a poor suitor binds himself to live with his parents-in-law and labour for them until he has paid for his wife. If it happens that he is unable all his life long to discharge the debt, it is trans- mitted to his children, who continue like their father in a state of bondage until the daughters, by the bride prices which are paid for them at their marriage, at last succeed in paying the sum which is still owing for the marriage of their mother.2 Among the Tshi-speaking peoples of the Gold Coast, when a man is too poor to pay even the smallest sum demanded for a wife, he will live with his wife without paying anything for her, unless it be a bottle or two of rum given to the family to drink; but in that case he generally resides with her family, and gives them his services towards their common support.3 Among the Makaranga of South Africa " a young man too poor to acquire a wife by the transfer of cattle would make an arrangement with the father of a girl to live with her and to serve him, when, as children do not belong to their father until the full ikazi (that is, the bride price consisting of cattle) has been transferred, the father of the woman had sole control over all that were born."4 Among some other peoples, also, although a man earns his wife by serving her Central Provinces, Kandhs). O'Malley, op. cit. p. 315 (Santals, Mundas, Ordons). Campbell, in Jour. Asiatic Soc. Bengal, vol. ix. pt. i. 603 (Limboos of Sikkim and Nepaul). Hodgson, Miscellaneous Essays, i. 402 (Kir^ntis of the Central Himalayas). Endle, op. cit. p. 44 sq. (Kachans). Gait, op. cit. p. 225 (Bodo group of tribes in Assam). Hodson, op. cit. p. 91 (Nagas ; when service is accepted in lieu of the bride price it is nevertheless the custom to insist on the transfer of some material object, such as a gong, or cloth, or spear). Dalton, op. cit. pp. 41, 104, no (Nagas, Limbus and KirAntis, Tipperahs). 1 Wilken, in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, xl. 162 (people of Lampong in the south of Sumatra). Junghuhn, op. cit. ii. 132 (Battas). Snouck Hurgronje, op. cit. p. 270 sq. (Gayos of Achin). Riedel, op. cit. p. 132 (people of Ceram). Lala, op. cit. p. 90 sq. (Philippine Islanders). 2 Wilken, in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, xl. 163. 3 Ellis, Tshi-speaking Peoples of the Gold Coast, p. 281. 4 Theal, op. cit. p. 220. XXIII MARRIAGE BY CONSIDERATION, ETC, 371 parents, he cannot earn his children without paying for them.1 Among the Mikirs of Assam, again, it appears that mar- riage by service is not a substitute for marriage by purchase, but an alternative to giving no consideration for a wife. It is said that the mode of marriage among them depends upon the wealth and standing of the parties. If the wedding is akejoi-that is, if no payment is to be made for the bride,- the girl goes with her husband next day to her new home ; whereas if it is akem^n, the lad stays in his father-in-law's house. He rests one day, and then works for his father-in- law for a year, or two years, or even it may be for life, according to agreement. There is no money payment in any case. If the girl is an heiress or only daughter, the marriage is usually akemin, but in the great majority of, cases it is akejoi.2 Among various peoples a man cannot obtain a wife by service alone, but, besides serving for her, has to pay a bride price. In some cases the service is a substitute for some part of the purchase-sum. Among the Basukuma, inhabit- ing a country bordering Lake Victoria on the south-east, the general rule is that some sixty sheep have to be paid as a first instalment, and the young man then goes to live with his father-in-law and serves him for two years in lieu of further instalment.3 Among the Wandorobbo, belonging to the Masai, a bridegroom who is too poor to pay the whole bride price makes up for the remaining portion by hunting for his father-in-law during a few months.4 Among the Californian Hupa a poor man " might pay half the usual sum and go to the home of the bride, where he served his father-in-law " ; but all offspring of the union belonged to the woman's people.6 In other cases, again, service is compulsory in addition to the payment of the bride price. Thus among the Achomawi and Atsugewi of Northern 1 Livingstone, op. cit. p. 622 sq. (Banyai and other South African peoples). Riedel, op. cit. p. 173 (people of Ceramlaut). Adriani and Kruijt, op. cit. ii. 24 sq. (Bare'e-speaking Toradjas of Central Celebes). Cf. Frazer, op. cit. ii. 356, 371. 2 Stack, op. cit. p. 18 sq. 8 Cunningham, op. cit. pp. 302, 304. 4 Merker, op. cit. p. 232. 8 Goddard, op. cit. p. 56. B B 2 372 THE HISTORY OF HUMAN MARRIAGE chap. California, after the affair has been settled by the payment of such property as has been agreed on, the suitor takes up his residence with the girl's family and stays there, hunting and working for them for a month or two ; if he has parents, he then takes his wife to their house, but in the contrary case he remains with his wife's family.1 Concerning the Baya of the French Congo, among whom one price is claimed for the virginity of a girl and another price for marrying her, M. Poupon writes :■-" La virginite pay£e, le prix du manage paye, ce ne sont pas la encore ies seuls versements. Avant le manage et en dehors des marchandises, le futur gendre a d6ja foumi a son beau-pere pas mal de journees de travail; aprds le manage, ces obligations subsistent pour le gendre pendant toute sa vie."2" In Ufipa3 and Ukinga4 in the former German East Africa a suitor must likewise work for his bride as well as pay for her. Among the Sango in the same part of Africa, after the bride price has been paid, the parents of the girl require that the suitor shall make a garden for them; they say that they want to see how he can work. If he is lazy or does not execute the work to their satisfaction, the bride price is restored to him and the intended marriage falls through.5 So also in the Philippine Islands, in former times, a suitor had not only to pay for his bride but was also obliged to serve for her in the house of her parents ; and it often happened that all his labour was thrown away, because, if he did not please them, they simply chose another man to work for the hand of their daughter.® Among the natives of South-Eastern Celebes a suitor goes to live with the girl's parents, and if after a period of probation they are satisfied with him and the girl returns his affection, he marries her; but he must pay for her a price which varies from fifty to a hundred guilders.7 Among the Yukaghir of the tundra a man is not allowed to carry off his wife to his own home unless he has served from 1 Dixon, in American Anthropologist, N.S. x. 217. 2 Poupon, in L'Anthropologic, xxvi. 124. 3 Fromm, in Mittheil. Deutsch. Schutzgeb. xxv. 96. * Fiilleborn, op. cit. p. 447. 6 Heese, in Archiv f. Anthrop. N.S. xii. 135. u Mallat, op. cit. i. 58. 7 Trefiers, loc. cit. p. 209 sq. xxin MARRIAGE BY CONSIDERATION, ETC. 373 one to three years in his father-in-law's house and besides paid a certain number of reindeer for her.1 The practice of serving for a. wife is no doubt in a large measure due to the unwillingness of a father to give his daughter in marriage for nothing; this is proved by the many cases in which service is a substitute for ordinary purchase. But from various statements just quoted it appears that it also has another meaning : the period of service is intended to test the young man's ability to work and to show whether he is an acceptable husband and son-in-law. This has been strongly emphasised by Dr. Jochelson. The Yukaghir themselves told him that the time of service was a test. Besides, he adds, there can be purchase by service only when the girl leaves the father's house and goes to live with her husband, depriving the father of a pair of working hands ; but among most of the Yukaghir the latter gains a new worker in the person of his son-in-law, who comes to live in his house.2 With reference to the Koryak, the same acute observer writes :-" A serving bridegroom is not an ordinary workman. The principal thought is not his usefulness, but the hard and humiliating trials to which he is subjected. The bridegroom is given a poor bed, he is ill-fed, he is not allowed to sleep late, he is sent on exhausting errands. As a herdsman he must pass his nights without sleep, while the proprietor of the herd and the bride's brothers are resting. In a word, during his term of service, his endurance, patience, and meekness, his adroitness as a hunter, and his zeal and frugality as a herdsman, are tested. The bride's father gives his assent to the marriage only after the bridegroom has stood the probation well. This view of the trial of a bridegroom, who must perform tests dangerous to his life, and win contests, is also found in Koryak tales."3 Among the Kamchadal, when the period of service has expired and the suitor requests leave to carry away his bride, 'the leave is granted him at once if he has earned the approbation of the parents of the bride and of her relations ; whereas if he has incurred their 1 Jochelson, Yukaghir, p. 93. 3 Ibid. p. 87. 8 Jochelson, Koryak, p. 740. 374 THE HISTORY OF HUMAN MARRIAGE chap. displeasure, they give him some small reward for his services and he departs.1 In Cambodia the period of service is likewise a period of probation, in which it is the youth's business to render himself agreeable to the young lady as well as to her parents.2 In Ponape, one of the Caroline Islands, the suitor who serves for his wife in the house of the prospective father-in-law " frequently has his pains for nothing."3 Of the Tubori on the border of the Cameroons it is said that the father of the girl often tests the suitor by making him work for him ; and if he is satisfied he fixes the price to be paid, which may amount even to eight or ten bullocks.4 Speaking of the service of the bridegroom among the Naudowessies, Carver observes that by this means " the father has an opportunity of seeing whether he is able to provide for the support of his daughter and the children that might be the consequence of their union."8 Among the South American Indians the services which a man has to render his intended father-in-law are obviously, to a large extent, meant to be a test of his ability.6 Spencer believed that the obtaining of wives by services rendered, instead of by property paid, constitutes a higher form of marriage, and is developed along with the industrial type of society. " This modification," he says, " practicable with difficulty among rude predatory tribes, becomes more practicable as there arise established industries affording spheres in which services may be rendered."7 This view, however, is not borne out by facts. Marriage by service occurs, nay flourishes, among hunting tribes; and in other caess also the services required of a suitor are not such as presuppose any higher development of economic culture. On the other hand, where there is marriage by purchase 1 Krasheninnikoff, op. cit. p. 212. 2 Aymonier, quoted by Hartland, op. cit. ii. 47. 3 Christian, op. cit. p. 74. , * Lamouroux, in L' Anthropologic, xxiv. 682. 5 Carver, op. cit. p. 373. • Nordenskiold, op. cit. p. 195. v. Martius, op. cit. i. 645. von den Steinen, op. cit. p. 332. Schmidt, in Zeitschr. f. vergl. Rechts- wiss. xiii. 307. 7 Spencer, Principles of-Sociology, i. 721 XXIII MARRIAGE BY CONSIDERATION, ETC. 375 there must be some accumulation of property. Hence service as a substitute for the normal form of purchase may be said to be found among more advanced savages alone. Most frequently the consideration given for a wife consists of property of some kind or other. The amount varies greatly among different peoples. In the lowest tribes it is very inconsiderable. Among the Veddas of Ceylon some food is presented to the parents of the girl,1 or nothing at all is, apparently, given for her.2 Among the Sakai of Perak the bridegroom makes a present to his prospective father-in-law of certain specified articles, for example, a knife, or a hatchet, or yams, " according to his means " ;3 whilst among the Orang Sletar of the Orang Laut a mouthful of tobacco and one chufiah of rice handed to the mother of the bride confirmed the hymeneal tie.4 A Negrito of Bataan, in Luzon, does not buy his wife ; he- only offers a small present to his future father-in-law, who in return gives to his daughter a few objects, which remain her personal property.5 In some Australian tribes, which have not the custom of exchanging females, the bridegroom supplies his father- or parents-in-law wdth game, or in addition presents the former with boomerangs and other weapons.6 Con- 1 Hartshorne, ' Weddas,' in Indian Antiquary, viii. 320. Lamprey, quoted by Dechamps, ' Les Veddas de Ceylan,' in L' Anthropologic, ii- 3i3- 2 Le Mesurier, ' VeddAs of Ceylon,' in Jour. Roy. Asiatic Soc. Ceylon Branch, ix. 340. Sarasin, Ergebnisse naturwissenschafflicker Forsckungen auf Ceylon, iii. 460 sq. 8 de Morgan, ' Mceurs, coutu mes et langages des Negritos de l'int6rieur de la presqu'ile Malaise,' in Bull. Soc. normande de Geo- graphic, vii. 421. Low, ' Karean Tribes or Aborigines of Martaban and Tavai,' in Jour. Indian Archipelago, iv. 431. [Maxwell,] ' Semang and Sakei Tribes,' in Jour. Roy. Asiatic Soc. Straits Branch, no. 1, p. 112. 4 Thomson, * Remarks on the Sletar and Sabimba Tribes,' in Jour. Indian Archipelago, i. 341*. 5 Montano, Voyage aux Philippines et en Malaisie, p. 71. 6 Howitt, Native Tribes oj South-East Australia, p. 220. Strehlow, Die Aranda- und Loritja-Stamme in Zentral-Australien, vol. iv. pt. i. 90, 100. Spencer and Gillen, Northern Tribes of Central Australia, p. 77 n. 1, Malinowski, op. cit. p 51. 376 THE HISTORY OF HUMAN MARRIAGE chap. ceming the Central African Pygmies, Grenfell says that marriage among them " is little else than the tendering of a gift of arrow heads or knives or other objects of local value to the father of the girl, who is thereby acquired by the purchaser; though no doubt inclination counts for much in the* bargain."1 According to Hutereau, again, a man acquires a girl for wife by offering to her father two animals killed in the chase and to her mother a dog.2 Among the Namib Bushmen some food, blankets, and furs are given to the prospective parents-in-law.3 Among many tribes of a more advanced type the con- sideration given for a wife is likewise of small value. It is often represented as a gift only, the offering and acceptance of which constitute no act of purchase ; whereas in other cases it is of a considerable amount and makes the conclusion of a marriage a genuine business transaction. The following facts may serve as illustrations of the quality and quantity of objects given for a bride among uncivilised peoples in different parts of the world. Among the Eskimo there is certainly no regular system of marriage by purchase.4 Among the Angmagsaliks on the east coast of Greenland, according to Holm, the young man may have to give to the father a harpoon or some similar article to get his daughter in marriage, but it also happens that good hunters are paid by fathers to marry their daughters.5 Among the Eskimo of the Ungava district, Hudson's Bay Territory, " the sanction of the parents is sometimes obtained by favor or else bought by making certain presents of skins, furs, and other valuables to the father and mother."6 Among the Eskimo round Repulse 1 Johnston, George Grenfell and the Congo, ii. 674. See also Casati, Ten Years in Equatoria, i. 157. 2 Hutereau, Notes sur la Vie familiale et juridique de quelques populations du Congo Beige, p. 4. 3 Trenk, ' Die Buschleute der Namib,' in Mittheil. Peutsch. Schutzgeb. xxiii. 168. 4 Cf. Gilbertson, Some Ethical Phases of Eskimo Culture, p. 65. 8 Holm, ' Ethnologisk Skizze af Angmagsalikeme,' in Meddelelser om Grdnland, x. 96. 6 Turner, ' Ethnology of the Ungava District, Hudson Bay Territory,' in Ann. Rep. Bur. Ethn. xi. 188. XXIII MARRIAGE BY CONSIDERATION, ETC. 377 Bay the father who selects a little girl for his young boy to be his daughter-in-law pays her father something-perhaps a snow-knife or a sled or a dog or, nowadays, a handful of powder and a dozen percussion caps.1 Of the Central Eskimo Dr. Boas simply says that the bride must be bought from the parents by some presents.2 Of the North American Indians in general it has been said that " the economic factor is everywhere potent," but that an actual purchase is not common.3 Among the Delaware, for instance, certain formal presents were made, the acceptance of which meant a favourable decision on the part of the parents of the girl.4 Among the Omaha " suitors may carry favor with parents and kindred of the girl by making presents to them, but parents do not sell their daughters. The presents made for such a purpose are generally given by some old man who wishes to get a very young girl whom he is doubtful of winning. Where a man courts the girl directly this is unnecessary. Then he gives what he pleases to her kindred, and sometimes they make presents to him."5 Of the Hidatsa, another Dakotan tribe, we are told that marriage is usually made formal by the distribution of gifts on the part of the man to the woman's relations.6 Among other North American tribes the con- sideration is of greater importance, and may have the character of a genuine bride price. In some parts of British Columbia and Vancouver Island the value of the articles given for the bride ranged from £20 to £40 sterling.7 The Indians of Oregon bought their wives for horses, blankets, or buffalo robes.8 Among the Shastika in California " a wife is purchased of her father for shell-money or horses, 1 Gilder, Schwatka's Search, p. 249 sq. 2 Boas, * Central Eskimo,' in Ann. Rep. Bur. Ethn. vi. 579. 3 Hodge, Handbook of American Indians, i. 809 sq. Cf. Kohler, ' Die Rechte der Urvolker Nordamerikas, ' in Zeitschr. f. vergl. Rechtswiss. xii. 380 sqq. 4 Harrington,' Preliminary Sketch of Len&pe Culture,' in. American Anthropologist, N.S. xv. 215. 6 Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethn. iii. 259 sq. c Matthews, Ethnography and Philology of the Hidatsa Indians, p. 52. 7 Macfie, Vancouver Island and British Columbia, p. 446. 8 Schoolcraft, Archives of Aboriginal Knowledge, v. 654. 378 THE HISTORY OF HUMAN MARRIAGE chap. ten or twelve cayuse ponies being paid for a maid of great attractions."1 Among the Californian Hupa " a man's standing in the world depended on the amount of money which had been paid for his mother at the time of her marriage. If the sum were large he was the peer of any in the tribe."2 The worth and attractiveness of an Apache woman are exalted in exact proportion to the number of horses given for her ; " if a girl is sold for one animal, no matter how good, she is deemed of little account." 3 Among the Navaho the value of the presents which the bridegroom's people must give the bride's family ranges from five to fifteen horses.4 Among the South American Indians the consideration given for a wife, if any, likewise varies greatly in different tribes. A Huitoto bridegroom gives a small quantity of coca or tobacco to the capital, or chief of the subtribe, to obtain his approval, and cuts a supply of firewood for his future mother-in-law; and shortly afterwards the girl is given to him.5 Among the Purls, Coroados, and Coroapos some game and fruits only are given by the bridegroom for his bride immediately before marriage, whereas the more civilised tribes of the Brazilian aborigines carry on an actual trade in women.® The price paid for a Goajiro girl chiefly consists of cattle.7 Among the Mocobis in the southern part of the Gran Chaco a girl was given in marriage either for some tiger furs or for one or two horses, a cow, and a few ornaments.8 The Patagonians give mares, horses, or silver ornaments for their brides ; even a hundred mares have been paid for an heiress who has animals of her own.9 1 Powers, Tribes of California, p. 247. 2 Goddard, op. oil. p. 56. 3 Cremony, Life among the Apaches, p. 247. 4 Stephen, ' Navajo,' in America^ Anthropologist, vi. 356. Cf. Schoolcraft, op. cit. iv. 214 ; Letherman, * Sketch of the Navajo Tribe of Indians,' in Smithsonian Report, 1855, p. 294. 5 Hardenberg, Putumayo, p. 154. 6 v. Martins, op. cit. i. 109 sq. 7 Simons,' Exploration of the Goajira Peninsula, U.S. of Colombia,' in Proceed,. Roy. Geograph. Soc. N.S. vii. 792. 8 Kohler, Pater Florian Baucke, p. 315. 9 Musters, ' On the Races of Patagonia,' in four. Anthr. Inst. i. 201. Falkner, Description of Patagonia, p. 124. Prichard, Through the Heart of Patagonia, p. 93. XXIII MARRIAGE BY CONSIDERATION, ETC. 379 Among African peoples the bride price is frequently paid in cattle or goats. Among the Kafirs it used to average ten or twelve cattle for ordinary people ; a chief would have to pay fifty or a hundred cattle or more, according to his riches, whereas a poor man often obtained a wife for an ox or a couple of cows.1 Among the Herero much less was given for a wife, and a rich man or a man of rank gave no more than a poor one; " custom required that when a woman went to live with a man he should transfer to her father or guardian a large ox, a heifer, a large fat sheep, a ewe with a lamb, and a young ewe, but the most valuable of these animals were at once strangled and eaten at the feast which was the only ceremony attending the alliance."2 Among the Banyoro the sum demanded by the wealthier people was from ten to twenty cows.8 In Lendu, in Uganda, the amount of the bride price fixed by general custom is sixteen cows and one hundred goats.4 Among the lower orders of the Baganda the ordinary price of a wife was either three or four bullocks, six sewing needles, or a small box of per- cussion caps; but Mr. Wilson was often offered one for a coat or a pair of shoes.6 Among the Batamba a chief has to give three or four cows for a wife, whereas an ordinary, peasant is required to give only six or seven goats.6 Among the Bantu Kavirondo the price to be paid for a wife is generally considered to be forty hoes, twenty goats, and one cow, though more cows are paid if the girl is the daughter of an important chief.7 The usual price for a Nandi girl is at the present time one bull, one cow, and ten 1 Kidd, The Essential Kafir, p. 214. v. Weber, Vier Jahre in Afrika, ii. 215. Barrow, Account of Travels into the Interior of Southern Africa, i. 206. 2 Theal, op. cit. p. 219. Cf. Chapman, Travels in the Interior of South Africa, i. 341 ; Viehe, in Steinmetz, Rechtsverhaltnisse von eingeborenen Vdlkern in Afrika und Ozeanien, p. 307. 3 Roscoe, Northern Bantu, p. 39. 4 Cunningham, Uganda, p. 328. 5 Wilson and Felkin, Uganda and the Egyptian Soudan, i. 187. Fei kin, * Notes on the Waganda Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xiii. 756. • Condon, ' Contribution to the Ethnography of the Basoga- Batamba, Uganda Protectorate,' in Anthropos, vi. 371. 7 Johnston, Uganda Protectorate, p. 747. 380 THE HISTORY OF HUMAN MARRIAGE chap. goats, though formerly the price was higher.1 Among the Akamba forty to fifty goats are on an average paid for a bride, besides some cattle, generally two cows and two bulls or oxen ; but a rich man may pay a hundred goats or even more.2 If a Kikiiyu wishes to buy a wife he must pay so many " goats," generally thirty, though the actual payment may take the form of cattle, sheep, and goats.3 Among the Bangala of the Upper Congo River, again, " a free man marrying a free woman would have to give her father and family -two male and two female slaves, and no money or goods would be taken in lieu of them."4 So also among the Mandingo of Senegambia, according to CailH6, no wife was to be had otherwise than by the presentation of slaves to the parents of the mistress.5 Among the Ewhe-speaking peoples of the Slave Coast the payment of " head-money," which constitutes a marriage, may be in actual coin or cowries, or, as is more commonly the case, in merchandise and rum.6 Among the Wayao of the former German East Africa, according to Dr. Weule, who have neither large herds of homed cattle nor abundance of sheep or goats, the whole " purchase " of a wife-were it correct, which it is not, to call the transaction by that name-is effected by handing over a moderate quantity of calico.7 Among the Ugro-Finnic and Turko-Tartar peoples of the former Russian empire the payment of a bride price, or kalym, has been, or still is, an essential condition of the conclusion of a marriage.8 Among the Votyak it rises to 1 Hollis, Nandi, p. 61. 2 Lindblom, Akamba, p. 70. Cf. Hobley, Ethnology of A-Kamba, p. 62. 3 Routledge, With a Prehistoric People, pp. 44, J23, 125. 4 Weeks, ' Anthropological Notes on the Bangala of the Upper Congo River,' in Jour. Roy. Anthr. Inst, xxxix. 440. 5 Cailli6, Travels through Central Africa to Timbuctoo, i. 348. 6 Ellis, Ebe-speaking Peoples of the Slave Coast, p. 199. 7 Weule, Native Life of East Africa, p. 306. 8 Hamalainen, op. cit. pp. 28-31, 103-105, 145, 186 sqq. v. Schroeder, Die Hochzeitsgebrauche der Esten und einiger anderer finnisch-ugrischer Volkerschaften, p. 27 sqq. Porthan, ' Anmark- ningar rorande Finska Folkets lage och tillstand,' in Kongl. Vitterhets, Historic och Antiquitets Academiens Handlingar, iv. 19. Topelius, De modo matrimonia jungendi apud Fennos quondam vigente, § 8 sqq. Castrdn, in Litterara Soirder, 1849, p. 13. XXIII MARRIAGE BY CONSIDERATION, ETC. 381 150 roubles, besides cattle and fowls j1 among the Mordvin to 200 roubles, besides com, clothes, and food ;2 among the Cheremiss even to 500 roubles.3 A rich Ostyak girl is not married without a gift of one hundred reindeer and an assortment of all kinds of furs.4 Among the Samoyed the kalym generally consists of a variety of clothes, household necessaries, reindeer, and small articles purchased from the Russians ;5 among the Tungus, of reindeer ;6 among the Buryat, of cattle and malykh, that is, calves still unborn.7 Rich Baskir pay sometimes even 3,000 roubles for a wife, but the poorest may buy one for a cart load of wood or hay.8 Among the aboriginal tribes of India the consideration given for a bride varies greatly. Sometimes the rate is very high ; the Lushais have been known to give as much as Rs.200 for their wives.9 Among the Morons of Assam a price of some Rs.20-100 is paid to the bride's parents, and a pearl is given both to them and to other relatives of the girl.10 Among the Kis an s of the Central Provinces " two baskets of rice and a rupee in cash constitute the com- pensatory offering given to the parents of the girl."11 Among the Mishmis a rich man gives for a wife twenty mithuns (a kind of oxen), but a poor man can get one for a pig.12 Among the Kunnuvans, a hill tribe of the Palnis in South India, the bride price is invariably fixed at Rs. 10^ in all cases, whether the parties be rich or poor.13 In the Indian Archipelago the bride price is paid in articles of value or in money.14 Among the Battas of Sumatra, for instance, an arm-ring or some other valuable object is given to the father of the bride.15 In Tenimber the price 1 Hamalainen, op. cit. p. 145. 2 Ibid. p. 29. 3 Ibid. p. 103 sq. 4 Miss Czaplicka, op. cit. p. 126. 6 Ibid. p. 124. 6 Ibid. p. 106. 7 Ibid. p. 118. 8 VAmbery, op. cit. p. 505. 8 Gait, Census of India, 1911, vol. i. (India) Report, p. 257. 10 Endle, op. cit. p. 89. 11 Dalton, op. cit. p. 132. 12 Griffith, Journals of Travels in Assam, Burma, Bootan, Afghani- stan and the Neighbouring Countries, p. 35. 13 Dahmen, ' Kunnuvans or Mannadis,' in Anthropos, v. 326. 14 Wilken, ' Plechtigheden en gebruiken bij verlovingen en huwelijken bij de volken van den Indischen Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, xxxv 209 sqq. 15 Ibid, p 210. 382 THE HISTORY OF HUMAN MARRIAGE chap. consists of gold ear-rings and elephants' tusks, which are indispensable.1 So also in Timorlaut " no wife can be purchased without elephants' tusks."2 Among the Subanu of Mindanao the bride price, or laxa, is generally paid in the form of cloth, Chinese jars, and brass gongs, seldom in money.8 In British New Guinea arm shells made from the shell of Conus millepunctatus, pearl shells, dogs' teeth and other ornaments, and pigs figure prominently among the presents given for a bride.4 From Dutch New Guinea we hear of tobacco as the most valued present.6 In the Bismarck Archipelago the bride price varies between fifteen and two hundred strings of shell money.6 In Florida, in the Solomon Group, it consists of " from fifty to a hundred rongo, coils of native money " ;7 in the Banks Islands of money and pigs;8 at Lepers' Island of pigs and mats.9 In Samoa the gifts of the bridegroom included canoes, pigs, and foreign property of any kind which might fall into the hands of the natives.10 In the Caroline Islands " the man 1 Anna Forbes, Insulinde, p. 170. 2 Forbes, ' On the Ethnology of Timor-laut,' in Jour. Anthr. Inst. xiii. n. 3 Finley and Churchill, op. cit. p. 29 sq. 4 Seligman, op. cit. pp. 77, 78, 89, 267 n. 2, 268, 506. Williamson, Mafulu Mountain People of British New Guinea, p. 173. 6 MOszkowski, ' Die Vblkerstamme am Mamberamo in Hollan- disch-Neuguinea,' in Zeitschr. f. Ethnol. xliii. 322. • Pfeil, Studien und Beobachtungen aus der Sudsee, p. 28. Thurn- wald, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 12 (Buin). . . ' Codrington, Melanesians, p. 238. 8 Ibid. p. 239. 9 Ibid. p. 242.-For marriage by consideration in Melanesia see also Romilly, Western Pacific and New Guinea, p. 27, and Powell, Wanderings in a Wild Country, p. 84 (New Britain); Elton, ' Notes on Natives of the Solomon Islands,' in Jour. Anthr. Inst. xvii. 95 ; Guppy, Solomon Islands, p. 45 ; Penny, Ten Years in Melanesia, p. 93 (Solomon Islands); Thurnwald, Forschungen auf den Salomo- Inseln und dem Bismarck-Archipel, iii. 18; Ribbe, Zwei Jahre unter den Kannibalen der Salomo-Inseln, p. 270; Macdonald, Oceania, p. 194, and Meinicke, Die Inseln des Stillen Oceans, i. 203 (New Hebrides); Robertson, Erromanga, p. 396; Moncelon, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. ix. 367 (New Caledonia); Thomson, Fijians, p. 201 sq., and Wilkes, Narrative of the United States Exploring Expedition, iii. 92 (Fiji). 10 Turner, Samoa, p. 93. Stair, Old Samoa, 172 sq. XXIII MARRIAGE BY CONSIDERATION, ETC. 383 makes a present to the father of the girl whom he marries, consisting of fruits, fish, and similar things." 1 Among some tribes the bride price is uniformly fixed by custom,2 but more frequently it varies according to the circumstances. It is greatly influenced by the rank and wealth of the families of the parties or of one of them.3 Sometimes each family has an unalterable long-established price for all its maidens,4 or the sum demanded for a bride is the same as was paid for her mother.6 In many cases the bride price is influenced by the personal qualities of the 1 v. Kotzebue, Voyage of Discovery into the South Sea and Beering's Straits, iii. 210. 2 Theal, op. cit. p. 219 ; Viehe, in Steinmetz, op. cit. p. 307 (Herero). Dahlgriin, * Heiratsgebrauche der Schambaa,' in Mittheil. Deutsch. Schutzgeb. xvi. 228. Dahmen, ' Kunnuvans or Mannadis, a Hill- Tribe of the Palnis, South India,' in Anthropos, v. 326. Crooke, Tribes and Castes of the North-Western Provinces and Oudh, i. p. ccv. 3 Sproat, Scenes and Studies of Savage Life, p. 97 (Nootka). Shooter, Kafirs of Natal and the Zulu Country, p. 50. Andersson, Notes on Travel in South Africa, p. 236 (Ovambo). Fiilleborn, op. cit. p. 344 (Konde people in the former German East Africa). Torday and Joyce, Les Bushongo, p. 113. Dennett, Notes on the Folklore of the Fjort {French Congo), p. 20. Nassau, Fetichism in West Africa, p. 7. Ellis, Ew>speaking Peoples of the Slave Coast, p. 199. Idem, Yoruba-speaking Peoples of the Slave Coast, p. 182. Nachtigal, Sahara und Sudan, i. 448 (Teda) ; ii. 177 (Baele). Mun- zinger, Ostafrikanische Studien, p. 240 (Marea). Georgi, op. cit. p. 431 (Buryat). Hamalaine 1, op. cit. p. 145 (Votyak). Neumann, Russian! und die Tscherkessen, p. 117 (Circassians). Scott Robertson, Kdfirs of the Hindu-Kush, p. 535. Hutchinson, op. cit. p. 97 (Chuk- mas). Shakespear, Lushei Kuki Clans, p. 51. Hodson, Naga Tribes of Manipur, p. 89 sq. (Tankhuls). Rowlatt, ' Report of an Expedition into the Mishmee Hills,' in four. Asiatic Soc. Bengal, vol. xiv. pt. ii. 488. Cole, loc. cit. p. 101 (Bagobo of Mindanao). Vetter, inNachrichtenuber Kaiser-Wilhelms-Land, 1897, p. 90 (Jabim). Pfeil, op. cit. p. 28 (natives of the Bismarck Archipelago). Ribbe, op. cit. p. 141; Penny, op. cit. p. 93 (Solomon Islanders), v. Kotzebue, op. cit. iii. 210 (Caroline Islanders). Post, Grundriss der ethnologischen furisprudenz, i. 292 sq. 4 Annandale and Robertson, Fasciculi Malayenses, ii. 74 (Patam Malays). 6 Rose and Brown, ' Lisu (Yawyin) Tribes of the Burma-China Frontier,' in Memoirs Asiatic Soc. Bengal, iii. 263. 384 THE HISTORY OF HUMAN MARRIAGE chap. girl, such as her beauty, strength, and ability.1 Among some Ibo-speaking people of Nigeria a tall, well-grown girl with a smooth skin and other desirable attributes may fetch from £25 to £40, whereas the father of an ugly girl may be glad to get three goats for his daughter ;2 and among the Californian Karok a girl belonging to an aristocratic family who is pretty and skilful in making acorn-bread and weaving baskets may cost two strings of dentalium shell, though other girls may be obtained for half a string.3 A virgin or a girl generally, though not always,4 commands a better price than a widow or a repudiated wife.5 But on the other hand a widower or an elderly man marrying a young girl may also have to pay much more than the usual amount for his wife.6 Among the Mordvin a higher price is paid for a girl below the age of twenty than for one above that age.7 The Baniyas of India, again, pay no bride price for 1 v. Weber, op. cit. ii. 215 sq. (Kafirs). Wiese, ' Beitrage zur Geschichte der Zulu im Norden des Zambesi, namentlich der Angoni,' in Zeitschr. f. Ethnol. xxxii. 191. Majerus, ' Brautwerbung und Hochzeit bei den Wabende (Deutsch-Ostafrika),' in Anthropos, vi. 894 sq. Dalton, op. cit. p. 43 (Nagas). Borheck, Erdbeschreibung von Asien, i. 540 (Tartars of Kazan). Lansdell, Through Siberia, ii. 225 (Gilyak). Hamalainen, op. cit. p. 145 (Votyak). Post, Grundriss der ethnologischen Jurisprudenz, i. 295. 2 Thomas, Anthropological Report on Ibo-speahing Peoples of Nigeria, i. 63. 3 Powers, op. cit. p. 22. 4 Hurel, ' Religion et Vie domestique des Bakerewe,' in Anthropos, vi. 287. Merker, op. cit. p. 46 (Masai). Supra, ii. 33. 5 Steinmetz, op. cit. p. 34 (Banaka and Bapuku of the Cameroons). Tellier, ibid. p. 152 (Bambara). Nassau, op. cit. p. 7 (West Africans). Felkin, ' Notes on the Madi or Moru Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xii. 323. Kovalewsky, Coutume contemporaine et loi ancienne, p. 163 (Ossetes of Caucasia). Gait, Census of India, 1911, vol. i. (India) Report, p. 257 (aboriginal tribes). Finley and Churchill, op. cit. p. 40 (Subanu of Mindanao). Post, Die Anfange des Staats- und Rechtslebens, p. 41 sq. Idem, Grundriss der ethnologischen Jurisprudenz, i. 293 sq. 6 Bainbridge, ' Saorias of the Rajmahal Hills,' in Memoirs Asiatic Soc. Bengal, ii. 57. Shortt, Hill Ranges of Southern India, ii. 40 (Malliallies, or hill men inhabiting the Shevaroys in the Salen district). 7 Hamalainen, op. cit. p. 30. xxin MARRIAGE BY CONSIDERATION, ETC. 385 a girl up to the age of eight, but after that payment is made at the rate of Rs.100 for every year of her age up to thirteen, which is regarded as the age of puberty.1 The consideration for a bride is in most cases given to her father, but he may have to share it with other members of the family or relatives of the girl, or special presents are given to them ;2 or the bride price, or some of it, is 1 Gait, Census of India, 1911, vol. i. (India) Report, p. 257. 2 Bancroft, Native Races of the Pacific States of North America, i. 92 (Aleut). Turner, in Ann. Rep. Bur. Ethn. xi. 270 (Hudson's Bay Indians). Powers, op. cit. p. 270 (Achomawi of Northern Cali- fornia). Loskiel, History of the Mission of the United Brethren among the Indians in North America, i. 57 (Delaware, &c.). La Flesche, ' Osage Marriage Customs,' in American Anthropologist, N.S. xiv. 128. de Alcedo, Geographical and Historical Dictionary of America and the West Indies, i. 416 ; Poppig, Reise in Chile, i. 383 sq. (Arau- canians). Kropf, Das Volk der Xosa-Kaffern im ostlichen Sudafrika, pp. 134, 145. v. Eberstein, * Ueber die Rechtsanschauungen der Kustenbewohner des Bezirkes Kilwa,' in Mittheil. Deutsch. Schutzgeb. ix. 181. Velten, Sitten und Gebrduche der Suaheli, p. 109. Kraft, in Steinmetz, op. cit. p. 286 (Wapokomo). Dundas, ' Notes on the Tribes inhabiting the Baringo District, East Africa Protectorate,' in four. Roy. Anthr. Inst. xl. 68 (Turkana). Dennett, At the Back of the Black Man's Mind, p. 39 (Bavili). Nassau, op. cit. p. 7 (West Africans). Spieth, Die EHje-Stamme, p. 62 ♦. Steinmetz, op. cit. p. 34 (Banaka and Bapuku). Baskerville, ibid. p. 18g ; Roscoe, Baganda, p. 89 (Baganda). Munzinger, Ostafrikanische Studien, p. 487 (Barea and Kunama). Kovalewsky, op. cit. p. 175 sq. (Ossetes of the Caucasus). Hamalainen, op. cit. p. 190 (Baltic Finns). Pallas, Merkwurdigkeiten der obischen Ostjaken, Samoyeden, &c., p. 66 (Samoyed). Miss Czaplicka, op. cit. pp. 100, 108, 124 (Gilyak, Yakut, Samoyed). Sumner, ' Yakuts,' from the Russian of Siero- shevski, in four. Anthr. Inst. xxxi. 85. Jochelson, Yukaghir, p. 95 (Yukaghir of the tundra}. Jagor, op. cit. p. 172 (Igorot of Ysarog). Cole, loc. cit. p. 102 (Bagobo of Mindanao). Seligman, Melanesians of British New Guinea, pp. 76, 77, 363 n. 1, 710 (Koita, Mekeo tribes, Southern Massim). Hagen, Unter den Papua's, p. 224 (natives of Bogadjim). Moszkowski, ' Die Volkerstamme am Mamberamo in Hollandisch-Neuguinea und auf den vorgelagerten Inseln,' in Zeitschr. f. Ethnol. xliii. 322. Thurnwald, ' Ermittlungen uber Eingeborenenrechte der Siidsee. A. Buin auf Bougainville (Deutsche Salomo-Inseln),' in Zeitschr. f. vergl. Rechtswiss. xxiii. 337 ; Idem, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 13, 18 (natives of Buin). Gason, ' Of the Tribes, Dieyerie, &c., in Jour. Anthr. Inst. xxiv. 169 (Dieri) VOL. II c C 386 THE HISTORY OF HUMAN MARRIAGE chap. paid to the whole clan of the bride.1 Her mother, in par- ticular, is a frequent recipient.2 In some cases the con- sideration, or a large part of it, is given to the maternal uncle,3 whilst in other cases he is specially mentioned among 1 Martin, Reisen in den Molukken, p. 289 (Waepote in Burn). 2 Francois, op. cit. p. 214 (Hottentots). Decle, Three Years in Savage Africa, pp. 232, 233, 347 (natives of the Portuguese Zambesi, Wanyamwezi). v. Eberstein, in Mittheil. Deutsch. Schutzgeb. ix. 181 (natives of Kilwa in the former German East Africa). Dahlgriin, ibid. xvi. 223 (Shambaa). Velten, op. cit. p. 109 (Swahili). Hollis, Masai, p. 302 ; Merker, op. cit. p. 45 (Masai). Hobley, Ethnology of A-Kamba, p. 63. Dundas, in Jour. Roy. Anthr. Inst. xl. 68 (Turkana of the Baringo district in the East Africa Protectorate). Paulitschke, Ethnographic Nordost-Afrikas, ii. 142 (Galla). Talbot, * Buduma of Lake Chad,' in Jour. Roy. Anthr. Inst. xli. 247. Thomas, Ibo-speaking Peoples of Nigeria, i. 63. Wilson, Western Africa, p. 113 (Kru people on the Grain Coast). Weeks, Among the Primitive Bakongo, p. 143. Grandidier, Ethnographic de Madagascar, ii. 185 (Betsimisaraka). Kovalewsky, op. cit. p. 163 sq. (Ossetes). Hamalainen, op. cit. p. 104 (Cheremiss). Miss Czaplicka, op. cit. p. 124 (Samoyed). Jochelson, Koryak, p. 743. Sherring, ' Notes on the Bhotias of Almora and British Garhwal,' in Memoirs Asiatic Soc. Bengal, i. 108 (Bhotias of the Darma pargana). Sarat Chandra Roy, Ordons of Chotd Nagpur, p. 365. Pardo de Tavera, ' Las costumbres de los tagalos de Filipinas segtin el Padre Plasencia,' in Revista contempordnea, Ixxxvi. 466 sq. n. 1. Worcester, Philippine Islands and their People, p. 493 (Tagbanuas of Culion and Busuanga). Sorge, in Steinmetz, op. cit. p. 406; Pfeil, op. cit. p. 28 (natives of the Bismarck Archipelago). Mrs. Langloh Parker, Euahlayi Tribe, p. 58. 3 Weeks, Among the Primitive Bakongo, p. 142 sq.Negersitten,' in Das Ausland, liv. 1026 (Negroes of Bondo). Munzinger, op. cit. p. 528 (Barea and Bazes). Neuhauss, Deutsch Neu-Gziinea, i. 160. Pfeil, op. cit. p. 28 (natives of the Bismarck Archipelago). Schnee, Bilder aus der Siidsee, p. 99 (natives of the Gazelle Peninsula of New Britain). Kubary, 'Die Bewohner der Mortlock Inseln,' in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, p. 262. Among the Goajiros of Colombia the maternal uncles are, according to one account (Nicholas, ' Aborigines of the Province of Santa Marta, Colombia,' in American Anthropologist, N.S. iii. 648), supposed to take charge of the bride price, whereas according to another account (Simons, in Proceed. Roy. Geograph. Soc. N.S. vii. 792) it is fixed by the father, who appropriates the best part of it for himself and his relations, the rest going to the mother's relatives. xxm MARRIAGE BY CONSIDERATION, ETC. 387 the recipients.1 The brothers of the bride may also play a prominent part in the transactions.2 Among the Tehuelches of Patagonia " a girl's value often depends upon the number of her brethren, who must receive two horses apiece." 3 In the southern Chin Hills the eldest brother takes the largest share, whilst " the remainder is divided amongst the parents, sisters, brothers, cousins, uncles, aunts, and the Chiefs of the tribe : even the slaves of the house expect presents."4 Among the Jabim of New Guinea the maternal relatives divide the bride price between themselves,6 in Ufipa in the former German East Africa they receive one half of it,6 among the Wabende they receive one third.7 At Bogadjim in the former German New Guinea, when a girl marries into another village, something must be paid not only to the bride's parents or family, but also to the other members of her village.8 In the Roro-speaking tribes of British New Guinea the bride's father shares the bride price among his relatives and the members of his local group.9 Among the Bayaka of the Congo her father must pay a goat to his chief, because the bride goes out of the village.10 Among the Turkana of the Baringo district in the East Africa Protectorate the bridegroom must give five head of cattle to each of his other wives.11 Among the Bagobo of Mindanao a man, though he serves for his first wife, renders no services to the father of the second wife, 1 Junod, Life of a South African Tribe, i. 253 (Thonga). Stigand, * Notes on the Natives of Nyassaland, &c.,' in Jour. Roy. Anthr. Inst, xxxvii. 122. Miss Werner, Natives of British Central Africa, p. 131. Kovalewsky, op. cit. p. 163 sq. (Ossetes). Ranga Rao, * Y4nidis of the Nellore District,' in the Madras Government Museum's BiMetin, iv. 98. Dalton, op. cit. p. 150 (Savaras). Shakespear, op. cit. p. 155 (Old Kuki clans). 2 Lumholtz, Unknown Mexico, i. 270 (Tarahumare). Shakespear, op. cit. p. 154 sq. (Old Kuki clans). 3 Prichard, op. cit. p. 93. 4 Carey and Tuck, Chin Hills, i. 190. 6 Vetter, loc. cit. p. 90. 4 Fromm, in Mittheil. Deutsch. Schutzgeb. xxv. 97. 7 Majerus, in Anthropos, vi. 895. 8 Hagen, op. cit. p. 224. 9 Seligman, op. cit. p. 268. 10 Torday, Camp and Tramp in African Wilds, p. 134. 11 Dundas, in Jour. Roy. Anthr. Inst. xl. 68. C C 2 388 THE HISTORY OF HUMAN MARRIAGE chap. but instead of this pays a double price for her; " for he not only pays her parents, but is forced also to give a like sum to his first wife, who, in turn, presents it to her father." And should a third wife be added to the family, a sum equal to her cost is divided among the earlier wives.1 Among some of the Indians of British Columbia the marriage presents, although nominally given to the parents of the girl, were never retained or used by them, but were divided among the girl's relatives ;2 whilst among the Tlingit the girl's father distributed among her mother and relatives everything received for her except a small amount reserved for himself.3 On the other hand, the paying of the bride price may, to some extent at least, devolve on other persons than the bridegroom or his father.4 Among the coast people of the Gazelle Peninsula in New Britain, if the young man cannot pay it himself, he asks his maternal uncle to help him.5 Among the Central African Madis the young man " collects together the cattle to be given for the bride. His friends all make him presents, the most substantial help coming from his father, mother, and father's brothers."6 Among the Toradjas of Central Celebes 7 and in Ceram 8 and Buru 9 a man's wife is purchased for him by, or with the assistance 1 Cole, loc. cit. p. 103. 1 Teit, ' Thompson Indians of British Columbia,' in Publications of the Jesup North Pacific Expedition, i. 322. Idem, ' Lillooet Indians,' ibid. ii. 267 sq. 3 Swanton, ' Social Condition, Beliefs, and Linguistic Relationship of the Tlingit Indians,' in Ann. Rep. Bur. Ethn. xxvi. 428. 4 Steinmetz, op. cit. p. 34 (Banaka and Bapuku of the Cameroons). Kraft, ibid. p. 286 (Wapokomo). Junod, Life of a South African Tribe, i. 263 (Thonga). Munzinger, Ueber die Sitten und das Recht der Bogos, p. 28. Thurnwald, Forschungen auf den Salomo-Inseln, iii. 12, 18 (natives of Buin). 5 Burger, Die Kristen- und Bergvolker der Gazellehalbinsel, p. 23. G Felkin, in Proceed. Roy. Soc. Edinburgh, xii. 320. 7 Adriani and Kruijt, op. cit. ii. 25. 8 Tauem, * Ceram,' in Zeitschr. f. Ethnol. xlv. 172. 9 Wilken, Bijdrage tot de kennis der Alfoeren van het eiland Boeroe, p. 17 (in Verhandelingen van het Bataviaasch Genootschap van kunsten en wetenschappen, xxxviii.). Martin, Reisen in den Molukken, p. 289. XXIII MARRIAGE BY CONSIDERATION, ETC. 389 of, his whole clan. Among the Mekeo tribes of British New Guinea the bride price is provided by contributions from both the paternal and maternal relatives of the bride- groom.1 Among the Roro-speaking tribes of the same colony, when a youth is of an age to be married, his father asks the members of his local group to help him to collect the bride price for his son.2 Among the Altaians, " to enable the bridegroom to pay the kalym, his bachelor friends help him by making each a small offering from his store." 3 The consideration for a wife may be offered by instal- ments.4 Among some Siberian peoples the man is allowed to have sexual intercourse with his future wife as soon as he has paid a certain portion of the kalym.5 In other cases the marriage may take place before the bride price has been paid in full.6 Among some West Africans the price is supposed to be completed in one or two years after the marriage,7 whereas among certain other tribes the payment 1 Seligman, op. cit. p. 363 n. 1. 2 Ibid. p. 267. Cf. ibid. p. 77 (Koita). 3 Miss Czaplicka, op. cit. p. 117. 4 Decle, op. cit. p. 78 (Barotse). Fiilleborn, op. cit. p. 346 (Konde people), Hobley, Eastern Uganda, p. 17 (Bantu Kavirondo). Gregory, Great Rift Valley, p. 343 (Wapokomo). Partridge, Cross River Natives, p. 254. Rossillon, ' Moeurs et Coutumes du peuple Kui, Indes Anglaises,' in Anthropos, vii. 100 sq. (Kandhs). Hagen, op. cit. p. 242 (people of Bogadjim in New Guinea). Speiser, Two Years with the Natives in the Western Pacific, p 234 (natives of Pentecost of the New Hebrides). 6 Miss Czaplicka, op. cit. pp. 105, 107, 108, 126 sq. (Tungus, Yakut, Ostyak). Hamalainen, op. cit. pp. 187, 194 (Ostyak, Vogul, Kazak-Kirghiz). 8 Kropf, op. cit. p. 145 ; Kidd, op. cit. p. 215 (Kafirs). Avon, ' Vie sociale des Wabende au Tanganika,' in Anthropos, x.-xi. 100. Hurel, ibid. vi. 288 (natives of Ukarewe, an island in the Victoria Nyanza). Hobley, Eastern Uganda, p. 28 (Nilotic Kavirondo). Idem, Ethnology of A-Kamba, p. 62 ; Lindblom, op. cit. p. 71 (Akamba). Steinmetz, op. cit. p. 34 (Banaka and Bapuku of the Cameroons). Shortt, op. cit. ii. 39 sq. (Malliallies of the Shevaroys). Rowney, Wild Tribes of India, p. 139 (Lepchas). Shakespear, op. cit. p. 52 (Lushais). v. Brenner, Besuch bei den Kapnibalen Sumatras, p. 248 (Battas of Sumatra). 7 Nassau, op. cit. p. 7. 390 THE HISTORY OF HUMAN MARRIAGE chap. of it may extend over many years. Among the Akikdyu the marriage will probably take place after twenty goats have been received, but the remainder is sometimes not paid over till the eldest child is eight or ten years of age.1 In the southern Chin Hills the debt usually hangs over the husband for the rest of his life, and " it is by no means rare to find men quarrelling over the still unpaid portion of the marriage price of their grandmothers and other female ancestors." 2 But it may be that until the bride price is fully paid the husband is little else than a drudge to his wife's male relatives ;3 or the parents may have a right to take back their daughter and consequently prefer that a portion of the price remains unpaid.4 In Timorlaut, again, so long as the bridal price is not paid in full, the wife is entitled to stay with her parents and is not completely subject to her husband, nor does the latter possess a right to the children.5 In Tenimber " the father of the girl has often to wait a long time for the ivory portion of her price ; but he hands her over, on the payment of the other items of the bargain, to her purchaser, who takes up his abode in her house, where she and her children remain as hostages till the full price is paid."6 Among various other peoples neither the wife nor her children would be allowed to leave her father's house until the bride price has been paid in full.7 Among the Central African Banyoro, according to Emin Pasha, when a poor man is unable to procure the cattle required for his marriage at once, he may, by agree- ment with the bride's father, pay them by instalments ; but the children born in the meantime belong to the wife's father, and each of them must be redeemed with a cow.8 1 Routledge, op. cit. p. 125. 2 Carey and Tuck, op. cit. i. 190 sq. 3 Robertson, Erromanga, p. 396. 4 Kropf, op. cit. p. 145 ; Kidd, op. cit. p. 215 (Kafirs). Marx, in Steinmetz, op. cit. p. 351 (Amahlubi). Hurel, in Anthropos, vi. 288 (Bakerewe). 6 Riedel, op. cit. p. 301. 8 Anna Forbes, op. cit. p. 170 sq. 1 Theal, op. cit. p. 220 n. * (Makaranga). Hildebrandt ' Ethno- graphische Notizen fiber Wakamba,' in Zeitschr. f. Ethnol. vii. 401. Scott Robertson, op. cit. p. 535 (Kafirs of the Hindu-Kush). Cooper, op. cit. p. 236 sq. (Mishmis). Vetter, loc. cit. p. 90 (Jabim of New Guinea). 8 Emin Pasha in Central Africa, p. 86. XXIII MARRIAGE BY CONSIDERATION, ETC. 391 Among the Zulus 1 and the Bakundu of the Cameroons 2 the issue of the marriage can be taken as mortgage until the payment agreed upon has been made. Among some tribes the bride price, like the service for a wife, gives the husband no claim over the offspring, for which a special payment has to be made ; or he may have to pay for his children although he paid no price for his wife. Thus among the Takelma Indians of South-Western Oregon " after the birth of the first baby an additional price was paid to the girl's father in the shape of a deerskin sack filled with Indian money. This payment was con- sidered as equivalent to the buying of the child and was metaphorically referred to as ' making its pillow.' "3 Among the Sakalava of Madagascar, the marriage feast being over, the young husband makes a present of an ox to his wife's parents and a further payment of four yards of cloth or a large bag of rice to each of her nearest relatives. " These must be presented before his wife gives birth to her first child, as they are regarded as the payment necessary to secure the child for himself, and, if not made in proper time, he loses his right to be considered the father of the child, which then belongs to his father-in-law and mother- in-law."4 Among the Shambaa in Usambara, when the wife becomes with child for the first time, the husband must give to his mother-in-law a goat, and, when the child is born, either a cow or five goats ; and should he fail to do so, the wife is taken back by her father.5 The Matabele, according to Mr. Decle, " do not buy the wife from her father,6 but 1 Tyler, Forty Years among the Zulus', p. ug. Grant, ' Magato and his Tribe,' in Jour. Anthr. Inst. xxxv. 270. 2 Bufe, ' Die Bakundu,' in Archiv f. Anthrop. N.S. xii. 236. 3 Sapir, ' Notes on the Takelma Indians of Southwestern Oregon,' in American Anthropologist, N.S. ix. 275. 4 Walen, ' Sakalava,' in Antananarivo Annual, 1884, p. 53 sq. 5 Storch, ' Gebrauche und Rechtspflege bei den Bewohnern Usambaras und Pares,' in Mittheil. Deutsch. Schutzgeb. viii. 312. Dahlgriin, ibid. xvi. 224 sq. 6 When the girl's father has given permission for the marriage to take place, however, the bridegroom kills an ox or a sheep, takes part of it to the town of the intended father-in-law, stops outside, and shouts, " Here is meat for your child." 392 THE HISTORY OF HUMAN MARRIAGE chap. after the first child is born the husband has to pay its value, or else the wife's father has the right to take the child away."1 Among various other African peoples, again, the paying of the bride price not only gives the husband a right to his children as well as to his wife, but, in case she dies before bearing a child, entitles him to reclaim the bride price 2 or to demand another woman, generally her sister, as a substi- tute.3 These two methods of compensation are sometimes represented as alternatives.4 In other cases it is said that if the wife dies soon after her marriage, her sister will take her place 5 or the bride price or part of it is to be restored.6 And the same may be the case if the wife proves barren ; 7 or the husband may stop his payments if his wife does not bear a child quickly.8 Where the paying of a bride price is a regular custom it would be considered highly disgraceful to a girl and her family if she were given in marriage for nothing.9 Ac- cording to Yakut ideas, " it would mean that she was not worth any price, was friendless, or an outcast. It can be understood, therefore, that the Yakut women look down 1 Decle, op. cit. p. 158. 2 Kropf, op. cit. p. 135 (Xosa Kafirs). Hollis, Nandi, p. 68. Ward, Voice from the Congo, p. 251. Torday, op. cit. p. 134 sq. (if the wife dies before giving birth to a girl). 3 Arbousset and Daumas, Narrative of an Exploratory Tour to the North-East of the Colony of Good Hope, p. 68 (Manteti). 4 Fiillebom, op. cit. p. 348 (Konde people). Weeks, ' Notes on some Customs of the Lower Congo People,' in Folk-Lore, xix. 413. Hobley, Eastern Uganda, pp. 18, 29 (Bantu and Nilotic Kavirondo ; among the latter, however, the man has to pay a reduced bride price for the sister of his deceased wife). 6 Northcote, * Nilotic Kavirondo,' in Jour. Roy. Anthr. Inst. xxxvii. 62. Tessmann, Die Pangwe, ii. 263. Dixon, Chimariko Indians and Language, p. 301. 6 Northcote, in Jour. Roy. Anthr. Inst, xxxvii. 62 (Nilotic Kavi- rondo). Torday and Joyce, Les Bushongo, p. 116. 7 Stuhlmann, Mit Emin Pascha ins Herz von Afrika, p. 81 (Wa- nyamwezi). Northcote, in Jour. Roy. Anthr. Inst, xxxvii. 62 (Nilotic Kavirondo). 8 Hobley, Eastern Uganda, p. 29. 9 Bancroft, op. cit. i. 277 (Indians of Columbia). Williamson, ' Some unrecorded Customs of the Mekeo People of British New Guinea,' in Jour. Roy. Anthr. Inst, xliii. 275. Vetter, loc. cit. p. 90 (Jabim). xxni MARRIAGE BY CONSIDERATION, ETC 393 upon the Russian women, who, as they say, pay somebody to take them."1 The Kafir women pour scorn and contempt on one who has not been duly bought by cattle ; they call such a woman an old cat, because the cat is the only animal that the natives consider unworthy of being sold.2 A woman for which no price has been paid, although she lives with a man, is regarded as a harlot,3 and her children are accounted no better than bastards, as children without a father.4 But she is a proud woman for whom a larger sum than usual is paid.5 Like other forms of marriage by consideration, the giving of material objects for a bride is in the first place due to the unwillingness of the person or persons who have a right to dispose of a girl's hand to part with her for nothing, and to the readiness of a man to give something in order to obtain a wife. This form of marriage has generally been called " marriage by purchase " ; but in many cases there is no justification at all for such a term, and in others it may be used only if it is understood that girls are not sold by their relatives like chattels. The gift may be an ex- pression of good-will or respect on the part of the bride- groom.6 It may be a proof of his ability to keep a wife.7 It may serve as a protection to the wife against ill usage and to the husband against misbehaviour on the side of the wife. Among the Hidatsa Indians, according to Mr. 1 Sumner, in Jour. Anthr. Inst. xxxi. 85 (Yakut). 2 Kidd, op. cit. p. 224. Cf. v. Weber, op. cit. ii. 215 sq. 3 Nassau, op. cit. p. 6 (West Africans). ♦ Powers, op. cit. pp. 22, 56 ; Goddard, op. cit. p. 56 (Californian tribes). Adriani and Kruijt, op. cit. ii. 25 (Toradjas of Central Celebes). 6 Weeks, Among the Primitive Bakongo, p. 142. Kidd, op. cit. p. 220 sq. (Kafirs). Roscoe, Northern Bantu, pp. 39, 209 (Banyoro, Busoga). Miss Czaplicka, op. cit. p. 101 (Gilyak). 6 Grinnell, Story of the Indian, p. 41 sq. (Pawnee). Kropf, op. cit. p. 140 (Xosa Kafirs). Fromm, in Mittheil. Deutsch. Schutzgeb. xxv. 97 (Wafipa). Paulitschke, op. cit. ii. 142 (Somal and Danakil). Dalton, op. cit. p. 28 (Padams). 7 v. Martius, op. cit. i. 109 sq. (various Brazilian tribes). Heese, Sitte und Brauch der Sango,' in Archiv f. Anthrop. N.S. xii. 137. Miss Werner, op. cit. p. 129 (Zulus). 394 THE HISTORY OF HUMAN MARRIAGE chap. Matthews, the gift "is a pledge to the parents for the proper treatment of their daughter, as well as an evidence of the wealth of the suitor and his relations.''1 Among the Xosa Kafirs the bride price makes the father take an interest in the behaviour of his daughter because, if she is divorced by her husband for a good reason, the bride price has to be returned ; whereas if the marriage is dissolved owing to ill treatment of the wife, the husband may lose what he paid for her.2 Among the Warega of the Belgian Congo " le mariage n'est pas un achat ; les biens que le fiance donne aux parents de la fiancee ne constituent qu'une caution que les beaux-parents doivent toujours etre en etat de restituer en cas de divorce."3 In many cases the price is said to be a compensation for the loss sustained in the giving up of the girl or a remuneration for the expenses incurred in her maintenance till the time of her marriage ; 4 and the gift to her mother is regarded as a reward for nursing her, as a" price of the mother's milk," 5 or sometimes, 1 Matthews, op. cit. p. 52. 2 Kropf, op. cit. p. 140. 3 Delhaise, Les Warega, p. 173. ' Paulitschke, op. cit. ii. 142 (Galla). Park, Travels in the Interior of Africa, p. 220 (Mandingo). Connolly, ' Social Life in Fanti- Land,' in Jour. Anthr. Inst. xxvi. 144. Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 182. Merolla da Sorrento, ' Voyage to Congo,' in Pinkerton, Collection of Voyages and Travels, xvi. 235 (Negroes of Sogno). Shooter, Kafirs of Natal and the Zulu Country, p. 49. Junod, op. cit. i. 262 (Thonga). Zache, ' Sitten und Gebrauche der Suaheli,' in Zeitschr. f. Ethnol. xxxi. 78. Hamalainen, op. cit. pp. 29, 190 (Mordvin, Baltic Finns). Castren, Nordiska resor och forskningar, iv. 126 (Ostyak). Ahlqvist, Die Kulturworter der westfinnischen Sprachen, p. 203 (Turkish and Finnish peoples). Vambery, Das Tiirkenvolk, p. 230 (Central Asiatic Turks). Forsyth, Highlands of Central India, p. 148 (Gonds). McNair, Perak and the Malays, p. 232. Carey and Tuck, Chin Hills, i. 189. Jagor, op. cit. p. 235 (Bisayans). d'Albertis, New Guinea, i. 395, 396, 414 (natives of Naiabui and Yule Island). 5 Francois, op. cit. p. 214 (Hottentots), v. Eberstein, in Mittheil. Deutsch. Schutzgeb. ix. 181 (natives of Kilwa in the former German East Africa). Grandidier, op. cit. ii. 185 (Betsimisaraka of Mada- gascar). Das, ' Marriage Customs of Tibet,' in Jour. Asiatic Soc. Bengal, vol. Ixii. pt. iii. 17, 21. Young, Kingdom of the Yellow Robe, p. 94 (Siamese). Pardo de Tavera, in Revista contempor&nea, Ixxxvi. 466 sq. n. 1 (Tagals of the Philippines in former times). XXIII MARRIAGE BY CONSIDERATION, ETC. 395 it seems, as a reward for guarding her virginity, as " the price of the virgin."1 Mercenary motives may be prominent: daughters may be bartered away to the highest bidders and be trained for the purpose of fetching a high price. But whatever be the reason for claiming a consideration for a bride and whether the consideration be large or small, it does not confer on the husband the right to do with the wife whatever he may please. He can only " buy " the rights which custom grants to a husband ; and however great these rights may be, I think we may safely say that they never are quite absolute, and that among no people a married woman is completely at the mercy of her husband.2 Among many African peoples the parents may in certain circumstances take back their daughter on restoration of the bride price.3 It has been suggested that marriage by purchase arose out of marriage by capture. We are told that abduction in spite of parents was the primary form ; then there came the offering of compensation to escape vengeance ; and this grew eventually into the making of presents beforehand.4 In support of this view instances have been quoted of peoples among whom a man who carries off or elopes with a woman afterwards has to pay compensation if she is to become his wife. But these cases merely show that marriage by consideration is the recognised form of marriage, although forcible abduction or elopement may be a preliminary 1 Decle, op. cit. 233 (natives of the Portuguese Zambesi). Miss Czaplicka, op. cit. p. 124 (Samoyed). 2 See Westermarck, Origin and Development of the Moral Ideas, i. ch. 27, ' The Subjection of Wives,' p. 629 sqq. 3 Gouldsbury and Sheane, Great Plateau of Northern Rhodesia, p. 166 (Awemba). Fromm, in Mittheil. Deutsch. Sch/utzgeb. xxv. 97 (Wafipa). Beltrame, Il Fiume Bianco, p. 88 sq. (Shilluk). Thomas, Anthropological Report on the Edo-speaking Peoples of Nigeria, i. 47. Nassau, op. cit. p. 7 (West Africans). Dennett, At the Back of the Black Man's Mind, pp. 38, 40 (Bavili). 4 Koenigswarter, Etudes historiques sur le developpement de la society humaine, p. 53. Spencer, Principles of Sociology, i. 625. In the earlier editions of the present work (p. 401) I expressed myself the opinion that " the transition from marriage by capture to marriage by purchase " was probably brought about in this way. 396 THE HISTORY OF HUMAN MARRIAGE chap. measure leading up to it. They by no means imply that the bride price was originally a ransom. Marriage by con- sideration prevails among a large number of peoples who have never been known to be in the habit of capturing women for wives ; and, as I said in the preceding chapter, marriage by capture cannot be proved to have been the usual form of marriage among any people. . Whilst the rendering of some consideration for the bride undoubtedly is the normal mode of securing marriage at all grades in the uncivilised world,1 it has, generally speak- ing, assumed increasing importance with the advance of economic culture. Messrs. Hobhouse, Wheeler, and Gins- berg observe that the increase is sharper in the pastoral as compared with the agricultural stages, and that " the same relations are even more strongly marked when we dis- tinguish purchase proper from other forms of consideration." 2 Among pastoral peoples marriage by consideration seems to be very nearly universal.3 Of some uncivilised peoples we are told that the commercial view of marriage is one of comparatively modern growth.4 That the giving of a consideration for a bride is not an act of ordinary purchase is also obvious in cases where the bride's people have to present the bridegroom or his people with a return gift. The exchange of presents at a marriage is in fact a very widespread practice,5 and fre- 1 Cf. Hobhouse, Wheeler, and Ginsberg, Material Culture and Social Institutions of the Simpler Peoples, p. 145. 2 Ibid. p. 155. See supra, it 314. 3 Cf. Grosse, Die Formen der Familie und die For men der Wirth- schaft, p. 104. 4 Grinnell, op. cit. p. 41 sq. (Pawnee). Tyler, op. cit. p. 118; Shooter, op. cit. p. 49 (Zulus). 5 Holmberg, ' Ethnographische Skizzen uber die Volker des russischen Amerika,' in Acta Societatis Scientiarum Fennicae, iv. 315 ; Swanton, in Ann. Rep. Bur. Ethn. xxvi. 428 (Tlingit). Idem, Haida, p. 51 sq. Hodge, Handbook of Indians of Canada, p. 275 sq. (Tsimshian, Coast Salish, Bellacoola, Delaware). Loskiel, op. cit. i. 57 (Delaware). Long, Voyages and Travels, p. 174; Keating, Narrative of an Expedition to the Source of St. Peter's River, ii. 15 (Ojibway). Lewis and Clarke, Travels to the Source of the Missouri XXIII MARRIAGE BY CONSIDERATION, ETC. 397 quently the amount of the return gift is fixed by custom. Among the Central African Bahima, when the bridegroom has built a house for himself and his wife, the latter is taken there by her father, who at the same time brings back three out of the ten head of cattle which were paid for her.1 Among the Bila-an of Mindanao, whatever the gift may River, p. 307 (Shoshoni). Farrand, ' Notes on the Alsea Indians of Oregon,' in American Anthropologist, N.S. iii. 243. Sapir, ' Notes on the Takelma Indians of Southwestern Oregon,' ibid. N.S. ix. 274. Stephen, ' Navajo,' ibid. vi. 356. Powers, op. cit. p, 354 (Miwok of California). Morelet, Reisen in Central-Amerika, p. 257 (Quich6). Nicholas, in American Anthropologist, N.S. iii. 647 (Goajiros). Nachtigal, op. cit. ii. 370, 448 (Budduma, Ted&). Hamalainen, op. cit. pp. 30, 104 (Mordvin, Cheremiss). Miss Czaplicka, op. cit. pp. 103, 118, 124 (Ainu, Buryat, Samoyed). Vhmbery, Das Turken- volk, p. 233 sq. (Central Asiatic Turks). Jochelson, Koryak, p. 743 sq. Muller, ' Uber die Wildenstamme der Insel Formosa,' in Zeitschr. f. Ethnol. xlii. 230. Strzoda, ' Die Li auf Hainan,' ibid. xliii. 203. Riedel, op. cit. p. 68 (natives of Amboyna). Finsch, Neu-Guinea, p. 102 (Papuans of Dorey). Seligman, op. cit. pp. 79, 504, 506, 510, 710 sq. (Koita, South Massim of British New Guinea). Williamson, ' Customs of the Mekeo People of British New Guinea,' in Jour. Roy. Anthr. Inst, xliii. 277. Haddon, in Reports of the Cambridge Anthropological Expedition to Torres Straits, v. 231 (Western Islanders of Torres Straits). Parkinson, Dreissig Jahre in der Sudsee, pp. 267, 394 (natives of New Hanover and Northern New Ireland, Admiralty Islanders). Pfeil, op. cit. p. 29 (natives of the Bismarck Archipelago). Burger, Die Kiisten- und Bergvblker der Gazellehalbinsel, p. 27 (coast people). Hahl, ' Ueber die Rechtsan- schauungen der Eingeborenen eines Theiles der Blanchebucht und des Innem der Gazelle Halbinsel,' in Nachrichten uber Kaiser Wil- helms-Land und den Bismarck-Archipel, 1897, p. 78. Thurnwald, Forschungen auf den Salomo-Inseln und dem Bismarck-Archipel, iii. 13 sq. (natives of Buin in Bougainville belonging to the Solomon Islands). Paton, quoted by Serbelov, ' Social Position of Men and Women among the Natives of East Malekula, New Hebrides,' in American Anthropologist, N.S. xv. 278. Rivers, History of Melanesian Society, ii. 105, 107. Turner, Samoa, pp. 93, 96; Schultz, ' The most important Principles of Samoan Family Law,' in Jour. Poly- nesian Soc. xx. 49; Stair, op. cit. p. 172 sq. ; Pritchard, Polynesian Reminiscences, p. 139 sq. (Samoans), v. Langsdorf, Voyages and Travels in various Parts of the World, i. 153 (Nukahivans). Marcuse, Die Hawaiischen Inseln, p. 108. Strehlow, op. cit. vol. iv. pt. i. 99 (Western Loritja in Central Australia). 1 Johnston, Uganda Protectorate, p. 632. 398 THE HISTORY OF HUMAN MARRIAGE chap. be, a return present equal to half its value must be made.1 Among many tribes it is the custom that the consideration given for a wife shall be returned in a present of equal value.2 Among the Pawnee Indians in Nebraska the girl's relatives would give the young man a horse or two, and it was the custom that those who had received horses from the rela- tives of the man should also return horses to him ; but it must be some time after the marriage, and the horses re- turned must not be the same as had been given by the boy's relations. Our informant, Mr. Grinnell, adds that, so far as he can discover, these presents did not in earlier times in any sense constitute a price paid for the girl, but were really wedding gifts, most of which ultimately came back to the young man.3 Among the Stlatlumh of British Columbia custom demands that the presents given by the parents and kinsfolk of the bride shall not be less in value than those made to themselves by the bridegroom and his party.4 We are told that " bartering women for cattle, as now practised in Natal and other parts of South Africa, is not an ancestral custom of the Zulus. Fifty years ago the bridegroom presented the bride's father with three or four cows to ratify the marriage contract, and he received from the bride's relatives an equivalent in cattle or some- thing else."5 Among the Siberian Yukaghir of the tundra, when the newly-married couple visit the bride's people, those of them who have received one reindeer each of the bride price present the couple, each in his turn, with one reindeer; this is called " taking back."6 Among some of the Southern Massim of British New Guinea, when the bride- 1 Cole, loc. cit. p. 144. 2 Teit, ' Indian Tribes of the Interior,' in Canada and its Provinces, vol. XXI. The Pacific Province, i. 309 (Salish). Sproat, op. cit. p. 98 (Nootka). Solberg, ' Gebrauche der Mittelmesa-Hopi (Moqui) bei Namengebung, Heirat und Tod,' in Zeitschr. f. Ethnol. xxxvii. 629, 630, 632. Musters, in Jour. Anthr. Inst. i. 201 (Tehuelches). Cooper, op. cit. p. 236 (Mishmis). 3 Grinnell, * Marriage among the Pawnees,' in American Anthro- pologist, iv. 279. 4 Tout, ' Report on the Ethnology of the Stlatlumh of British Columbia,' in Jour. Anthr. Inst. xxxv. 132. 6 Tyler, op. cit. p. 118. 6 Jochelson, Yukaghir, p. 95 sq. XXIII MARRIAGE BY CONSIDERATION, ETC. 399 groom announces his approaching marriage to his father, the latter takes one or more pearl shells or other valuables to the bride's folk, and later he takes them a present of food ; but " after this the bride's father gives a return preseift, probably of the same kind and value, and also an equivalent amount of food."1 In his recent essay on the natives of Mailu, Dr. Malinowski observes that the regular exchange of gifts is one of the main features of the Papuo-Melancsian and Massim cultures, as has been fully recognised by Pro- fessor Seligman. He thinks "it is beyond doubt that this exchange is in nine cases out of ten initiated as the original price of the bride paid by the bridegroom at marriage, and subsequently returned to him. Then, in turn, he has to give a pig when the next opportunity offers, and so on. Although the balance seems to be always in favour of the girl's family-in the sense that they get more-there is no doubt that this system differs strongly from marriage by purchase pure and simple, and that one should speak of bride-price, etc., only in a qualified sense." 2 Sometimes the return gifts even exceed the original gifts in value. This is the case among some Indians of Canada,3 the Badagas of the Nilgiris in South India,4 and the Bogos inhabiting the outlying spurs of the Abyssinian mountains towards the north.5 In Sarae the girl's father, at the wedding, has to return five times the price which he received from the bridegroom's father at the espousals, the return gift, however, becoming the common property of the married couple.6 Bensen observes that among the Herero the consideration received by the bride's father can hardly be called a bride price, as he has to spend a much larger amount on the wedding feast.7 1 Seligman, op. cit. p. 506. See also ibid. p. 510. 2 Malinowski, ' Natives of Mailu,' in Trans. Roy. Soc. South Australia, xxxix. 564 sq. 3 Hodge, Handbook of Indians of Canada, p. 275. 4 Harkness, Description of a Singular Aboriginal Race inhabiting the Neilgherry Hills, p. 116 sq. 6 Munzinger, Ueber die Sitten und das Recht der Bogos, p. 57 sq. 6 Idem, Ostafrikanische Studien, p. 387. 7 Bensen, quoted by Kohler, ' Das Recht der Herero,' in Zeitschr. f. vergl. Rechtswiss. xiv. 302. 400 THE HISTORY OF HUMAN MARRIAGE chap. Speaking of the marriage customs of the Yukaghir of the tundra, Dr. Jochelson remarks that the exchange of gifts " is designed to bring the members of both families into close contact " j1 and Professor Hobhouse rightly suggests that this practice may, generally, be " a method of cementing the union of the two families.'' 2 Mr. Crawley connects it with savage ideas of gifts : "a gift," he says, " means far more to primitive man than it does to us ; it is part of himself." And he even maintains that marriage by purchase is a development from a previous custom of exchanging gifts, the so-called bride price being originally " a pledge, a part of one's self, given to another and received from him."3 No doubt, superstitious ideas are often con- nected with the transference of gifts ; 4 and it may very well be that there are such ideas also in connection with the exchange of gifts at a marriage, even apart from those which are exchanged between bride and bridegroom. Yet Mr. Crawley has given no direct evidence in support of his theory, and in no case can it be accepted as a general explanation of the practice we arc now discussing. Nor do I find any reason whatever to suppose that marriage by purchase has developed out of an earlier exchange of gifts. In some cases the exchange is obviously connected with a feeling of shame at the idea of making a daughter an article of traffic. Among the Bagobo of Mindanao, according to Mr. Cole, the price for a bride varies according to the wealth of the interested parties and the accomplishments of the bride, but whatever the sum paid, the father of the girl must make a return present equal to one-half the value of the marriage gift " so that he does not sell his daughter like a slave." 5 In Florida, of the Solomon Group, when fifty rongo, or coils of native money, are offered by the bridegroom's party, the bride's party give in return five pigs, and when a hundred coils are offered they give ten 1 Jochelson, Yukaghir, p. 95. 2 Hobhouse, Morals in Evolution, p. 154. 3 Crawley, Mystic Rose, p. 386 sqq. 4 See Westermarck, Origin and Development of the Moral Ideas, i- 593 sq. r' Cole, loc. cit. p. 101. XXIII MARRIAGE BY CONSIDERATION, ETC. 401 pigs ; " and they say that the money buys the pigs and not the damsel."1 Again, among the Western Islanders of Torres Straits, according to Dr. Haddon, " the return of presents on the father-in-law's part appears to be the result of a feeling that a wife costs too much, and that the husband should not be impoverished."2 But sometimes the return gift depends on the behaviour of the husband, and thus serves as a protection for the wife. Among the Hidatsa, whilst marriage is usually made formal by the dis- tribution of gifts on the part of the man to the woman's relatives, presents of equal value are afterwards commonly returned by the latter, " if they have the means of returning them and are satisfied with the conduct of the husband." 3 Among the Lower Thompson Indians " wealthy people, if pleased with the new son-in-law, returned the marriage presents to him." 4 Schadenberg states that among the Bagobos of Southern Mindanao, when six months have elapsed, the father of the wife gives one-half of the bride price back to the husband, if the newly married couple are satisfied with each other.5 Of the Chukchee we are told that when the son-in-law takes his wife home without quarrelling with her father, he is usually given some reindeer, the number of which depends partly upon the quality of work the young man has done while serving for his bride.6 The return gift may take the shape of a dowry given to the bride by her father or parents or other relatives but also directly or indirectly benefiting her husband. The practice of giving a dowry to a daughter prevails among many uncivilised peoples, although among others nothing of the kind is said to be found. The dowry often consists of some food, clothes, ornaments, household goods, or 1 Codrington, op. cit. p. 238. 2 Haddon, in Results of the Cambridge A nthropological Expedition to Torres Straits, v. 231. 3 Matthews, op. cit. p. 52. 4 Teit, in Publications of the fesup North Pacific Expedition, i. 322. 5 Schadenberg, ' Die Bewohner von Sud-Mindanao und der Insel Samal,' in Zeitschr. f. Ethnol. xvii. 12. 6 Bogoras, op. cit. p. 586. VOL. II D D 402 THE HISTORY OF HUMAN MARRIAGE chap. other objects,1 but among many peoples it includes or consists of a certain number of domestic animals.2 Among the Nilotic tribes of the Kavirondo the father of the bride presents his daughter with a goat.3 Among the Bahima he gives her a present of a number of cows, " never less than six, to ensure her having food."4 Among the Zulus he gives her a blanket and cattle according to his rank ; " but no girl ever goes to her husband without an ox, which is ever looked upon afterwards as the ox of the amahlozi (ancestral spirits), the loss of which by death would be considered a token of desertion by the protecting spirits of her father's house, and the slaughter of which, in the event of any calamity (such as disease or barrenness), is an acceptable sacrifice." 5 Among the Todas the bride may receive from her father a dowry of several buffaloes, together with neck- 1 Nordenskiold, Den andra Dicksonska expeditionen till Gronland, p. 508 (Greenlanders), v. Martins, op. cit. i. 115 (Brazilian abori- gines). Bove, op. cit. p. 132 (Yahgans of Tierra del Fuego). Thomas, Ibo-speaking Peoples of Nigeria, i. 66. Bargy, ' Notes ethnographiques sur les Birifons,' in L'Anthropologic, xx. 172. Waitz, Anthropologic der Naturvolker, ii. 522 (Somal). Marshall, A Phrenologist amongst the Todas, p. 212 ; Rivers, Todas, p. 504. Prejevalsky, Mongolia, i. 70 (Mongols). Pallas, Merkwurdigkeiten der Morduanen, Kasaken, Kalmucken, &c., p. 262 (Kalmucks). Pfeil, op. cit. p. 29 (natives of the Bismarck Archipelago). Thurn- wald, in Zeitschr. f. vergl. Rechtswiss. xxiii. 337 (natives of Buin in Bougainville). Post, Die Anfange des Staats- und Rechtslebens, p. 54 sq. Idem, Grundriss der ethnologischen Jurisprudent, i. 304. 2 Francois, op. cit. p. 214 (Hottentots). Kropf, op. cit. p. 142 (Xosa Kafirs). Schinz, Deutsch-Sud-W est-Afrika, p. 311 (Ovambo). Decle, op. cit. p. 158 (Matabele). Roscoe, Northern Bantu, p. 39 sq. (Banyoro). Baumann, Durch Massailand zur Nilquelle, p. 187 (Wambugwe). Last, ' Visit to the Masai People living beyond the Borders of the Nguru Country.' in Proceed. Roy. Geograph. Soc. N.S. v. 532. Metz, Tribes inhabiting the Neilgherry Hills, p. 87 ; Ward, in Grigg, Manual of the Nilagiri District in the Madras Presi- dency, Appendix, p. Ixxi. (Badagas). Miss Czaplicka, op. cit. p. 118 (Buryat). Jochelson, Yukaghir, p. 93 (Yukaghir of the tundra). Dittmar, loc. cit. p. 25 (Koryak). Vetter, loc. cit. p. 90 (Jabim of the former German New Guinea). 3 Hobley, Eastern Uganda, p. 29. 4 Roscoe, Northern Bantu, p. 120. 6 Tyler, op. cit. p. 202 sq. See also Kidd, The Essential Kafir, p. 216. XXIII MARRIAGE BY CONSIDERATION, ETC. 403 laces, armlets, and ear-rings.1 The dowry of a Samoyed bride consists of a tent, some reindeer, sledges, harness, clothes, and meat.2 Among the Maori of New Zealand some of the wives of men belonging to distinguished and rich families brought good dowers with them in the shape of lands and slaves.3 Sometimes the dowry given to the bride by her father is represented as a return gift to her husband ;4 and where, as among the Marea of North-Eastern Africa, it becomes the exclusive property of the husband,5 it is really nothing else. In other instances the wife gets back her dowry in case of separation or divorce, though the husband may have the usufruct of it so long as the marriage lasts.6 Among various peoples there is an interesting connection between the bride price and the dowry in so far that the bride price, or a part of it, is given to the bride by her father as a dowry, or is looked upon as a settlement or provision for the wife.7 1 King, Aboriginal Tribes of the Nilgiri Hills, p. 24. 8 Miss Czaplicka, op. cit. p. 124. 3 Shortland, Traditions and Superstitions of the New Zealanders, p. 118. * Sapir, ' Indian Tribes of the Coast,' in Canada and its Provinces, vol. XXI. The Pacific Province, i. 341 (coast tribes of British Columbia). Prichard, op. cit. p. 93 (Tehuelches of Patagonia). Miss Czaplicka, op. cit. p. 108 sq. (Yakut). 5 Munzinger, Ostafrikanische Studien, p. 240. 6 Richardson, op. cit. i. 407 (Kenai). Holmberg, in Acta Societatis Scientiarum Fennicae, iv. 315 (Tlingit). Bancroft, op. cit. i. 197 (Nootka). Hawkins, ' Notes on the Creek System of Government,' in Trans. American Ethn. Soc. vol. iii. pt. i. 66. Prichard, op. cit. p. 93 (Tehuelches). Baumann, op. cit. p. 187 (Wambugwe). Rochon, ' Voyage to Madagascar and the East Indies,' in Pinkerton, op. cit. xvi. 747 (Malagasy). Lewin, op. cit. p. 254 (Kukis). 7 Prichard, op. cit. p. 93 (Tehuelches). Elmslie, Among the Wild Ngoni, p. 58. Monrad, Bidrag til en Skildring af Guinea-Kysten og dens Indbyggere, p. 47 (Negroes of Accra). Henry, Les Bambara, p. 203. Hamilainen, op. cit. p. 104 (Cheremiss). Miss Czaplicka, op. cit. pp. 117, 121 (Altaians, Kalmucks). Patkanov, Die Irtysch- Ostjaken und ihre Volkspoesie, i. 139. Georgi, op. cit. pp. 55, 103, 182, 324 (Votyak, Tartars of Orenburg, Bashkir, Tungus). Vd.m- b6ry, op. cit. p. 433 (Tartars of Kazan). Cooper, op. cit. p. 236 (Mishmis). Crawfurd, History of the Indian Archipelago, i. 90 (Javanese). Pardo de Tavera, in Revista contempordnea, Ixxxvi. D D 2 404 THE HISTORY OF HUMAN MARRIAGE chap. Among many tribes the suitor gives a present to his prospective wife, or a bridegroom to his bride either before or immediately after the marriage.1 Among some of the Eskimo the lover presents clothes to the lady, who puts them on, and is thenceforth his wife.2 Among the Dakota of the Upper Mississippi, according to Prescott, men ask for consent to marriage by sending the price of the girl, and in addition they may give small presents to the object of their esteem.3 Among the Pima Indians of Arizona " the groom presented the bride with a new blanket and his parents gave her presents, but there was no idea of purchase and no gifts were made to the bride's parents." 4 Speaking of the Guanas of Paraguay, Azara says, " Toutes les ceremonies du manage se reduisent a un petit present que le mari fait a sa prdtendue."5 Among the West African Fanti the bridegroom is obliged to make presents of cloths, Manchester stuffs, and silks to the bride, after which nothing more is to be expected of him.6 In Benin " the bridegroom 466 (Tagals of the Philippines in former times). Dahlgren, ' Om Palau-oarna,' in Ymer, iv. 333 (Pelew Islanders). Post, Grundriss der ethnologischen Jurisprudenz, i. 305. 1 Voth, ' Oraibi Marriage Customs,' in American Anthropologist, N.S. ii. 238 (Hopi Indians of Arizona). Trenk, in Mittheil. Deutsch. Schutzgeb. xxiii. 168 (Namib Bushmen). Fritsch, Die Eingeborenen Sild-Afrika's, p. 192 (Bechuanas). Dundas, ' Wawanga and other Tribes of the Elgon District, British East Africa,' in Jour. Roy. Anthr. Inst, xliii. 40. Caillie, op. cit. i. 349 (Mandingo). Partridge, Cross River Natives, p. 254. Tessmann, op. cit. ii. 255 (Pangwe). Ellis, Ewe-speaking Peoples of the Slave Coast, p. 157 ; Klose, Togo unter deutscher Flagge, p. 252 ; Spieth, op. cit. p. 62 * (Ewhe). Proyart, ' History of Loango,' in Pinkerton, op. cit. xvi. 569. Johnston, George Grenfell and the Congo, ii. 679 (Bakongo). Waitz, op. cit. ii. 522 (Somal). Munzinger, Ostafrikanische Studien, p. 324 (Beni- Amer). Bainbridge, ' Saorias of the Rajmahal Hills,' in Memoirs Asiatic Soc. Bengal, ii. 58. Snouck Hurgronje, Achehnese, i. 325. Seligman, op. cit. p. 76 (Koita of British New Guinea). Hahl, loc. cit. p. 80 (natives of the Gazelle Peninsula of New Britain). 2 Bancroft, op. cit. i. 66. Seemann, Narrative of the Voyage of Herald, ii. 66. 3 Prescott, in Schoolcraft, op. cit. iii. 238. 4 Russell, ' Pima Indians,' in Ann. Rep. Bur. Ethn. xxvi. 184. 6 Azara, Voyages dans l'Amerique meridionals, ii. 92 6 Connolly, in Jour. Anthr. Inst. xxvi. 144. xxni MARRIAGE BY CONSIDERATION, ETC. 405 covers his future bride with a rich suit of clothes, necklaces and bracelets."1 Among the Buduma of Lake Chad " the groom gives four cows to the father of his bride and five Maria Theresa thalers to her mother. The bride herself receives from him one cow, which must be in milk." 2 Among the Bhotias of the Darma pargana (Almora district), " the young man, either personally or through his friends, offers the girl a sum of money varying from five rupees to one hundred bound up in a piece of cloth. Generally the young lady is not directly approached, but the gift is handed over to her intimate associates . . . , and they promise to exercise their influence with her. Her answer is not obtained without a family consultation of her relations, and should the match appear a suitable one the gift is retained, otherwise it is returned." 3 Among the Maori the expression whakawhere- where is applied to conciliation of a desired woman by means of gifts ; a man will give or send to such a woman some present, which he hopes will cause her to like and desire him.4 In Lifu a present given to induce a woman to marry was called june hmalaf Again, among the Bateso, when the bride first enters her new home, the husband presents her with a goat which she keeps alive for breeding.6 Among the Baganda, when the bride reached her husband's house, she refused to enter until he had given her a few cowry-shells, and when she had entered, she would not sit down until he had again given her a few shells. When the evening meal was dished up, she would not touch the food until her husband gave her a few more cowry-shells ; and again, when it was bed- time, she would not move until a further sum had been given her. On the day after the consummation of the marriage the husband gave her a goat " as a token of his affection." 7 Among the Shilluk, when the bride has 1 Nyendael, quoted by Ling Roth, Great Benin, p. 38. 2 Talbot, in Jour. Roy. Anthr. Inst. xli. 247 sq. 3 Sherring, in Memoirs Asiatic Soc. Bengal, i. 106 sq. 4 Best, in Trans, and Proceed. New Zealand Inst, xxxvi. 36. 5 Ray, ' People and Language of Lifu, Loyalty Islands,' in Jour. Roy. Anthr. Inst, xlvii. 286. Roscoe, Northern Bantu, p. 264. 7 Idem, Baganda, p. 90 sq. 406 THE HISTORY OF HUMAN MARRIAGE chap. entered the bridegroom's hut, she refuses to lie down. But metal ornaments are brought to cause her to lie down, and then she does so. The next morning a goat is brought and killed, but the bride refuses to eat; and metal orna- ments are again brought, and then she eats.1 Among the Mpfumo clan of the Thonga the first night when the bride sleeps in the hut with her husband she may refuse to allow him his conjugal rights. He then goes to his father and asks him what he ought to do under the circumstances. The father says, " Give her sixpence or one shilling " ; and then she consents.2 Marriage by consideration is not only prevalent among most uncivilised races, but is also found among peoples who have reached a higher degree of culture. In China a present is given by the father of the suitor, the amount of which is not left to the good-will of the parties, as the term " present " would suggest, but is exactly stipulated for by the negotiators of the marriage. In ordinary cir- cumstances it varies from twenty-five to forty dollars, in- creasing to a hundred and over according to the condition of the bridegroom ; and Until it is paid the marriage does not take place. And besides money, other presents, some- times costly, consisting of silks and satins and rice and fruits or other articles, are sent to the parents or guardians of the bride.3 The people will not hear of the marriage presents being called a " price " for the girl; but Mr. Jamieson thinks that they are a survival from a time when the transaction was one of ordinary bargain.4 Among the poor it occurs that parents actually purchase a young girl and bring her up as a daughter until she is marriageable, thus securing her services in the household and reducing the expenses of a wedding.5 In Japan marriage by sale and marriage by capture are said to have been common in 1 Westermann, Shilluk People, p. no sq. 2 Junod, op. cit. i. 113. 3 Gray, China, i. 193. Doolittle, Social Life of the Chinese, i. 70 sqq. Wells Williams, The Middle Kingdom, i. 785 sqq. 4 Jamieson, ' Translations from the General Code of Laws of the Chinese Empire ; vii. -Marriage Laws,' in China Review; x. 78 n. 5 Wells Williams, op. cit. i. 789. xxin MARRIAGE BY CONSIDERATION, ETC. 407 ancient days; but from the time of the Taiho-ryo (701- 1192) customs gradually became more refined.1 Marriage by consideration has prevailed in all branches of the Semitic race. In Babylonia a suitor had to give to the father of his intended wife a bride price or present (tirhAtum or terhatu), the amount of which varied, according to the rank of the parties ; and if the young man did not himself possess the sum which was required, his parents were expected to provide him with it.2 But from the Laws of Hammurabi we learn that although it was the general custom for a man to pay a bride price to his father- in-law, this payment was not universal,3 and also that the bride price reverted to the husband if the wife died without having granted him children.4 In Genesis we read that Jacob served Laban, his mother's brother, for seven years for each of his cousins Leah and Rachel ;5 but the usual method of obtaining a wife in ancient Israel was by paying a bride price, called mohar or mahr* According to Talmudic law the mutual consent of the parties to marry each other has to be legally manifested by a special formality, which gives validity to the marriage contract. The usual form- ality is that called kaseph, or " money." In the presence of two witnesses the man gave to his chosen bride a piece of money-even a f>eruta, the smallest copper coin used in 1 Nakajima, * Marriage (Japanese and Korean),' in Hastings, Encyclopedia of Religion and Ethics, viii. 460. 2 Koschaker, Rechtsvergleichende Studien zur Gesetzgebung Qam- murapis, p. 130 sqq. Meissner, Beitrage zum altbabylonischen Privat- recht, p. 13 sq. 3 Laws of gammurabi, § 138 sq. (Johns' translation, p. 27; Winckler's translation, p. 41). Koschaker (op. cit. p. 153 sqq.) explains this by suggesting that marriage by purchase had ceased to exist according to Sumerian law, whereas it was retained in Babylonian law. 4 Laws of jlammurabi, § 163 sq. ( Johns' translation, p. 32 sq. ; Winckler's translation, p. 47), 6 Genesis, xxix. 20 sqq. 6 Ewald, Antiquities of Israel, p. 200. Gans, Das Erbrecht in weltgeschichtlicher Entwickelung, i. 128. Robertson Smith, Kinship and Marriage in Early Arabia, p. 79 Wellhausen, ' Die Ehe bei den Arabern,' in Nachrichten von der Kbnigl. Gesellschaft der Wissen- schaften zu Gottingen, 1893, p. 433. Benzinger, Hebraische Archd- ologie, p. 106. Nowack, Lehrbuch der hebraischen Archaologie, i. 155. THE HISTORY OF HUMAN MARRIAGE chap. 408 Palestine, was sufficient for the purpose-or any object of equal value, with the words, "Be thou consecrated to me"; but during the Middle Ages it became customary for the act of betrothal by kaseph to be performed by means of a plain ring, instead of a piece of money, and this custom has ever since prevailed up to our time.1 The betrothal by kaseph may be a survival of the old marriage by purchase, but it has also been supposed to be an imitation of the Roman co emtio.2 * Among the ancient Arabs a bride price, mahr, was given by the bridegroom to the father or guardian of the bride ; and this bride price has survived in Muhammadan law, where it has been confounded with the sadaq, which was a gift offered to the bride by the bridegroom.3 Although the Koranic law presumes that the mahr, or sadaq, shall be the property of the bride, this rule is not always'followed in practice. In the rural districts of Palestine, for instance, the marriage contract is avowedly an act of purchase, most part-or at least one-half 4-of the sadaq going to the girl's father;5 this makes the birth of a girl so much more welcome among the fellahin than among the townspeople, who do not appropriate the payment given for their daughters.6 In Morocco the girl's father in some places spends the whole of the sum paid down on the trousseau of his daughter, whereas in other cases he so spends only a part of it, keeping the rest for himself ;7 then the sadaq is hardly, as Sidi Halil puts it,8 merely " similar '' to a selling price. But even 1 Mielziner, Jewish Law of Marriage and Divorce in Ancient and Modern Times, p. 77 sqq. 2 Ibid. p. 78 n. 2. 8 Robertson Smith, op. cit. pp. 76-78, 91. Wellhausen, loc. cit. P- 433 sq. * v. Miilinen, ' Beitrage zur Kenntnis des Karmels,' in Zeitschr. des Deutschen Palastina-Vereins, xxx. 170. 8 Klein, * Mittheilungen uber Leben, Sitten und Gebrauche der Fellachen in Palastina,' ibid. vi. 90. Van-Lennep, Bible Lands, p. 540. Wilson, Peasant Life in the Holy Land, p. 109. Jaussen. Coutumes des Arabes au pays de Moab, p. 49. 6 Wilson, op. cit. p. 109. 7 See Westermarck, Marriage Ceremonies in Morocco, p. 70 sqq. 8 Sidi galil, Muhtasar, §151 (Russell and Abdullah al-Ma'mun Suhrawardy, ' A Manual of the Law of Marriage ' from the Mukhtasar of Sidi Khalil, p. 54). XXIII MARRIAGE BY CONSIDERATION, ETC. 409 when the woman has the full benefit of the sadaq, the marriage may nevertheless be an act of purchase. Besides the sadaq another payment, often of a much higher amount, is in many tribes given to the girl's father to be retained by him; this custom is found particularly among Berber tribes or their Arabic-speaking neighbours, and is no doubt the old Berber marriage by purchase.1 Moreover, in many tiibes a payment is also made to other members of the girl's family than her father, particularly her eldest brother, who in case her father is dead is her recognised guardian. This payment is sometimes represented as a " bribe," the object of which is to induce the other relatives of the girl to try to influence the father.2 It has been supposed that wife purchase was the basis of Indo-European marriage before the separation of peoples took place.3 In Vedic times brides were won by rich presents to their fathers,4 though a certain discredit would seem to have attached to the sale of daughters.5 In the Mahabharata we are told that Pandu paid the Madra king in gold, jewels, elephants, horses and cars, and various other articles for the hand of his sister,6 and that the purchase of women was the family practice of the king.7 Still more was the custom prevalent among the lower classes of Hindus. One of the eight forms of marriage mentioned by Manu-the asura form-was marriage by purchase, and he admits that some allowed the two lower castes, the Vaidyas and the Audras, to practise it. Manu himself, however, forbade it altogether.8 0 No father who knows the law," he says, " must take even the smallest gratuity for his daughter ; for a man who, through avarice, takes a gratuity, is a seller of his offspring." 9 But the so-called arsha form, which implied that the bridegroom sent a cow and a bull 1 See Westermarck, Marriage Ceremonies in Morocco, p. 75 sqq. 2 See ibid. p. 79 sqq. 3 Wintemitz, * On. a Comparative Study of Indo-European Customs,' in Transactions of the International Folk-Lore Congress, 1891, p. 287. Schrader, Reallexikon der indogermanischen Alter- tumskunde, p. 109. 4 Zimmer, Altindisches Leben, p. 310 5 Macdonell and Keith, Vedic Index of Flames and Subjects, i. 482. 0 Mahabharata, i. 113. 14 sq. 7 Ibid. i. 113. 9 sqq. 8 Laws of Manu, iii. 24 sq. 9 Ibid. iii. 51. Cf. ibid. ix. 93, 98. 410 THE HISTORY OF HUMAN MARRIAGE chap. or two pairs to the bride's father, was counted by Manu and other lawgivers as one of the legitimate modes of marriage.1 It was expressly denied that this gift was a gratuity,2 but there can be little doubt that the arsha form was the survival of a transaction which might be called a purchase. This is borne out by references in the older Grihyasutras of Paraskara 3 and Sankhayana4 to the practice of giving the father-in-law a hundred cows with a chariot, and by the recognition in the Grihyasutras of the Kathaka and the Manava schools of a usage by which the bride price was paid in money to the father.5 Not- withstanding the prohibition in the 'Laws of Manu,' marriage by purchase occurs to this day even among high castes, and is frequently practised among the Audras.6 Among the Hindus of South India, according to Mr. Padfield, it is not uncommon for the bride's parents to demand a sum of money, sometimes comparatively large, from the boy's friends in addition to the jewels offered to the bride. The name given to the arrangements for this money gift to the girl's parents is one which means bargaining ; and where there are several applicants for her hand, it often, becomes very much like an auction in which the highest bid is held out for.7 Aristotle tells us that in the primitive ages of Greece men bought their wives.8 In heroic times a suitor gave e^va, consisting of cattle, to the father of the bride elect, and a maid was called aX^ea-i/Soia, that is, one who yields her parents many oxen as presents from her suitor,9 or by some other name compounded from the Greek word for an 1 Ibid. iii. 29. 2 Ibid. iii. 53. 3 Pdraskara-Grihya-Sdtra, i. 8. 18. 4 Sdhkhdyana-Grihya-S'Atra, i. 14. 16. 5 Keith, ' Marriage (Hindu),' in Hastings, op. cit. viii. 451. See also Jolly, Recht und Sitte, p. 51 sq. (in Buhler, Grundriss der Indo- Arischen Philologie und Altertumskunde, vol. ii.). 6 Gait, Census of India, 1911, vol. i. (India) Report, p. 257. Dubois, Description of the Character, Manners, and Customs of the People of India, p. 102. Mayne, Treatise on Hindu Law and Usage, p. 99. Jolly, op. cit. p. 52. Keith, in Hastings, op. cit. p. 451 7 Padfield, The Hindu at Home, p. 99 sq. 8 Aristotle, Politica, ii. 8. 9 Ilias, xviii. 593. xxiit MARRIAGE BY CONSIDERATION, ETC. 411 ox, the gold currency of the time.1 Contrary to other writers, Hruza maintains that the e^va were not a bride price, but merely a douceur intended to prevail upon the father ; but he admits that in a yet earlier age marriage by purchase existed in Greece, although, as he rightly points out, the transaction could not have been the purchase of a chattel but only of the rights of a husband.2 Marriage by purchase cannot with equal certainty be established as a form of marriage on Roman soil; but a reminiscence of it is supposed to have been preserved in the symbolical process of coemtio, which was the ordinary form in which any Roman citizen, whether patrician or plebeian, might contract a marriage. It was a traditional ceremony representing a purchase of the bride, the intending husband " purchasing " the bride from the person in whose power she was, with a view to thereby acquiring that marital power without which marriage as a legal relationship was considered impossible.3 On the other hand, there are also writers who deny that coemtio can be regarded as a survival of ancient bride purchase. Marquardt maintains that it was an artificial and comparatively late form of marriage, whilst confarreatio, the specifically patrician kind of marriage which suggested no idea of purchase, was the oldest form of marriage in Rome.4 Karlowa thinks that if marriage by purchase prevailed at ancient Rome or among the ancestors of the Romans, a survival of it remained not in the coemtio but in the arrha. sponsalitia, which in his- torical times was given to the bride.5 Marriage by consideration was a custom of all Teutonic peoples. There is a trace of marriage by service in the ' Eyrbyggja Saga ' : Vigstyr says to the berserk Halli, 1 Murray, Rise of the Greek Epic, p. 186. z Hruza, Beitrage zur Geschichte des griechischen und romischen Familienrechtes, i. 8 sqq. For a criticism of Hruza's view see Beauchet, Histoire du droit prive de la Rdpublique Athenienne, i. 113 ^7- 3 Rossbach, Untersuchungen uber die romische Ehe, pp. 80, 87 sqq. Sohm, Institutes, p. 452 sq. Leist, Alt-arisches Jus Gentium, pp. 128, 129, 133. Schrader, op. cit. p. no. 4 Marquardt, Das Privatleben der Rdmer, p. 38. 5 Karlowa, Romische Rechtsgeschichte, ii. x66. 412 THE HISTORY OF HUMAN MARRIAGE chap. who asked for the hand of his daughter Asdi, " As you are a poor man, I shall do as the ancients did and let you deserve your marriage by hard work."1 But the general con- sideration given for a bride was money paid down by the bridegroom. Originally, we may presume, the amount of it depended on agreement, but during the period of the law-books, both in England, and on the Continent, it was generally fixed by custom or by statute.2 The bride price was called wittum or widem ; the Anglo- Saxons called it weotuma, the Langobardians meta or mundium, the Frisians mundsket, the Scandinavians mundr, and in Latin it was termed pretium nuptiale. or pretium emtionis.3 Betrothal was mercatio, and to marry was uxorem emere or feminam vendere* The Kentish law of king Aethelbirht speaks of a man buying a maiden with cattle, and the transaction is called a " bargain."5 In Germany the expression " to purchase a wife " was in use till the end of the Middle Ages,6 and we find the same term in Christian IV.'s Norwegian Law of 1604.7 In Holland the bride is still, in the language of the common people, repre- sented as verkocht, that is " sold."8 But here again we should notice that marriage by purchase did not imply the purchase of a piece of property: the ancient Teutons bought the mund, or protectorship over -the woman9 1 Weinhold, Altnordisches Leben, p. 242. 2 Sohm, Das Recht dev Eheschliessung, p. 23 sq. Schroeder, Lehrbuch dev deutschen Rechtsgeschichte, p. 299. Howard, History of Matrimonial Institutions, i. 265. 3 Schroeder, op. cit. pp. 70, 299. v. Amira, * Recht,' in Paul Grundriss der germanischen Philologie, iii. 161. 4 Brunner, Deutsche Rechtsgeschichte, i. 74. Schroeder, op. cit. p. 299. 5 Laws of King Aethelbirht, ch. 77. Cf. ibid. ch. 31. 6 Laband, ' Die rechtliche Stellung der Frauen im altromischen und germanischen Recht,' in Zeitschr. fur Volkerpsychologie und Sprachwissenschaft, iii. 154. 7 Olivecrona, Om makars giftoratt i bo, p. 150. R Brunner, op. cit. i. 74. 9 Friedberg, Das Recht der Eheschliessung in seiner geschichtlichen Entwicklung, p. 18. v. Amira, loc. cit. p. 162. Pollock and Mait- land, History of the English Law before the Time of Edward I., ii. 364. Howard, op. cit. i. 260. Cf. Brunner, Grundzuge der deutschen Rechtsgeschichte, p. 222 sq. xxin MARRIAGE BY CONSIDERATION, ETC. 413 and other rights which marriage conferred on the husband. Marriage by purchase prevailed among the early Slavs.4 The bride price was called veno, in Polish wiano ;2 and in old Russian a marriageable girl was called a kunka, from kuna, " marten," because her parents might exchange her for marten-skins, the usual medium of payment in ancient Russia.3 " To this day among the Russian peasantry," says Schrader, " the first act of the nuptials is the suit or proposal (svdtanie), which is a purely commercial trans- action. The father of the suitor, usually accompanied by a relative, visits the girl's parents and says, ' We have a purchaser ; you a commodity : will you sell your ware ? ' Then follows the bargaining, which, as our informants state, differs in no respect from a negotiation about the sale of a cow."4 Among the Southern Slavs the marriage by purchase still partially prevails, or did so recently. In Serbia, at the beginning of the nineteenth century, the price of girls reached such a height that Black George limited it to one ducat.5 In High Albania " marriage is entirely by purchase, except for the occasional forcible capture of a girl." 6 The ancient Celts paid a price for their brides. In Ireland it consisted of various objects, such as articles of gold, silver, or bronze, clothes or horse-bridles, cattle or swine, 1 Ewers, Das attests Recht der Russen, p. 226. Hartknoch, Alt- und neues Preussen, p. 177. Macieiowski, Slavische Rechtsgeschichte, ii. 195 (Pomeranians and Bohemians). Krauss, Sitte und Brauch der Sudslaven, p. 273. Kovalewsky, ' Marriage among the Early- Sia vs,' in Folk-Lore, i. 463, 465, 478 sq. Idem, Modern Customs and Ancient Laws of Russia, p. 26 sqq. Volkov, ' Rites et usages nup- tiaux en Ukraine,' in L' Anthropologic, ii. 168. 2 Piprek, Slawische Brautwerbungs- und Hochzeitsgebrauche, p. 151. 3 Schrader, ' Family (Teutonic and Balto-Slavic),' in Hastings, op. cit. v. 750. 4 Ibid. 750. 5 Krauss, op. cit. p. 275. Wesnitsch, ' Die Blutrache bei den Sudslaven, ' in Zeitschr. f. vergl. Rechtswiss. ix. 50 sq. 6 Miss Durham, * High Albania and its Customs in 1908,' in Jour. Roy. Anthr. Inst. xl. 459. See also ibid. p. 460 ; GopCevid, Oberalbanien und seine Liga, p. 444. 414 THE HISTORY OF HUMAN MARRIAGE chap. land or houses.1 It often had the form of a yearly payment from the husband after marriage, and we find it laid down in the Brehon Law that the woman's father was entitled to the whole of the first year's bride price.2 In Ireland it was called coibche, though there were also other names for it,3 and in Wales gobcr (gobyr) or amober (amobor, amobyr}^ Among all these peoples, however, marriage by con- sideration has in the course of time been subject to modi- fications, which are very similar to certain customs already noticed among some uncivilised tribes ; and it has led to institutions totally different from the original practice. The general trend of this process is that the parents of the woman more or less lose the economic advantages they derived from her marriage, and that greater regard is paid to the interests of the contracting parties. Here, also, we meet with the practice of offering a return gift; and although this practice, as we have seen, may serve different purposes, it seems in some cases at least to be a mitigation of marriage by purchase. In China the parents of the bride, or her guardians, accept only a part of the proffered presents, returning the balance, to which they add some articles for the parents of the bridegroom; but they accept all the money, and all the silks and satins designed for the use of the girl.5 The exchange of presents forms, in fact, the subject of a long section in the old penal code ; for " the marriage articles and betrothal presents once exchanged, the parties are considered irrevocably engaged." 6 As to the presents exchanged at the time of sending the engagement cards without which no betrothal 1 Sullivan's ' Introduction,' in O'Curry, Manners and Customs of the Ancient Irish, vol. i. p. clxxiv. sq. Joyce, Social History of Ancient Ireland, ii. 4. 2 Ancient Laws and Institutes of Ireland, ii. 347 ; iii. 315. Joyce, op. cit. ii. 4 sq. 3 Joyce, op. cit. ii. 4. * d'Arbois de Jubainville, Cours de literature celtique, vii. 234. Venedotian Code, ii. 19. 1 ; Dimetian Code, ii. 8. 73 (Ancient Laws and Institutes of Wales, pp. 92, 223). 6 Doolittle, op. cit. ii. 71. 6 Medhurst, ' Marriage, Affinity, and Inheritance in China,' in Trans. Roy. Asiatic Soc. China Branch, iv. 11 sq. xxiii MARRIAGE BY CONSIDERATION, ETC. 415 is regarded as binding, it is said that in the Chinese view they are " omens of good to the parties most intimately concerned."1 In India, besides the arsha form of marriage, another mode of preserving the symbol of sale, while re- jecting the reality, appears to have been the receipt of a gift of real value, such as a chariot and a hundred cows, which was immediately returned to the giver. This arrangement is said by Apastamba to have been prescribed by the Vedas " in order to fulfil the law "-that is, appa- rently, the ancient law by which the binding form of marriage was a sale.2 In ancient Greece there were at the conclusion of a marriage not only the e8va given for the bride but also the pecKca or presents given by her father to the bridegroom.3 Tacitus, after speaking of the gifts for which a German obtains his wife, says that the wife " in her turn brings her husband some gift of arms. This," he adds, " represents to them our marriage bond, the mystic cele- brations, and all the gods of matrimony." 4 The Welsh agweddi was, strictly speaking, a payment made by the kindred or parent of the bride to the bridegroom, although the word sometimes seems to have been used to include the marriage portion of the bride as well.5 The dowry which the bride brought with her may also be partly regarded as a return gift to the husband.6 In this group of peoples we also meet with gifts offered to the bride by the bridegroom; and although these gifts no doubt may have an independent origin, they may also be a survival of the old bride price. And we know that in many cases the price paid for the bride, instead of being appro- priated by her parents or guardian, became wholly or in part her own property. In China special presents are sent for the bride, and the 1 Doolittle, op. cit. ii. 66 sqq. 2 Apastamba, ii. 6. 14. 12. Mayne, op. cit. p. 97. 3 Hruza, op. cit. i. 11. Beauchet, op. cit. i. 113. 4 Tacitus, Germania, ch. 18. 6 Rhys and Brynmor-Jones, The Welsh People, p. 211. 8 Cf. Meissner, op. cit. p. 14 (Babylonians) ; Smith, Waytej and Marindin, Dictionary of Greek and Roman Antiquities, i. 691 ; Grimm, Deutsche Rechtsalterthumer, p. 429. 416 THE HISTORY OF HUMAN MARRIAGE chap. money given to her parents is generally spent in out- fitting her.1 In Japan the proposed husband sends certain prescribed presents to his future bride, and this sending of gifts forms one of the most important parts of the marriage ceremony. In fact, when once the presents have been sent and accepted, the contract is completed and neither party can retract. Mr. Kiichler says that he has been unable to find out the exact meaning of these presents : the native books on marriage are silent on the subject, and the Japanese themselves have no other explanation to give than that the custom has been handed down from ancient times.2 But considering that marriage by purchase once prevailed in Japan, it is reasonable to suppose that the sending of presents is a relic of that custom. There is, how- ever, an exchange of gifts : the bride also gives certain conventional presents to her future husband and his parents and relatives, and as to the value of these presents she should always be guided by the value of those brought by the bridegroom.3 In the Laws of Hammurabi we read not only of a bride price, but also of presents (nudanndm) which the bridegroom gave to the bride.4 Similar gifts, called mattcm, were also offered to the bride in ancient Israel.5 Abraham's servant " brought forth jewels of silver, and jewels of gold, and raiment, and gave them to Rebekah : he gave also to her brother and to her mother precious things." 6 In ancient Arabia the bridegroom offered to the bride a gift, called sadaq, at their wedding.7 But over and above 1 Doolittle, op. cit. i. 68, 70 sq. 2 Kiichler, ' Marriage in Japan,' in Trans. Asiatic Soc. Japan, xiii. 120. 3 Ibid. p. 123. Cf. Nakajima, in Hastings, op. cit. viii. 460 ; Rein, Japan, i. 584. 4 Laws of Hammurabi, §§ 150, 171 sq. (Johns' translation, pp. 30, 35 sq.; Winckler's translation, pp. 43, 49 sqq.). Koschaker, op. cit. p. 164 sqq. 5 Benzinger, op. cit. p. 106. 6 Genesis, xxiv. 53. Cf. ibid, xxxiv. 12. 7 Robertson Smith, op. cit. p. 76. Wellhausen, loc. cit. p. 434. For presents given by bridegrooms to their brides in Muhammadan countries see Westermarck, Marriage Ceremonies in Morocco, p. 82 sq. XXIII MARRIAGE BY CONSIDERATION, ETC. 417 these gifts the bride price, or a part of it, became in the course of time the property of the Semitic bride. Herodotus was probably well informed when he wrote that among the Babylonians " the marriage portions were furnished by the money paid for the beautiful damsels." 1 Dr. Koschaker believes that in Sumerian law the old bride price survived as a present, called nig-mussa, which the bridegroom gave to the bride.2 Among the Israelites the mohar was sooner or later, wholly or partly, given to the bride ; 3 Laban's daughters complained that their father had sold them as slaves and wasted their mohar.In Arabia the mahr was already in pre-Muhammadan times, at least occasionally, given to the wife as her property ; and under Islam the distinction between mahr and sadaq disappeared alto- gether.5 To this day the custom of the husband providing the wife with a dowry is found both among Jews and Muhamma- dans. In order to protect the wife in the event of her becoming widowed or divorced, it was established by the Jewish Law that before the nuptials the husband was to make out an obligation in writing, which entitled her to receive a certain sum from his estate in the case of his death or in the case of her divorcement. This obligation was termed kethubhah (the marriage deed). As minimum of this obligation was fixed the sum of two hundred silver denarii at the marriage of a virgin and one hundred at the marriage of a widow. For the security of the wife's claim to the amount fixed in the kethubhah all the property of the husband, both real and personal, was mortgaged. The kethubhah is still retained in most Jewish marriages, though it has little legal significance in many countries.6 It is said that the institution in question was originated or regulated by Simon ben Shatach, about 100 b.c,7 But 1 Herodotus, i. 196. 2 Koschaker, op. cit. p. 161 sqq. 3 Saalschiitz, Das mosaische Recht, ii. 736. Mayer, Die Rechte der Israeliten, Athener und Romer, ii. 342 sq. Benzinger, op. cit. p. 106. 4 Genesis, xxxi. 15. 6 Wellhausen, loc. cit. p. 434 sq. fi Abrahams, ' Marriage (Jewish),' in Hastings, op. cit. viii. 462. 7 Mielziner, op. cit. p. 85 sq. VOL. II EE 418 THE HISTORY OF HUMAN MARRIAGE chap. it seems very probable that it is in some way connected with the old custom of marriage by consideration.1 This is undoubtedly the case with the Muhammadan mahr or saddq, which, though handed over to the father of the bride, is presumed by the Koranic law to become the property of the bride herself.2 Islam requires the giving of a saddq for the contraction of a valid marriage. It is true that a man may legally marry a woman without mentioning a saddq, but in such a case the law presumes a consideration in her favour by virtue of the contract itself.3 As to the amount of the saddq there is no maximum fixed by law, though an excessive amount is considered improper both by the Sunnis and the Shi'ahs. As the minimum the early Hanaii lawyers fixed ten dirhems, equal to about four or five shillings, whereas the Malikis, inhabiting a poorer and less populous country than that in which the Hanafi doctrine flourished, regarded three dirhems as the lowest sum which could be given by way of saddq ; but these minimums have been abandoned long ago.4 In some parts of Morocco the saddq is fixed once for all by custom, although the amount may vary greatly even in the same tribe ; in one subdivision of the tribe Ulad Bu'aziz in Dukkala, for instance, it is twenty mctqal, nominally eight Spanish pesetas, whereas in another subdivision it is as much as four hundred metqal. As a rule, however, the saddq varies according to circumstances ; at Fez people who are not considered well-off pay seventy to a hundred dollars for a virgin and thirty to forty for a widow or a divorced woman, whilst the saddq may be as much as six hundred dollars if the parents of the parties are wealthy.6 At Cairo, according to Burckhardt, " among the first-rate merchants the price is from 200 to 300 dollars ; among those of the second class, from sixty to eighty ; and the lower 1 Cf. Kohler, ' Rechtsphilosophie und Universalrechtsgeschichte,' in v. Holtzendorft, Enzyklopddie der Rechtswissenschaft, i. 29. 2 Koran, iv. 3. Wellhausen, loc. cit. p. 435. 3 Ameer Ali, Mahommedan Law, ii. 472 sq. 4 Ibid. ii. 473 sq. Sautayra and Cherbonneau, Droit Musulman, i. 95. Sidi Halil, op. cit. § 168 (p. 61). 5 See Westermarck, Marriage Ceremonies in Morocco, p. 66 sqq. XXIII MARRIAGE BY CONSIDERATION, ETC. 419 classes often pay no more than from three to five dollars."1 In Mecca the mahr varies between a couple of dollars and some hundred.2 In Medina four hundred dollars would be considered a fair average sum among respectable citizens.3 Among the Bedouins of Mount Sinai the price of a girl " is from five to ten dollars, but sometimes amounts to thirty, if the girl is well connected and very handsome."4 The sadaq is commonly smaller if the woman is a widow or a divorced wife ;5 but in Morocco, at least, it is not invariably so.6 It is not necessary that the sadaq should be paid in money ; Sidi Halil points out that it is lawful to contract a marriage for a sadaq consisting of household furniture, or a certain number of camels or slaves or the like.7 There is nothing in the Koran or in the traditions tending to show that the whole of the sadaq must be paid prior to the consummation of the marriage ; hence later jurists have held that only a portion of it should be considered payable at once or on demand, and the remainder within a certain stipulated period or on the dissolution of the contract, whether by divorce or the death of either of the parties.8 At Cairo two-thirds 9 or one-half10 of the money is usually paid immediately. In Morocco the sadaq is sometimes paid in full before the consummation of the marriage, sometimes one-half of it or a smaller part is left unpaid. The " de- ferred " portion may be paid by instalments or compensated for by a present given to the wife by the husband ; but in 1 Burckhardt, Arabic Proverbs, p. 113. 2 Snouck Hurgronje, Mekka, ii. 158 sq. 3 Burton, Personal Narrative of a Pilgrimage to Al-Madinah and Meccah, ii. 23 n. 2. 1 Burckhardt, Notes on the Bedouins and Wakdbys, p. 152. 5 Lane, Arabian Society in the Middle Ages, p. 230 (Cairo). Burckhardt, Notes on the Bedouins and Wahdbys, p. 153 (Bedouins of Mount Sinai). Westermarck, Marriage Ceremonies in Morocco, pp. 67-69, 71 sqq. 6 Westermarck, op. cit. pp. 66, 73. 7 Sidi Halil, op. cit. § 154 (p. 55). 8 Sautayra and Cherbonneau, op. cit. i. 98 sqq. Ameer Ali, op. cit. ii. 482. Sidi Halil, op. cit. §§ 156, 173 (pp. 56, 62). 9 Lane, Arabian Society in the Middle Ages, p. 230. 10 Burckhardt, Arabic Proverbs, p. 113. 420 THE HISTORY OF HUMAN MARRIAGE chap. many cases it is only paid on the death of the latter or if he divorces his wife without sufficient reason, and in some tribes these are the only circumstances in which the sadaq or a part of it is paid at all, nothing of it being given when the marriage is contracted.1 Among the Indo-European peoples the transformation of the bride price is indicated in their languages by the fact that the words used for it subsequently assumed the meaning of dowry.2 In India the price originally due to the parents or guardian of the bride who surrendered her to the bridegroom became in after times a wedding present, which the bride received from the bridegroom either directly or through her parents.3 Manu says, " When the relatives do not appropriate for their use the gratuity given, it is not a sale ; in that case the gift is only a token of respect and of kindness towards the maidens." 4 This gift was called culka, or the bride's fee ; but its close connection with a previous purchase appears from the fact that it passed in a peculiar course of devolution to the woman's own brothers, and one rendering of the text of Gautama which regulates this succession even allowed the fee to go to her brothers during her life.5 In modern India, according to Dubois, men of distinction do not appropriate the money acquired by giving a daughter in marriage, but lay it out in jewels which they present to the bride on the wedding day.6 Among the Greeks of the heroic age the bridegroom gave not only to the father of the bride but b&pa, or presents, to the bride herself.7 Moreover, the father did not always keep the ebva for his own use, but bestowed them wholly or in part on the daughter; and we are also told that the bridegroom himself gave presents to his wife, 1 See Westermarck, Marriage Ceremonies in Morocco, p. 66 sqq. 2 Schrader, Reallexikon der indogermanischen Alter lumskunde, p 544. Piprek, op. cit. p. 151 (Slavs). 3 Jolly, in Institutes of Vishnu, p. 69 n. 18. 4 Laws of Manu, iii. 54. 5 Mayr, Das indische Erbrecht, p. 170. Mayne, op. cit. p. 97. Gautama, xxviii. 25 sq. 6 Dubois, Description of the Character, Manners, and Customs of the People of India, p. 103. 7 Hruza, op. cit. i. 14. XXIII MARRIAGE BY CONSIDERATION, ETC. 421 when he saw her unveiled for the first time or after the vv^ /iva-TiKt].1 In Rome the bridegroom gave to the bride a betrothal present, called arrha sponsalitia, which may, or may not, have been the survival of an earlier bride price.2 Among the Teutonic peoples a similar process of develop- ment took place. In the time of the ' folk-laws, from the sixth to the ninth century, the bride price is no longer paid to the father or guardian of the bride, but to the bride herself,3 the right of the guardian being practically limited to the receipt of the handheld, that is, to a merely formal fulfilment,4 a solidus and a denarius according to Salic law.5 This, however, does not mean that the former bride price was actually paid to the bride at the conclusion of the mar- riage : since the wife's property was subject to the husband's control during his lifetime, the bride price was really trans- formed into a provision for the widow, payable only after death from the husband's goods.6 " When light begins to fall upon the Anglo-Saxon betrothal," say Pollock and Maitland, " it is not a cash transaction by which the bride's 1 Rossbach, op. cit. p. 220 sq. Hermann-Blumner, Lehrbuch dev gviechischen Privatalterthumer, pp. 262, 266. Becker-Goll, Charikles, iii. 377. 2 Karlowa, op. cit. ii. 166. 3 Grimm, Deutsche Rechtsalterthilmev, p. 423. Sohm, Das Recht der Eheschliessung aus dem deutschen und canonischen Recht ge- schichtlich entwickelt, p. 33. Brunner; Grundzuge der deutschen Rechtsgeschichte, p. 223. Laboulaye, Histoire du droit de propridte foncibre en Occident, p. 403 sq. Ginoulhiac, Histoire du regime dotal, p. 187 sq. Friedberg, Das Recht der Eheschliessung in seiner ge- schichtlichen Entwicklung, p. 19. Lehmann, Verlobung und Hochzeit nach den nordgermanischen Rechten des friiheren Mittelalters, p. 59. Olivecrona, op. cit. pp. 57, 152. Weinhold, Die deutschen Frauen in dem Mittelalter, i. 325. v. Amira, loc. cit. p. 163. Gudmundsson and Kalund, ' Sitte. Skaninavische Verhaltnisse,' in Paul, Grundriss der germanischen Philologie, iii. 418 sq. Schroeder, op. cit. p. 298. Howard, op. cit. i. 266. Hoops, Reallexikon der Germanischen Altertumskunde, i. 513. 4 Sohm, Das Recht der Eheschliessung, p. 33. 5 Grimm, op. cit. p. 424. Friedberg, op. cit. p. 19 n. 7. 8 Sohm, Das Recht der Eheschliessung, p. 33. Brunner, Grundzuge der deutschen Rechtsgeschichte, p. 223. Friedberg, op. cit. p. 19. Howard, op. cit. i. 266 sq. 422 THE HISTORY OF HUMAN MARRIAGE chap. kinsmen receive a price in return for rights over their kinswoman ; rather we must say that the bridegroom cove- nants with them that he will make a settlement upon his future wife. He declares, and he gives security for, the morning-gift which she shall receive if she ' chooses his will ' and the dower that she shall enjoy if she outlives him. Though no doubt her kinsmen may make a profit out of the bargain, as fathers and feudal lords will in much later times, the more essential matter is that they should stipulate on her behalf for an honourable treatment as wife and widow."1 As to the morning gift-which has survived very long in Europe, indeed in Germany2 and Switzerland3 up to our own time-various writers have expressed the opinion that it originated in the bride price or formed a part of it.4 Schroeder, again, maintains that it has de- veloped out of an ancient " adoption gift," although it afterwards became a pure provision for the widow and, when it consisted of immovable property, was almost entirely amalgamated with the former bride price. He connects it with the dos spoken of by Tacitus 5 in the state- ment that among the ancient Germans the husband brought a dowry (dos) to the wife, not the wife to the husband.6 It has, further, been regarded as a -pretium virginitatis ;7 but against this view the argument has been adduced that the morning gift was also sometimes given to widows.8 The very name of this gift, however, certainly suggests that it has something to do with the consummation of the marriage. But it is quite possible that some other idea than that of mere compensation was connected with it. In Morocco and 1 Pollock and Maitland, op. cit. ii. 365. 2 Eichhorn, Einleitung in das deutsche Privatrecht, p. 726. 3 Bluntschli, Staats- und Rechtsgeschichte der Stadt und Land- schaft Zurich, ii. 164 sq. 4 Schlyter, Juridiska afhandlingar, i. 201. Schlegel, ' Om Mor- gengavens Oprindelse,' in Astrea, ii. 189 sq. Koenigswarter, Histoire de Vorganisation de la famille en France, p. 123. 5 Tacitus, Germania, ch. 18. 6 Schroeder, op. cit. pp. 70, 71, 310 sqq. 1 Ginoulhiac, op. cit. p. 202. Warnkoenig and Stein, Franzosische Staats- und Rechtsgeschichte, ii. 257. 8 Weinhold, op. cit. i. 402. Schroeder, op. cit. p. iiq. xxin MARRIAGE BY CONSIDERATION, ETC. 423 other Muhammadan countries a present of money is given by the bridegroom to the bride immediately before, or sometimes after, the consummation of the marriage ; and I have found reason to suggest that its original object was to serve as a safeguard against evil influences.1 Among the Great Russians, according to Professor Kovalewsky, the father, as a rule, disposes of the money received from the bridegroom in favour of his daughter, giving her as dowry a larger or smaller sum, according to what he has himself received.2 In the Brehon law it is stipulated that the woman's father, though entitled to the whole of the first year's coibche, gets only two-thirds of the second year's, one-half of the third year's, and so on ; whilst in each case what is left of the coibche belongs to the wife.3 In ancient Wales the husband had to give to his wife a present, called cowyll, " for her maidenhood " on the morning after the consummation of the marriage. Its amount, which was fixed by the law, depended on the status of her father.4 From marriage by consideration we have thus reached the practice of providing the bride with a marriage portion, which in part consists of the price paid for the bride. The marriage portion serves different ends, often indissolubly mixed up together. It may have the meaning of a return gift. It may imply that the wife as well as the husband is expected to contribute to the expenses of the joint house- hold. It is very often intended to be a settlement for the wife in case the marriage is dissolved through the husband's death or otherwise. But as in such cases the husband generally has the usufruct of the portion as long as the marriage lasts, it may be a return gift to the man at the same time as it is a settlement for the woman. And it may also be, practically, a means of buying a husband. 1 Westermarck, Marriage Ceremonies in Morocco, p. 262 sqq. 2 Kovalewsky, in Folk-Lore, i. 479. 3 Ancient Laws and Institutes of Ireland, ii. 347 ; iii. 315. 4 Venedotian Code, ii. 1. 32, 39 ; Dimetian Code, ii. 8. 73 (Ancient Laws and Institutes of Wales, pp. 42, 44, 223). 424 THE HISTORY OF HUMAN MARRIAGE chap. Speaking of newly-married people among the ancient Mexicans, Acosta says, " When they went to house they made an inventory of all the man and wife brought together, of provisions for the house, of land, of iewells and ornaments, which inventories every father kept, for if it chanced they made any devorce (as it was common amongest them when they agree not), they divided their goods according to the portion that every one brought." 1 In ancient Babylonia the bride usually brought a dowry from her father's house, which, remained her property, although the husband had the usufruct of it. 2 According to the Laws of Hammurabi, it is returned to her if the husband puts her away ;3 or "if she has been economical and has no vice, and her husband has gone out and greatly belittled her," and she in consequence leaves him ;4 or if she does so because a sickness has seized her and her husband takes another wife.5 On her death it passes to her children,6 and, in case she leaves no children, to the house of her father-, if the bride price has been returned.7 But if the husband does not get back the bride price, he shall deduct it from the marriage portion and give the rest to the father-in-law.8 This shows that the marriage portion might -exceed the bride price. And the custom of providing the bride with a marriage portion, which gradually assumed great import- ance in Babylonian law, remained after the practice of paying a bride price had fallen into desuetude.9 Among the Muhammadans the father of the bride often gives her something in addition to the sadaq or mahr provided by the bridegroom. Among the Muhammadans of India 1 Acosta, Natural and Moral History of the Indies, ii. 370. 2 See Kohler and Peiser, Aus dem Babylonischen Rechtsleben, iv. 11. 3 Laws of ^ammurabi, § 138 (Johns' translation, p. 27 ; Winckler's translation, p. 41). ' Ibid. § 142 (Johns' trans, p. 28 ; Winckler's trans, p. 41). ' Ibid. § 148 sq. (Johns' trans, p. 29 ; Winckler's trans, p. 43). 6 Ibid. § 162 (Johns' trans, p. 32 ; Winckler's trans, p. 47). 7 Ibid. § 163 (Johns' trans, p. 32 sq. ; Winckler's trans, p. 47). 8 Ibid. § 164 (Johns' trans, p. 33 ; Winckler's trans, p. 47). 9 Kohler, in Holtzendorfi, op. cit. i. 29. xxiii MARRIAGE BY CONSIDERATION, ETC 425 " the bride's father is bound to furnish her with a wedding outfit, unless he is in straitened circumstances, when the obligation rests with the husband. When a man advances a sum of money towards the trousseau of his future wife, it is obligatory on the father, if he be possessed of means, to do the same or contribute such an amount as is customary." 1 In Egypt not only the dowry paid by the bridegroom but an additional sum supplied by the bride's family, which is often more than the dowry itself, is expended in purchasing the articles of furniture, dress, and ornaments for the bride.2 At Aleppo in Syria " le prix paye est . . . . employe, avec une pareille somme que donne le pere de la future, en meubles, ustensiles, effets d'habillement."8 At Fez I was told that custom requires the father of the bride to spend on her trousseau from his own money at least the same amount as the sadaq-the so-called mltsdl* In Andjra, in Northern Morocco, he is expected at his own expense to provide her with a sivdr of wearing apparel ;5 and among the Ait Tameldu, a Berber tribe of the Great Atlas, it is a matter of pride to a father to give his daughter a good liqqdma, consisting for example of a cow, some sheep, silver orna- ments, silk kerchiefs, and other articles, worth perhaps from fifty to two hundred dollars, although a poor man may have nothing to give but some clothes bought with the dmerwas; or sadaq. Among the Vedic people dowries were not infrequently given by fathers or brothers in order to secure the marriage of daughters or sisters.6 It may be assumed that in such cases the husband appropriated the dowry, as well as her earnings, if any ; for even in the Epic 7 the rise of the recog- 1 Ameer Ali, op. cit. ii. 508. 2 Lane, Manners and Customs of the Modern Egyptians, p. 175. Cf. Trumbull, Studies in Oriental Social Life, p. 20. 3 Guys, Un Dervich algerien en Syrie, p. 19g. 4 Cf. Leo Africanus, History and Description of Africa, ii. 450. 5 Cf. Michaux-Bellaire, ' Quelques tribus de montagnes de la region du Habt,' in Archives marocaines, xvii. 128. 6 Macdonell and Keith, op. cit. i. 482. Keith, in Hastings, op, cit. viii. 449. 7 Satapatha Brdhmana, iv. 4. 2. 13. 426 THE HISTORY OF HUMAN MARRIAGE chap. nition of women's property as their own (stridhand) is only slow.1 The stridhana is mentioned by Gautama2 but first described in detail in the Artha^astra3 and by Vishnu.4 It included any presents from parents, sons, brothers, or other relatives, the marriage gifts, the bride price (gulkd) when given to the woman by her father, and the fine paidi by her husband if she was degraded from her position as chief wife in favour of another.5 This property fell on her death to her daughters, if she had any ;6 if not, apparently, to her sons, who according to some shared it with their sisters in any event.7 In case she died without issue, it belonged to her husband if she had married according to one of the four superior forms of marriage, but otherwise to her father.8 The Hindu law recognises the dominion of a married woman over this property, but the husband has nevertheless power to use and consume it in case of distress.9 In ancient Gaul, according to Caesar, the wife brought a dowry, but the husband also added an equal amount taken from his own property ; and on the death of either party the survivor received both portions, along with the revenues accumulated after marriage.10 Among the Irish the fact that the husband paid the bride price did not prevent the bride bringing goods or valuables of her own, if she had them ; very frequently, in fact, she brought with her jewels or gold or herds or lands, which continued to be her own special property.11 Moreover, in many cases the friends of the young couple made a collection for them, which was 1 Macdonell and Keith,-op. cit. i. 484. 2 Gautama, xxviii. 24 sq. 3 Arthahdstra, iii. 2. 59, quoted by Keith, loo. cit. p. 454. 4 Institutes of Vishnu, xvii. 18 sqq. 5 Ibid. xvii. 18. Laws of Manu, ix. 194. Mandlik, Vyavahdra Maydkha, p. 91 sq. Keith, loc. cit. p. 454. 8 Gautama, xxviii. 24. Institutes of Vishnu, xvii. 21. 7 Laws of Manu, ix. 192, 195. Mandlik, op. cit. p. 95. Keith, loc. cit. p. 454. 8 Institutes of Vishnu, xvii. 19 sq. 9 Macnaghten, Principles of Hindu Law, p. 33 sq. Steele, Law and Custom of Hindoo Castes, p. 67. 10 Caesar, Commentarii de bello Gallico, vi. 19. Cf. d'Arbois de Jubainville, op. cit. vii. 240 ; Jullian, Histoire de la Gaule, ii. 408. 11 Joyce, op. cit. ii. 6. Sullivan's ' Introduction,' in O'Curry, bp. cit. vol. i. p. clxxii. XXIII MARRIAGE BY CONSIDERATION, ETC. 427 called tin 61, that is, " collection," of which two-thirds belonged by law to the man and one-third to the woman.1 If the wife separated in consequence of injury inflicted on her by the husband and proved her case, she was entitled to her dowry, or that part of it that remained with her after marriage.2 If the couple separated by mutual consent, the woman took away with her all she had brought on the marriage day ; whilst the man retained what he had con- tributed. Supposing the joint property had gone on increasing during married life, then at separation the couple divided the whole in proportion to the original contributions.3 In ancient Wales the daughter seems to have been entitled to a marriage portion or settlement (gwaddol) from her father or kindred, which usually included not only things of utility for a new household but also articles for her own use.4 In case the couple separated before the end of seven years (less three days), the wife was to receive this portion back, whereas if she left her husband before the seventh year without good cause she lost all her property except the cowyll and her right to any fine due from the husband for having committed adultery. If the separation took place after this period, the property of the couple was divided into two portions.5 At Athens it was the general rule that a woman at her marriage brought with her a dowry (7rpoi£), usually con- sisting of money and often also of movable objects, rarely of real estate.6 It was the wife's contribution towards the expenses of the marriage, and at the same time served as an obstacle to the dissolution of the union for frivolous reasons.7 The amount of the dowry was left to the dis- 1 Ancient Laws and Institutes of Ireland, ii. 347, 350 sq. Joyce, op. cit. ii. 6. 2 Ancient Laws and Institutes of Ireland, v. 293. 3 Ibid. ii. 397. 4 Rhys and Brynmor-Jones, op. cit. p. 209. 5 Ibid. p. 213. Lewis, Ancient Laws of Wales, p. 8 sq. 6 Hruza, op. cit. i. 24. Beauchet, op. cit. i. 289 sqq. Hermann- Bliimner, op. cit. p. 263. ' Beauchet, op. cit. i. 299. Hruza, op. cit. i. 24 sq. Becker-Goll, op. cit. iii. 312. Hermann-Blumner, op. cit. p. 264. Isaeus, Oratio de Pyrrhi hereditate, § 36, p. 41. 428 THE HISTORY OF HUMAN MARRIAGE chap. cretion of her father,1 and there is no evidence that it was a legal obligation for him to provide his daughter with any dowry at all.2 At the same time the dowry became almost a criterion of honourable marriage as distinguished from concubinage ;3 Isaeus says that no decent man would give his legitimate daughter less than a tenth of his property.4 The husband enjoyed the usufruct of the dowry ;5 hence Euripides, transferring to the heroic age the practice of his own time, makes Medea complain that her sex had to purchase husbands with great sums of money.6 But the dowry remained the wife's property; and as the husband might have to pay it back some day, he was generally required to mortgage real estate as security.7 While the union continued the dowry could not be withdrawn; but upon dissolution of marriage on the initiative of either party or by mutual agreement it must be refunded to the woman's father or guardian.8 Rupture by death of either party compelled restitution if there was no issue. On the other hand, if there were children and the widow continued to live in the conjugal domicile, the dowry became their property ; but she might also choose to go back with her dowry to her father's house. The dowry thus followed the wife and went to her children.9 It may be added that in Greece the privilege of being provided with a marriage portion was not restricted to the women of Athens. In the time of Aristotle nearly two-fifths of the whole territory of Sparta were supposed to belong to women as their dowers.10 In Rome, even more than in Greece, the marriage portion became a mark of distinction for a legitimate wife. A 1 Beauchet, op. cit. i. 287 sqq. 2 Ibid. i. 262 sqq. 3 Cauvet, ' De 1'organisation de la famille & Athenes,' in Revue de legislation et de jurisprudence, xxiv. 154. 4 Isaeus, Oratio de Pyrrhi hereditate, § 51, p. 43. 5 Cauvet, loc. cit. p. 154. Beauchet, op. cit. i. 303 sqq. 0 Euripides, Medea, 231 sqq. 7 Cauvet, loc. cit. p. 155. Meier and Schomann,- Der attische Process, p. 518 sq. Mayer, op. cit. ii. 345 sq. Hermann-Bliimner, op. cit. p. 265. Beauchet, op. cit. i. 331 sqq. Smith, Wayte, and Marindin, op. cit. i. 692. 8 Beauchet, op. cit. i. 317 sqq. ? Ibid. i. 311 sqq. W Aristotle, Politica, ii. 9. ij. xxni MARRIAGE BY CONSIDERATION, ETC. 429 woman had a legal right to demand a dos, or dower, from her father, but it was to be given to her husband, not to herself, as a contribution towards the defrayal of the expenses of the joint household-ad matrimonii oner a ferenda-although it was also intended as a provision in the interests of the wife.1 According to the ancient law of the Republic the husband had all the rights and remedies incident to ownership as such, including amongst others the right to alienate and mortgage ; and he was recognised as the sole owner of the dos not only during the marriage but also after its dissolution.2 But in the course of time the husband's rights were subjected to important restrictions. The * Lex Julia de adulteriis ' of the year 18 b.c. prohibited him from alienating or mortgaging any fundus Italicus comprised in the dos ;.and Justinian extended this prohibition to any dotal land whatever. Not even the wife's consent could validate a mortgage or a sale of the dotal land, which was thus in all circumstances preserved intact for the wife.3 The husband's right to the use and the fruits of the dos was restricted to the time during which the marriage lasted. In the law prior to Justinian his obligation to restore the dos was still a very limited one : he continued to be acknow- ledged as the true owner of it even after the dissolution of the marriage, and it was only within certain limits that the law allowed the wife, or the person who provided the dos, to maintain an obligatory claim for a return of it side by side with the ownership of the husband. But Justinian's law imposed upon the husband the duty to return the dos in all circumstances, except when the dissolution of the marriage was caused by the misconduct of the wife.4 These restrictions*in the husband's right to dispose of his wife's marriage portion were probably, to some extent, connected with the loosening of the marriage tie ; the confusion of the dos with the husband's patrimony was of comparatively little moment as long as marriage was contracted for life, but must have become intolerable when divorces grew frequent. But it seems that the lines upon which the 1 Sohm, Institutes, p. 465 sq. 2 Ibid. p. 467 sq. 3 Ibid. p. 467. 4 Ibid. p. 470 sq. 430 THE HISTORY OF HUMAN MARRIAGE chap. development of the Roman law of dos proceeded were also largely determined by the influence of Greek law, according to which the wife was the owner of her marriage portion.1 The general tradition of the Roman dos was carried on by the Church, the practical object of the marriage portion being to secure for the wife a provision of which the husband could not wantonly deprive her and which would remain to her after his death.2 The Justinian principle that the wife's dotation remains her property, although the husband administers and has the use of it, underlies the later legisla- tion on the subject, though it has been more or less modified in the laws of the different countries. Already Justinian declared, in several constitutions, that the giving of a dos is obligatory for persons of high rank only ;3 but the old custom did not fall into desuetude. According to many later laws a daughter is entitled to demand a dowry at her marriage.4 The Prussian ' Landrecht ' still prescribes that the father, or eventually the mother, shall arrange about the wedding and fit up the house of the couple.5 According to the ' Code Napoleon,' on the other hand, parents are not bound to provide their daughter with a marriage portion ;6 and the same principle has generally been adopted by modern legislation. Yet there is still a strong feeling, especially in the so-called Latin countries, in favour of dotation. This feeling, as Sir Henry Maine remarks, is the principal source of those habits of saving and hoarding which characterise the French people, and is probably descended, by a long chain of succession, from the obligatory provisions of the Roman law.7 In our days there is particularly one factor that tends 1 Mitteis, Reichsrechte und Volksrecht in den ostlichen Provinzen des romischen Kaiserreichs, p. 231 sqq. Sohm, Institutes, p. 472 n. 5. 2 Maine, Early History of Institutions, p. 338. 3 Ginoulhiac, op. cit. p. 103. 4 For dos necessaria in Germany see Mittermaier, Grundsatze des gemeinen deutschen Privatrechts, ii. 3. 5 Eccius, ' Dotationspflicht,' in Holtzendorff, Enzyclopadie der Rechtswissenschaft, pt. ii. vol. i. 414. 6 Code Napoleon, art. 204. 7 Maine, Early History of Institutions, p. 339. XXIII MARRIAGE BY CONSIDERATION, ETC. 431 to preserve the marriage portion as a social institution of some importance. In a society where monogamy is prescribed by law, where the adult women outnumber the adult men, where many men never marry, and where married women too often lead an indolent life-in such a society the marriage portion not infrequently becomes a purchase-sum by means of which a father buys a husband for his daughter, as formerly a man bought a wife from her father. In India the difficulty of finding a husband for a daughter has led to undisguised purchase of bridegrooms. Whilst the low castes ordinarily pay for the bride, the high castes pay for the bridegroom ; and in some cases very large sums are paid, especially where hypergamy prevails, that is, where girls must marry in a caste equal or superior to their own, or where there is a great shortage of women. In recent times the bridegroom price has been affected very largely by the educational qualifications of the bridegroom. A Kayastha graduate in Bengal usually fetches from Rs.500 to Rs.1,000, and in some instances even Rs.10,000 have been recorded.1 1 Gait, Census of India, 1911, vol. i. (India) Report, p. 257. Subramhanya Aiyar, ibid. vol. xxiii. (Travancore) Report, p. 149. Elliot, Memoirs on the History, Folk-Lore, and Distribution of the Races of the North Western Provinces of India, i. 250. Crooke, Things Indian, p. 316. Risley, People of India, p. 165 sqq. CHAPTER XXIV MARRIAGE RITES Even when the consent which is necessary for the con- clusion of a marriage has been given and other conditions mentioned above have been fulfilled, something may still be required to make the union valid or the marriage, from the legal point of view, complete. Moreover, though not indispensable for the conclusion of a marriage, certain ceremonies are often observed in accordance with old- established custom. The rites connected with the conclusion of a marriage often form a long sequence of practices and taboos, which may commence at the moment when the marriage is first thought of and last till after it has been concluded. They are particularly prominent at the betrothal and the wedding and during the period lying between these events. The length of this period, or of the interval between the cele- bration of the betrothal and the wedding, varies indefinitely -it may last for years or months or days or only a few hours ; or there may be no such period at all. In the West of Ireland only a day or two intervenes between the festive betrothal and the nuptial ceremony.1 Among the Jews it was already in the eleventh century customary to solemnise both the betrothal and the marriage proper on the same day, either contemporaneously or with an interval of a few hours, during which the bridal party feasted merrily 1 Blake, ' Matrimonial Customs in the West of Ireland,' in Folk- Lore, xviii. 80. XXIV MARRIAGE RITES 433 at the new husband's cost.1 In ancient Rome the be- trothal, or sponsalia, although the proper and usual pre- liminary of marriage, was not legally necessary; and in the ritual of the Christian Church, which is largely derived from the Roman sponsalia and miptice, the betrothal and nuptials were from early times combined in practice.2 But in popular customs they still remain separate, though the introduction of the ecclesiastic ceremony has led to a great confusion of the rites practised on those occasions.3 In a monograph on the marriage rites of a single people or group of related peoples it is natural to deal with them in the order in which they follow upon one another. This is what I have done in my essay on ' Marriage Ceremonies in Morocco.' But in the present treatise I shall adopt another method. Similar rites will be grouped together even though found at different stages of the ritual. This, of course, does not imply that no notice is taken of their place in the sequence ; the occasion when a rite occurs is often of the greatest importance for its interpretation. But the same rite may occupy a different place in different cases ; some- times, for instance, it may be practised at the betrothal and at other times at the wedding, or it may occur on both occasions among the same people. My chief object will be to find out the meaning of the rifes. The conclusions, however, must often be hypothetical. For very often the meaning of a rite is not mentioned at all, or the inter- pretation given of it shows that the idea originally under- lying it has been forgotten ; and rites which externally resemble each other may serve very different purposes. The most general social object of marriage rites is to give publicity to the union. " Publicity," says Miss Burne, " is everywhere the element which distinguishes a recognised marriage from an illicit connection."4 In order to be recog- nised as valid, the union may have to be sanctioned by an 1 Abrahams, Jewish Life in the Middle Ages, p. 178. 2 Foley, * Marriage (Christian),' in Hastings, Encyclopedia of Religion and Ethics, viii. 436. 3 See Bachtold, Die Gebrduche bei Vevlobung und Hochzeit, par- ticularly i. 72. 4 Miss Burne, Handbook of Folklore, p. 203. VOL. II F F 434 THE HISTORY OF HUMAN MARRIAGE chap. official. This is the case not only in modem civilised countries. Among the ancient Peruvians the king convoked annually, or every two years, at Cuzco all the marriageable young men and maidens of his family. After calling them by name he joined their hands and delivered them to their parents. Such marriages among that class were alone denominated lawful; and the governors and chiefs were by their offices obliged to marry, after the same formalities, the young men and women of the provinces over which they presided.1 In Nicaragua, also, marriage was " a civil rite, performed by the cacique."2 Among the savage Porno of California, who have two chiefs, a " war-chief " and a " peace-chief," the latter, as being a kind of censor morum, has to perform the marriage ceremonies, so far as they ex- tend, that is, he causes the parties to enter into a simple covenant in presence of their parents and friends.3 Among the Subanu of Mindanao " marriages are performed by the timuai or chief of a settlement, and he may be rewarded for his services if the groom is able to make a gift."4 Among the Sakai of the Malay Peninsula the chief formally declared the parties married.5 In certain tribes no marriage is said to be permitted without the chief's approval. Publicity may be achieved in other ways as well. Among the Maori belonging to the inferior classes " it was cus- tomary, if the marriage had not been arranged at the annual meeting, for a girl intending matrimony to call an assembly of her friends the night before her marriage, and, standing up, proclaim to them, ' I am going to take a husband. So and So is his name.' This was sufficient for the purpose."6 Among the Kubus of Sumatra the chief part of the whole 1 Garcilasso de la Vega, First Part of the Royal Commentaries of the Yncas, i. 306 sq. 2 Squier, ' Observations on the Archaeology and Ethnology of Nicaragua,' in Trans. American Ethn. Soc. vol. iii. pt. i. 127. 3 Powers, Tribes of California, p. 157. 4 Finley and Churchill, Subanu, p. 29 sq. 5 de Morgan, ' Moeurs, coutumes et langages des Negritos de 1'interieur de la presqu'ile Malaise,' in Bull. Soc. normande de Geogr. vii. 422. 6 Tregear, The Maori Race, p. 293. XXIV MARRIAGE RITES 435 marriage ceremony is the simple announcement of the union.1 Among the pagan tribes of the Malay Peninsula in general the act of purchase alone is said to be regarded as suffi- ciently binding so long as it is performed before proper witnesses.2 Most Manipuris " regard cohabitation and public acknowledgment as sufficient," provided that due regard has been paid to their endogamous and exogamous rules.3 Among the Ekoi of Southern Nigeria the bride's acceptance of the bridegroom's wedding gift " must be followed by public proclamation oi the marriage before chiefs and people, after the bell has been rung round the town for the purpose,"4 Under the Muhammadan Sunni law it is required that there should be at least two witnesses present to attest the conclusion of the contract of marriage-to testify that it was properly entered into and in accordance with the conditions laid down for the contractual performance of marriage ;5 and the Malikis insist that the witnesses should be " men of established reputation."6 The Talmud recom- mended that a " congregation " should be constituted for the purpose of celebrating a wedding, that is, the presence of ten adult males was regarded as desirable, and in the Middle Ages many Jewish communities converted this desire into a binding statute.7 In classical Sanskrit litera- ture Agni (the fire) is often called the "witness" of marriages, and a marriage witnessed by the fire, according to Hindu ideas, cannot be annulled.8 Nowadays five of the gods are invoked and requested to be present at a Brahman wedding, namely Indra, Varuna, Chandra, Yama, and 1 Hagen, Die Orang Kubu auf Sumatra, p. 133. 2 Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. 55. 3 Hodson, Meitheis, p. 116. 4 Talbot, In the Shadow of the Bush, p. 105 sq. 5 Ameer Ali, Mahommedan Law, ii. 325. 6 Sldi JHalil, Muhtasar, 5 (Russell and Abdullah al-Ma'mun Suhrawardy, ' Manual of the Law of Marriage 'from the Mukhtasar of Sidi Khalil, p. 2). 7 Abrahams, op. cit. p. 199. 8 Wintemitz, ' On a Comparative Study of Indo-European Customs, with special reference to the Marriage Customs, ' in Trans- actions of the International Folk-Lore Congress, 1891, p. 287. Idem, Das altindische Hochzeitsrituell, p. 61. F F 2 436 THE HISTORY OF HUMAN MARRIAGE chap. Brahma.1 Among the Parsees a marriage must be cele- brated before an assembly of at least five persons who have been summoned for this special occasion.2 The Roman confarreatio, the patrician form of marriage, needed the presence of the Pontifex Maximus, the Flamen Dialis, and ten other witnesses.3 In Teutonic countries the betrothal has up to modern times taken place in the presence of witnesses.4 Publicity may have to be given to the sexual consumma- tion of the marriage. Among the Bantu Kavirondo the bridegroom consummates it in the presence of a large number of girls and women.6 Among the Lower Congo people witnesses have to see if the bridegroom can consummate the marriage ; and if through impotency he cannot, the marriage is broken off.6 Among the Koita of British New Guinea, when the young man is ready to marry he sends a message about it to the bride's mother, and " that night the bride prepares a sleeping mat for the bridegroom, who now comes to her publicly, before the house is quiet for the night."7 Among the lyca, a Chibcha tribe in Northern Colombia, the consummation of the marriage takes place in the presence of a respectable elderly Indian, though not actually under his eyes, and is then at once publicly announced.8 Among the Salinan Indians of California, " whenever a youth and maid appear in com- pany, both marked by the scratches of finger-nails, they are thus known to have contracted matrimony on the 1 Padfield, The Hindu at Home, p. 102. 2 Jivanji Jamshedji Modi, 'Marriage (Iranian).-1. Zoroastrian,' in Hastings, op. cit. viii. 456. 3 Rossbach, Untersuchungen uber die romische Ehe, p. 117 sqq. 4 Nordstrom, Bidrag till den svenska samhalls-forfattningens historia, ii. 15. Bachtold, Die Gebrauche bei Verlobung und Hochzeit, i. 76 sqq. Troels-Lund, Dagligt Liv i Norden i det i6dt A ar hundreds, ix. 131. Hylt6n-Cavallius, War end och Wirdarne, ii. 439. 5 Johnston, Uganda Protectorate, p. 747. 8 Weeks, ' Notes on some Customs of the Lower Congo People,' in Folk-Lore, xix. 413. 7 Seligman, Melanesians of British New Guinea, p. 77 sq. 8 Bolinder, Ijca-indianernas kultur, p. 252 sq. xxiv MARRIAGE RITES 437 preceding night. This alone is considered proof and they are publicly known as man and wife throughout the ran- cheria."1 In Teutonic countries, where a marriage was formerly regarded as legally valid only when it could be proved that the couple had been together under the same blanket, the bride and bridegroom went to bed in the presence of witnesses.2 This custom has survived up to quite modem times ;3 indeed, in some places in Skcine, in the south of Sweden, they were still in the beginning of the nineteenth century undressed in the presence of all the guests,4 and in Bohuslan, in the same country,5 and in Mark Brandenburg in Germany,6 it was not long ago the custom for the guests to enter into the bridal room after the couple had gone to bed. Among all Slavonic peoples bride and bridegroom are solemnly conducted into the nuptial chamber, and among most of them the bride goes to bed with her husband in the presence of witnesses.7 It seems that a very similar custom prevailed in ancient India.8 In Rome the pronuba accompanied the couple into the bridal room.9 An extremely frequent method of giving publicity to the union is to celebrate it with feasting, the guests being, in a way, regarded as witnesses.10 Sometimes the wedding takes place in the house of the bride's parents, sometimes in that of the bridegroom ; but feasts may also be held in 1 Fages, quoted by Mason, Ethnology of the Salinan Indians, p. 163. 2 Aldre V dstgotalagen, edited by Sjoros, Giptaer bolkaer, ix. 2, p. 59. Grimm, Deutsche Rechtsalterthilmer, p. 440 sq. Nordstrom, op. cit. i. 28 sq. Weinhold, Die deutschen Frauen in dein Mittelalter, i. 399. Lehmann, Verlobungund Hochzeit nach den nordgermanischen Rechten des friiheren Mittelalters, p. 80 sqq. 3 Troels-Lund, op. cit. xi. 181 sqq. 4 Nicolovius (Lovdn), Folklifvet i Skytts Harad i Skdne vid borjan af delta drhundrade, p. 144. 8 Holmberg, Bohusldns historia och beskrifning, i. 243. 8 Kuhn, Markische Sagen und Mdrchen, p. 358. 7 Piprek, Slawische Brautwerbungs- und Hochzeitsgebrauche, p. 174. 8 Haas, ' Die Heirathsgebrauche der alten Inder,' in Weber, Indische Studien, v. 278, 279, 401. Schrader, Reallexikon der indo ger manischen Altertumskunde, p. 358. 9 Rossbach, op. cit. p. 329. 10 Cf. Demosthenes, Oratio I. adversus Onetorem, 21. 438 THE HISTORY OF HUMAN MARRIAGE chap. both places. In some cases, however, the expenses are defrayed by the bridegroom, although the feast is celebrated in the bride's home.1 At the same time the wedding does not merely serve the object of making the marriage public, it brings together the families of the bride and the bride- groom and makes them more friendly to each other ; in this respect its social importance must be particularly great in countries where a common meal is looked upon almost as an act of covenanting.2 The various marriage rites are performed for many different' purposes. Some of them have direct reference to the separation of the bride from her old home. To these belong the ceremonial resistance made by her relatives or herself and the official crying of the bride, which have been discussed in a previous chapter. There are also, no doubt, other rites which might be classified among the so-called rites de separation, spoken of by M. van Gennep; but the list he gives of these rites3 is distinctly overloaded. Some of them are mainly prophylactic, and others have reference not to separation pure and simple but to the transition from one state of life to another. This, for instance, may be said of the custom of changing clothes, as when an English " peasant-bride about to dress for her wedding, first strips herself of every article of clothing, and begins absolutely de novo to attire herself in new and unwashed garments, rejecting even pins that have ever been used before."4 Again, the rites d'agregation, mentioned by M. van Gennep, belong to a larger class of marriage rites, namely, those which have a bearing upon the new condition brought about by the marriage. For many of these the term " rites of aggregation " would be entirely inappropriate. Some of the most frequent marriage rites symbolise the 1 E.g., among the Li in Hainan (Strzoda, ' Die Li auf Hainan,' in Zeitschr. f. Ethnol. xliii. 204), Burmese (Shway Yoe, The Burman, P- 57)- 2 See Westermarck,' Marriage Ceremonies in Morocco, p. 364 sq. Cf. van Gennep, Les rites de passage, p. 189 ; Junod, Life of a South African Tribe, i. 123. 3 van Gennep, op. cit. p. 185 sq. 4 Miss Burne, Shropshire Folk-Lore, p. 289. XXIV MARRIAGE RITES 439 union between the parties or, rather, are originally intended to strengthen the marriage tie. First, there is the joining of hands. In Africa this custom is found among the Wabondei,1 whilst in Abyssinia bride and bridegroom crook their little fingers together.2 At Limestone, in Queens- land, when a marriage is contracted, " the parties join hands in the presence of the tribe."3 Among the Nufors of New Guinea an old man puts the right hand of the bridegroom in that of the bride.4 Among the Benua of Malacca one of the elders of the tribe makes an address in which he says, " Listen, all ye that are present, those that were separated are now united " ; the young couple then approach each other and join hands, and the sylvan ceremony is concluded.5 Among the Sakai, again, " the little finger of the right hand of the man is joined to that of the left hand of the woman."6 Among the Gonds of the Eastern Ghauts the deesari, a sort of priest of indigenous origin, joins the little fingers of the couple, " apparently in recognition of their union in a holy wedlock."7 The old custom in Burma was " that the bride and bridegroom should join their right hands together, palm to palm, in the presence of all the assembled guests."8 The joining of hands, or the bridegroom's taking the bride by the hand, has of old been one of the most important marriage ceremonies among all Indo-European peoples.9 1 Dale, ' Account of the Principal Customs and Habits of the Natives inhabiting the Bondei Country,' in Jour. Anthr. Inst. xxv. 199. 2 Parkyns, Life in Abyssinia, ii. 54. 3 Lang, Cooksland in North-Eastern Australia, p. 394. Idem, Queensland, p. 337'. 4 van Hasselt, ' Die Noeforezen,' in Zeitschr. f. Ethnol. viii. 181. 5 Newbold, Account of the British Settlements in the Straits of Malacca, ii. 407. 6 Low, ' Karean Tribes,' in Jour. Indian Archipelago, iv. 431. 7 Hayavadana Rao, * Gonds of the Eastern Ghauts,' in Anthrofos, v. 794- 8 Shway Yoe, op. cit. p. 57. 9 Schrader, op. cit. p. 355 sq. Leist, Alt-arisches Jus Gentium, pp. 150, 156 sq. Haas, loc. cit. p. 310 sq. Schmidt, Beitrage zur Indischen Erotik. Das Liebesleben des Sanskritvolkes, p. 499 sqq. Sohm, Das Recht der Eheschliessung aus dem deutschen und canonischen Recht geschichtlich entwickelt, p. 57. Brand, Observations on Popular 440 THE HISTORY OF HUMAN MARRIAGE chap. In the Veda the husband is called hastagrabha, " hand- taker," and panigrahana or hastagrahana, " hand-taking," is a common name for " wedding " in Sanskrit.1 According to the Grihyasutras, the bridegroom with his right hand takes the right hand of the bride, reciting the verse from the Rig-Veda, " I take thy hand for the sake of happiness, that thou mayst live to old age with me, thy husband ; the gods Bhaga, Aryaman, Savitri, and Purandhi have given thee to me for householdership (or, for the sake of keeping the domes- tic fire)." 2 In Rome the bride, under the guidance of a pronuba who must be a matron only once married, placed her right hand in the right hand of the bridegroom.3 The joining of hands may undoubtedly be an expression of several different ideas. By the dextrarum junctio the bride came under the manus of the husband, or was " handed over " to him. The joining of hands is also from early times the outward sign of a troth that two persons give to each other ; Handschlag, Hand in Hand geloben, Handgelubde, are familiar legal phrases in Germany.4 But it is obvious that the rite in question very frequently at least is a symbolic act of union; and this may be the case even when it is something else besides. In some European countries and in many parts of India the hands of the bridal pair are not only joined but tied together. This custom occurs in Poland and Bulgaria.5 In Portugal the priest ties the hands of the pair with the end of his stole, and a similar custom is found among the Saxonians in Transylvania. Among several castes in Antiquities, p. 345 sq. Piprek, op. cit. pp. 5, 23, 89, 102, 109, 166 (Slavs). Machal, ' Marriage (Slavic),' in Hastings, op. cit. viii. 471. Bachtold, op. cit. p. 117 sqq. Wintemitz, in Transactions, p. 282. Idem, Das altindische Hochzeitsrituell, p. 48 sqq. On the joining of hands among Ugro-Finnic peoples see Hamalainen, Mordvalaisten, tseremissien ja votjakkien kosinta- ja hddtavoista, p. 218 sqq. 1 Wintemitz, in Transactions, p. 282. Idem, Das altindische Hochzeitsrituell, p. 48. Cf. Zimmer, Altindisches Leben, p. 311 sq. 2 Grihya-Sdtras, i. 35, 167, 282, 381, 383 ; ii. 47, 189, 259. 3 Festus, De verborum significatione quae supersunt, p. 242b. Rossbach, op. cit. pp. 308, 329. 4 Wintemitz, in Transactions, p. 282. Bachtold, op. cit, i. 113 sqq. s Piprek, op. cit. pp, 73, 146, XXIV MARRIAGE RITES 441 Southern India the bridegroom takes the bride by the hand, whereupon the hands are bound together with a handker- chief. In Bengal and in Lahore the hands of the young couple are tied together with a string of flowers, among the Parsees of Bombay with a delicate twine.1 Among the Sinhalese a maternal uncle of the bride or some other principal member of her family joins together with a chain the little fingers of the right hands of the bride and bride- groom, who thereupon turn three times round to the right; the chain is then taken off.2 According to another account of the same people, however, bride and bridegroom have generally their thumbs tied together; but ''when it is desired to make the marriage as firm and indissoluble as the nature of their manners will allow, the parties are joined together with a long piece of cloth, which is folded several times round both their bodies." 3 Among the Veddas the bride ties a thin cord of her own twisting round the bridegroom's waist, and they are then husband and wife ; this string is emblematic of the marriage tie, and " as he never parts with it, so he clings to his wife through life." 4 In Burma a ribbon is sometimes stretched round the couple, and their hands are clasped.6 Amongwthe Chukmas of Chittagong " the bride and bridegroom are made to sit together, and two of their relations, a man and a woman, will with the consent of all present bind the couple together with a white cloth."6 Among the Hindus of South India, after the bride's family priest has asked the bridegroom if he is willing to take So-and-so to wife and he has answered in the affirma- tive, the ends of the upper garments of the pair are tied together in what is called the Brahman knot. In tying this knot the priest says, "You must trust and be a prop to each other." They sit thus tied together until it may be necessary for them to move away from the place. This 1 Wintemi'tz, in Transactions, p. 283. 2 Niti-Nighanduva, p. 18. 3 Percival, Account of the Island of Ceylon, p. 180. 4 Bailey, ' Account of the Wild Tribes of the Veddahsof Ceylon,' in Trans. Ethn. Soc. N.S. ii. 293 sq. See also Seligman, Veddas, p. 97. 6 Fielding Hall, The Soul of a People, p. 187 sq. 6 Hutchinson, Account of the Chittagong Hill Tracts, p. 97. 442 THE HISTORY OF HUMAN MARRIAGE chap. tying of the cloths is an important part of the marriage ceremony, and is repeated at various stages of the pro- ceedings.1 Among the Moriori of the Chatham Islands " in the darkness of evening the young pair were set together in the centre of a house that had been swept and garnished, and their friends formed a circle about them, holding a thin rope of plaited grass, which was placed round the shoulders of the wedded pair and knotted into a ring (henga) binding them together."2 Among the Tarahumare of Mexico the bridal couple are covered with blankets, and in some cases their right hands are tied together.3 Among the Orang Mamaq of Sumatra the bride and bridegroom are sitting on the ground face to face with the feet touching each other.4 In the Andaman Islands the bridegroom is made to sit down on the bride's thighs, which are straightened out for the purpose by her female attend- ants.5 At Khasi weddings " the couple about to be married merely sit together in one seat, and receive their friends, to whom they give a dinner or feast."6 Among some of the Negritos of Zambales an old man of the tribe knocks the heads of bride and bridegroom together.7 Among the Negritos of Bataan, in .Luzon, " les fiances doivent grimper au sommet de deux arbres flexibles et Voisins ; le chef incline les arbres 1'un vers 1'autre, et, quand les fronts des deux futurs sont arrives au contact, le mariage est devenu un fait accompli."8 In China, after the couple have sipped some wine and changed glasses, the hair of the bridegroom is fastened to that of the bride, and vice versa ; hence the phrase " hair-fastened " couple, meaning in- separable or faithful and loving.9 1 Padfield, op. cit. p. 104 sq. 2 Tregear, The Maori Race, p. 580. 3 Lumholtz, Unknown Mexico, i. 870. 4 Hagen, Die Orang Kubu auf Sumatra, p. 165. 5 Man, ' Aboriginal Inhabitants of the Andaman Islands,' in Jour. Anthr. Inst. xii. 137. 6 Steel, 'On the Khasia Tribe,' in Trans.-. Ethn. Soc. London, N.S. vol. vii. 308. 7 Reed, Negritos of Zambales, p. 59. See also Lala, Philippine Islands, p. 96. 8 Montano, Voyage aux Philippines, p. 71. 9 Leong and Tao, Village and Town Life in China, p. 107. XXIV MARRIAGE RITES 443 The union between the bride and bridegroom may also be represented by the tying of something to each of them separately. Among the Nandi they bind a sprig of sekutiet grass (Vernonia) on to each other's wrists, after which much feasting and dancing is indulged in.1 Among the Southern Galla each person present at the wedding ties a knot in the fringe of the bridegroom's toga and in the fringe of the bride's garments; and these knots are never undone.2 Among the Basuto the bride's father shows that the bargain is completed by killing the fattest ox he possesses and cutting off the dewlap and dividing it into two strips, one of which is bound round the girl's wrist and the other sent to the bridegroom to be bound round his wrist; this signifies that they are now bound to each other.3 It seems that betrothal and wedding rings, partly at least, serve a similar purpose. Betrothal rings are found among the natives of the Portuguese Zambesi4 and the Mikirs of Assam.5 Among the Fors of Central Africa, when the bride, in the afternoon of the seventh day after the wedding, meets the bridegroom at the door of her hut, he presents her with a marriage ring, and she makes him in return a present of a love token and a ring.6 Among the Khasis it was formerly the custom for the bridegroom to place a ring-usually of silver, but amongst the rich of gold-- upon the bride's finger, and for the bride to place a similar ring upon the bridegroom's finger.7 Among the Ahoms of Assam, after bride and bridegroom for a few minutes have had their thumbs tied together with a blade of the cusha grass, they present each other with a ring.8 The wedding ring was in use among the ancient Hindus,9 1 Hollis, Nandi, p. 62. 2 Wakefield, ' Marriage Customs of the Southern Gallas,' in Folk-Lore, xviii. 321. 3 Minnie Martin, Basutoland, p. 83. 4 Decle, Three Years in Savage Africa, p. 232. 5 Stack,. Mikirs, p. 17. 6 Felkin, ' Notes on the For Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xiii. 228. 7 Gurdon, Khasis, p. 128. 8 Barua, Notes on the Marriage Systems of the Peoples of Assam, P' 47 9 Haas, loc. cit. p. 299. 444 THE HISTORY OF HUMAN MARRIAGE chap. and the betrothal ring in ancient Rome, where the man presented it to his fiancee.1 The same custom prevailed in Christian Europe throughout the Middle Ages and later, but was subsequently mostly succeeded by an exchange of rings.2 The ring, however, was only slowly introduced in the northern countries, replacing the old Teutonic customs of tying a knot or breaking a gold or silver coin, one-half of which was kept by the woman and the other half by the man ;3 and the exchange of rings is not known to have existed in Scandinavia until the end of the seventeenth century.4 Various superstitions connected with the marriage ring indicate that it is regarded as a symbolic tie between the couple. To lose it or break it means death or the dissolution of the union or some other misfortune.5 In the north-east of Scotland people say that if a woman loses her marriage ring " she will lose her man."6 The Jewish marriage ring, which is placed on the forefinger of the bride's right hand, no doubt, originated in mediaeval times ; it must be made of pure gold and be devoid of gems, its purity being said to be symbolic of conjugal fidelity.7 1 Pliny, Historia naturalis, xxxiii. 12. Marquardt, Das Privat- leben'der Romer, p. 41. Sohm, op. cit. p. 54 sq. 2 Bachtold, op. cit. p. 155 sqq. 3 Troels-Lund, op. cit. ix. 159 sq. Brand, op. cit. pp. 347, 349 sq, 4 Troels-Lund, op. cit. ix. 163. 5 Kuhn and Schwartz, Norddeutsche Sagen, Marchen und Gebrauche, p. 434. Strackerjan, Aberglaube und Sagen aus dem Herzogthum Oldenburg, i. 29. Bartsch, Sagen, Marchen und Gebrauche aus Mecklenburg, ii. 70. Raaf, Samlingar 'och Anteckningar till en beskrifning dfver Ydre hdrad i Ostergbthland, i. 116. S6billot, Coutumes populaires de la Haute-Bretagne, p. 122. Sauv6, Le Folk- Lore des Hautes-Vosges, p. 99. Dalyell, Darker Superstitions of Scotland, p. 288. Marie Trevelyan, Folk-Lore and Folk-Stories of Wales, p. 271. Bachtold, op. cit. p. 174 sq. In Narpes, a Swedish- speaking community in Finland, it is believed that if the wedding ring is too small there will be misfortune (Tegengren, ' Magi och vidskepelse, hanforande sig till trolovning, brollop o.s.v.,' in Hem- bygden, iii. 41). 6 Rorie, ' Stray Notes on the Folk-Lore of Aberdeenshire and North-East of Scotland,' in Folk-Lore, xxv. 349. 7 Rosenau, Jewish Ceremonial Institutions and Customs, pp. 160. 162. Abrahams, Jewish Life in the Middle Ages, p. 183 sq. XXIV MARRIAGE RITES 445 The union between the couple may also be strengthened by certain ceremonies in which an article of dress belonging to or presented by one of the partners is worn by the other. The following practices are found in different parts of Morocco. In Andjra the bridegroom sends his betrothed a new hayek-a long, rectangular, and seamless piece of white woollen cloth, worn without fastenings by men and women'- and while she is dressed in it a married woman, who is her husband's first and only wife and much beloved by him, plaits its fringe ; and this hayek, and also other clothes, which have been presented to the young man by his fiancee, are worn by him when he proceeds to his house to meet the bride.1 Among the Ath Ubahthi, after the marriage has been consummated, the belt of the bride is tied round the bride- groom's hayek over the crown of his head, where it is left as long as the bride remains ungirdled ;2 and in Fez the bride gives him two handkerchiefs, one of which he ties round his waist, and a cord, which she threads through his trousers.8 In the Hiaina, at a feast preliminary to the wedding, the bridegroom has on his feet a pair of slippers bought with money out of the bride's share of the dowry ;4 and on the fortieth day after her arrival the young wife takes off her husband's slippers, puts them over her own, and slips a new pair on his feet, which is supposed to make the couple friendly to one another and prevent divorce.5 Similar superstitions may also to some extent underlie the frequent practice of the fiance giving a present to his betrothed or the exchange of presents between them. In European folk- custom the present of a handkerchief plays a prominent part.6 The union between the parties is sometimes symbolised, or effected, by means of blood. In some parts of Brittany the bride makes an incision under the left breast immediately the ceremony in church is over, and the bridegroom then 1 Westermarck, Marriage Ceremonies in Morocco, pp. 127, 142, 143, 260, 261, 354. 2 Ibid. pp. 249, 260, 354. 3 Ibid. pp. 226, 227, 260, 354. 4 Ibid. pp. 102, 260, 354. 5 Ibid. pp. 305, 354. 6 See Bachtold, op. cit. i. 123 sqq. ; Brand, op. cit. p. 348 sq. 446 THE HISTORY OF HUMAN MARRIAGE chap. applies his lips and sucks a drop of her blood.1 This closely resembles certain marriage ceremonies practised in India. Among the Kewats of Bihar a tiny scratch is made on the little finger of the bridegroom's right hand and of the bride's left; blood is drawn from each and mingled with a dish of boiled rice and milk, and either party then eats the food containing the other's blood.2 It is reported that among the Dharhis and Dosadhs of Monghyr " the fingers are lanced by a barber, and the blood is soaked in red cotton wool, which is enclosed in pan. The bride chews the pan containing the bridegroom's blood, and the bridegroom that containing the bride's blood. A similar custom prevails among the Gulgulias ; only in their case the wool which absorbs the blood is used to dye the feet of the bride- groom and the bride."3 Among the Haris 4 and Birhors 5 of Bengal a bride and bridegroom are smeared with each other's blood, which has been extracted from their fingers. In Singbhum " they touch and mark each other with blood as a sign that they have become one flesh." 6 It has been suggested, or even taken for granted, that the custom, prevalent in several aboriginal tribes of India, of the bridegroom marking the bride, or the bride and bridegroom marking each other, with red lead7 is a later transformation of the custom of marking with blood ; 8 but although this 1 Conybeare, A Brittany Marriage Custom,' in Folk-Lore, xviii. 448. 2 Risley, Tribes and Castes of Bengal. Ethnographic Glossary, i. 456. 3 O'Malley, Census of India, 1911, v. (Bengal, Bihar and Orissa and Sikkim) Report, p. 320 sq. 4 Ibid. p. 320. 5 Dalton, Descriptive Ethnology of Bengal, p. 220. Risley, op. cit. i. 138. 0 Dalton, op. cit. p. 319. 7 Ibid. pp. 131, 160, 216, 220, 252, 273, 317, 319, 231. Risley, op. cit. passim. Shaw, ' On the Inhabitants of the Hills near Rajama- hall,' in Asiatick Researches, iv. 71. Dehon, ' Religion and Customs of the Uraons,' in Memoirs Asiatic Soc. Bengal, i. 163. Bradley- Birt, Chota Nagpore, p. 49. 8 Dalton, op. cit. pp. 220, 319. Hartland, Legend of Perseus, ii. 334 sqq. Crooke, ' Hill Tribes of the Central Indian Hills,' in four. Anthr. Inst, xxviii. 241 sq. O'Malley, op. cit. p. 320. Russell, Tribes and Castes of the Central Provinces of India, iv. 108. XXIV MARRIAGE RITES 447 may be the case, we can hardly be positive that it is so, considering how often, in India as well as elsewhere, th» red colour is used in marriage rites in circumstances which do not allow us to presume that the use of it is the survival of an earlier practice of using human blood.1 Not infrequently animal blood is employed in marriage rites. Among the Wadders, Dravidian earthworkers of South India, a fowl is sacrificed at the threshold of the bride's room, and the foreheads of the couple are marked with its blood ;2 and among the Kayans of Borneo a cock and a hen are slaughtered, their blood is received in a cup, and the bride and bridegroom are marked with it from hand to foot.3 Dr. Hartland's conclusion that in these cases the blood of a fowl takes the place of that of the parties4 does not seem to me fully justified. Among the Orang Mamaq of Sumatra, when bride and bridegroom are sitting on the ground with their feet touching each other, the chief 1 On the use of red in wedding rites see Zachariae, ' Zuin altin- dischen Hochzeitsritual,' in Vienna Oriental Journal, xvii. 230 sqq.; Russell, op. cit. iv. 107; Mary Billington, Woman in India, p. 74 ; Hartland, op. cit. ii. 335 sq. ; Samter, Geburt, Hochzeit und Tod, p. 186 sq. ; Idem, Familienfeste der Griechen und Romer, p. 47 sqq. ; Wachsmuth, Das alte Griechenland im neuen, p. 89 sq ; Sakellarios, Die Sitten und Gebrauche der Hochzeit bei den Neugriechen, p. 17 ; Piprek, op. cit. pp. 16, 42, 43, 73, 91, 100, 113, 124, 130-134, 148 (Slavs) ; Volkov, ' Rites et usages nuptiaux en Ukraine,' in L'Anthro- pologic, iii. 545 ; Franzisci, Cultur-Studien uber Volksleben, Sitten und Brattche in Kdrnten, pp. 9, 67 ; Meyer, Badisches Volksleben im neunzehnten Jahrhundert, p. 285 ; Aina Wadstrom, ' Frieri- och brollopsbruk fian Dagsmark i Lappfjard ' (Finland), in Hembygden, ii. 83 ; Mattans, ' Brollopsseder i Korsnas ' (Finland), ibid. vi. 137 , Doolittle, Social Life of the Chinese, i. 73, 78, 80, 86 sq. ; Gray, China, i. 200-202, 204 sq. ; Stewart Lockhart, ' Marriage Cere- monies of the Manchus,' in Folk-Lore, i. 487, 489 ; Westermarck, Marriage Ceremonies in Morocco, pp. 148, 284, 343; Tessmann, Die Pangwe, ii. 261 ; Johnston, George Grenfell and the Congo, ii. 679 ; Thumwald, Forschungen auf den Salomo-Inseln, iii. 13 (natives of Buin) ; Strehlow, Die Aranda- und Loritja-Stdmme in Zentral- Australien, vol. iv. pt. i. 102 (Loritja). See also infra, ii. 466 sq. 2 Fawcett, ' On Some of the Earliest existing Races of the Plains of South India,' in Folk-Lore, v. 2f. 3 ' Memoirs of Malays,' in Jour. Indian Archipelago, ii. 359. 4 Hartland, op. cit. ii. 343. 448 THE HISTORY OF HUMAN MARRIAGE chap. who officiates on the occasion kills a cock and then a hen over their legs, which become sprinkled with the blood.1 Among the Reindeer Chukchee the bride and bridegroom are anointed with the blood of a sacrificial reindeer, which has been killed for this purpose ; this, Dr. Bogoras says, is the most important part of their marriage rite and is " obviously intended to strengthen marriage-ties."2 So, too, among the Yukaghir a reindeer is killed, by the father of the bride, and while the latter is dressing for the journey to the bride- groom's place, her mother and the match-maker's wife smear her joints with the fresh blood. " Thus," they say, " is a child to be washed before it is sent away from the house to live with strangers."3 As I have shown elsewhere, blood is used at weddings as a prophylactic or means of purification.4 An extremely frequent and widespread marriage rite is the eating together of bride and bridegroom. It has been found among some American tribes. Among the Navaho the marriage ceremony merely consisted in eating maize pudding from the same platter ;5 and among the Pawnee the girl takes a dish containing food, which she places before the bridegroom, they both eat, and she is now his wife.6 In Morocco it is a very common custom that the pair partake of some food together before they have intercourse. Sometimes the bridegroom eats first and then puts some food into the bride's mouth,7 and sometimes, among the Berbers of Southern Morocco, they both push a little food into each other's mouths. In Aglu, for instance, the woman who attends the bride brings a dish of meat, a dish of seksu, and a third dish containing bread, honey, and salt butter, and from each of these dishes the bridegroom pushes a little food three times into the mouth of the bride ; 1 Hagen, op. cit. p. 165. 2 Bogoras, Chukchee, p. 595 sq. 3 Jochelson, Yukaghir, p. 94. 4 Westermarck, Marriage Ceremonies in Morocco, pp. 210, 219, 262, 263, 294, 295, 305, 327. 5 Waitz, Anthropologic der Naturvolker, iii. 105. 6 Grinnell, Story of the Indian, p. 46. 7 Westermarck, op. cit. p. 231. XXIV MARRIAGE RITES 449 she bashfully tries to prevent it, but then does the same to him.1 Among the Niam-Niam women eat with their hus- bands at the marriage ceremony but on no other occasion.2 Among the Sakalava of Madagascar " the bride and bride- groom receive their food served up on one dish, from which they take alternate mouthfuls to signify their union and affection."8 The rite of eating together has also been found among other natives of the same island,4 and in Mangaia of the Hervey Group,5 Fiji,6 Lepers' Island of the New Hebrides,7 and various tribes of New Guinea.8 In the Malay Archipelago,9 the Malay Peninsula,10 and South- 1 Ibid. p. 251. See ibid. p. 258 sq. and Index, s.v. * Meal or meals, bride and bridegroom having a common.' 2 Schweinfurth, Im Herzen von Afrika, ii. 28. 3 Walen, ' Sakalava,' in Antananarivo Annual, no. viii. 1884, p. 53. 4 Sibree, The Great African Island, p. 251 (Hovas). Grandidier, Ethnographic de Madagascar, ii. 182 (Merina). 6 Gill, ' Report on Mangaia (Hervey Islands),' in Report of the Second Meeting of the Australasian Association for the Advancement of Science, p. 330. 6 Williams and Calvert, Fiji and the Fijians, p. 146. 7 Codrington, Melanesians, p. 242. 8 Chalmers, ' Report on New Guinea. Toaripi and Koiari Tribes,' in Report of the Second Meeting of the Australasian Associa- tion for the Advancement of Science, pp. 314, 319. Guise, ' Tribes inhabiting the Mouth of the Wanigela River,' in four. Anthr. Inst. xxviii. 209. van Hasselt, ' Die Noeforezen,' in Zeitschr. f. Ethnol. viii. 181 ; Finsch, Neu-Guinea, p. 102 sq. (Nufors of Geelvink Bay). Krieger, Neu-Guinea, p. 394. v. Rosenberg, Der malayische Archipel, p. 455. 9 Wilken, ' Plechtigheden en gebruiken bij verlovingen en huwe- lijkenbij devolken van den Indischen Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Nederlandsch-Indie, ser. v. vol. iv. p. 393 sqq. Schmidt, Liebe und Ehe im alten und modernen Indien, p. 411 sqq. Crawley, Mystic Rose, p. 379 sqq. Raffles, History of fava, i. 356. van Hasselt, Volksbeschrijving van Midden-Sumatra, p. 280. 10 Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. 57, 77. Martin, Die ' Inlandstdmme der Malayischen Halbinsel, p. 868. Bourien, ' On the Wild Tribes of the Interior of the Malay Peninsula,' in Trans. Ethn. Soc. London, N.S. vol. iii. 82 (Mantras). Logan, ' Orang Binua of Johore,' in four. Indian Archipelago, i. 270. Low, ' Karean Tribes,' ibid. iv. 431 (Sakai, &c.). VOL. II G G 450 THE HISTORY OF HUMAN MARRIAGE chap. Eastern Asia and India in general,1 it is an extremely preva- lent custom that bride and bridegroom eat together from the same dish. In Southern Mindanao2 and among the Negritos of Zambales 3 they feed each other with rice, and the same is done at Malay weddings elsewhere.4 Among the Chukmas the bride has to place cooked rice and a prepared betelnut in the mouth of the bridegroom, and he has to do the same to her.5 Among the Gonds and Korkus the marriage rites consist in part of " eating together, tying the garments together, dancing together round a pole, being half-drowned together by a douche of water, and the interchange of rings,-all of which may be supposed to symbolise the union of the parties."6 Among Hindus of every rank and caste it is the custom for bride and bridegroom to take food to- gether from the same leaf or the same plate.7 In ancient India the newly-married couple had to offer a burnt oblation of a pancake or a mess of boiled rice on the wedding night, and they partook together of that dish and of some liquid food besides.8 In ancient Greece they partook together of a sesamum-cake.9 The cake made of the old Italian grain called far, from which the patrician marriage in ancient Rome received its name of confarreatio, was offered to Jupiter Farreus and partaken of by bride and bridegroom 1 Fielding Hall, op. cit. p. 188 (Burmese). Woodthorpe, in Jour. Anthr. Inst. xxvi. 22 (Shans). Barua, op. cit. p. 54 (Miris). Hutchinson, op. cit. p. 119 sq. (Magh tribes of Chittagong). Dalton, op. cit. p. 216 (Santals). Hayavadana Rao, ' Irulans of the Gingee Hills,' in Anthropos, vi. 811. Crooke, ' Hill Tribes of the Central Indian Hills,' in Jour. Anthr. Inst, xxviii. 241. Idem, Things Indian, p. 319. 2 Cole, ' Wild Tribes of Davao District, Mindanao,' in Field Museum of Natural History, Anthropological Series, xii. 102, 144, I92- 3 Reed, op. cit. p. 58. 4 Skeat, Malay Magic, p. 383. 5 Hutchinson, op. cit. p. 97. 6 Forsyth, Highlands of Central India, p. 149. 7 Padfield, op. cit. p. in sq. Dubois, Description of the Character, Manners, and Customs of the People of India, p. 107. Pandit Harikishan Kaul, Census of India, 1911, vol. xiv. (Punjab) Report, p. 276. 8 Grihya-Sutras, ii. 49. Wintemitz, Das altindische Hochzeits- rituell, p. 79 sq. Oldenberg, Die Religion des Veda, p. 330. 9 Rossbach, op. cit. p. 107. XXIV MARRIAGE RITES 451 in the presence of witnesses.1 At the present day the custom of eating together-usually from the same plate or dish, or of the same loaf of bread, or with the same spoon-at the betrothal or, more often, at the wedding, is found in many parts of Europe.2 Yet in spite of its occur- rence among so many peoples of the Indo-European race we cannot be certain that it was a primitive Indo- European custom. As Wintemitz points out, it is possible that it originated among different branches of that race independently. " It would be different," he says, "if we found exactly the same kind of dish-say, the wedding-cake -used among all the different Indo-European peoples, or if the ceremony had its fixed place in the marriage ritual, like the joining of hands, and some other customs."3 As for the meaning of the rite in question, there can be no doubt that it was originally something more than a mere symbol. In Warmland, in Sweden, it was a popular belief that if a girl and a youth ate off one morsel, they would fall in love with each other.4 In Germany it is believed that if the couple eat the " morning soup " with the same spoon, they will have a peaceful married life.5 In an Arabic-speaking 1 Gaius, Institutionum juris civilis commentarii, i. 112. Rossbach, op. cit. p. 106 sqq. Marquardt, op. cit. p. 50. De-Marchi, La religions nella vita domestica, p. 155 sqq. Fowler, Social Life at Rome in the Age of Cicero, p. 136. 2 Placucci, Usi e pregiudizj dei contadini della Romagna, pp. 52, 58. Bustico, ' Il matrimonio nel Bellunese,' in Provenzal, Usanze e feste del popolo italiano, p. 25. Bresciani, Dei costumi dell' isola di Sardegna, ii. 153. Laisnel de la Salle, Croyances et legendes du centre de la France, ii. 46. de Nore, Coutumes, mythes et traditions des provinces de France, p. 9. Bachtold, op. cit. i. 105 (Switzerland). Wuttke, Der deutsche Volksaberglaube der Gegenwart, §560, p. 369 ; Sartori, Sitte und Brauch, i. 73 sq. (Germany). Raaf, op. cit. i. 112 ; Nicolovius (Lov6n), op. cit. p. 133 (Sweden). Thiers, Traitd des superstitions qui regardent les sacremens, iv. 473 (Russia). Franzisci, op. cit. p. 69. (Slovenes of Carinthia). Ida von Diiringsfeld and von Reinsberg-Diiringsfeld, Hochzeitsbuch, p. 81 (Serbia), v. Schroeder, Die Hochzeitsgebrduche der Esten, p. 82. Hamalainen, op. cit. pp. 94, 137, 176, 272 sq. (various Ugro-Finnic peoples in Europe). 3 Wintemitz, in Transactions, p. 281. 4 Femow, Beskrifning bfver Warmeland, p. 254. 5 Wuttke, op. cit. §560, p. 369. Sartori, op. cit. i. 73 sq. G G 2 452 THE HISTORY OF HUMAN MARRIAGE chap. mountain tribe in Morocco, the Tsui, it is the custom for bride and bridegroom to eat together the Ever of the sheep which was slaughtered for the occasion when the bridegroom was for the first time painted with henna, for the purpose, I was told, " of making them dear to one another?'1 But the rite of eating together may also lay a mutual constraint on the couple. All over Morocco a common meal is a frequent form of 'ahd, or covenant, which ultimately derives its binding force from the idea that the parties transfer con- ditional curses to one another by a material medium-in this case the food-as appears from the usual saying that the food will repay him who breaks the contract.2 To the same class of ceremonies belong in part the bride's and bridegroom's partaking of food in common: it is a means of sealing their union by an act which has naturally been suggested by one of the most prominent features of married life, the husband's sharing of food with his wife. At the same time the common meal may serve a prophylactic purpose by bringing bride and bridegroom nearer each other previously to the great event by which the union is completed.8 And in certain cases some particular effect is obviously expected from the kind of food partaken of; as when in modem Greece the young man's mother makes the couple eat honey from the same vessel that their marriage shall be sweet,4 and the Livonians obliged them to eat the testicles of a goat or a boar with the avowed intention of rendering the pair prolific.6 Besides, and sometimes combined with, the rite of eating together there is the rite of drinking together. Among the Maynas of the Upper Maranon bride and bridegroom sealed their union by drinking from the same bowl.6 1 Westermarck, Marriage Ceremonies in Morocco, p. ioi. 2 Idem, ' L-'dr, qt the Transference of Conditional Curses in Morocco,' in Anthropological Essays presented to E.B. Tylor, p. 373 sq. 3- See Idem, Marriage Ceremonies in Morocco, p. 259. 4 Wachsmuth, op. cit. p. 95. 6 Meletius, De religions et sacrificiis veterum Borussorum epistola, p. D 3. Cf. Frazer, Totemism and Exogamy, ii. 262 sq. n. 4. 0 Chantre y Herrera, Historia de las Misiones de la Campania de Jesds en el Mar art tin espanol, p. 71. XXIV MARRIAGE RITES 453 Among some Brazilian Indians they drink brandy together.1 In Humphrey's Island the priest gave the man a cocoa- nut, and he, after sipping the milk, gave it to the woman, who did the same.2 Among the Mandaya of Mindanao the marriage ceremony consists not merely of the young couple feeding each other with rice, but of their drinking from a common cup as well.8 Among the Hos4 and Lepchas5 the bride and bridegroom drink beer together or out of the same cup. Among the Jansens, one of the Kuki tribes, " a glass of liquor is handed to the husband, who prinks half, and gives the rest to his wife."6 Among the Tipperahs of the Chittagong Hills, again, " the girl's mother pours«out a glass of liquor and gives it to her daughter, who goes and sits on her lover's knee, drinks half and gives him the other half; they afterwards crook together their little fingers."7 In Annam the bride and bridegroom exchange cups and then put them back one on the other.8 In China it was the ancient custom for bride and bridegroom to eat together of the same sacrificed animal, and to drink out of cups made of the two halves of the same melon, the bride drinking from the bridegroom's half and he from hers; they thereby showed, according to the ' Li Ki,' " that they now formed one body, were of equal rank and pledged to mutual affection."9 At present, in China, the couple, before the consummation of the marriage, drink wine or a mixture of wine and honey alternately from two goblets, which are sometimes tied together by a red string, the bridegroom, after having sipped from his goblet, hand- ing it over to the bride, and the bride handing hers over to her husband, and so repeatedly.10 So, too, in Japan they 1 v. Eschwege, Journal von Brasilien, i. 96. 2 Turner, Samoa, p. 276. 3 Cole, loc. cit. p. 192. Cf. ibid. p. 102. 4 Dalton, op. cit. p. 193. 5 Risley, op. cit. ii. 8. 8 Soppitt, Short Account of the Kuki-Lushai Tribes, p. 17. 7 Lewin, Races of South-Eastern India, p. 202. 8 Dumoutier, quoted by Hartland, op. cit. ii. 350. 9 Lt Kt, ix. 3. 11; xli. 2. Legge, in Sacred Books of the East, xxviii. 429 n. 3. 10 Doolittle, op. cit. i. 86 sqq. Arendt, in Folk-Lore, i. 491. Leong and Tao, op. cit. p. 107. 454 THE HISTORY OF HUMAN MARRIAGE chap. drink wine together, exchanging cups nine times; this constitutes the entire ceremony, after which the couple are introduced to relatives and friends at a wedding dinner.1 The rite of drinking together out of the same vessel is found in Europe from Italy to Norway, from Brittany to Russia, and there are traces of it in Scotland too.2 It was even adopted into the rites of the Greek Church in Russia, where the priest blesses a small silver ladle, called the Common Cup, which contains wine mingled with water, and then holds it to the lips of the pair, who sip it alternately each three times, "as a type that the husband and wife must share everything in joy or grief."3 The custom of drinking together forms part of the nuptial ceremony among the Jews of all countries.4 That this rite, like that of eating together, is in the first place a symbol of, or a means of strengthening, the union of the couple, is beyond a doubt; but, like the latter rite, it may at the same time serve some other purpose. In Humphrey's Island the man's sister took a second cocoa-nut, opened a hole in it, and threw it up to the supposed god above them, " and prayed as the priest had just done before that the woman might be prolific."5 In India, at least, a cocoa-nut is a " symbol of fertility," and all through Upper India it is kept in shrines and presented by the priest to women who desire children.6 1 Nakajima, ' Marriage (Japanese and Korean),' in Hastings, op. cit. viii. 460. Kiichler, ' Marriage in Japan,' in Trans. Asiatic Soc. Japan, xiii. 115. 2 v. Schroeder, op. cit. p. 83 sq. Wintemitz, in Transactions, p. 280 sq. Hartland, op. cit. ii. 347 sqq. Bachtold, op. cit. i. 96 sqq. Sartori, op. cit. i. 74. Raaf, op. cit. i. 112 ; Hammarstedt, * Kvar- levor av en Fros-ritual i en svensk brollopslek,' in Festskrift til H. F. Feilberg, p. 504 (Scandinavia)'. Bresciani, Dei costumi dell' isola di Sardegna, ii. 153. Hamalainen, op. cit. pp. 94, 137, 176, 272 sq. (Ugro-Finnic peoples). 3 Romanoff, Rites of the Greek Church, p. 204. Goar, Euchologion sive Rituale Gracorum, pp. 394, 396. Thiers, op. cit. iv. 473. 4 Wintemitz, in Transactions, p. 281. Abrahams, op. cit. p. 208. Rosenau, op. cit. p. 160 sqq. See infra, ii. 461 sq. 6 Turner, op. cit. p. 276. 6 Crooke, Popular Religion and Folk-Lore of Northern India, i. 227 ; ii. 106. XXIV MARRIAGE RITES 455 At Bogadjim, in New Guinea, the bridegroom splits a betel-nut into two pieces, one of which is then chewed by himself and the other by the bride.1 The mountaineers of the Gazelle Peninsula in New Britain regard a man and a woman as husband and wife only after they have exchanged betel-nuts ; this is the case even when the woman has been carried' off by force.2 The exchange or partaking together of betel is also a marriage rite in various islands of the Malay Archipelago.3 Among the Ahoms of Assam, after the couple have presented a pinch of food to each other's lips and have washed their hands, the bridegroom offers a betel-nut to the bride, who accepts it and reciprocates the offer; they are then tied together by their thumbs with a blade of the cusha grass.4 Among the Rabhas, living in the same country, the essential features of the marriage ceremony are, first, the exchange of betel-leaves and areca- nuts, and, secondly, the formal sacrifice of a cock and hen, the latter being made into a curry of which the bride and bridegroom partake together.5 Among the Irulans of the Gingee Hills the would-be husband must smoke a tobacco cheroot and then hand it over to the bride, who should smoke it a little and pass it back to him ; then comes the dinner, which the parties must take out of a common dish, and these things done, the marriage is completed and consummation takes place immediately.6 Among the Ahom Chutiyas bride and bridegroom are made formally to inhale the smell of turmeric together, and this is followed by the tying of the nuptial knot.7 To the same class of rites belongs the following one, which is practised by the Winamwanga and Wiwa in East Africa :-The headman who officiates at a wedding " takes two short bits of stick, and gives one to the bridegroom and one to the bride. He puts a stick into each 1 Hagen, Unter den Papua's, p. 243. See also Guise, in Jour. Anthr. Inst, xxviii. 209 (tribes inhabiting the mouth of the Wanigela River in New Guinea). 2 Burger, Die Kiisten- und Bergvolker der Gazellehalbinsel, p. 55. 3 Crawley, op. cit. p. 384. 4 Barua, op. cit. p. 47. 6 Endle, Kacharis, p. 85. 6 Hayavadana Rao, in Anthropos, vi. 811. 7 Endle, op. cit. p. 95. 456 THE HISTORY OF HUMAN MARRIAGE chap. of their mouths. They then exchange these sticks, the bridegroom putting the bride's in his mouth, and the bride the bridegroom's in hers."1 There are also ceremonies of a different nature which are supposed to ensure the durability of the marriage. In Boeotia, on reaching the bridegroom's house, the axle of the car on which he fetched the bride was burnt to symbolise, or to effect, the irreversibility of the step taken.2 Among the Manchus, when the bride comes out of the chair in which she has been carried to the bridegroom's house, she has to step over a miniature saddle, as a sign that she will never marry a second husband, in accordance with the saying :-" Just as a good horse will not carry two saddles, a chaste maiden will not marry two husbands."3 The Merina of Madagascar believe that if the bride's parents do not accompany her to the bridegroom's house as evidence of their complete consent to her marriage, the couple will not long remain together.4 In some parts of Morocco the person who paints the bride with henna must be a married woman who has been married only once, as it is believed that if she had been divorced by a former husband the same thing would also happen to the bride.5 In the same country the bride is frequently pelted with stones when she leaves her parents' old home, and one explanation given of this custom was that it makes the husband love his wife and prevented him sending her back to her parents; but it was also said to rid her of evil, or to cause her to take her evil with her or to remove all the evil from the village.6 Among the Ait Warain, a Berber tribe in the neighbourhood of Fez, when the bride has arrived at the bridegroom's place, his mother takes the bride's kerchief and ties it very firmly on the head of a donkey, and it is believed that if 1 Chisholm, ' Notes on the Manners and Customs of the Winam- wanga and Wiwa,' in Jour. African Soc. ix. 382. 2 Plutarch, Quaestiones Romanae, 29. 3 Stewart Lockhart, in Folk-Lore, i. 487 sq. 4 Grandidier, op. cit. ii. 183. 5 Westermarck, Marriage Ceremonies in Morocco, pp. 141, 149, 150, 161, 353. 6 Ibid. pp. 169-171, 173, 176, 177, 179, 180, 190, 323, 326, 353, 357. XXIV MARRIAGE RITES 457 it remains there till the following morning, it is a sign that the wife, too, will remain with her husband j1 but it is more than probable that the rite was originally meant to be, not a mere act of divination, but a means of strengthening the marriage tie. In a neighbouring Berber tribe, when the bride after her arrival is sitting on a mat in the bridegroom's tent, a bachelor lifts up one of the vertical tent-poles and puts it into her lap, in order that she shall remain in her new home and support it by becoming a mother of sons, as the pole supports the tent.2 There are various rites that are intended to ensure or facilitate the consummation of the marriage. Among the Aith Yusi, a Berber tribe south of Fez, after the bride has been painted with henna, an egg enveloped in a kerchief is tied round her forehead ; it is then broken by the woman who painted her, and is left there till she is washed. This is done in order that her hymen shall be broken by her husband as easily as was the egg. Next morning she is washed with water containing some henna, while seated on a weaving-stool and a pack-saddle-on the former because when a little girl she was, in accordance with the customs of her tribe, taken three times underneath the two upper cross-bars of a weaving-stool after the web was ready, in order that no man should be able to destroy her virginity, and the magical effect of this procedure must naturally be cancelled before her wedding.3 In the same tribe, when the bride is taken to her new home on the back of a mare, she holds in front of her a cane, with or without a flag, which is fired at by the men of the procession ; they want to blow it to pieces so that the bridegroom shall be able to break the hymen of his bride that night, but she makes some efforts to prevent it by waving the cane to and fro.4 In another Berber tribe, in the Rif, the bridegroom's mother places a mug upside down with a so-called didli (an ornament consisting of dollar or half-dollar pieces threaded on a string of horsehair and worn by women round 1 Ibid. p. 208 sq. 2 Ibid. pp. 202, 349 sq. 3 Ibid. p. 151 sq. 4 Ibid. p. 179. For a similar ceremony in a neighbouring tribe see ibid. p. 177. 458 THE HISTORY OF HUMAN MARRIAGE chap. the forehead) and an egg on the top of it, and the bridegroom then breaks both the mug and the egg with a kick, as I was told, " so as to destroy the evil."1 It is probable that " the evil " in this case meant any possible impediment to the con- summation of the marriage. There is much fear in this part of Northern Africa of magic obstacles to sexual intercourse.2 The ceremonial breaking of eggs at weddings is found in other countries than Morocco. In a book on the ' Customs and Manners of the Persians ' we read :-" Dadeh Bazm Ara says, the bride should take a hen's egg in her hand, and on getting up throw it against the wall to break it, keeping her face towards the Kibleh, or in the direction of Mecca. Kulsum Naneh thinks that a useless proceeding, and recom- mends a needle to be presented to her on her marriage."3 Among the Tenggerese in East Java the bridegroom on the last day of the wedding breaks an egg which has been placed on a stone, after which the bride smears her feet with its contents.4 Among the Sundanese in West Java a hen's egg is placed before the door of the newly wedded pair; which appears to imply a similar rite of breaking it.5 In Bali an egg and a cocoa-nut are offered to the bride and bridegroom, who throw them on the ground so that they break, and then disperse the pieces in different directions as offerings to the kalas, or spirits.6 In France, in the seventeenth century, a bride, in order to be happy in her marriage, trod upon and broke an egg when she entered her new home on the wedding day.7 At Avola, in Sicily, on the bride's arrival at her new home, the bridegroom breaks two eggs with his foot.8 The use of eggs in marriage rites may, as we shall see, serve different purposes ; but when the breaking of an egg plays a prominent part in the rite and the 1 Ibid. p. 131. 2 See DouttA Magie st religion dans l'Afrique du nord, p. 288 sqq.; Mouli6ras, Le Maroc inconnu, pp. 53, 499 sqq. 3 Customs and Manners of the Women of Persia, trans, by James Atkinson, p. 71. 4 Domis, De Residentie Passoeroeang op het eiland Java, p. 155. 5 Hartland, Primitive Paternity, i. 58. 6 Tonkes, Volkskunde von Bali, p. 30. 7 Thiers, op. cit. iv. 471. 8 Pitrd, Usi e costumi, credenze e pregiudizi del popolo siciliano, ii. 75. XXIV MARRIAGE RITES 459 marriage is consummated shortly after, there is some reason to suspect that the original intention of the ceremony was to ensure the defloration of the bride, though the idea of pro- moting her fertility or other ideas1 may very well have been combined with it. Earthenware vessels or objects of glass are often cere- monially broken at weddings both in Morocco and else- where.2 Thus in Andjra, after the bridegroom has been painted with henna, his best-man takes the bowl containing the rest of the henna-mixture, lifts it on his head, and begins to dance before the bridegroom. After a while he hands the bowl to another bachelor, who does the same ; and thus all the bachelors present dance in turn with the bowl on their heads till the last one lets it drop down on the ground and break, which is supposed to remove the bas, or evil.3 In another tribe the girl who painted the bridegroom with henna puts the bowl on her head and dances with it, till at last she throws it on the ground so that it breaks, and thereby, it is thought, rids the bridegroom of his bas.11 Among the Bogos of North-Eastern Africa the bridegroom, before he has intercourse with the bride, breaks an earthen- ware pot.5 In Armenia a plate is offered to the bridegroom, who throws it on the ground and tramples it to pieces.6 At Bajar, when the marriage contract had been made, it was the custom for the guests to throw the bottles of rose-water which they had brought with them against the wall.7 The breaking of an earthenware vessel is a marriage ceremony among the Gypsies in Turkey, Moldavia, Tran- sylvania, Spain, and Germany.8 The Basque Gypsies con- 1 Cf. Westermarck, op. cit. pp. 194, 195, 218. 2 For European weddings cf. v. Schroeder, op. cit. p. 84 sqq. 3 Westermarck, op. cit. p. 98 sq. 4 Ibid. p. 101. 5 Munzinger, Ueber die Sitten und das Recht dev Bogos, p. 63. 8 Bodenstedt, quoted by Lobel, Hochzeitsbrauche in dev Turkei, p. 119. 7 Fraenkel, ' Aus orientalischen Quellen,' in Mitteil. d. Schlesischen Gesellsch. f. Volkskunde, xix. 28. 8 Thompson, ' Ceremonial Customs of the British Gipsies,' in Folk-Lore, xxiv. 338. 460 THE HISTORY OF HUMAN MARRIAGE chap. tracted their marriages by throwing a jar towards the sun and counting the fragments.1 In Sardinia it is the custom, while a marriage procession is passing, to cast on it grain from windows and doors and then to throw on the pavement the vessels in which the grain was kept so that they break, " in order to disperse every evil augury.''2 In the Sette Comuni "it is always necessary to break something at the nuptial banquet.''8 In Upper Brittany, " pour que le mariage soit chanceux, il faut qu'il y ait pendant la noce quelque objet cassA S'il n'y a rien eu de casse par accident, on en casse un exprds."4 In the Ardennes a glass is broken " so that the couple shall have no quarrel."5 In Argyllshire, " if a glass is accidentally broken during a marriage feast, it foretells misfortune to the bridal pair, but, when the health of bride and bridegroom is drunk, someone must throw a glass over their shoulder and break it ' for luck.' "6 At Newburgh, after the marriage ceremony and just as the newly-married couple are leaving the bride's house, a plate containing salt is at some marriages stealthily broken over the head of the bridegroom.7 At Guisborough in Cleveland the bridegroom took the plate which, with a small cake upon it, had been presented to the bride on the arrival at the door of her home, and threw it over his left shoulder, " their hope of future happiness depending upon it being broken on falling to the ground."8 In the Saalfeld country, after the wedding, one of the bridesmaids hurries home first, gets beer or brandy, and offers a glass to the bridegroom, who empties it. and tosses it 1 Plants, Une race maudite, reviewed in Journal of the Gypsy Lore Society, N.S. iv. 297. 8 Faggiani, * Feste ed usanze della Sardegna,* in Provenzal, op. cit. p. 232'. 3 Frescura, ' Fra i Cimbri dei Sette Comuni Vicentini,* per lo studio delle tradizioni popolari, xvii. 46. 4 Sdbillot, Coutumes populaires de la Haute-Bretagne, p. 136. 5 Ibid. p. 136. 6 Minnie Cartwright, ' Scraps of Scottish Folklore,' in Folk-Lore, xxi. 89. 7 Simpkins, County Folk-Lore, vol. vii. Fife, p. 163. 8 Mrs. Gutch, County Folk-Lore, vol. ii. North Riding of Yorkshire, p. 295. XXIV MARRIAGE RITES 461 behind his back. If the glass breaks, it is good; if not, not.1 At the German Poller abend-which is celebrated/ on the eve of the wedding day-all sorts of old pottery are broken outside the bride's house, and the North Germans say, "The more shards the more luck.''2 In the Upper Palatinate, on the same night, a window is broken in her house, and many shards are said to indicate wealth.3 Among the Slovaks, while the bride and bridegroom are in bed, the guests sing and make noise outside their room, and an old pot is thrown at the door.4 Among the Little Russians the bride's mother offers to the bridegroom a bowl of water; he sips from it a few drops and then throws it over his shoulder, and if it breaks the marriage will be blessed with children.6 Or she offers him a cup containing oats and water, and when he throws it over his shoulder his best-man, who is standing behind him, tries to break it; in case he succeeds, the marriage will be happy.6 Among some Slovenes, when the bridegroom and his friends have come to fetch the bride, she at last appears, holding in the hand a beaker of wine covered with a red kerchief. She offers him the kerchief, and is then taken three times round the bridegroom. She drinks a little of the wine and gives him the beaker, which he empties at once and throws at the wall; and it is considered a great shame for him if it does not break.7 In his description of Jewish marriages in the seventeenth century, Addison states that in many places " the young Men who wait upon the Bridegroom, at the hearing of the Husband giving the Missal Tob, or wishing happiness to his Wife, break certain small Earthem Pots which for that purpose they hold in their hands. And thereby signifie their good wishes of prosperity and health to the New- married Couple."8 In a narrative of a fifteenth century 1 Grimtn, Teutonic Mythology, iv. 1797. 8 Sartori, op. cit. i. 72. Samter, Geburt, Hochzeit und Tod, p. 60. Kuhn and Schwartz, Norddeutsche Sagen, M&rchen und Gebrauche, P- 434- 8 Schdnwerth, Aus der Oberpfah, i. 74. 4 Piprek, op. cit. p. 107. 8 Ibid. p. 38. 8 Ibid. p. 37. 7 Ibid. p. 113. 8 Addison, Present State of the Jews, p. 51. 462 THE HISTORY OF HUMAN MARRIAGE chap. Jewish wedding we are told that the Rabbi, at the end of the marriage ceremony, passed wine to the bridegroom and then to the bride. " He retained the glass in his hand while they sipped its contents, but he now gave it to the bridegroom, who turned round, faced the north, and threw the glass at the wall, breaking it. Thereupon the assembled company rushed at the bridegroom, uttering expressions of joy, and conveyed him-before the bride-to the wedding- house."1 Up to the present day the breaking of a glass has remained one of the most characteristic features of Jewish weddings; the bridegroom breaks it with his foot, or it is broken by the Rabbi.2 Various fanciful explanations have been suggested for this ceremony, but its true meaning, as I understand it, has to my knowledge never been recog- nised. The Jews of Morocco have another custom which, partly, belongs to the same class of rites : a fortnight before the nuptial ceremony, on a Thursday, a pot filled with corn is broken at the door of the bridal chamber " with the symbolical meaning that the bride shall be fruitful."3 Many of the facts just quoted strongly suggest that both this and other rites consisting in the breaking of some fragile object was intended to ensure the consummation of the marriage, which in Europe,4 as elsewhere, has been supposed to be impeded by malign magic influences. That this intention has generally been more or less disguised is not to be wondered at considering the nature of the subject. But among some of the Southern Slavs it is stated with a frankness which leaves no room for doubt. Among the 1 Abrahams, op. cit. p. 208. 2 Ibid. pp. 187, 188, 208 n. 2. Andree, Zur Volkskunde der Juden, p. 144 sq. Rosenau, Jewish Ceremonial Institutions and Customs, p. 162 sq. Jansen, ' Mitteilungen liber die Juden in Marroko,' in Globus, Ixxi. 360. 3 Jansen, in Globus, Ixxi. 359. 4 Sandys, Relation of a Journey, p. 7 (Zante). Piprek, op. cit. p. 42 ; Volkov, ' Rites et usages nuptiaux en Ukraine,' in L'Anthro- pologic, ii. 579 sqq. (Little Russians and other peoples). Liebrecht, Zur Volkskunde, p. 322; Grimm, Deutsche Mythologie, p. 1127 (Germany). Dalyell, op. cit. pp. 302, 306 sq. (Scotland). Frazer, Taboo, p. 299 sq. In the Atharva-Veda (vi. 138, p. 108 sq.) there is a charm for depriving a man of his virility. XXIV MARRIAGE RITES 463 Serbs of Syrmia the head of the family, about midnight* conducts the bride into the bridegroom's room and closes the door. " He now takes a glass filled with wine, drinks its contents to the health of the young couple, and throws it then at the door so that it is shattered to pieces; this is regarded as a symbol of the soon happening loss of virginity." While the bride and bridgroom are together, the guests are making an uproar, breaking glasses and pots, and they are also trying to break an egg which has been put into a sack, "as a sign of the marriage having been consummated."1 Among various peoples a staff or rod or tree is broken, and it is reasonable to suppose that the object of this cere- mony is the same as the Berber custom of blowing to pieces a cane. The last part of the marriage ceremony of the Zulus, called ukuhlambisa, is described as follows :- " The bride comes in with her party of girls, carrying in her hand a spear, which, by the way, she has carried all the time. One girl bears a dish of water and a calabash, and another some beads. Then coming up, singing and dancing, the bride throws the water over her husband. She also sprinkles her brother and sister in law, striking the latter as a symbol that from that time she assumes authority over the girls in her husband's household. After this is done she breaks the staff of the spear, and makes a run for the gate of the kraal, as a last effort to get away. If she is not stopped by a young man appointed for the purpose, it is a great disgrace, and the husband has to pay a cow to get her back. The marriage rites are then finished. No widow, remarried, breaks the staff of the spear."2 In a somewhat similar ceremony among the Matabele the bride crushes the calabash brought by her, " and the marriage is sealed."3 Among the Yakut,4 when the bride arrives at the bridegroom's hut, the- entrance is barred by two thin dry rods held by two girls ; the bride breaks them against her breast and, after picking up the pieces, lights them on the hearth, " to 1 Rajacsich, Das Leben, die Sitten und Gebrauche, der im Kaiser- thume Oesterreich lebenden Sudslaven, p. 161. 2 Tyler, Forty Years among the Zulus, p. 203 sq. 3 Decle, op. cit. p. 158 sq. See infra, ii. 509. 4 Shklovsky, In Far North-East Siberia, p. 52 sq. 464 THE HISTORY OF HUMAN MARRIAGE chap. show -it is said-that the Spirit of Fire now has another priestess." In the island of Skarpanto (Carpathos), halfway between Rhodes and Crete, when the bridegroom goes to the bride's house on the wedding morning, her mother places on the threshold a rug or blanket folded, with a stick resting on one of the corners. The bridegroom ad- vances his right foot, breaks the stick, and passes in; this is said to symbolise the future submission of the wife.1 In certain parts of Anjou, " lorsque le cortege revient de l'6glise, on trouve un arbre plants la veille au centre d'une roue recouverte de terre. On invite tous les j eunes gens, mari6 et gargons d'honneur en tete, & venir essayer de l'arracher pour montrer leur vigueur ; apr^s de vains efforts, on le brise en le tprdant et on danse autour du trongon."2 In Esthonia, at the betrothal feast, the bridegroom and the bride eat soup with the same spoon. The bride then throws the spoon on the ground, and the bridegroom treads on it; and if it does not break, it is considered a bad omen signifying that the marriage will not last long. Subsequently, before the bride leaves for her new home, a small plate con- taining butter is offered to her ; after she has eaten the butter the plate is broken and the wish is expressed that she shall suffer no greater damage than this.3 It is possible that the breaking of a loaf of bread or a cake over the heads of the couple, or over the head of the bride, which is found both among the Serbs4 and in England, Scotland, and Ireland,5, also belongs to the class of rites which we are now discussing. There are other rites which are apparently intended to help the bridegroom to consummate the marriage. In some parts of Morocco, after he has entered the room where the bride is waiting, he cuts with his sword the rope which has been tied from wall to wall in front of the bed, thus cutting off the bas* He takes off the bride's right slipper, 1 Rodd, Customs and Lore of Modern Greece, p. 102. In Cyprus the bridegroom, before entering, cuts off the head of a fowl which is held down by head and feet upon the threshold {ibid. p. 101). 2 Sdbillot, Le Folk-Lore de France, iii. 401. 3 Wiedemann, qu'oted by v. Schroeder, op. cit. p. 84 sq. 4 Piprek, op. cit. p. 136. 5- See infra, ii. 476 sq. 6 Cf. infra, ii. 466 sq. (Bashkir). XXIV MARRIAGE RITES 465 removes the needle which her mother has put into it as a protection against evil spirits, and throws it away, at the same time throwing away the bas, and then puts the slipper back on her foot.1 Or he removes the cover from her head, crushes its cane-work and throws it on the floor, thus destroying the bas.2 In Andjra the bridegroom, before he can have intercourse with the bride, must untie seven knots which have been made in the cord of her drawers by the women attending her ;3 and in ancient Rome the girdle worn by the bride was tied up in a so-called " Herculean knot " (nodus Herculaneus)-particularly difficult to loose- which the bridegroom untied in bed.4 In these cases the " tying up " of the bride may be, or have been, a sham attempt to protect the bride by laying obstacles in the bridegroom's way, or a means of protecting the bride from being by magic deprived of her virginity before the conjugal intercourse, or a means of ensuring the consummation of the marriage by compelling the bridegroom to untie the knot; but in any case we may presume that the untying of it was regarded as a necessary preliminary to coition. Knots and tangles are frequently looked upon as magical impediments and therefore avoided at weddings as well as child-births.5 In many tribes in Morocco the bride comes to the bride- groom's place with her hair dishevelled and without a belt,6 and in some places the bridegroom also has no belt round his waist.7 Among Slavonic peoples, too, the bride has her hair dishevelled,8 and in some parts of Poland it is the bridegroom who opens her plaits.9 In Ukrainia, when the marriage is going to be con- summated, songs are sung in which God is invoked to give his assistance. Other songs, to which perhaps a magical 1 Westermarck, op. cit. p. 237. 2 Ibid. p. 235. 3 Ibid. p. 232. 4 Festus, op. cit. p. 63. Cf. Marquardt, op. cit. p. 45 n. 2; Smith, Wayte, and Marindin, op. cit. ii. 142. 6 Volkov, in L'Anthropologic, ii. 579 sqq. Frazer, Taboo, p. 294577. Samter, Gcburt, Hochzeit und Tod, p. 121 sqq. 6 Westermarck, op. cit. pp. 261, 263 sq.; see Index, s.v. 'Hair,' and ' Belt.' 7 Ibid. p. 264; see Index, s.v. 'Belt.' 8 Piprek, op. cit. pp. 61, 75, 77, 124, 128. 9 Ibid. p. 75. VOL. II H H 466 THE HISTORY OF HUMAN MARRIAGE chap. efficacy is ascribed, contain sentences like these :-" Et vous, boiards, allez dans les buissons d'osiers et abattez des j eunes tiges pour un joug, nous irons def richer le sol, et labourer la terre vierge. . . . Celui qui l'aura cultivee en deviendra le proprietaire." Or, " La charrue ne va pas, le fer n'enforce pas-le sol est dur-et cette terre vierge n'est pas A moi: elle appartient a la mdre ; pretez-moi un fouet pour faire marcher le taureau - faites-le pousser . . . nous labourerons la terre vierge."1 It is quite possible that the red colour, which is so fre- quently used in marriage rites, is not only a sign of virginity, as it is sometimes expressly said to be,2 but is regarded as a means of ensuring defloration. The Chinese of Canton suspend from the top of the nuptial bed three long strips of red paper containing good wishes, such as, " A hundred sons and a thousand grandsons be your portion 1 "3 In Greece and Rome the nuptial bed was covered with red cloths ;4 and at Ekenas, in Finland, the bridal blanket must be red " in order that the bride shall be happy."5 In Rome the bride wore a red veil (flammeum),6 and this is still the case in Albania,7 Bulgaria,8 and other parts of South-Eastern Europe.9 In some parts of Russia the bridegroom's companions, the so-called boyary, carry red ribbons.10 In Ukrainia the horses pulling the car in which the bride is, late at night, taken to the bridegroom's hoiise, are adorned with red ribbons ; her mother seizes hold of their bridles, leads them to the road, and wishes her daughter a " good night," while the other women who accompany her sing, " We have laid her upon a white bed, she has herself desired a red beet-root for her white body."11 Among the Bashkir the bridegroom, before 1 Volkov, in L' Anthropologic, ii. 573. 2 Piprek, op. cit. p. 42. In Ukrainia, if the bride is known not to be a virgin, " on ne voit point de rouge " (Volkov, in L' Anthropologic, iii. 546). 3 Gray, op. cit. i. 205. 4 Volkov, in L'Anthropologic, iii. 545. 6 Wessman, ' Folktro i Ekenas,' in Hembygden, ix. 51. 6 Marquardt, op. cit. p. 45. 7 v. Hahn, Albancsischc Studien, p. 145. 8 Lobel, op. cit. p. 218. 9 Volkov, in L'Anthropologic, iii. 545. 10 Piprek, op. cit. p. 16. 11 Volkov, in L'Anthropologic, ii. 562 sq. XXIV MARRIAGE RITES 467 entering the room where the guests are assembled, " doit briser du pied un fil rouge que tiennent deux femmes en travers de la porte ; s'il ne voit pas le fil et tombe, tout le monde se moque de lui. Puis il s'assied, et les hotes s'en vont l'un apres 1'autre ; quand il est seul, les amies lui am^nent la fiancee, et s'en vont."1 Many rites are practised with a view to making the wife fruitful or the mother of male offspring. First, prayers are often offered to this effect by a relative or a priest; of this instances will be given below. In Morocco, when the bride has been dressed in her wedding-costume, the bridegroom's mother, and subsequently the other women who are present as well, sing :-" Go out, may you give birth to male twins; even one son I wish you to give birth to."2 Or she carries a sieve,3 or a bundle of her son's old clothes,4 on her back, as if it were a baby ; or the bride's mother is put into a net by the bachelors, and swung to and fro in the same manner as a child is rocked to sleep.5 When the bride is taken to the bridegroom's place the animal on which she rides must sometimes be a mare, on account of its fruitfulness,6 and sometimes a stallion, that she shall give birth to male offspring.7 It is, in certain cases at least, for the same purpose that a little boy rides behind her on the mare ;8 and the custom which sometimes requires that the animal shall also be ridden by a little boy when it is taken to the bride's place9 seems partly to serve a similar object, although the chief thing may be that it should be ridden by somebody, since an empty saddle is looked upon as a bad omen and should therefore be avoided at a wed- ding,10 just as are empty dishes, trays, and tables.11 1 van Gennep, op. cit. p. 172. Cf. supra, ii. 464. 2 Westermarck, op. cit. pp. 154, 155, 348 sq. 3 Ibid. pp. 195, 199. 349- 4 Ibid. pp. 127, 128, 230, 349. 6 Ibid. PP.-246, 247, 349. 6 Ibid. pp. 175, 180, 191, 349. 7 Ibid. pp. 168, 191, 349. In Wales it was believed that " if the bridegroom rode a mare to the wedding, his wife would have daughters, but no sons " (Marie Trevelyan, op. cit. p. 271). 8 Westermarck. op. cit. pp. 172, 174, 176, 179, 181, 185, 191, 349. 9 Ibid. pp. 172, 175, 191, 192, 349- lfl Ibid. pp. 172, 180, 182, 192, 320 sq. 11 Ibid. pp. 24, 25, 34, 239, 241, 249, 279, 287, 321. H H 2 468 THE HISTORY OF HUMAN MARRIAGE chap; Very similar customs are found in other countries. Among the Manchus, when the bridal sedan chair is sent to the bride's house, it is sometimes occupied " by a child of two years old, whose presence is supposed to be an omen of a numerous progeny of male children."1 Miss Durham tells me that among the Roman Catholics of Scutari, in order that the bride shall give birth to male offspring, a little boy is put into the carriage in which she drives from the church to the bridegroom's house. In Southern Albania, again, as I am informed by M. Mehmed Bey of Konitza, a little boy is for the same purpose put into the nuptial bed and rolled to and fro before the couple lie down; and a similar custom prevails among the Slovaks.2 Among the Muhammadans of Visegrad and Cainica a boy is rolled on a mattress on which the bride is placed before she is taken to the nuptial chamber; 3 whilst in some parts of Sweden she should have a boy-baby to sleep with her on the night preceding the wedding day in order that her first-bom shall be a son.4 Among the Esthonians5 and many or all Slavonic peoples6 a boy is offered to the bride or is put to sit on her lap ; and in Bulgaria the bridegroom has to hold a boy and the bride a girl.7 In Corsica, also, a child is given to the bride, after which the whole assembly pronounce the following blessing upon the couple :-" Dio vi dia buona fortuna, Tre di maschi e femmin' una."8 The 1 Stewart Lockhart, in Folk-Lore, i. 486. 2 Piprek, op. cit. p. 107. 3 Ibid. p. 132. 4 Gaslander, ' Beskrifning, Om Allmogens Sinnelag, Seder vid de arliga Hogtider etc. i Jonkopings Lahn och Wassbo Harad,' in Nyare bidrag till kdnnedom om de svenska landsmalen och svenskt folklif, Bihang i. p. 276. Raaf, op. cit. p. i. 113. Lloyd, Peasant Life in Sweden, p. 85. 5 Boeder-Kreutzwald, Der Ehsten dberglaubische Gebrauche, Weisen und Gewohnheiten, p. 38. Among the same people a suitor must not ride on a mare to the house of the girl he is courting, lest he should get only daughters (ibid. p. 26). 6 Piprek, op. cit. pp. 18, 19, 46, 106, 114, 129, 130, 133, 136, 145, 173. Krauss, Sitte und Brauch der Sudslaven, pp. 386, 398, 428, 430, 147 sq. Rajacsich, op. cit. pp. 159, 178. Mannhardt, Mythologische Forschungen, p. 357. 7 Piprek, op. cit. p. 146. Lobel, op. cit. p. 221. 8 Ida v. Diiringsfeld and v. Reinsberg-Diiringsfeld, op. cit. p. 257. XXIV MARRIAGE RITES 469 custom may have belonged to the primitive Indo-European marriage ritual; for we learn from the Grihyasutras that in ancient India, on the bride's entering her new home, a little boy was placed on her lap, as an omen of male progeny.1 The same rite is for the same purpose practised in Nias.2 Among the Masai, when bride and bridegroom are together in their new hut, the mother of the latter brings there a little child, which the bridegroom puts to sit on the bride's lap, and she gives it milk to drink from a bottle ; this, also, is done for the purpose of obtaining children.8 Among the Suk of East Africa the bride refuses to enter the bridegroom's house until the child of a neighbour is put in her arms.4 The following practices may also serve as instances of the- many methods by which marriages are intended to be rendered fruitful or productive of male offspring. Among the Manchus, when bride and bridegroom are sitting upon the bed face to face, an " offspring dumpling " is brought in and handed first to the bridegroom, who eats a mouthful, and then to the bride, who takes a small piece into her mouth and afterwards spits it out, " as an omen that the marriage will be productive of a numerous offspring. "5 In Gardenston, in Scotland, the bridal bed was made up by a woman giving suck, as it was believed that if any other woman did so there would be no family.6 Among some of the Serbs a man's belt was formerly tied round the hand of the bride so that her first child should be a boy.7 Among the Slovaks of Gemer, when the bride is taken to the nuptial chamber, the young men, who are waiting at the door, for the same purpose touch her with their hats.8 Among the old Prussians no castrated animal was slaughtered to provide meat for 1 Grihy a-Sutras, i. 42 ; ii. 50, 263. 2 Schmidt, Liebe und Ehe im alten und modernen Indien, p. 417. 3 Merker, Die Masai, p. 48. 4 Macdonald, ' Notes on the Ethnology of Tribes met with during Progress of the Juba Expedition 1897-99/ in Jour. Anthr. Inst. xxix. 238. 8 Stewart Lockhart, in Folk-Lore, i. 488. 6 Gregor, Notes on the Folk-Lore of the North-East of Scotland, p. 100. 7 Piprek, op. cit. p. 135. 8 Ibid. p. 107. 470 THE HISTORY OF HUMAN MARRIAGE chap. a wedding, because the slaughter of such an animal was supposed to make the marriage childless.1 A marriage ceremony which has been represented as a fertility rite is the custom of the bridal pair, or the bride alone, being seated on the skin of an animal. This custom has prevailed, or still prevails, among several Indo-European peoples, and has therefore been supposed to have originated in primitive times.2 In India the bride, on her arrival, was made to sit down on a red bull's hide ; this custom is mentioned already in the Atharva-Veda.3 But in two Grihya- sutras the rule is given that both bride and groom shall sit down on the hide.4 In Rome, again, the ordinary custom was to place the bride on a sheepskin, although at the marriage of priests the custom was to cover two chairs with the skin of the sacrificial sheep, and to make the bridal pair sit down on this skin.5 Among the Slavonic peoples the couple sit on a fur, cushion, or cloth.6 In Esthonia the bride is placed on a blanket or fur ; but in one place she is made to stand on a man's coat,7 and this Dr. Hartland considers to favour the view8 that the rite was intended to promote the bride's fertility.9 Yet it may very well be that the bride, or the groom as well, had to sit on a skin because it was considered dangerous for them to sit on the ground,10 and that the use of a male animal's hide, which is not mentioned in the descriptions of the European rites, or of a man's coat, was a local modification of the prophylactic rite combining with it a reproductive purpose. Another marriage rite which has been traced to the primitive Indo-Europeans is the custom of throwing some kind of cereals or fruit on the bride.11 This custom, or the 1 Hartknoch, Alt- und Neues Preussen, p. 180. 2 Wintemitz, in Transactions, p. 273 sq. 3 Ibid. p. 273. 4 Grihya-Stitras, ii. 193, 194, 263. 6 Servius, Commentarii in Virgilii Aeneidos, iv. 374. Festus, op. cit. p. 114. Rossbach, op. cit. p. 112 sq. 6 Piprek, op. cit. pp. 13, 14, 36, 185 sq. 7 v. Schroeder, op. cit. p. 89. 8 For this view see, e.g., Piprek, op. cit. p. 186 sq. ; Crooke, 'Lifting of the Bride,' in Folk-Lore, xiii. 244; Frazer, Totemism and Exogamy, iv. 210. 9 Hartland, Primitive Paternity, i. 133. 10 See infra, ii. 530 sqq. 11 Wintemitz, in Transactions, p. 277 sqq. XXIV MARRIAGE RITES 471 custom of throwing grain or fruit on the bridal pair or on the bridegroom separately or even on the wedding company, has been found to prevail from India, Indo-China, and the Indian Archipelago in the East to the Atlantic Ocean in the West.1 In India we can trace it from the Grihyasutras through the classical Sanskrit literature down to the present day. The poet Kalidasa describes how Prince Aja and his bride, sitting on a golden chair, were strewn with wet grains of barley, first by young Brahmans, then by the King and all the relations, and lastly by noble women.2 In Siam bride and bridegroom are sprinkled with rice, scented oil, and flowers.3 Among the Karens of Martaban " the head person of the village, or an elder selected for the occasion, takes some rice and places part on the head of the bride- groom and part on that of the woman."4 Among the Gonds of the Eastern Ghauts the assembled females throw yellow-coloured rice from all directions on the couple.5 Among the Hindus of South India all present throw rice on to the heads of the married pair.6 Among the Mundas the bridegroom throws three handfuls of rice at the forehead of the bride and the bride next throws three handfuls of rice at the forehead of the bridegroom;7 and among the Yan&dis of North Arcot, also, each of them pours a handful of rice upon the other's head three times.8 Among the 1 Idem, Das altindische Hochzeitsrituell, p. 75 sqq. Idem, in Trans- actions, p. 277 sqq. Crooke, Popular Religion and Folk-Lore in Northern India, ii. 26 sq. Schmidt, Liebe und Ehe im alten und modernen Indien, p. 417 sqq. Idem, Beitrdge zur indischen Erotik, p. 504 sqq. Samter, Familienfeste der Griechen und Romer, p? 1 sqq. Mannhardt, Mythologische Forschungen, p. 354 sqq. v. Schroeder, op. cit. p. 112 sqq. de Gubematis, Storia comparata degli usi nuziali, p. 174. 2 Wintemitz, Das altindische Hochzeitsrituell, p. 75. Idem, in Transactions, p. 278 sq. 3 Young, Kingdom of the Yellow Robe, p. 95. 4 Low, in Jour. Indian Archipelago, iv. 418. 6 Hayavadana Rao, in Anthropos, v. 794. 6 Padfield, op. cit. p. 108. 7 Sarat Chandra Roy, Mundas, p. 447. 8 Cox, Madras District Manuals: North Arcot, i. 250. Ranga Rao, ' YdnAdis of the Nellore District,' in the Madras Government Museum's Bulletin, iv. 99. 472 THE HISTORY OF HUMAN MARRIAGE chap. Coorgs, at the feast in the bride's house the bridegroom strews some grains of rice upon her head, gives her a little milk to drink, and makes her a present of some coin; and then his parents and relatives salute her in the same manner.1 In Bihar, again, " when the bridegroom arrives at the door of the bride's house, the women of her family receive him, and scatter over him uncooked rice, the dung of a heifer, balls of cooked rice, and other articles."2 So also in Dar- distan the bridegroom, and his friends as well, standing round the door of the bride's house, are sprinkled with flour.8 Contact with grain may also be brought about in some other way than by throwing it. Thus among the Orions the bride- groom marks the bride with red lead while both are standing on a curry-stone, under which a sheaf of com lies upon a plough-yoke.4 Among the Berads of Bombay the bride is made to stand in a basket of millet.5 At Maratha marriages bride and bridegroom have to stand in baskets filled with unhusked wheat.6 Among high-caste Hindus of the Punjab the bridegroom,on entering the bridal chamber, finds a sieve hanging on the doorpost.7 Similar rites are found among more northerly peoples in Asia. In Tibet, when the bride arrives at the bride- groom's house, his mother receives her with some barley flour mixed with butter in her right hand, and with a jar full of milk in her left hand ;8 or when the bride goes to the bridegroom's house accompanied by the relatives and friends of both families, each of these takes grain and scatters it over the bride ;9 or a measure full of grain, in which is stuck an arrow with a lump of clarified butter at the top, is at the wedding placed between bride and bridegroom.10 1 Moegling, Coorg Memoirs, p. 39. 2 Grierson, Bihar Peasant Life, p. 364 sq. 3 Biddulph, Tribes of the Hindoo Koosh, p. 79 sq. 4 Dalton, op. cit. p. 252. 6 Crooke, in Folk-Lore, xiii. 235. 6 Shastri, in Panjab Notes and Queries, i. 99. 7 Maya Das, ibid. i. 98. 8 Sarat Chandra Das, * Marriage Customs of Tibet,' in Jour. Asiatic Soc. Bengal, vol. Ixii. pt. iii. 19. 8 Rockhill, ' Tibet,' in Jour. Roy. Asiatic Soc. N.S. xxiii. 229. 10 Ahmad Shah, Four Years in Tibet, p. 57. XXIV MARRIAGE RITES 473 Among the Buryat " the bridegroom enters the yurta to put some grease in the fire; and when the bride and her party follow him in, grains of com are thrown upon their heads."1 Among the Manchus " laichees, lung ugan, chestnuts, and dates " are placed on the four comers of the bridal bed.2 At Foochow in China five coins, belonging to the reigns of five different emperors, are usually scattered around on the bottom of the bridal bedstead ; five bunches of boiled rice, each consisting of five bundles, are hung up from the frame provided for suspending the curtain of the bed; and a square wooden vessel, half filled with uncooked rice, is placed on the centre of the bedstead, and on the top of the rice is spread a sheet of red paper on which are arranged various articles, such as ten pairs of chop- sticks, a small case containing money-scales, and five kinds of dried fruits. When the bride is seated in the sedan in which she is to be taken to her future home, but before she starts, her parents or some other members of her family take a bed-quilt by its four comers and hold it thus before the bridal chair, while one of the bride's assistants tosses into the air, one by one, four bread-cakes, in such a manner that they will fall into the bed-quilt. When the sedan, on its arrival at the bridegroom's place, has been carried into the reception-room, a sieve is put on the top of it, over its door; and when the bride is then led towards the door of her room, this sieve is either held over her head or is placed in front of the door of the sedan, so that, on stepping out, she will step into it.3 In ancient Greece the bride, on entering the bridegroom's house, was taken by him to the hearth and was there showered with dates, figs, nuts, little coins, and so forth; or sweetmeats were poured upon the wedded pair at the entrance to the bridegroom's house.4 In ancient Rome, again, the bridegroom scattered nuts for the boys in the 1 Miss Czaplicka, Aboriginal Siberia, p. ng. 2 Stewart Lockhart, in Folk-Lore, i. 488. 3 Doolittle, op. cit. i. 76, 79, 83 sq. 4 Samter, Familienfeste der Griechen und Romer, p. 1 sq. Smith, Wayte, and Marindin, op. cit. ii. 136. 474 THE HISTORY OF HUMAN MARRIAGE chap. crowd j1 but Mannhardt suggests that the nuts were originally thrown by him over the bride, and then only gathered up by the boys.2 In modem Greece coins, rice, cotton-seeds, sweetmeats, and also nuts, are thrown upon the bridal procession from the windows of relatives and friends, or on bride and bridegroom on their departure from the bride's home or on their arrival at the bridegroom's.3 In Crete, at the threshold of the bridegroom's house, a maiden receives the bride with honey and nuts mixed with sesame, and when the bride comes to the interior of the house a pomegranate is offered her; she breaks it into pieces, and strews the pips on the ground.4 At Scutari, according to Miss Durham, when the bride arrives at the bridegroom's house, the wedding guests throw on her flowers, sweets, and holy water blessed by the priest in church whilst in South Albania, according to M. Mehmed Bey, the bridegroom's mother throws rice over her.5 In Slavonic countries bride and bridegroom are sprinkled with corn, or corn and hops, when they enter the house of the latter ;6 or corn or almonds and coins as well are thrown over the bride,7 or hops or wheat over the bridegroom,8 or hops, grain, and nuts over the nuptial procession.9 In some parts of Russia the ceremony takes place in church : when 1 Vergil, Bucolica, Ecloga viii. 30. Catullus, Carmina, Ixi. 131 sqq. Marquardt, op. cit. p. 54. 2 Mannhardt, Mythologische Forschungen, p. 361. 3 Wachsmuth, op. cit. p. 92. Sakellarios, op. cit. p. 20. Rodd, op. cit. pp. 95, 99. In Cappadocia it is the custom to throw a mixture of com and coins over the heads of the couple (Naumann, quoted by Stem, Medizin, Aberglaube und Geschlechtsleben in der Tiirkei, ii. 103. 4 Wachsmuth, op. cit. p. 95 sqq. 5 Hahn states (op. cit. p. 146) that in Albania, on the arrival of the bridal procession at the bridegroom's house, his mother throws rice over the couple and then on the whole company. Among the Kil- meni, a large tribe living north of Scutari, the bride is led three times round the bridegroom's house, an apple is thrown over the roof, and she is given com (Miss Durham, High Albania, p. 86). 6 Piprek, op. cit. p. 12 (Great Russians), 64 (White Russians). 7 Ibid. pp. 133, 144 (Southern Slavs). Rajacsich, op. cit. p. 159. 8 Piprek, op. cit. pp. 18 (Great Russians), 45 (Little Russians). 9 Ibid. p. 16 (Great Russians). XXIV MARRIAGE RITES 475 bride and bridegroom kneel down on a carpet to receive the blessing of the bridegroom's parents, the mother scatters grains of hops over the bride's head j1 or when the priest has tied the nuptial knot at the altar, the clerk or sexton sprinkles on her head a handful of hops.2 Among the Serbs of Syrmia the bridegroom's mother receives the bride with a loaf of bread and a plate containing grain or rice ;3 and a sieve figures prominently in the marriage ceremonies of the Serbs, Croats, and Bulgars.4 Among the old Prussians the bride was taken to all the doors of her new home, and at each door she was sprinkled with barley, wheat, and pulse.5 In Italy grain is, or used to be, thrown over bride and bridegroom or on the bridal company, or confetti is used instead ;6 or sweets, boiled chestnuts, nuts, and walnuts are distributed to the spectators.7 In some parts of France hemp-seed or wheat is showered over bride and bridegroom ;8 or the bride, besides treading upon an egg, had wheat thrown over her on entering her new home-a custom which was expressly condemned by a Synod in 1626.9 In the same country the bride is also received with three loaves of bread, which she immediately gives to some poor people,10 or with a loaf of bread and a lump of butter, which she distributes among the young people accompanying her.11 In Germany the bride and bridegroom are frequently received with bread and drink,12 and in Silesia and Bohemia peas or grains of peeled barley are thrown on them.13 But 1 Mannhardt, op. cit. p. 355. 2 Lady Hamilton, Marriage Rites, Customs, and Ceremonies, of the Nations of the Universe, p. 99. 3 Rajacsich, op. cit. p. 159. 4 Ibid. p. 179. Piprek, op. cit. pp. 125, 129, 136, 143, 145 sq. 6 Hartknoch, op. cit. p. 180. 6 Pitrd, Usi e costumi del popolo siciliano, ii. 72 sqq. Bresciani, op. cit. ii. 154 ; Faggiani, in Provenzal, op. cit. p. 232 (Sardinia). 7 Canziani, ' Courtship, Marriage, and Folk-Belief in Vai d'Ossola (Piedmont),' in Folk-Lore, xxiii. 460. 8 Laisnel de la Salle, op. cit. ii. 47. 9 Thiers, op. cit. iv. 471. 10 de Nore, op. cit. p. 238. 11 Ogde, Dictionnaire historique et geographique de la province de Bretagne, ii. 357. 12 Sartori, op. cit. i. 90 sq. 18 Mannhardt, op. cit. p. 360. 476 THE HISTORY OF HUMAN MARRIAGE chap. the custom of throwing corn on bride or bridegroom seems to be rare in Germany and unknown at peasant weddings in Scandinavia, although grains of wheat or com are sometimes put into their stockings.1 On the other hand, the custom in question is found among some Ugro-Finnic peoples, who have probably borrowed it from their Slavonic neighbours.2 In England other things than rice were formerly, or are still in some places, thrown upon the bride, apart from the modern adoption of confetti. We are told that in the seventeenth century wheat was cast on her head when she came from church.3 In the north of England one of the oldest inhabitants of the neighbourhood, who has been stationed on the threshold of the bride's new home, throws a plateful of short-bread over her head, so that it falls outside ; and a scramble ensues for the pieces, as it is deemed very fortunate to get a piece of the shortbread.4 At Siston, in Gloucestershire, at the beginning of the eighteenth century, a large cake was broken over the heads of the couple.5 In the north-east of Scotland, when the bride passed over the threshold, there was held over her head a sieve containing bread and cheese, which were distributed among the guests or sometimes scattered around her, in which case there was a rush made by the young folks to secure a piece. At times an oatmeal cake was broken over her head, but in later days a thin cake of short-bread, called the bride-cake, was substituted for it; and this, too, was distributed among the guests, who carefully preserved it, particularly the unmarried. In some districts, when the sieve was in the act of being placed over her head, or the bread was broken, it was the bridegroom's duty to snatch her from below it. She was led first to the hearth and at last to the girnal, or mehl-bowie, and her hand was pressed into the meal as far as possible.6 In Rosehearty, in Aberdeenshire, barley is thrown over the bridal pair as they come to the feasting- 1 Ibid. p. 361. 2 Hamalainen, op. cit. p. 251 sq. ' MoSet, Health's Improvement, p. 218. 4 Henderson, op. cit. p. 36. 6 Gloucestershire Notes and Queries, ii. 276. 8 Gregor, op. cit. p. 92 sq. XXIV MARRIAGE RITES 477 place.1 In Shetland the bride's mother and one or two female relatives meet the bride carrying in a napkin az cake baked with seeds and sugar, broken into small pieces, which are thrown over the head of the bride.2 In the west of Ireland the bridegroom's mother breaks an oaten cake on the head of the bride as the young woman passes the doorway of her future home.3 In the description of a rustic marriage festival in the wilds of Kerry about the year 1830 it is said that " behind the pair followed two attendants, bearing high over the heads of the young couple a sieve filled with meal."4 Customs of the same sort have been found among Jews of all countries. Already in Talmudic times nuts and wheat were cast about the path on which bride and bride- groom strode ; and barley was sown in a flower-vase a few days before the wedding, and was thrown over the young couple as in modem times.5 At a Jewish wedding in May- ence at the end of the fourteenth century the bridegroom Took the bride's hand at the door of -the synagogue while the two were showered with wheat and coins, which were afterwards given to the poor.6 But as the custom is not known to have existed among the ancient Hebrews, it has been suggested that the Jews borrowed it from Indo- Europeans.7 Among the Maronites " the bride, conveyed to her new home, is pelted with corn and raisins, and on arriving she flings a pomegranate amid the party, which is greedily seized and divided by the bridegroom's com- panions."8 Among some Algerian Berbers, after the bride 1 Idem,' Some Marriage Customs in Cairnbulg and Inverallochy,' in Folk-Lore Journal, i. 120. 2 Black, County Folk-Lore, vol. iii. Orkney and Shetland Islands, p. 209 sq. 3 Blake, ' Matrimonial Customs in the West of Ireland,' in Folk-Lore, .xviii. 81. 4 Wood-Martin, Traces of the Elder Faiths of Ireland, ii. 33. 5 Abrahams, op. cit. p. 196. 6 Grunwald, * Marriage Ceremonies,' in Jewish Encyclopedia, viii. 340. 7 v. Schroeder, op. cit. p. 210 n. 1. 8 Conder, Heth and Moab, p. 293. 478 THE HISTORY OF HUMAN MARRIAGE chap. has drunk the milk and water offered her, " on lui donne une poignee de ble, d'orge et de sei, qu'elle doit jeter, a droite et & gauche, par dessus ses epaules."1 In Morocco, when the bride arrives at the bridegroom's place, she is received with grain, flour, bread, seksu, or something else made of grain, or red raisins; and very commonly dried fruits-raisins, figs, dates, walnuts, or almonds-are thrown over her, or over the box in which she has been brought, by the bridegroom's mother or other people of his family or the bridegroom himself or, in one tribe visited by me, by a woman of the bride's family.2 It is quite possible that the peoples of North Africa have received the custom from Indo-Europeans, but it may have been a genuine Berber custom; the Guanches of the Canary Islands are said to have thrown grain in the faces of the newly-wed.3 It is interesting to note that among the Wabondei in Central Africa rice and maize are strewn over the heads of bride and bridegroom, and maize is put on their feet.4 In Ponape, one of the Caroline Islands, when the girl has arrived at the house of the bridegroom, his mother rubs coconut oil vigorously into her back and shoulders, and a garland of flowers is placed on her head.5 Mannhardt suggests that the custom of throwing grain or seeds or dried fruit over the bride undeniably takes its rise " from the feeling of a sympathetic connection between mankind and seed-bearing grasses and the comparison between the fruit of the body and of com " ;6 and some later writers likewise assume that it was intended to pro- 1 F6raud, ' Moeurs et coutumes kabiles,' in Revue africaine, vi. 43i. 2 Westermarck, Marriage Ceremonies in Morocco, pp. 194-198, 201, 204-207, 209, 211-214, 216. 3 Alice C. Cook, ' Aborigines of the Canary Islands,' in American Anthropologist, N.S. ii. 478. 4 Dale, in Jour. Anthr. Inst. xxv. 199. Baumann, Usambara, p. 134. Among the Wasambara, according to Farler (' Usambara Country in East Africa,' in Proceed. Roy. Geograph. Soc. N.S. i. 92), the women go round the bride with baskets of Indian com, dropping some before her as they pass, until a large heap is made. 5 Christian, Caroline Islands, p. 73. G Mannhardt, Mythologische Forschungen, p. 365. XXIV MARRIAGE RITES 479 mote fecundity.1 It is also evident that in certain cases this or kindred customs are looked upon in such a light by the peoples who practise them. In Bohemia and Silesia, where peas or grains of peeled barley are thrown on the couple, it is believed that the number of grains afterwards found lying on the bride's dress indicates the number of children she will have.2 The Great Russians are said to throw corn on bride and bridegroom " in order that their married life shall be fruitful" ; 3and the Russian clerk or sexton who sprinkles hops on the bride's head wishes that she may be as fruitful as that plant.4 Addison wrote in the seventeenth century that among the Jews of some coun- tries " the Guests bring with them handfuls of corn, which they cast at the New-Married, saying, Increase and Multiply. By which they also wish them Peace and Abundance."^ And the same custom is still practised by West Russian Jews.6 The Cretan customs mentioned above are repre- sented by Wachsmuth as fertility rites.7 At Licata, in Sicily, the barley thrown on the couple augurs the birth of male offspring and the wheat the birth of female.8 Of the Manchus it is said that the chestnuts, dates, and other things that are placed on the four corners of the bridal bed " are intended to indicate that the pair will produce an early, numerous, and intelligent offspring."9 The corn which the Wabondei throw over the heads of bride and bridegroom is represented as a symbol of fertility.10 Mannhardt also re- garded the sieve, of which so frequent use is made in marriage ceremonies, as a symbol of fecundity; 11 and this view is supported by certain practices I found in Morocco.12 For in- 1 v. Schroeder, op. cit. p. 120. Wintemitz, in Transactions, p. 278 sq. Idem, Das altindische Hochzeitsrituell, p. 75 sqq. JHartland, Primitive Paternity, i. 109 sq. Frazer, Totemism and Exogamy, ii. 260 n. 1. Piprek, op. cit. pp. 173, 187. 2 Mannhardt, op. cit. p. 360. 3 Piprek, op. cit. p. 12. 4 Lady Hamilton, op. cit. p. 99. 5 Addison, Present State of the Jews, p. 52. Cf. supra, ii. 462. c Andree, Zur Volkskunde der Juden, p. 144. 7 Wachsmuth, op. cit. pp. 96, 98. 8 Pitrd, op. cit. ii. 73. 9 Stewart Lockhart, in Folk-Lore, i. 488. 10 Baumann, op. cit. p. 134. 11 Mannhardt, op. cit. pp. 357, 358, 360, 361, 364. 12 Westermarck, op. cit. p. 195 sq. 480 THE HISTORY OF HUMAN MARRIAGE chap. stance, among some Berbers of the Rif, when a person buys sheep or goats, he puts a sieve at the entrance of the yard and makes the animals step over it so that they shall multiply. At the same time we also find other ideas connected with the throwing of grain and other kindred marriage rites. In many cases they are said to be means of ensuring pros- perity as well as offspring, or prosperity or abundance only. The object of placing coins, rice, and other articles on the bridal bed at weddings in Foochow " is to secure prosperity to the couple after their marriage, especially with regard to the bearing of children in their family in successive generations. The five cash of five successive emperors, etc., are good omens of such fruitfulness on the part of the expectant bride, or of general prosperity to the family. 1 Among the old Prussians the person who sprinkled the bride with corn and pulse said, " May our gods give you a sufficiency of everything as long as you remain in the faith in which your forefathers died and you manage your house with industry and due care."2 In Sicily3 and Sardinia4 the throwing of grain over the bridal pair or the nuptial procession is considered to be an augury or symbol of abundance. M. Mehmed Bey tells me that the South Albanians throw rice over the bride to make the marriage prosperous, whilst, according to v. Hahn, the Albanian custom of throwing rice on the pair and their company is a " symbol of fertility and wealth."5 Among the Parsees rice is cast towards the young couple to signify plenty and prosperity.6 In Surrey, also, as old people have told me, the rice which is thrown 6n the bride means prosperity, and at Hackness in Yorkshire the casting of rice after the wedding-party when it comes out of church is a sign of the wish, " May plenty strew their path."7 The Irish custom of bearing a sieve filled with meal over the heads of the bridal pair was "a sign of the plenty that would be in their house, 1 Doolittle^ op. cit. i. 77. 2 Hartknoch, op. cit. p. 180. 3 Pitrd, op. cit. ii. 72 sq. 4 Faggiani, in Provenzal, op. cit. p. 232. 5 v. Hahn, op. cit. p. 146. 6 Mary Billington, Woman in India, p. 85. 7 Mrs. Gutch, County Folk Lore, vol. ii. North Riding of York- shire, York and the Ainsty, p. 295. XXIV MARRIAGE RITES 481 and an omen of good luck and the blessing of children."1 The pressing of the Scotch bride's hand into the meal was believed to secure "in all time coming abundance of the stuff of life in the household."2 In Montenegro a basket filled with fruits is offered to the bride, who distributes them among those who are present, as a symbol of prosperity having entered the house together with the bride.3 Among the Ruthenians of Bukovina bread and salt are thrown into the bosom of the bride, in order that these articles shall never fail her.4 The Algerian Berbers say that corn and salt are given to the bride and thrown by her over her shoulders " pour faire descendre la benediction et 1'abondance dans la famille."5 In Andjra, in Morocco, I was told that the bridegroom's mother throws bread and fruit over the bridal box so that the married couple shall have much to eat, and what falls on the ground is partly taken by persons who want to impart the bride's baraka, or holiness, to their com by putting it underneath the heap on the threshing- floor.6 To connect com or fruit with prosperity or abund- ance may, it seems to me, be just as primitive as to connect them with fertility. And when the rite also includes the throwing of coins, we may suppose that the idea of promot- ing the material welfare of the couple must have been present in the minds of those who initiated the rite. Among the Moors I have found yet other ideas attached to the rite we are discussing, though never that of ensuring fertility. Sometimes the raisins, figs, or dates are said to bring good luck,7 or to make everything sweet,8 or to make the bride sweet to the bridegroom's family ;9 and the cake, barley, or wheat which is offered her and then thrown by her on the people is said to give them the benefit of her baraka10 or to make the year good.11 In one tribe, where it 1 Wood-Martin, op. cit. ii. 33. s Gregor, op. cit. p. 93. 3 Lobel, op. cit. p. 261. 4 Kaindl, ' RuthenischeHochzeitzgebrauche in der Bukowina,' in Zeitschr. d. Vertins f. Volkskunde, xi. 163. 6 F^raud, in Revue africaine, vi. 431. 0 Westermarck, Marriage Ceremonies in Morocco, p. 195. 7 Ibid. p. 204. 8 Ibid. p. 206 sq. 9 Ibid. p. 209. 10 Ibid. pp. 196, 211. 11 Ibid. p. 198. VOL. II I I 482 THE HISTORY OF HUMAN MARRIAGE chap. is the custom for the bridegroom, sitting on the roof or upper floor of the house, to throw down dried fruit and bread on the blanket held over the bride and on the people round her when she walks into his house, it is said that he does so for the object of averting the evil eye from the bride, as the things thrown would attract the looks of the people.1 In other tribes the wheat, flour, and so forth, which the bride throws over her head is represented as a means by which she rids herself of evil influences,2 and sometimes the dried fruit which she throws on the people is said to remove bas, or evil, from those who partake of it.3 In other countries, also, customs of this sort are not infre- quently regarded as prophylactics or means of purification. In Syria4 and Palestine5 the throwing of grain and salt on the people at a wedding is said to be a charm against the evil eye. Among the Tobads the bridal couple are sprinkled with peeled rice as a protection against evil spirits.6 Among the Muhammadans of the kingdom of Acheh, on the east coast of Sumatra, the bridegroom is at the end of the wedding-feast smeared behind the ears with yellow glutinous rice to be protected from evil in- fluences.7 The customs of showering the bridegroom with rice and the dung of a heifer and of sprinkling him with flour, before he enters the bride's house,8 undoubtedly look like purification rites, cowdung being regarded in India as a purifying substance. On the north-east coast of Aber- deenshire, when the bride returns to her father's house after the marriage, broken bread of various sorts is thrown over her before she enters, and a similar ceremony is gone through with the bridegroom at his father's door.9 It is also note- worthy that the bride is subject to the rite chiefly on her arrival at her new home-an occasion when purification rites are of frequent occurrence. In ancient Greece dried 1 Ibid. p. 213. 2 Ibid. pp. 197, 207 sq. 3 Ibid. p. 208. 4 yan Kasteren, ' Aus dem " Buche der Weiber," ' in Zeitschr. d. Deutschen Palaestina- Vereins, xviii. 49. 6 Lydia Einszler, ' Das bose Auge,' ibid. xii. 208. 6 Schmidt, Liebe und Ehe im alten und modernen Indien, p. 431. 7 Snouck Hurgronje, Achehnese, i. 326. 8 Supra, ii. 472. 9 Gregor, in Folk-Lore Jozirnal, i. 119 sq. XXIV MARRIAGE RITES 483 fruits were thrown not only over a bride but over a newly- bought slave as well,1 and this, too, suggests the idea of protection against evil influences. It may be added that the Gypsies, in England at least, think that the scattering of bread on a person, or his carrying a grain of wheat, protects him against dangers, both natural and supernatural.2 Mr. Crawley maintains that " the practice of throwing rice originated in the idea of giving food to the evil influences to induce them to be propitious and depart," and that this idea, in particular, underlies the custom of flinging flour, sweetmeats, and similar things among the onlookers.3 So also Dr. Samter believes that the rite which we are now. discussing was originally meant as a food-offering to spirits which required propitiation ;4 and the same opinion has been expressed by M. Reinach5 and by Sartori.6 All dangerous influences at marriages, however, are not thought of as spirits, and it is very doubtful whether corn or fruits when used as prophylactics have always been originally regarded as offerings. In Morocco flour is on various occasions used as a means of purification,7 and among the Achehnese the most effective method of averting evil in- fluences consists in besprinkling the person or thing with water mixed with a little rica-flour.8 Mr. Crooke says that rice or wheat is poured over the heads of bride and bride- groom with the object " to annul the influences which would prevent the fertility of the union " ;9 but he also observes that one object of this rite " would seem to be to keep in the soul which is likely to* depart at such a crisis in life as 1 Samter, Familienfeste der Griechen und Romer, p. 2. 2 Thompson, in Folk-Lore, xxiv. 336. 3 Crawley, op. cit. p. 325. 4 Samter, Familienfeste, p. 7 sq. Idem, Geburt, Hochzeit und Tod, p. 172 sqq. 5 Reinach, Cultes, mythes et religions, i. 117. 6 Sartori, op. cit. i. 91. 7 Westermarck, ' Popular Ritual of the Great Feast in Morocco,' in Folk-Lore, xxii. 142. Idem, Ceremonies and Beliefs connected with Agriculture, Certain Dates of the Solar Year, and the Weather in Morocco, p. 50. Idem, Marriage Ceremonies in Morocco, p. 119. 8 Snouck Hurgronje, op. cit. i. 305. 0 Crooke, Things Indian, p. 319. I I 2 484 THE HISTORY OF HUMAN MARRIAGE chap. marriage."1 A similar observation, based, on the beliefs of the Macassars and Buginese of Celebes,2 had previously been made by Wilken ;3 but it can certainly not be accepted as a general explanation of the rite. Considering how many different explanations of the cus- tom in question are given by the people practising it, even in the same country, there can be no doubt that its real origin has, in certain cases at least, been forgotten and a new interpretation substituted for the idea from which it rose. But at the same time, as has been said before,4 we should be on our guard against the assumption, only too common in Anthropology, that similar ceremonies necessarily have their roots in similar ideas, even when practised by different peoples. Objects like corn and dried fruit may certainly be used for a variety of purposes. And if similar ceremonies may have sprung from different motives in different cases, it is obvious that the same ceremony in a given case may also be intended to serve more than one purpose; nay, there is no reason to deny the possibility of mixed motives from the beginning. It is an unwarranted assumption, then, that the custom of throwing grain, seeds, or dried fruit at weddings, wherever it is found, originated in a rite the exclusive object of which was to promote fecundity. To ensure prosperity and abundance and to avert evil may have been equally primitive motives for it. Fish are frequently used for reproductive purposes,5 and figure, partly at least, on that account in marriage rites. Thus the Brahmans of Canara take the married pair to a pond and make them throw rice into the water and catch min- nows ; they let all go save one, with the scales of which they mark their brows.6 Among oriental Jews the newly-wedded couple immediately after the religious ceremony jump three times over a large platter filled with fresh fish or over a vessel containing a live fish, or step seven times backwards and for- 1 Idem, Popular Religion and Folk-Lore of Northern India, ii. 26. 2 Matthes, Bijdragen tot de Ethnologie van Zuid-Celebes, p. 33. 8 Wilken, loc. cit. p. 419 sqq. Cf. Crawley, op. cit. p. 325. 4 See supra, i. 15 sq. 8 Hartland, Primitive Paternity, i. 48 sqq. « Crooke, Things Indian, p. 222. XXIV MARRIAGE RITES 485 wards over a fish ; and the ceremony is expounded to be the symbol of a prayer for children.1 So also it was as a symbol of fertility that fish was formerly eaten on the second day of the wedding week among German Jews.2 At Tangier the bride in the evening of the seventh day after her arrival at her new home, when she has been belted by two little boys, makes a tour of inspection in the house; when she comes to the kitchen door, says the Shareefa of Wazzan, " a fish is produced, and a pretence is made of scraping it on her foot-a symbol of plenty in the culinary department."3 That the fish is scraped on her foot, however, suggests that the ceremony was in its origin a fertility rite. The same was probably the case with a custom mentioned by Leo Africanus in his description of marriages at Fez some four hundred years ago. He writes:-" So soone as the new married man goeth foorth of the house (which is for the most part on the seuenth day after the mariage) he buieth great plentie of fishes, which he causeth his mother or some other woman to cast vpon his wiues feete; and this they, from an ancient superstitious custome, take for a good boading."4 A somewhat similar custom still prevails at Fez, although another meaning is attributed to it. I was told that on the ninth day after the actual wedding day the young wife must make bread in order that there shall always be much bread in the house, and her husband buys some fish, which he gives her to prepare, in order that they shall become pros- perous ; to eat fish is generally considered to be lucky. It would seem that the roe of the fish might suggest not only fertility but abundance. Eggs, also, are frequently used as means of promoting fecundity.8 In Andjra in Morocco, for instance, a woman who is anxious to become a mother sits down over a new bowl in which she has put a raw egg and some rain-water 1 Lobel, Hochzeitsbrauche in dev Tiirkei, p. 286. Abrahams, op. cit. p. 196. Grunwald, in Jewish Encyclopedia, viii. 341. Notes ar^d Queries, ser. vi. vol. viii. 513, and vol. ix. 134. 2 Grunwald, in Jewish Encyclopedia, viii. 341. 3 Emily, Shareefa of Wazan, My Life Story, p. 137 sq. Leo Africanus, History and Description of Africa, ii. 451. 5 Hartland, Primitive Paternity, i. 57 sqq. 486 THE HISTORY OF HUMAN MARRIAGE chap. which has fallen on 27th April (Old Style)-supposed to contain much benign virtue ; after a while she drinks the water and puts the egg underneath a hen to be hatched, and it is believed that if the peeper is a cock she will give birth to a boy, and if it is a hen to a girl. In the same tribe a man who, wants to increase his capacity of reproduction eats the yolk of an egg every morning before breakfast for forty days in succession, and after eating it fills the shell with oil, which he drinks. Eggs are also frequently used in marriage rites, and in some such cases they are expressly said to hint at offspring.1 But in Morocco, where they figure at weddings as prominently as anywhere, I have never heard of them being represented as fertility charms on such occasions. They are usually, like various other white things, said to make the couple's or the bridegroom's future bright and happy,2 and on account of their whiteness they are also used for magical purposes in circumstances when they could not possibly be meant to ensure fertility.8 In one case the ceremonial use of an egg before the wedding was said not only to make the bridegroom's life bright, but also to make the weather fine during the wedding,4 and in another case to give good luck to him and a good year to the community.5 Among the Jews of Morocco, again, according to Addison, the bridegroom on the marriage day " takes a raw Egg, which he casts at the Bride ; intimating thereby his desire that she may have both an easie and joyful Child- birth."8 And the West Russian Jews, particularly the strict sect of the Chasidim, have the custom of setting a raw egg before a bride as a symbol of fruitfulness and that she may bear as easily as a hen lays an egg.7 There are other ceremonies which are likewise meant to facilitate the delivery of the young wife. In some parts of Sweden a bride must leave the laces of her shoes untied, " so that she may bear children as easily as she 1 Lasch, ' Einige besondere Arten der Verwendung des Eies im Volksglauben und Volksbrauch/ in Globus, Ixxxix. 104 sq. 2 Westermarck, Marriage Ceremonies in Morocco, pp. 89, 90, 101, 113, 115, 124, 164, 194, *95. 260, 347. 3 Ibid. pp. 218 sq. n. 2, 261. 4 Ibid. p. 89. 5 Ibid. p. 113. 6 Addison, op. cit. p. 52. 7 Andree, op. cit. p. 145. XXIV MARRIAGE RITES 487 removes the shoe,"1 and on returning from church she should dismount quickly from her horse, snatch off the bridle, hit the animal on the nose, and loosen the saddle-girths in order to have an easy labour.2 In Esthonia, when the bridegroom comes riding, someone at once rushes to meet him and undoes his saddle-girth, because this tends to facilitate child-birth in the future wife.3 In the same country, when the young wife is brought into the husband's house, the fence on both sides of the entrance is pulled down, that she may drive in swiftly without hindrance ; then, it is thought, her confinements will come off quickly and easily.4 In the Ansbach country it is believed that if the bridegroom ties the bride's garters for her, she will have easy labours.3 Some marriage rites are supposed to influence the external appearance or the behaviour of the future offspring. Among some Southern Slavs a bowl of milk and two spoons are put into the nuptial room " in order that the couple shall have beautiful children."6 In Esthonia the bridegroom's atten- dant cuts a small piece off a whole loaf, butters it, and puts it in the bride's mouth ; her children will then have a small smooth mouth.7 In the same country, when the bride is fetched in, she must wear no chains or bells, but be led in in solemn silence ; then she will have quiet children.8 Various ceremonies are performed for the purpose of making the couple prosperous.9 In Morocco dates are eaten to make them wealthy, in accordance with the com- mon blessing, Allah itdmmar u i'dnmar, " May God give dates and plenty."10 In a Berber tribe, after the bride has arrived at the bridegroom's place, his mother throws a silk 1 Gaslander, loc. cit. p. 276. Raaf, op. cit. i. 115. 2 Raaf, op. cit. i. 115. See also Tomer, quoted by Wikman, ' Magiska bindebruk,' in Hembygden, iii. 66 ; and, for Denmark, Kristensen, Gamle folks fortcellinger om det jyske almueliv, iv. 69. 3 Boecler-Kreutzwald, op. cit. p. 30. 4 Grimm, Teutonic Mythology, iv. 1843. 5 Ibid. iv. 1806. 6 Piprek, op. cit. p. 128 sq. 7 Boecler-Kreutzwald, op. cit. p. 40. 8 Ibid. p. 31 sq. 9 See Westermarck, Marriage Ceremonies in Morocco, p. 347 and the references in the foot-notes. 10 Ibid. pp. 25, 27, 28, 139, 164, 347. 488 THE HISTORY OF HUMAN MARRIAGE chap. kerchief round her neck and leads her by it to the four comers of the tent, saying at each comer one of the follow- ing sentences:-" Here is the threshold of boys," " Here is the threshold of sheep and cattle," " Here is the threshold of safety and quietness," " Here is the threshold of cheap- ness."1 In another Berber tribe a lamb is handed to the bride by the bridegroom's mother, or, if she has no lamb, by some other person, after which the bride hurls the animal over the tent, so that there shall be many sheep in the village.2 In several tribes a woman of the bridegroom's family puts into the bride's hand some butter, with which she smears the horizontal pole supporting the roof of the tent, so that there shall be an abundance of butter or " grease " in the household.8 Perhaps it was partly for a similar purpose that the Roman bride anointed the door- posts of the bridegroom's house with fat and oil and hung wool on them,4 though the idea of averting evil may also have been connected with the custom.6 Among the Valakhs of Acarnania butter or, sometimes, honey is offered to the bride, who smears the door with it, " thus indicating that her arrival will bring. into the house sweetness and joy."® Moffet wrote in the seventeenth century:- " English People, when the Bride comes from Church, are wont to cast Wheat upon her Head, and the Grecians to anoint the Door-posts with fat Lard; so when their Brides and Bridegrooms return homeward from Church, one presents them, as presaging Plenty, and Abundance of all good things, with a Pot of Butter."7 In some parts of Sweden the bride on the same occasion ran into the larder, and drank some milk and ate some food, so that there should 1 Ibid. p. 201 sq. 2 Ibid. p. 204. 3 Ibid. pp. 201, 205-207, 221. For a similar practice in Algeria see Villot, Mceurs, coutumes et institutions des indigenes de l'Algirie, p. 105. 4 Rossbach, op. cit. pp. 351, 356 sqq. Samter, Familienfeste de? Griechen und ROmer, p. 80. 5 Cf. Fowler, Religious Experience of the Roman People, pp. 83, 90. 6 Heuzey, Le Mont Olympe et l'Acarnanie, p. 278. For a very similar custom in Rumania see Flachs, Rumanische Hochzeits- und Totengebrduche, p. 37 sq. 7 Moffet, op. cit. p. 218. XXIV MARRIAGE RITES 489 always be plenty of milk and food in the house.1 In Nerike, in the same country, the bride must always have a little to eat while she is dressed, in order that she shall never be without food.2 In Esthonia beer is wilfully wasted and spilt about at the wedding meal, so that superfluity may house with the happy pair.8 Among some of the Little Russians, on the return of bride and bridegroom from church, the bride's mother sprinkles the bridegroom three times with wheat and evergreen, puts some wool at his chest, offers him two cakes, and smears his mouth three times with honey, saying, " May your life become as sweet as the honey, may you become as rich as the sheep and as warm as the wool."4 Honey or other sweet things are often used in wedding ceremonies, in order to make the marriage happy.5 Among the Ruthenians of Bukovina the face of the bride is smeared with honey, or sugar is thrown into her bosom, so that her future shall be sweet.6 In Bulgaria one of the women anoints the bridegroom's face with honey, saying, " Be fond of each other as the bees are fond of the honey."7 In Rhodes, on arriving at the new dwelling which constitutes the dowry of the bride, the husband dips his finger in a cup of honey and traces a cross over the door, while those present cry aloud, " Be good and sweet as this honey isl "8 In the neighbourhood of Sparta, when the couple arrive at their new home, the bridegroom's mother stands waiting at the door holding a glass of honey in her hand. From this glass the bride must drink, that the words of her lips may become sweet as honey ; while the lintel of the door is smeared with the remainder, that strife may never enter in.9 1 Gaslander, loc. cit. p. 277. Raaf, op. cit. i. 116. 2 Djurklou, Ur Nerikes folksprdk och folklif, p. 48. 3 Boecler-Kreutzwald, op. cit. p. 38. 4 Piprek, op. cit. p. 45. 5 Ibid. pp. 106, 107 (Slovaks), 135 (Serbs)? Hartknoch, op. cit. p. 180 (old Prussians). Lobel, op. cit. p. 118 (Armenians), v. Seid- litz, ' Die Abchasen,' in Globus, Ixvi. 41. Sakellarios, op. cit. p. 22 (Cretans). Hahn, op. cit. p. 146 (Albanians). Pitrd, op. cit. ii. 74 sqq. (Sicilians). 6 Kaindl, loc. cit. pp. 163, 284. See also ibid. p. 285. 7 Piprek, op. cit. p. 144. 8 Rodd, op. cit. p. 99. 8 Ibid. p. 95 sq. 490 THE HISTORY. OF HUMAN MARRIAGE chap. There are marriage rites that have reference to the par- ticular position of the husband and the wife. Among the South Australian Narrinyeri the girl is said to signify her consent to the marriage by making a fire for her husband.1 The Negroes of Loango contract their marriages by the bride- groom's eating from two dishes which the bride has cooked for him in his own hut.2 Among the Mundas of Chota Nagpur " the bride goes to a stream or well near by with her pitcher, and, having filled it, raises it aloft on her head, steadying it with her hand. The bridegroom comes behind as she turns homewards, and, resting his hand on her shoulder, shoots an arrow along the path in front of her, through the loophole formed by her uplifted arm. The bride then walks on to where the arrow lies and picks it up with her foot, still balancing the pitcher on her head. Transferring it gracefully to her hand, she restores it to the bridegroom, thus showing that she can perform her domestic duties well, with hand and foot at his service. He, in turn, by shooting the arrow in front of her, has shown his ability to protect her and clear her path of any danger that may beset it."3 This rite, however, may originally have had a different meaning; the shooting of an arrow through the loophole formed by the bride's uplifted arm may be either a rite of impregnation or a mode of ensuring a safe delivery. We know that an arrow is sometimes regarded as the symbol of an embryo. In an ancient Hindu rite the husband fastened an arrow to his wife, after saying to her;-" May a male embryo enter thy womb, as an arrow the quiver; may a man be born here, a son after ten months."4 And among the polyandric Todas the husband who is to be the putative father of the first child presents the wife with a bow and arrow in or about the seventh month of her pregnancy,5 1 Taplin, Folklore, &c. of the South Australian Aborigines, p. 35. Idem, ' Narrinyeri,' in Woods, Native Tribes of South Australia, p. 12. 2 Soyaux, Aus West-Afrika, p. 161. 3 Bradley-Birt, op. cit. p. 50 sq. The same ceremony has been described by Dalton, op. cit. p. 194 sq. 4 Hymns of the Atharva-Veda, pp. 97, 356. Grihya-SHtras, i, 45. 5 Rivers, Todas, p. 319 sqq. Breeks, Account of the Primitive Tribes and Monuments of the Nilagiris, p. 19. XXIV MARRIAGE RITES 491 which certainly looks like a delivery rite.1 Moreover arrows are used at weddings to dispel evil spirits.2 In East MaUicolo in the New Hebrides, again, the bridegroom at the wedding sticks a poisoned arrow into the mat the bride has about her shoulders ; this is said to signify that " he has the power of life and death over her, and also that he must defend her with his life."3 Among the East African Wafipa the bridegroom boxes the bride's ear and puts into her mouth a small piece of wood or some straw; by the former act he shows that she will have to obey him, by the latter that it is his duty to support her.4 In Morocco the bridegroom tries in various ways to gain power over his wife. For this purpose, I was told, he taps her three or seven times on her head or shoulder with his sword,5 or beats her three times between her shoulders with the cord of his dagger,6 or smacks or kicks her gently,7 or drinks first from a bowl which he then holds for her to drink from.8 In Croatia the bridegroom boxes the bride's ears in order to indicate that henceforth he is her master.9 In Russia, as part of the marriage ceremony, the father took a new whip, and after striking his daughter gently with it, told her that he did so for the last time, and then presented the whip to the bridegroom.10 Among many 'Slavonic peoples the bridegroom gently beats the bride three times,11 "as a sign that she owes him obedience," or in order that she shall forget her earlier sweethearts and be afraid of her husband.12 It is also the custom for the bride to pull off 1 Cf. Frazer, Totemism and Exogamy, ii. 256 sqq. ; iv. 259 sq. 2 See infra, ii. 499. 3 Paton, quoted by Serbelov, ' Social Position of Men and Women among the Natives of East Malekula,' in American Anthropologist, N.S. xv. 278. 4 Fromm, ' Ufipa,' in Mittheil. Deutsch. Schutzgeb.-xxv. <yj. 5 Westermarck, Marriage Ceremonies in Morocco, pp. 235, 244, 256. 6 Ibid. pp. 250, 256. 7 Ibid. pp. 158, 159, 162, 242, 256. 8 Ibid. p. 232. 9 Krauss, Sitte und Brauch der Sildslaven, p. 385. 19 Meiners, Vergleichung des cittern und neuern Russlandes, ii. 167 sq. 11 Piprek, op. cit. pp. 41, 145, 149. Kaindl, loc. cit. p. 285 (Huzuls of Bukovina). 12 Piprek, op. cit. pp. 83, 19. See also ibid. p. 45 ; Volkov, in L'Anthropologie, ii. 562. 492 THE HISTORY OF HUMAN MARRIAGE chap. the bridegroom's boots,1 and in Russia the bridegroom formerly used to beat the bride on the head with the boot-leg to show that she now was in his power and had to obey him ; but among the Slovenes the bride nowadays beats the bride- groom with the boot-leg, ,so as to make him understand that she is not always going to pull off his boots.2 In some parts of Galicia the bride is seated on a yoke when the bridegroom puts on her the head-gear which she is to wear as a married woman, and she protests by throwing it off twice.3 Among the Hindus an ox-yoke is brought in, and a cord, supposed to represent one which is placed round the neck of an ox when it is yoked, is tied round the wrist of the bride by the bride- groom. The yoke is then held over the bride in such a manner that one of the holes in it shall come right over her head. As our informant observes, it is easy to see the significance of this ceremony.4 On the other hand, there are also rites that are intended to make the husband considerate or subject to his wife. Among the Brahmans of Eastern Bengal the bride's mother " places a weaver's shuttle between the young man's hands, and binds him hand and foot to symbolise the fetters he has imposed upon himself, and he then indicates his humility in the situation by bleating like a sheep " ; and the bride " lays upon his lips a padlock and turns the key, so showing that the door of unkind speech has been closed."5 Among the Chukmas, when the cloth with which the couple have been bound together is loosened, " both spring up, and if the wife is first on her feet she will always possess un- bounded influence over the affections of her husband."6 In many parts of Germany, when the priest joins the hands of the couple, the bride tries, in a literal sense, to get the upper hand, the bridegroom trying to do the same, and often a struggle of hands ensues, which is sometimes settled by the priest placing the man's hand uppermost; and one of the pair, generally the bride, also tries, for the same purpose, to put the foot on the top of the foot of the other 1 Piprek, op. cit. pp. 13, 42, 107, 128, 175. 2 Ibid. p. 175. 3 Ibid. p. 47. 4 Padfield, op. cit. p. 106. 5 Mary Billington, op. cit. p. 75 sq. 0 Hutchinson, op. cit. p. 97. XXIV MARRIAGE RITES 493 party.1 In Warinland, in Sweden, the bride must endeavour to see the bridegroom before he sees her, and to sit down first in the bridal chair, so as to have the mastery ; 2 and in many parts of Sweden3 and in Swedish-speaking communities in Finland4 she, for the same purpose, tries to place her foot before his during the nuptial ceremony. In Pellinge, in Finland, the bride's mother, on the same occasion, used to throw her daughter's skirt over the heels of the bridegroom, if she wanted her to rule over him.6 Among the Slovenes the bride endeavours during the nuptial ceremony or at the wedding dinner to put her knee on the skirt of the bride- groom's coat so as to become the ruler,6 and this custom is also found in Wiirtemberg.7 Among the Great Russians, when a glass of brandy has been offered to the bridegroom and another glass to the bride, each of them tries to pour a little of the brandy into the other's glass, and it is believed that the one who first succeeds in doing so will have more influence and power in the married fife.8 In Wales the bride should always buy something as soon as she is married, and before the bridegroom can make 1 Sartori, op. cit. i. 87 sq. Woeste, ' Aberglaube und Gebrauche in Siidwestfalen, ' in Jahrbuch d. Vereins f. niederdeutsche Sprach- forschung, 1877, p. 136. Temme, Die Volkssagen der Altmark, p. 73. Tetzner, ' Die Drawehner im hannbverschen Wendlande um das Jahr 1766,' in Globus, Ixxxi. 372. Wintemitz, in Transactions, p. 282. There is a similar scramble with the feet for the mastery in the household at Esthonian (Boecler-Kreutzwald, op. cit, p. 28) and Jewish weddings (Grunwald, in Jewish Encyclopedia, viii. 342). 2 Fernow, op. cit. p. 254. 3 Ibid. p. 254. Gaslander, loc. cit. p. 276. Raaf, op. cit. p. 112. Djurklou, op. cit. p. 49. Lithberg, ' Brollopsseder pa Gottland,' in Fataburen, 1907, p. 173. For other methods by which Swedish brides try to obtain the mastery see Raaf, p. 114. 4 Allardt, Nyldndska folkseder och bruk, p. 24. Tegengren, ' Magi och vidskepelse, hanforande sig till trolovning, brollop o.s.v. (FrAn Osterbotten),' in Hembygden, iii. 41. 8 Lindroos and Andersson, ' Ett brollop i Pellinge, Borga skargard, for 100 Ar tillbaka,' in Hembygden, 1910, p. 157. 6 Piprek, op., cit. p. 113. Franzisci, Cultur-Studien uber Volksleben, Sitten und Brduche in Karnten, p. 69. 7 Grimm, Teutonic Mythology, iv. 1803. 8 Piprek, op. cit. p. 12. 494 THE HISTORY OF HUMAN MARRIAGE chap. a purchase. " Then she'll be master for life 1 " say the old women.1 In Morocco, too, the bride in various ways tries to make herself the ruler. For this purpose she mounts the ram which is to be slaughtered for the occasion when she is painted with henna and boxes its ears, the ram represent- ing the husband ;2 she hangs on it a necklace to make him weak and harmless like a woman; and when its stomach has been removed, she puts her right foot on it.3 She is washed seated on a weaving-stool and a pack-saddle, the riding of the latter being supposed to give her power over her husband.4 She tries to smack the bridegroom when he, together with two other men, is running through the tent, so as to become his mistress.5 She waves her right slipper seven times towards the door of the nuptial chamber when she hears his steps outside ;6 or she throws at him one of her slippers when he enters ;7 or she beats him three times on his body with her slipper when he is going to have connection with her, though in this case it is said that she will be the ruler of the house only if he cries out, whereas otherwise he will rule over her.8 We must not presume, however, that all these and similar rites are practised for the purpose, or the sole purpose, of gaining mastery; and even when a rite is so now, it may very well have originated in a different idea. We shall see that bride and bridegroom are also beaten for purificatory pur- poses ; and this may be the case even when the bride is beaten by the bridegroom, or the bridegroom by the bride. In the Hiaina in Morocco the bridegroom, being alone with the bride, gently slaps her on her forehead and shoulders with the flat of his sword, "so as to expel evil spirits ";9 whilst among the Kabyles of Algeria, according to Hanoteau and Letourneux, " le mari, avant de la conduire a sa couche, la frappe legerement trois fois sur les epaules, avec le dos 1 Marie Trevelyan, op. cit. p. 274. 2 Westermarck, Marriage Ceremonies in Morocco, pp. 146, 157, 158, 355. 3 Ibid. pp. 158, 355 sq. 4 Ibid. pp. 151, 356. 5 Ibid. pp. 198, 199, 356. 0 Ibid. pp. 235, 256, 356. 7 Ibid. pp. 244, 256, 356. 8 Ibid. pp. 252, 256, 257, 356. 9 Ibid. p. 237. XXIV MARRIAGE RITES 495 d'un sabre ou d'un poignard, pour conjurer les effets du mauvais ceil."1 At Demnat, in the Great Atlas, again, th^ bride beats the bridegroom sometimes with a piece of rock- salt,2 which is much feared by the spirits, and sometimes with her slipper ;3 and this certainly suggests that the slipper ceremonies mentioned above as means of getting power over the other partner may at the same time be, or have been, methods of purification. 1 Hanoteau and Letoumeux, La Kabylie et les coutumes kabyles, ii. 219. 2 Said Boulifa, Textes bevberes en dialecte de VAtlas marocain, p. 19. 3 Westermarck, op. cit. p. 258. CHAPTER XXV MARRIAGE RITES (Continued) Besides marriage rites which, are supposed to confer positive benefits upon bride or bridegroom or both, there are others that are intended to protect them from evil influences or to rid them of such influences, that is, prophylactic or cathartic rites. There is a very general feeling or idea that bride and bridegroom are in a state of danger, being par- ticularly exposed to other persons' magical tricks or evil looks, or to the attacks of evil spirits, or to some impersonal mysterious cause of evil, which the Moors call l-bas, and therefore are standing in particular need of protection or purification. Moreover, the bride is considered to be not only herself in danger but also a source of danger to others.1 Customs that have direct reference to her may therefore at the same time be looked upon as safeguards against evils which threaten the bridegroom-as is evidently the case with various rites which immediately precede the con- summation of the marriage2-and also other persons less intimately connected with her. Purificatory ceremonies are thus of frequent occurrence on the bride's arrival at the bridegroom's place. In my book on ' Marriage Ceremonies in Morocco ' I have pointed out that a very large proportion of the marriage 1 See, e.g., Wintemitz, Das altindische Hochzeitsrituell, p. 41 sq. ; Sartori, op. cit. i. 79. 2 See Westermarck, Marriage Ceremonies in Morocco, p. 325. 496 ch. xxv MARRIAGE RITES 497 rites practised in that country are meant to serve a pro- phylactic or cathartic purpose,1 and the same is evidently the case in many other countries as well.2 In the present connection it will be impossible to give anything like a full account of practices belonging to this class of rites. Sometimes care is taken to shut out evil influences from the place where the marriage is celebrated. Thus in Russia all doors, windows, and even the chimney, are closed at a wedding, to prevent malicious witches from flying in and hurting the bride and bridegroom.3 Very frequently guns are fired off at a wedding,4 and in many cases at least the object of this is, or has been, to dispel evil spirits or other evil influences ; and the same may be said of the terrific noise or loud music which so often forms a part of the mar- riage ritual.5 Among the Siberian Yukaghir some relatives of either of the parties, following the train of sledges, " shoot their guns on both sides to drive away the evil spirits that might attempt to attack the bride " ; this is termed " shoot- ing into the eyes of the evil spirits."6 In Morocco, on the 1 See ibid. p. 321 sqq. and the references in the foot-notes. 2 See, e.g., Fra Paolino da S. Bartolomeo, Viaggio alle Indie Oriental!, p. 209 ; Atkinson, ' Notes on the History of Religion in the Himalaya of the N. W. Provinces,' in Jour. Asiatic Soc. Bengal, vol. liii. pt. i. 95 ; Sarat Chandra Das, ' Marriage Customs of Tibet,' ibid. vol. Ixii. pt. iii. 18 ; Stewart Lockhart, ' Marriage Ceremonies of the Manchus,' in Folk-Lore, i. 487. 3 Ralston, op. cit. p. 381. 4 Samter, ' Hochzeitsbriuche,' in Neue Jahrbucher fur das klas- sische Altertum, xix. 140 sq. Clozel and Villamur, Les coutumes in- digenes de la Cdte d'Ivoire, p. 309. Tessmann, Die Pangwe, ii. 261. Wilson, Western Africa, p. 267. Baumann, Usambara, p. 134 (Wabondei). 5 On the making of noise for the purpose of frightening away evil spirits see Samter, in Neue Jahrbucher filr das klassische Altertum, xix. 139 sqq. ; Idem, Geburt, Hochzeit und Tod, p. 58 sqq. ; Sartori, op. cit. i. 71, 72, 84 ; Feilberg, ' Hochzeitsschiisse, Neujahrsschusse,' in Archiv f. Religionswiss. iv. 172, 287 sq. In country villages in Burma, " on the night of the marriage a band of the young bachelors of the place come and shower stones and sticks on the roof of the house where the happy couple are " (Shway Yoe, The Burman, P- 58). 0 Jochelson, Yukaghir, p. 94. VOL. II K K 498 THE HISTORY OF HUMAN MARRIAGE chap. occasion when the bridegroom is painted with henna, the con- stant firing of guns, the loud music, and the peculiar noise called zghdnt made by women, especially the one who keeps hold of the bridegroom's hood, obviously serve the purpose of purifying the atmosphere and frightening away evil spirits by the noise, and, in the case of the powder-play, also by the smell of powder, of which the jnun (jinn) are believed to be much afraid.1 So also, when the bride is taken to her new home, guns are repeatedly fired off in front of the animal on which she rides, and there is again loud music and zghdnt,2 and the same is repeated on her arrival; on the latter occasion the purificatory or protective character of the firing of guns is particularly obvious when it is done so close to the bride that she is enwrapped in the smoke, or when a shot is fired inside the room which she is going to occupy.3 Gun-fire is a frequent practice at country wed- dings in Europe, and sometimes it is expressly said to drive away evil spirits.4 In rural parts of the County of Durham the bridal party is escorted to church by men armed with guns, which they fire again and again close to the ears of bride and bridesmaids; and at Guisborough in Cleveland guns are fired over the heads of the newly-married couple all the way from church.5 In Germany there is much shooting and noise-making on the night before the wedding (Pollerabend) and on the way to church.6 Among Slavonic 1 Westermarck, Marriage Ceremonies in Morocco, p. 122 sq. 2 Ibid. p. 188. 3 Ibid. p. 218. On the firing of guns, the loud music and singing, and the zghdrit of women at weddings in Morocco, see ibid. p. 322 nn. 10-13. 4 Feilberg, in Archiv f. Religionswiss. iv. 171-175, 274 sqq. Samter, Geburt, Hochzeit und Tod, p. 43 sqq. In Languedoc, when bride and bridegroom leave for church " on tire . . . force coups de pistolet dans le but eloigner les mauvais ge'nies " (Laisnel de la Salle, op. cit. ii. 36). 6 Henderson, op. cit. p. 38. Cf. Marie Balfour, County Folk- Lore, vol. iv. Northumberland, p. 92 ; Gregor, Notes on the Folk-Lore of the North-East of Scotland, p. 91 sq. 6 Sartori, op. cit. i. 71, 72, 84. Of the terrific shooting at a Swedish country wedding in former days a vivid description is given in Byskomakaren Jonas Stolts minnen frdn 1820-talet, p. 86. XXV MARRIAGE RITES 499 peoples the wedding guests make a terrific uproar out- side the bridal chamber while the marriage is being consummated.1 Other weapons are used for a similar purpose.2 At a wedding in ancient India a Brahman shot arrows into the air, saying, " I pierce the eye of the demons which are sneaking around this bride."3 Among the Oraons, "when the bride first leaves her parents' home for her husband's, her father puts an iron-headed arrow into her hands. The girl carries it to prevent the evil spirits of her parents' village from following her."4 Among the Manchus, before the bride is taken out of her sedan chair on arriving at the bridegroom's house, the latter fires three arrows at the blinds.5 Among the Bechuanas the bridegroom throws an arrow into the hut before he enters to take his bride.6 As we have seen before, however, the use of arrows at weddings may serve other than prophylactic purposes.7 In Morocco, again, it is the custom for the bridegroom to carry a sword, dagger, or pistol, and on the occasion when he is painted with henna swords are crossed over his head or in front of him to ward off jnun, who are afraid of steel and, especially, of weapons made of this metal. So also swords are sometimes crossed over the head of the bride at her henna-ceremony, and she, too, may carry a dagger. And in order to drive away evil spirits or other evil in- fluences, the bridegroom sends his sword in advance to be put on the bridal bed, or puts it there himself, or hangs it on the wall, or lays a pistol underneath the pillow.8 In other Muhammadan countries there are in the bridal 1 Piprek, op. cit. pp. 130, 131, 175. See supra, ii. 461. 2 Samter, Geburt, Hochzeit und Tod, p. 41 sqq. Crawley, op. cit. p. 323 sqq. Frazer, Folk-Lore in the Old Testament, i. 521 sq. 3 Winternitz, Das altindische Hochzeitsrituell, p. 60. 4 Sarat Chandra Roy, Oraons of Chota Nagpur, p. 363 sq. 5 Stewart Lockhart, in Folk-Lore, i. 487. 6 Conder, ' Present Condition of the Native Tribes in Bechuana- land,' in Jour. Anthr. Inst. xvi. 83. 7 Supra, ii. 490 sq. 8 Westermarck, Marriage Ceremonies in Morocco, pp. 123, 162, 255, 322 nn. 16-18. 500 THE HISTORY OF HUMAN MARRIAGE chap. procession two men with drawn swords escorting the bride ;l and from the ' Song of Solomon' it appears that the bridal procession of a Hebrew wedding also contained armed men, holding their swords upon their thighs " because of fear in the night."2 Among high-caste Hindus of the Punjab " the bridegroom always carries an iron weapon with him to drive away the evil spirits which haunt him, especially at the marriage ceremony."3 At Maratha marriages, while bride and bridegroom are standing in the baskets filled with unhusked wheat, the maternal uncles of each, or any other fit persons, stand behind them with naked swords in their hands.4 Among the Bhils of Central India the bride- groom touches the " marriage-shed " with a sword.5 In Bombay he keeps a dagger in his hand day and night, from the beginning to the end of the marriage rites, for the purpose of averting evil spirits.6 Among certain castes in the Central Provinces of India the bridegroom's mother passes in front of him, or touches him, with a pestle in order to keep off or drive away such spirits, either when he starts for his wedding or when the newly-married couple return to his house.7 At Foochow, in China, the black sedan in which the bride is carried to her parents on the third day after her marriage has painted upon the outside a charm consisting of a picture of a grim-looking man sitting on a tiger, with one of his hands raised up, holding a sword, as if in the act of striking. The object of this charm "is to 1 Mornand, La Vie arabe, p. 58 (Constantine, in Algeria). Van- Lennep, Bible Lands, p. 551 (Palestine). Lynch, Narrative of the United States' Expedition to the River Jordan and the Dead Sea, p. 448 (Jaffa). 2 Song of Solomon, iii. 6 sqq. Cf. Samter, Geburt, Hochzeit und Tod, P- 45- 3 Maya Das, in Panjab Notes and Queries, i. 98. See also Pandit Harikishan Kaul, Census of India, 1911, vol. xiv. (Punjab) Report, p. 274. 4 Shastri, in Panjab Notes and Queries, i. 99. 6 Kincaid, ' Bheel Tribes of the Vindhyan Range,' in Jour. Anthr. Inst. ix. 401. 6 Munshi, in Panjab Notes and Queries, i. 125. 7 Russell, Tribes and Castes of the Central Provinces of India, iv. 284 (Naodas), 407 (Rajjbars). xxv MARRIAGE RITES 501 keep off evil and unpropitious influences from her. It is said that, in former times, whenever a new bride in her chair passed by a certain place, evil spirits would invariably approach and injure her, causing her to be sick."1 In Armenia, where bride and bridegroom are believed to be par- ticularly exposed to evil influences both at and after the wedding ceremony,they carry as talismans a locked door-lock and a closed clasp-knife; they are, further, constantly attended by a man armed with a sword for their protection, and when- ever they pass through a door their guardian makes a cross with the sword over the lintel, since doorways are thought to be the abode of spirits.2 Or, according to another account, a sword is offered to the bridegroom, who, standing at the door of the house, lets the bride enter underneath it.3 In Zante the bridegroom, entering the church, stuck his dagger in the door as an antidote to enchantment, since it was a common practice to bewitch bridegrooms to make them impotent.4 In some parts of Bulgaria it was formerly the custom for the dever to guard the couple with a sword in his hand while the marriage was consummated.6 In various parts of Germany the bridesmen protect the bride with drawn swords.6 According to a Netherlandish belief, a bride who desires to have good luck and prosperity in her wedded state must, on coming out of the church, enter her house under two sabres laid crosswise over the door.7 In some parts of Esthonia two swords were stuck into the wall over the place where the bride and bridegroom sat, and it was believed that the one whose sword kept up the longest vibration would live longest.8 In France, in the seventeenth century, the couple had on the wedding day to 1 Doolittle, op. cit. i. 94 sq. 2 Abeghian, Der armenische Volksglaube, p. 91. 3 Bodenstedt, quoted by Lobel, Hochzeitsbrauche in dev Tiirkei, p. 118. 4 Sandys, op. cit. p. 7. 6 Piprek, op. cit. p. 149. 6 Meyer, Deutsche Volkskunde, p. 177. Sartori, op. cit. i. 83. Sepp, Volkerbrauch bei Hochzeit, Geburt und Tod, p. 56 sq. Bir- linger, Volksthumliches aus Schwaben, ii. 376. 7 Thorpe, Northern Mythology, iii. 330. 8 Boecler-Kreutzwald, op. cit. p. 35 sq. 502 THE HISTORY OF HUMAN MARRIAGE chap. pass under two drawn swords forming an Andrew's cross.1 In Normandy, when the bridegroom joined the bride in the marriage chamber, one of his friends cracked a whip in order to drive away the evil spirits who might otherwise molest the couple.2 In some tribes in Morocco a needle in the right slipper of the bride, or of the bridegroom as well, is used as a charm against evil spirits or witchcraft.3 Various other charms against jnun are worn by them,4 and salt, which is regarded as a powerful safeguard against those spirits, is in very frequent use at Moorish weddings.5 It is also used on such occasions in India,6 Germany,7 France,8 Scotland,9 Slavonic countries,10 and elsewhere,11 and is in some cases said to serve as a charm against witchcraft, evil spirits, or the evil eye. In Morocco incense is burnt, to dispel evil spirits, or " to please the spirits of the place."12 But the most impor- tant of all prophylactic or cathartic rites at Moorish weddings is the custom of painting the bride and bridegroom with henna, a colouring matter produced from the leaves of the Lawsonia inermis, or Egyptian privet, which is considered to contain much baraka, or benign virtue, and is therefore used as a means of purification or protection on occasions when people think they are exposed to supernatural dan- 1 Thiers, op. cit. iv. 470. 2 de Nore, op. cit. p. 240. 3 Westermarck, Marriage Ceremonies in Morocco, pp. 150, 162, 237, 256, 290. 4 Ibid. pp. 104, 105, 123, 147, 162, 163, 290. 5 Ibid. pp. 123, 162, 187, 256, 322 n. 14. 0 Crooke, Popular Religion and Folk-Lore of Northern India, ii. 23. 7 Sartori, ' Her Schuh im Volksglauben,' in Zeitschr. d. Vereins f. Volkskunde, iv. 172. Schell, ' Das Saiz im Volksglauben,' ibid. xv. 146. Schonwerth, Aus der Oberpfalz, i. 76. 8 Thiers, op. cit. iv. 447. Tuchmann, ' La fascination,' in Melusine, vii. 234. 9 Simpkins, County Folk-Lore, vol. vii. Fife, p. 163. Tuchmann, loc. cit. p. 235. 10 Piprek, op. cit. pp. 35, 86. 11 Tremearne, Hausa Superstitions and Customs, p. 84. 12 Westermarck, Marriage Ceremonies in Morocco, pp. 235, 237, 244. 255, 305. XXV MARRIAGE RITES 503 gers, as, for example, at religious feasts.1 The henna is applied to the bride's hands and feet, occasionally also to her legs below the knees, her arms, face, and hair; and she is thus painted either privately or publicly, sometimes on two festive occasions, but in most tribes in the presence of women and children only. The bridegroom, again, has the henna applied sometimes to the palm or fingers or little finger of his right hand, sometimes to both hands, and sometimes to his feet as well; the act is mostly performed before the arrival of the bride, in some cases privately but more often publicly, and may be repeated more than once.2 These rites are extremely prevalent in the Muhammadan world, particularly the custom of painting the bride.3 It is also a widespread custom among Muhammadans that the bridegroom has a bath before meeting the bride ; and the bride also is purified by bathing or water-pouring, to which, in Morocco, much importance is attached by towns- folk.4 At 'Fez, five days before *the wedding, she goes in the afternoon to the hot bath, accompanied by some women of her family, one of whom carries a candlestick with a wax candle. When they enter the bath the candle is lighted, and the women make zghdrit-obviously to ward off jnun, who are generally supposed to be haunting the hot baths. The same ceremony is again repeated on the following two days. On the latter of these occasions, that is, three days before the wedding, seven buckets of lukewarm water are poured over the girl by seven women, " so that she shall have no quarrel with her husband," in accordance with the saying, L-md amdn u sh-sharr ma kan, " The water is safety and quarrel there is none." This ceremony is called taqbib, which means " pouring." According to Arabic writers, 1 Westermarck, ' Popular Ritual of the Great Feast in Morocco,' in Folk-Lore, xxii. 132 sqq. 2 Idem, Marriage Ceremonies in Morocco, ch. iii. sq. ; see particu- larly pp. 118, 119, 160 sq. 3 For their prevalence see my Marriage Ceremonies in Morocco, pp. 118 n. 1, 160 n. 1. 4 For these customs in Morocco see my Marriage Ceremonies in Morocco, pp. 120, 161, 162, 321 n. 9, and among Muhammadans elsewhere ibid. pp. 120 n. 2, 161 n. 3. 504 THE HISTORY OF HUMAN MARRIAGE chap. " it is a sunneh ordinance that the bride wash her feet in a clean vessel, and sprinkle the water in the comers of the chamber, that a blessing may result from this."1 At Tunis, when the bride has arrived at the bridegroom's place, a woman sometimes washes her feet; " c'est un rite de purification au moment ou elle p&i&tre dans un nouveau foyer."2 Among the Bedouins of Sinai the girl is on her betrothal sprinkled with the blood of a sheep sacrificed for the occasion, and remains for three days in the tent erected for her in front of her father's habitation ; at the end of this period " she is conducted by a procession of women to ' a spring of living water ' (that is, a perennial spring), and after performing her ablutions is led home to the house of her husband."3 In his description of the marriage customs of the Jews, particularly those of Morocco, Addison says that the bride bathes daily for eight days before the wedding.4 Among the ancient and modern Hindus, as also among other Indo-European peoples, it has been, or still is, con- sidered an essential preparation for a wedding that the bride, or frequently the bridegroom also, should have a bath.6 In the Atharva-Veda there are prayers referring to the bath of the bride, and priests are requested to fetch the water for the bath, that it may be auspicious for the welfare of the future husband.8 Among the Parsees bride and bridegroom take a sacred bath in the morning and afternoon of the marriage day, and formerly it was the custom for their feet to be washed after the marriage ceremony.7 In Persia, when the bridegroom has entered the bridal chamber, " a 1 Lane, Arabian Society in the Middle Ages, p. 234. 2 Bertholon and Chantre, Recherches anthropologiques dans la Berbdrie orientate, i. 579. 3 Palmer, Desert of the Exodus, p. 90. 4 Addison, op. cit. p. 46. 8 Colebrooke, Miscellaneous Essays, i. 208. Haas, loc. cit. pp. 198, 278, 294 sqq., 304, 381, 382, 411. Wintemitz, Das altindische Hoch- zeitsrituell, p. 43 sqq. Idem, in Transactions, p. 274 sq. Grierson, op. cit. p. 370 (Hindus of Bihar). Padfield, op. cit. p. 103 (Hindus of South India), de Gubernatis, op. cit. p. 139 sqq. (peoples in Europe and India). 6 Winternitz, in Transactions, p. 274 sq. 1 Jivanji Jamshedji Modi, ' Marriage (Zoroastrian),' in Hastings, op. cit. viii. 455 sq. xxv MARRIAGE RITES 505 basin and ewer and water . . . are brought, and the right leg of the bride and the left leg of the bridegroom are placed together and washed, and their hands also in the same manner."1 In ancient Greece, on the wedding day, bride and bridegroom bathed in water drawn from a particular fountain of running water, which at Athens was the fountain Callirrhoe ;2 and in modem Greece the bride's bath still forms part of the nuptials.8 In Rumania the bride must bathe in running water, although in the winter the water may be brought into the house from the nearest brook.4 In Sweden it was formerly the custom for the bride to have a bath before she was dressed.5 Among several Slavonic peoples, especially the Great Russians, the bride has a bath on the day before the wedding, and among the Serbs flowers are thrown into the water.6 Among the old Prus- sians the bride's feet were washed in the bridegroom's house after she had been taken round the fire, and the water was afterwards sprinkled over the guests, the bridal bed, the cattle, and the whole house.7 In Cromarty the bride has her feet washed on the day before the wedding, money is put into the water, and when she sits with her feet in the basin or tub, her girl friends scramble for the coin.8 In the north-east of Scotland, on the evening before the marriage, there was the ceremony of " feet-washing " : a few of the bridegroom's most intimate friends assembled at his house, a large tub was brought forward and nearly filled with water, and the groom was stripped of shoes and stockings and his feet and legs were plunged in the water.9 The same custom prevailed in Northumberland, and there the bride, too, had her feet washed though in a more private way.10 1 Customs and Manners of the Women of Persia, trans, by Atkinson, p. 43- 2 Becker-Goll, Charikles, iii. 364. Thucydides, Historia belli Peloponnesiaci, ii. 15. 3 Sakellarios, op. cit. p. 13. 4 Flachs, op. cit. p. 30. 5 Norlind, Gamla brbllopsseder hos svenska allmogen, p. 68. 6 Piprek, op. cit. pp. 7, 178 sq. ' 7 Hartknoch, Alt- und Neues Preussen, p. 179 sq. 8 Ashton-Rigby, ' Marriage Customs in Cromarty,' in Folk-Lore, xxvii. 433. 9 Gregor, op. cit. p. 89 sq. 10 Balfour, op. cit. p. 92. 506 THE HISTORY OF HUMAN MARRIAGE chap. Marriage rites of a similar kind are also found among some backward races. In South Celebes the bride and bridegroom bathe in holy water, and the bride is besides fumigated.1 In some tribes of Northern Rhodesia the bride is carried to a stream on the shoulders of an old woman and is there im- mersed, and the bridegroom is also bathed at the stream.2 Among the Edo-speaking peoples of Nigeria, on the arrival of the bride at her new home, the bridegroom brings water in a basin, a member of the bride's family washes her hands, and the bridegroom wipes them with a towel.3 In Egbaland, a division of Yorubaland in Southern Nigeria, before the bride enters the bridegroom's home, some of his relatives wash her feet in gin and water, " to signify that she enters he husband's house purified."4 There may, no doubt, be other than superstitious reasons for the bathing or washing of bride and bridegroom, but the ceremonial character of the act cer- tainly suggests a purificatory object. And so do other water ceremonies so frequently connected with weddings. In a Berber tribe in the Great Atlas, when the bride is taken to her new home, the procession first goes to a river, which the bride crosses on her mule three times to and fro, while the others are waiting on the bank.5 It is a frequent custom in Morocco that when she arrives at the bridegroom's place purifying substances, like milk, water, and henna, are offered her or sprinkled on her; and although the sprinkling of water sometimes is said to bring about rain, I have no doubt that the primary object of the rite is purification.6 In Andjra the bridegroom, on entering the room where the bride is waiting for him, takes a new bottle filled with water and a new bowl, brought there among other articles from the bride's house, goes to the four comers 1 Matthes, op. cit. p. 21. 2 Gouldsbury and Sheane, Great Plateau of Northern Rhodesia, p. 160 sq. 3 Thomas, Anthropological Report on the Edo-speaking Peoples of Nigeria, i. 48. 4 Partridge, ' Native Law and Custom in Egbaland,' in Jour. African Soc. x. 426. 5 Westermarck, Marriage Ceremonies in Morocco, p. 185. 6 Ibid. pp. 215, 216, 217, 326 nn. Ti-13, 347 sq. xxv MARRIAGE RITES 507 of the room, and sprinkles a little water in each corner, takes hold of the bride's hands, first one and then the other, and dips the tips of her fingers into the water which he has poured out into the bowl. He drinks from the bowl and then holds it for the bride to drink from ; and touching her temples with the thumb and middle finger of his right hand he recites the " s&ra of the jinn." In the Hiaina, in the same country, on the fortieth day after the bride's arrival at her new home, her sister-in-law, or some other unmarried girl who lives in the house, .sprinkles water over her and puts on her a pair of new slippers and a new girdle instead of the old ones, which she has been wearing since her wedding ; and I was told that the object of this ceremony is to remove her bas, or evil. The Shi'ahs have a tradition that the Prophet, before he gave his daughter in marriage to 'Ali, commanded her to fetch water and then sprinkled both her and 'Ali with it, invoking God to protect them and their offspring against the devil.1 In ancient Rome it was the custom to receive a bride " with water and fire "-aqua et igni accijere ;2 and water ceremonies of some sort or other have been, or still are, practised at weddings in many other European countries,3 as also in ancient4 and modern5 India. In Epirus two children, whose parents are alive, sprinkle bride and bridegroom with water when they enter their new home, apparently- we are told-as a protection against witchcraft.6 In Albania, when the bridegroom comes to fetch the bride, 1 Goldziher, ' Wasser als Damonen abwehrendes Mittel,' in Archiv f. Religionswiss. xiii. 31 sq. 2 Rossbach, op. cit. pp. 351, 361 sqq. Samter, F amilienfeste der Griechen zind Romer, p. 14 sqq. 3 v. Schroeder, op. cit. p. 133 sqq. Piprek, op. cit. p. 171 sq. (Slavs). 4 Haas, loc. cit. pp. 341, 342, 358, 365, 367, 373. 5 Mary Billington, op. cit. p. 74 (Brahmans of Eastern Bengal). Hayavadana Rao, in Anthropos, v. 794 ; Hislop, Papers relating to the Aboriginal Tribes of the Central Provinces, p. 9 (Gonds). Crooke, ' Hill Tribes of the Central Indian Hills,' in Jour. Anthr. Inst, xxviii. 245. Butler, Travels and Adventures in the Province of Assam, p. 83 (Kukis). 6 Sakellarios, op. cit. p. 21. 508 THE HISTORY OF HUMAN MARRIAGE chap. her mother meets him at the door holding in the hand a basin containing clean water, with which she sprinkles the bridegroom.1 Among the Little Russians the bridal pair are frequently sprinkled with water,2 and the same is the case among the Transylvanian Gypsies.3 Among the Southern Slavs a vessel containing water is often offered to the bride on her arrival at the bridegroom's house.4 In the Himalaya of the North-Western Provinces the bridegroom asperses his bride with water.5 In the Magh tribes of Chittagong an old man sprinkles the couple five or seven times with water, while the groomsman links the right-hand little finger of the bridegroom with the corresponding left-hand finger of the bride.6 Among the Tulu people of South Canara " the essen- tial element of the marriage ceremony is dhdre, or pouring water over the joined hands of the bride."7 Among the Sinhalese a priest pours water upon the bride and bride- groom after they have been joined together with a long piece of cloth,8 or throws a little water over their right thumbs, which have been put together.9 In Tibet a lama sprinkles the bridal pair with holy water.10 In Siam " the couple kneel down to be bathed with holy water. The chief elder pours it first over the head of the bridegroom, and then over the head of the bride, at the same time pro- nouncing a blessing upon them both."11 Among the Aetas of the Philippines the father of the bride throws a cocoanutshell full of water over the pair, who have been made to kneel down; he then bumps their heads together, and they are adjudged man and wife.12 Through- out Madagascar, at the moment the bride is going to leave her old home, her parents bless her, either by sprinkling the 1 v. Hahn, op. cit. p. 145. 2 Piprek, op. cit. p. 47. 3 Wlislocki, Vom wandernden Zigeunervolke, p. 189. 4 Piprek, op. cit. p. 171. 5 Atkinson, in Jour. Asiatic Soc. Bengal, vol. liii. pt. i. 98. 8 Hutchinson, op. cit. p. 119. 7 Sturrock, Madras District Manuals : South Canara, i. 143. 8 Percival, Account of the Island of Ceylon, p. 180. 9 Joinville, * Religion and Manners of the People of Ceylon,' in Asiatick Researches, vii. 427. lc Ahmad Shah, op. cit. p. 57. 11 Young, op. cit. p. 95. 72 Reyes Lala, op. cit. p. 96. XXV MARRIAGE RITES 509 tips of her fingers with some drops of water, or, as was the custom in ancient times, by projecting on her some drops of saliva or of the blood of an animal sacrificed for this purpose.1 Among the Matabele, on her arrival at the bridegroom's house, " the bride, who has brought a calabash filled with water, at the bottom of which are strings of beads, pours some of the contents over the bridegroom, and sprinkles his people and his friends with the remainder. She then puts the beads on her head, and placing the calabash on the ground in front of her husband she crushes it with her foot."2 Among the New Caledonians of the tribe Belep " les epoux qui se marient pour la premiere fois doivent se soumettre i une cer^monie le matin des deux jours qui suivent leur union, soit legitime, soit illegitime ; sans cela, ils se croiraient exposes a certaines maladies. La personne chargee de cette purifica- tion a prepare un petit paquet oblong qui renferme la sciure d'un bois special et une eau determinee ; apres avoir place successivement 1'une des extremites du paquet dans la bouche des 6poux, elle arrose 1'autre extremite avec 1'eau prescrite. La c^remonie accomplie, on enterre prds de la case des 6poux les objets qui ont servi & cette purification."3 Besides water, fire is a frequent means of dispelling evil influences at a wedding. In Morocco burning candles play a prominent part on such occasions, and partly, no doubt, for superstitious reasons ;4 for the jnun love darkness and are terrified by light.5 At Fez, in former days, the bride was taken to her new home in a wooden cage " with a great noise of trumpets, pipes, and drums, and with a number of torches."6 In modern Palestine lanterns and torches are held aloft in front and along the flank of the bridal procession ;7 and among the ancient Jews, when the 1 Grandidier, op. cit. ii. 184 sq. 2 Decle, op. cit. p. 158 sq. 3 Lambert, Moeurs et Superstitions des Ndo-Caledoniens, p. 99 sq. 4 Westermarck, Marriage Ceremonies in Morocco, pp. 121, 122, 162, 187, 322 n. 8. 6 Idem, ' Nature of the Arab Ginn, illustrated by the Present Beliefs of the People of Morocco,' in Jour. Anthr. Inst. xxix. 253. Idem, Belief in Spirits in Morocco (Acta Academiae Aboensis, Humaniora, i. no. 1), p. 49. 6 Leo Africanus, op. cit. ii. 450. 7 Van-Lennep, op. cit. p. 551. 510 THE HISTORY OF HUMAN MARRIAGE chap. bridegroom, attended by a company of young men,1 brought home his bride from her father's house, the young female friends of both parties joined the train, lighting their torches.2 In ancient Greece and Rome the bride was always taken to her new home with torches,3 and in Rome one of these torches was made of whitethorn,4 which was believed to keep away evil influences.5 In modem Greece the bridal procession is not only headed by torch-bearers, but the bride and bridegroom themselves carry torches.6 Brand thinks it doubtful whether the custom of carrying torches in the bridal procession ever prevailed in England, although there are indications that it did ;7 but among the Scandi- navian peoples torches have played a prominent part at their weddings up to recent times, and, as Troels-Lund observes, their object was no doubt to keep away the powers of darkness.8 In Sk&ne, in the south of Sweden, on the arrival of the bride at her new home, the bridegroom's mother met her holding in the hand a burning candle, and the bride had to touch the flame before she was allowed to enter.9 At Clenze in the Hanoverian Wendland the bride was received at the door of her new home by a woman carrying four candles and was then accompanied to all the corners of the house.10 In the government of Kharkov the bridal pair are received with torches, or pitch and tar are lighted on their arrival.11 At Hindu weddings lights and other objects are waved round the heads of the bride and bridegroom as a protection against evil spirits.12 Among the Javanese of Surakarta, 1 Judges, xiv. n. 2 St. Matthew, xxv. i sqq. 3 Sticotti, ' Zu griechischen Hochzeitsgebrauchen,' in Festschrift fiir Otto Benndorf, p. 182. Rossbach, op. cit. p. 335 sqq. Samter, Geburt, Hochzeit und Tod, p. 72 sqq. 4 Festus, op. cit. p. 245 a. 6 Samter, op. cit. p. 73 sq. 0 Wachsmuth, op. cit. p. 93. 7 Brand, op. cit. p. 391. 8 Hylt^n-Cavallius, Wdrend och Wirdarne, p. 435 sqq. Troels- Lund, op. cit. xi. 23 sqq. 8 Nicolovius (Lov6n), Folklifueti Skytts Hdrad i Skane, p. 14657. 10 Tetzner, ' Die Drahwener im hannoverschen Wendlande um das Jahr 1700,' in Globus, Ixxxi. 271. 11 Piprek, op. cit. p. 41. 12 Crooke, Popular Religion and Folk-Lore of Northern India, ii. 24. XXV MARRIAGE RITES 511 i who believe that newly-wedded pairs are exposed on the first night of their marriage to the injuries and outrages of evil spirits, it is the custom that when the young couple have been in the bridal chamber for about an hour, their friends enter with burning torches and poke about the room with them, as if they were looking for something ; this they do to frighten away the demons by the glare of the torches, or at least to disturb and thwart them in their hellish design to turn the couple's love to hate.1 The wooden vessel which at Foochow, near Canton, is placed on the centre of the bridal bedstead contains among other things a glass lamp filled with oil and two candles, which are lighted and, together with the vessel, are left untouched on the bedstead until they have burned out; " the light of the lamp and of the candles, although in broad daylight, is re- garded as peculiarly efficacious in keeping away evil spirits."2 But fire is used at weddings as a means of dispelling evil influences not only on account of its light but because it burns. In Swatow, in China, when the bride arrives at the bridegroom's home, she steps over a flare-up fire on the ground, made by burning a few wisps of dry grass, to be purified from " the contamination of any devils or other dangers that she may have come across on the road " ; 3 whilst-at Canton she is placed on the back of a female servant and carried over a slow charcoal fire,4 and at Peking the chair in which she is carried to the bridegroom's place is there lifted over a vessel containing glowing charcoal for the purpose of expelling evil influences.5 The Manchus, who have the same custom, say that the bride has to pass over a charcoal pan " as a sign that the happy pair will be as brilliant as fire " ;6 but this explanation is evidently an afterthought. Among the White Russians, before a wedding, straw is burned inside the houses of both bride and 1 Winter, ' Instellingen, gewoonten en gebruiken der Javanen te Soerakarta,' in Tijdschrift voor NeMands Indie, 1843, vol. i. 4$5- Frazer, Folk-Lore in the Old Testament, i. 521. 2 Doolittle, op. cit. i. 76 sq. See also ibid. i. 90 sq. 3 Ball, Things Chinese, p. 423. 4 Gray, China, i. 205. 6 Grube, quoted by Samter, Geburt, Hochzeit und Tod, p. 27 sq. 0 Stewart Lockhart, in Folk-Lore, i. 487. 512 THE HISTORY OF HUMAN MARRIAGE chap. bridegroom to drive away evil spirits ; the bridegroom, when fetching the bride from her home, must ride or drive over a burning fire, and so also the bride, when arriving at the house of her parents-in-law, must pass a fire, into which she throws coins.1 In North Germany, when a bridal pair are going to church, it is the custom, before they leave the house, to throw a firebrand on the threshold over which they must pass.2 In many parts of Europe and in India the bride, on her arrival at her new home is, or formerly was, taken three times round the fire of the hearth.3 Among the Vedic people the joining of hands was followed by the ceremony of leading the bride round the fire, and the same custom is practised all over India to the present day. In the Grihyasutras we read that the bridegroom shall lead the bride round the fire so that their right sides are turned to it, that is, from left to right, and the ceremony is to be repeated three times.4 In ancient Rome, after the ceremony of the dextrarum junctio, a sacrifice was made, and bride and bridegroom walked round the sacrificial altar, likewise from left to right. Among the Croats the bridesman leads the bride three times round the hearth, on which a fire is burning, and each time the bride bows before it.5 In many parts of Germany the bride is led three times round the fire by the bridegroom or his mother.6 1 Kupczanko, ' Hochzeitsgebrauche der Weissrussen,' in Am Urquell, ii. 138. In some parts of Russia the bridal pair have to pass a burning fire before they leave for the bridegroom's home, generally at the door of the bride's house; and outside the bridegroom's house they also find some burning straw which they have to pass over before they can enter (Piprek, op. cit. p. 171). 2 Kuhn and Schwartz, Norddeutsche Sagen, Marchen und Gebrauche, P- 434- . 3 v. Schroeder, op. cit. p. 127 sqq. Samter, Familienfeste der Griechen und Romer, p. 20 sqq. Haas, loc. cit. pp. 316 sqq., 396. Winternitz, Das altindische Hochzeitsrituell, p. 56 sqq. Idem, in Trans- actions, p. 283 sq. Crooke, Popular Religion and Folk-Lore of Northern India, i. 11; Padfield, op. cit. p. 108 sq. (modern Hindus). 4 Grihya-Sutras, i. 168, 283, 384 ; ii. 46, 191, 260 sq. 5 Krauss, op. cit. p. 386. Rajacsich, op. cit. p. 146. G Sartori, op. cit. i. 115. Kuhn and Schwartz, op. cit. p. 433. Schell, ' Bergische Hochzeitsgebrauche,' in Zeitschr. d. Vereins f. Volkskunde, x. 165 sq. Tetzner, in Globus, Ixxxi. 271. xxv MARRIAGE RITES 513 But in Westphalia she is simply led to the hearth and the tongs are put into her hands to make up the fire j1 and the same custom is found in the north-east of Scotland.2 Among the Aith Sadden, near Fez, it was formerly, when they lived in tents, the custom for the bridegroom's mother to take the bride to the fireplace ; and among a neighbouring Berber tribe, the Ait Warain, she not only takes the bride to the hearth but washes her right foot and hand over one of the fire-stones, in order, I was told, that she shall be as permanent in the house as these stones, which the Ait Warain, unlike many other Berbers, never change. Among the Merina in Madagascar, when the bridal procession has arrived at the bridegroom's place, it goes three times round the little wall surrounding his house, then three times round the house itself, and at last three times round the fireplace ; this, we are informed, is done to strengthen the ties which attach the young wife to her new home and to prevent her leaving it.3 Among the Koryak, again, " when the bride approaches the house of her bridegroom's parents, the latter come out with fire-brands taken from the hearth to meet her. This re- ception," Dr. Jochelson adds, " symbolises the acceptance of the bride into the family cult which the hearth repre- sents."4 So also the Indo-European rites have been inter- preted as aggregation rites. Yet I doubt that these ex- planations, whether given by the peoples practising the rites or by the writers describing them, disclose their original meaning, at least in full. They seem too similar to other fire rites, which obviously serve a purificatory purpose, to be dissociated from them.0 Among the Koryak, when the newly-married couple go to visit the wife s parents, they are also met with fire-brands from the hearth.6 At Brackel, near Dortmund in Germany, while the bride is taken 1 Winternitz, in Transactions, p. 284. 2 Gregor, op. cit. p. 93. 3 Grandidier, op. cit. ii. 183. 1 Jochelson, Koryak, p. 743- 5 Of the Vedic rite Oldenberg (Die Religion des Veda, p. 463) says that it was looked upon as a veneration of Agni, the god of Ure. But he adds that it originally seems to have been a magical lustration rite. " Jochelson, op. cit. p. 744. VOL. II L L 514 THE HISTORY OF HUMAN MARRIAGE chap. round the hearth, fire is thrown after her.1 At Bockum, near Kaiserswerth, she is placed on a chair, and live coals are shovelled underneath it.2 And in some cases the circumambulation of the hearth already takes place in the bride's old home.3 How could it in such cases be regarded as an aggregation rite ? The circumambulation does not always take place round the hearth. Among some Southern Slavs the bride is taken three times round the church before she is allowed to enter it.4 In the Isle of Man, according to Waldron (who wrote in 1726), when the bridal company arrived at the church-yard, they walked three times round the church before entering ;5 and in Perthshire, according to Sinclair (who wrote at the end of the same century), they walked round the church after leaving it, keeping the church walls upon the right hand.6 In Warend, in Sweden, it was the custom for the bridal procession, before entering the church, to walk three times round a certain stone outside it, which was therefore called the " bridal stone."7 In modern Greece bride and bridegroom are led three times round the altar.8 Dr. Wintemitz has no doubt that the circum- ambulation of the church is a survival of an older custom of leading the bride round the sacrificial fire ;9 and the great antiquity of this rite together with its prevalence among so many Indo-European peoples certainly speaks in favour of his suggestion. In any case the circumambulation of a holy place, like a church or something connected with it, decidedly has the appearance of a cathartic or prophylactic rite. Among the Poles, again, the bridal pair walk three times round the table in the bridegroom's house,10 and 1 Woeste, in Jahrbuch d. Vereins f. niederdeutsche Sprachforschung, 1877, p. 139. 2 Schell, 'Nachtrage zu den " Bergischen Hochzeitsgebrauchen," ' in Zeitschr. d. Vereins f. Volkskunde, x. 430. 3 v. Schroeder, op. cit. p. 127 sqq. 4 Piprek, op. cit. p. 128. 5 Waldron, Description of the Isle of Man, p. 60. 8 Sinclair, quoted by Brand, op. cit. p. 381. 7 Hylt^n-Cavallius, op. cit. ii. 437. 8 Wachsmuth, op. cit. p. 91. 9 Wintemitz, in Transactions, p. 284. 1,1 Piprek, op. cit. p. 82 sq. XXV MARRIAGE RITES 515 among the Great Russians the bride is taken round it.1 But among the last-mentioned people the bridegroom is also taken three times round it before he mounts the carriage by which he is going to fetch the bride ;2 and among the Little Russians both bride and bridegroom, before leaving for the house of the latter, are taken three times round a bread-trough covered with a blanket with bread and salt on the top of it.3 Circumambulation may by itself be regarded as a safe- guard against evil influences, apart from the purifying nature of the place or object round which it is performed. It may serve as a protection for the inhabitants of the place by allowing such influences to evaporate from a dangerous person, or dangerous persons, who are about to enter it, and it may also be a safeguard for the latter by neutralising the danger of entering a strange place.4 In Morocco the bride, on her arrival at the bridegroom's village, is in some cases taken three or seven times (from right to left) round the mosque of the place,5 but in other cases round the whole village6 or, more often, round the bridegroom's house or tent.7 Among the Aith Yusi, a Berber tribe near Fez, for example, the bride is not- allowed to enter the village until she has been taken three times round it; but if, as is often the case, there are several brides whose weddings are celebrated■ on the same occasion, they must wait for each other so as to, perform this ceremony all together, and if any of them is from that village itself, she also comes out and joins them, riding on a mare. In case there is in the village a woman, who was married within a month previously, she leaves the place before the brides are taken round, since it is believed that otherwise all the evil would fall on her. This obviously shows that the rite is supposed to have the effect of ridding the brides of the evil influences which they are carrying with them ; and 1 Ibid. p. 170. 2 Ibid. p. 170. 3 Ibid. p. 170. 4 Cf. Hamalainen, Mordvalaisten, fteremissien ja votjakkien kosinta- ja haatavoista, p. 271. 6 Westermarck, Marriage Ceremonies in Morocco, pp. 199, 200, 203, 208, 215. fl Ibid. pp. 196-198, 200, 206, 209, 215. 7 Ibid. pp. 203, 215. L L 2 516 THE HISTORY OF HUMAN MARRIAGE chap. the same is suggested by the fact that the circumambulation both there and elsewhere in Morocco is connected with other purificatory rites. Among the Bedouins of Mount Sinai the bride, mounted on a camel, is taken three times round the bridegroom's tent, while her companions utter loud exclamations.1 Professor Seligman informs me that among the Kababish, an Arabic-speaking nomad tribe on the Kordofan-Darfur border, the slave or slaves who bring the bride to the tent where she is going to meet the bridegroom carry her three times round the outside of it.2 Among the Touareg of the Ahaggar " les hommes amenent le marie a la tente. On lui en fait fair le tour trois fois."3 Among the Mundas of Chota Nagpur, when the bride and bride- groom arrive at the latter's house, they are carried three times round the courtyard, the bridegroom from the right to the left and the bride from the left to the right.4 Among the tundra Yukaghir,on reaching the tent of the bridegroom's parents, the train of sledges with the bridal pair makes three rounds about the t(jnt.5 So also among the Siberian Tartars the bridal company drive three times round the bridegroom's tent ; whilst among the Samoyed, Chuvash, and Cheremiss a similar ceremony is performed round the bride's as well as the bridegroom's dwelling.6 In some parts of Denmark, when the bridal company come from church to the place where the wedding is held, the bridegroom rides three times round it before he dismounts, or, in the winter, drives his sleigh, in which the bride is sitting with him, three times round the place.7 Among the Kilmeni tribe of High Albania the bride is led three times round the bridegroom's house, and subsequently round the hearth.8 Among the Berbers of the Aith Sadden and Aith Yiisi the bride, after her tour round the bridegroom's tent or 1 Burckhardt, Notes on the Bedouins and Wahdbys, p. 150. 2 This fact is mentioned in Professor Seligman's essay ' The Kababish, a Sudan Arab Tribe,' in Harvard African Studies (ii. 134), which has been published since the above was written. 3 Benhazera, Six mois chez les Touareg du Ahaggar, p. 16. 4 Sarat Chandra Roy, Mundas, p. 454. 5 Jochelson, Yukaghir, p. 94. 0 Hamalainen, op. cit. p. 270 sq. 7 Kristensen, op. cit. iv. 69. 8 Miss Durham, High Albania, p. 86. XXV MARRIAGE RITES 517 village or the mosque in it, beats the tent three times with a cane, as I was told, in order that the evil shall go away from it, or to remove any evil which may be in the bride- groom's family and to expel death from the domestic ani- mals ; it would be very unpleasant for the young wife if a child or animal should die shortly after her marriage, as its death would naturally be associated with her presence. But in Morocco bride and bridegroom are also themselves beaten or tapped for purificatory purposes.1 When the bridegroom gets up after having been painted with henna, some men who are carrying swords unsheath them to beat him,2 or the bachelors who have been standing round him try to beat him gently with their hands or with sticks, while his best-man defends him, beating the aggressors in return ; and I was expressly told that this ceremony is calculated to drive away evil influences.3 In Cairo, immediately before the bridegroom enters his house in order to meet the bride, " his friends leave him, but at parting strike him many times with their hands upon his back ; these blows he endeavours to avoid by running in as fast as possible."4 In ancient India the bridegroom was chaffed or beaten.5 In some parts of Germany he is beaten by the wedding-guests, especially the unmarried ones.6 Among the White Russians, when the bridal pair have gone to bed and covered themselves up, the bride- groom's best-man beats him three times with his whip, saying, " Look at each other, kiss, and embrace fast ! "7 A very similar ceremony is found among the lyca Indians of Colombia : the man who accompanies the pair to the hut in which the marriage is to be consummated says to the bridegroom, " Take the woman " ; and then enforces his command by beating him with a smell whip, similar to that which, for purificatory purposes, is used at funerals.8 The bride, also, may be ceremonially beaten, and not only 1 Westermarck, Marriage Ceremonies in Morocco, pp. 120, 121, 162,256-258,32311.7. 2 Ibid. p. 104. 3 Ibid. p. 107 sq. 4 Burckhardt, A rabic Proverbs, p. 116. 5 Haas, loc. cit. pp. 300, 301, 411. 0 Sartori, op. cit. i. 88 sq. Woeste, loc. cit. p. 137. 7 Piprek, op. cit. p. 64. 8 Bolinder, op. cit. pp. 252, 254, 255, 258 sq. 518 THE HISTORY OF HUMAN MARRIAGE chap. by the bridegroom, as in some cases mentioned above. Thus at Amzmuz, in the Atlas, the bride's brother, after he has placed a silver coin in one of his sister's slippers and then put them on her feet, taps her three times with his own slipper J1 the original object of this ceremony was presumably to drive away evil influences. In Denmark and Sweden it was formerly the custom that all the witnesses or guests at the ceremony of " hand-fasting," when the father of the bride solemnly joined the hands of the couple, afterwards tried to thump both bride and bridegroom on the back.2 In France, in the seventeenth century, they were ceremonially struck, with the hand or with a stick, in church after the nuptial ceremony, as Thiers says, " dans la pensee que ces outrages leur produiront quelque bien."3 Mannhardt, who gives many instances of the custom of beating a bridegroom or bride, suggests that its object is to expel evil spirits which might otherwise prevent fecundity.4 But there is no reason to think that the rite is intended to expel evil influences merely for the purpose of making the union fruitful. It should, moreover, be remembered that the ceremonial beating of the bride by the bridegroom and of the bridegroom by the bride is, nowadays at least, connected with the idea of gaining power ;5 whilst the beating of the bridegroom by his bachelor friends may be a ceremonial punishment inflicted on him because he is deserting their class.® Besides marriage rites which are intended to expel evil spirits or other evil influences, there are rites that are intended to safeguard bride or bridegroom by deception. Disguises at marriages are of frequent occurrence,7 and * Westermirck, Marriage Ceremonies in Morocco, p. 157. 2 Olaus Magnus, Historia de gentibus septentrionalibus, xiv. 9, p. 48c. Troels-Lund, op. cit. ix. 151 sq. 3 Thiers, op. cit. iv. 465. 4 Mannhardt, Wald- und Feldkulte, i. 299 sqq. 5 Supra, ii. 49 r, 492, 494. 0 See infra, ii. 585. 7 v. Schroeder, op. cit. p. 68 sqq. Crawley, op. cit. p. 371 sq. Farnell, ' Sociological Hypotheses concerning the Position of Women in Ancient Religion,' in Archiv f. Religionswiss. vii. 75, 89 sq. Frazer, Totemism and Exogamy, iv. 255 sq. Fehrle, Die kultische Keuschheit iin Altertum, p. 92. Samter, Geburt, Hochzeit und Tod, p. 91 sqq. XXV MARRIAGE RITES 519 many writers have suggested that their object is to deceive malignant spirits who lie in wait for the young couple.1 Thus, among some sections of the Brahmans of South India, especially the Tamil sections, the bride is on the fourth day dressed up as a boy, and another girl is dressed up to represent the bride. They are taken in procession through the street, and, on returning, the pseudo-bridegroom is made to speak to the real bridegroom in somewhat insolent tones, and some mock play is indulged in. The real bride- groom is addressed as if he were the groom or clerk of the pseudo-bridegroom, and is sometimes treated as a thief, and judgment is passed on him by the latter. It is also said that on the " shaving day " six months after marriage, in cases where the Brahman bridegroom is a young boy, he is dressed up as a girl, and the bride's party, when they detect the fraud, jeer at him and his relations for having deceived them.2 In, ancient Cos, according to Plutarch, the bridegroom was dressed in women's clothes when he received his bride ;3 whilst in Sparta, after the bride had been carried off by her husband, " the bridesmaid received her, cut her hair close to her head, dressed her in a man's cloak and shoes, and placed her upon a couch in a dark chamber," where she had to wait for the entrance of the bridegroom* Among the Egyptian Jews in the Middle Ages the bridegroom donned feminine attire, whilst the bride wore a helmet and, sword in hand, led the procession and the dance.5 Among the Swahili, when the guests have arrived at the bride- groom's father's house, the young man is dressed in two fine female garments and is then seated on the bridal bed, which has been placed in the room where the guests are assembled ; and every one of the other young men who 1 Gruppe, Griechische Mythologie und Religionsgeschichte, p. 903. Nilsson, Griechische Feste von religibser Bedeutung, p. 372. Schwally, Der heilige Krieg im alien Israel, p. 76. Reinach, op. cit. i. 116. Samter, Geburt, Hochzeit und Tod, p. 93 sqq. Frazer, Totemism and Exogamy, iv. 257. Piprek, op. cit. p. 191. 2 Thurston, Ethnographic Notes in Southern India, p. 2 sq. 3 Plutarch, Quaestiones Graecae, 58. 1 Idem, Lycurgus, xv. 4. 5 Abrahams, op. cit. p. 193. 520 THE HISTORY OF HUMAX MARRIAGE chap. sits down by his side has to pay a fine.1 At Fez, when the betrothal of a young man is celebrated in his parents' house, some negresses whose profession is to assist women on festive occasions dress him up as a bride with garments which they have brought with them. He is then placed on cushions on a mattress opposite the door, and sits there with his eyes closed as if he were a bride. After the young man's friends have entered the room, one of the negresses gives him some milk to drink, which is supposed to make his life " white," and another, with a plate of dates in her hand, puts into his mouth a date, representing wealth. After this they give milk and a date to each of his friends, who in their turn put money on his forehead, fixing it there with spittle.2 On the other hand, in some country places in Morocco, the bride imitates the appearance of a man by having designs resembling whiskers painted on her face,3 or by wearing her shawl thrown over her left shoulder and a dagger slung over her right,4 or by leaving her old home clad in a man's cloak.5 So also among the Bedouins of Mount Sinai, if the bride belongs to another camp than the bridegroom and is bride for the first time, one of her relatives throws over her an 'abba, or man's cloak, before she is placed upon the camel which is going to carry her to her new home ;6 and among the peasants of Palestine the bride is likewise covered with an 'aba when she is taken to the bridegroom's place.7 At Klovborg, in Denmark, on the first day of the wedding bride and bridegroom dress themselves in old clothes, she in men's and he in women's ; they then hide themselves from each other.8 It is also the custom in Den- mark,9 and in Esthonia10 and Russia11 as well, to put the 1 Zache, ' Sitten und Gebrauche der Suaheli/in Zeitschr. f. Ethnol. xxxi. 81 sq. 2 Westermarck, Marriage Ceremonies in Morocco, p. 25. 3 Ibid. p. 153. 4 Ibid. p. 152. 5 Ibid. p. 174. G Burckhardt, Notes on the Bedouins and Wahabys, pp. 150, 152. 7 Klein, ' Mittheilungen uber Leben, Sitten und Gebrauche der Fellachen in Palastina,' in Zeitschr. d. Deutschen Palaestina-Vereins, vi. 98. 8 Kristensen, op. cit. iv. 75. 9 Ibid. Tilkegsbind, iv. 59. 10 v. Schroeder, op. cit. p. 93 sqq. 11 Ida v. Diiringsfeld and v. Reinsberg-Duringsfeld, op. cit. p. 36 sq. XXV MARRIAGE RITES 521 bridegroom's hat or cap on the bride. Among the people of Southern Celebes the bridegroom at one stage of the pro- ceedings puts on the garments which have just been put off by the bride.1 I doubt, however, whether all these practices can be explained as attempts to deceive evil spirits. The fiance or bridegroom can hardly be protected against such spirits by being dressed up as a bride, as he is at Fez, or by putting on the garments which have been worn by his bride, since the bride is supposed to be haunted by evil spirits as much as, or even more than, the bridegroom himself; nor does the bride seem to be particularly well protected by pretending to be the bridegroom, as in South India, or by wearing his cap or hat. Facts of this sort seem better to agree with Mr. Crawley's theory of " inoculation," according to which the bride or bridegroom assumes the dress of the opposite sex in order to lessen the sexual danger by wearing the same kind of clothes as " the loved and dreaded person,"2 and the greatest possible assimilation between them would best serve the purpose of neutralising that danger. In some parts of Esthonia the bride is on the wedding day girded with a man's girdle and the bridegroom has a woman's girdle tied round his hat.3 Similar customs may, as already said, spring from different motives, or there may be mixed motives for the same custom. It should be added that when the bride imitates the appearance of a man, she may do so to be protected not only against evil spirits but against the evil eye. In many cases some other person or persons than the bride or bridegroom imitate his or her costume or in some way or other personate one of them ; and it has been said that persons so disguised may be supposed to serve as dummies to attract the attention of the demons or to divert to them- selves the envious glance of the evil eye and so allow the real bride or bridegroom to escape unhurt.4 It is a common 1 Matthes, op. cit. p. 35. 2 Crawley, op. cit. p. 371 sq. 3 v. Schroeder, op. cit. p. 94 sq. 4 Crooke, Popular Religion and Polk-Lore of Northern India, ii. 8. Sartori, op. cit. i. 75. Frazer, Totemism and Exogamy, iv. 257 sq. Samter, Geburi, Hochzeit und Tod, p. 106. 522 THE HISTORY OF HUMAN MARRIAGE chap. custom among Slavonic, Teutonic, and Romance peoples, as also among the Esthonians, that when the bridegroom or his representative comes to fetch the bride from her home, a false bride is substituted for the real one, another woman, frequently an ugly old one, or a little girl, or even a man being palmed off on him as the bride.1 In Brittany the substitutes are first a little girl, then the mistress of the house, and lastly the grandmother.2 In the Samerberg district of Bavaria a bearded man in woman's clothes personates the bride ;3 in Esthonia, the bride's brother or some other young man.4 Sometimes the substitution takes place already at the betrothal, and sometimes only at the wedding-feast. The custom is not restricted to Europe. Among the Beni-Amer in North-East Africa, when women with a camel are sent to fetch the bride, her people often sub- stitute a false bride for the true one, and it is only when the procession is well outside the village that the substitute reveals herself and runs back laughing.5 Professor Weber has suggested that a certain passage in the Kauiika-Sutra may possibly refer to the occurrence of the False Bride in ancient India.6 But Dr. Winternitz maintains that the interpretation of that passage is very doubtful, and that it at all events is too rash an assertion to say, as v. Schroeder 1 Usener, Ttalische Mythen,' in Rheinisches Museum, xxx. 183 sqq.; Piprek, op. cit. pp. 94, 98, 108, 112, 113, 131, 191 (most Slavonic peoples). Franzisci, op. cit. p. 68 (Slovenes in Carinthia). Tetzner, Die Slawen in Deutschland, p. 317. Ida v. Diiringsfeld and v. Reinsberg-Diiringsfeld, op. cit. pp. 53 (Rumania), n^(Switzerland), 150, 151 (Hesse), 179, 183, 190 sq. (Bohemia), de Nore, ap. cit. p. 135 (Landes, in France). Drechsler, op. cit. i. 245, 246, 256 (Silesia). Hepding, ' Die falsche Braut,' in Hessische Blatter f. Volkskunde, v. 161 sqq. Sepp, op. cit. p. 74 sqq. Gertrude M. Godden, ' The False Bride,' in Folk-Lore, iv. 142 sqq. v. Schroeder, op. cit. p. 68 sqq. Crawley, op. cit. p. 337. Sartori, op. cit. i. 75. Samter, Geburt, Hochzeit und Tod, p. 98. 2 Ida v. Diiringsfeld and v. Reinsberg-Diiringsfeld, op. cit. p. 246. 3 Ibid. p. 126. 4 v. Schroeder, op. cit. p. 218. For "the false bride" among the Cheremiss and Votyak see Hamalainen, op. cit. pp. 132, 171, 281 sq. 5 Munzinger, Ostafrikanische Studien, p. 324. 6 Weber, in Indische Studien, v. 393 n. XXV MARRIAGE RITES 523 does,1 that this is " undoubtedly" a primitive Indo-European custom.2 As for the meaning of the custom in question, I think it must be admitted that it allows of more than one explanation. The attempt to palm off on the bridegroom a mock bride may be another of those rites, already men- tioned in connection with marriage by capture, by which the girl and her .relatives show opposition to her marriage and till the last put obstacles in the bridegroom's way.3 Sometimes effigies are substituted for the bride and bride- groom. A traveller in Java noticed two painted wooden figures, one of a man and the other of a woman, standing at the foot of the " family nuptial couch." These figures had been placed there to cheat the devil, who, according to the belief of the people, during the wedding night hovers round the bed with a view to carrying off one of the happy pair ; for it was thought that he, deceived by their resem- blance, would carry off the figures instead of the sleeping lovers.4 Much more frequently there is an inanimate substitute for one of the parties only. Mock marriages with things or trees or animals are often resorted to in India for the purpose of averting some threatened evil from the bride or bridegroom or both.5 Tree-marriages, in par- ticular, occur widely throughout Northern India ; and, as 1 v. Schroeder, op. cit. p. 72. 2 Wintemitz, in Transactions, p. 269. 3 Cf. van Gennep, op. cit. p. 187 :-" Je crois que le plus sou vent le rite a pour but d'6viter Taffaiblissement des groupements intSresses (classe d'age sexuelle, famille, etc.) en tachant de livrer ou d'unir des individus de moindre valeur sociale g6n6rale, et surtout dconomique." 4 d'Almeida, quoted by Frazer, Folk-Lore in the Old Testament, i. 520 sq. 6 Crooke, Popular Religion and Folk-Lore of Northern India, ii. 115 sqq. Idem, ' Hill Tribes of the Central Indian Hills,' in Jour. Anthr. Inst, xxviii. 242 sq. Idem, Things Indian, p. 317 sq. Dalton, Descriptive Ethnology of Bengal, pp. 194, 319. O'Malley, Census of India, 1911, vol. v. (Bengal, Bihar and Orissa and Sikkim) Report, p. 323 sq. Russell, Tribes and Castes of the Central Provinces of India, iii. 188 sq. ; iv. 28, 506. Schmidt, Liebe und Ehe im alten und modernen Indien, p. 405 sqq. Thurston, Ethnographic Notes in Southern India, pp. 35, 36, 81, Ac. Frazer, Totemism and'Exogamy, iv. 210 sqq. Idem, Folk-Lore in the Old Testament, i. 525 sqq. Crawley, op. cit. p. 340 sq. 524 THE HISTORY OF HUMAN MARRIAGE chap. Mr. Crooke observes, the idea that the tree itself is supposed to die soon after the ceremony " seems to point to the fact that the marriage may be intended to divert to the tree some evil influence, which would otherwise attach to the wedded pair."1 In the Punjab mock marriages, owing to fear of ill luck, take place either when a widower wishes to marry a third wife, or when the horoscope of a girl shows that the influence of certain stars is likely to lead to early widowhood. In cases of the former kind the mock marriage is celebrated sometimes with a certain tree or bush and sometimes with a sheep, which is dressed up as a bride and is led by the bridegroom round the sacrificial fire while the real bride sits by. And the fear of ill luck is due partly to the suspicion, caused by the death of the two former wives, that the wife of the man is destined to die and particularly the wife taken by the third marriage, which is considered to be peculiarly inauspicious ; but partly also to the belief that the jealousy of the spirit of the first wife is instrumental in causing the death of subsequent wives, although in the case of a fourth marriage the evil influence of the first wife is supposed to have spent itself and therefore no mock marriage is usually deemed necessary. In mock marriages of the second kind, again, a pitcher full of water is dressed like a boy, and the girl is taken through the ceremonies of marriage with this pseudo-bridegroom. The. ceremonies are then repeated with the real bridegroom by way of an informal marriage ; and it is supposed that the effect of the evil star will befall the pitcher and not the bridegroom, thus averting the disaster of early widowhood*2 Similar mock marriages are found in other parts of India, for instance in the Central Provinces. Thus among the Kawars, if a widower marries a girl for his third wife, an earthen image of a 1 Crooke, Popular Religion and Folk-Lore of Northern India, ii. 120. See also O Malley, op. cit. p. 324- Mr. Crooke, however, also suggests (in Jour. Anthr. Inst, xxviii. 242) that the custom of tree- marriage in India is partly based on the desire to bring the wedded pair into intimate connection with the reproductive powers of nature. 2 Pandit Harikishan Kaul, Census of India, 1911, vol. xiv. (Punjab) Report, p. 283 sq. Cf. Thurston, op. cit. p. 44. XXV MARRIAGE RITES 525 woman is made and he goes through the marriage ceremony with it ; he then throws the image to the ground so that it is broken, when it is considered to be dead and its funeral ceremony is performed, and the widower consequently may marry the girl, who now becomes his fourth wife.1 Among the Barais a bachelor espousing a widow must first go through the ceremony of marriage with a swallow-wort plant.2 Among the Gonds in some parts of Bastar a widow whose husband has been killed by a tiger is formally married, not to her new husband, but either to a dog or to a lance, axe, or sword, because they believe that the ghost of the deceased has entered into the tiger and in that form will seek to devour the man who marries his widow, but now will carry off the dog or perish by the weapon.3 In the Himalayas, when the conjunction of the planets por- tends misfortune at a marriage, or when a boy or girl on account of some bodily or mental defect cannot find a spouse, the unattractive or luckless person is first wedded to an earthen pot, the marriage-knot being tied by a string which unites the neck of the bridegroom or bride to the neck of the pot; and the dedicatory formula sets forth that the ceremony is undertaken in order to counteract the malign influence of the adverse planets or of the bodily or mental blemish of the husband or wife.4 There are, further, cases in which the bridegroom or the bride, instead of assuming the appearance of somebody else or being represented by a substitute, is sheltered by some person or persons who are dressed up to resemble him or her, so that there apparently are two or more bride- grooms or brides. Thus at Fez, when the bride is taken to her future home, she is accompanied not only by the bride- groom's people who have come to fetch her, some men of her own family, and a crowd of boys, but by some-perhaps six or eight-women relatives, who are dressed exactly like herself so that no one can distinguish between them ; i Russell, op. cit. iii. 395- 2 Ibid- iL W 3 Ibid- iiL 8l- 4 Atkinson, quoted by Frazer, Totemism and Exogamy, iv. 211, and Crooke, Popular Religion and Folk-Lore of Northern India, ii. 117. 526 THE HISTORY OF HUMAN MARRIAGE chap. this was said to protect her from magic and the evil eye. The men and boys of the bridegroom's party head the pro- cession, behind them walk the women, including the bride, and last come her male relatives, both men and boys.1 In Egypt, again, when the bridegroom goes to the mosque before meeting the bride, he walks between two friends dressed like himself ;2 whilst in Palestine the bridegroom leads by the hand a little boy dressed pre- cisely like himself and called " the mock bridegroom," who imitates his slightest movements and thus diverts the company.3 Among the Abyssinians, when a princess is married, she is accompanied in the procession by her sister in an attire exactly similar to her own.4 In South Celebes during the first days of the wedding the bride is accompanied by a woman of her own age dressed as far as possible like her, and the bridegroom by a young man of his own age dressed like himself.5 In some parts of France and Italy6 and among the White Russians in the neighbourhood of Minsk7 the bridegroom must at the betrothal feast or on the wedding day find out his bride from among a crowd of girls. Among the Livonians two bridesmaids are dressed exactly as the bride herself.8 So also in Belford " the bride and her maids are dressed alike " ,9 and I am told that this has been the custom also elsewhere in England. The functions of bridesmaids, bridesmen, and groomsmen have been not only to attend upon bride and bridegroom but to protect them from evil influences, even when no attempt is made to imitate their dress ; people always feel safer in company.10 In Shetland the best-man must 1 Westermarck, Marriage Ceremonies in Morocco, p. 165 sq. 2 Lane, Modern Egyptians, p. 180 sq. 3 Van-Lennep, op. cit. p. 551 sq. 4 Harris, Highlands of Aethiopia, ii. 225. 5 Matthes, op. cit. p. 29. 8 de Nore, op. cit. p. 308 (Vosges). Usener, in Rheinisches Museum, xxx. 187 sq. Liebrecht, Zur Volkskunde, pp. 407, 408, 511. 7 Piprek, op. cit. p. 67. 8 Sjogren, Livische Grammatik nebst Sprachproben, p. Ixxix. " Marie Balfour, County Folk-Lore, vol. iv. Northumberland, p. 97. 10 Cf. Crawley, op. cit. p. 338. XXV MARRIAGE RITES 527 sleep with the bridegroom during the night before the wedding.1 Among the White Russians he lies down on the nuptial bed before the bride and the bridegroom.2 Among other peoples he or some bridesmen are present when the marriage is consummated, or bridesmen and bridesmaids have to prevent the speedy consummation of it.3 An effective method of protecting the bride against ex- ternal influences, particularly the evil eye, is to shut her up in a box when she is taken to her new home. This was formerly done at Fez, where she was carried in " a wooden cage or cabinet eight-square covered with silk,"4 and it is still done in the north of Morocco, where the bride is trans- ported to the bridegroom's house in a so-called 'ammartya on the back of a mule or a horse ; and in one tribe this box is made of oleander branches, which are supposed to afford 'particularly good protection against the evil eye.5 In other parts of that country, and in the Muhammadan world in general, she is taken to the bridegroom's home with her face well covered, and the same is the case elsewhere, also in many uncivilised tribes.6 The Chinese bride inside the marriage sedan chair is completely secluded from profane gaze, and when she makes her appearance her features are concealed by a piece of red silk.7 The veiling of the bride is referred to in Genesis.8 It has been common in Europe ;9 and the importance which the ancient Romans attached to this custom appears from the ordinary use of the word nubere or obnubere to denote a woman's marrying. Its primary object was in all probability to protect the bride, 1 Black, County Folk-Lore, vol. iii. Orkney and Shetland, p. 209. 2 Piprek, op. cit. p. 64. 3 Supra, i. 205 n. 3. Infra, ii. 547, 550, 551, 555, 557 sq. 4 Leo Africanus, op. cit. ii. 450. 6 Westermarck, op. cit. p. 167 sqq. 6 E.g., among the Banyoro, Bakene (Roscoe, Northern Bantu, pp. 39, 41, 150), Baganda (Idem, Baganda, p. 89), Wafipa (Fromm, in Mittheil. Deutsch. Schutzgeb. xxv. 97), Zulus (Tyler, op. cit. p. 205), Li of Hainan (Strzoda, ' Die Li auf Hainan,' in Zeitschr. f. Ethnol. xliii. 203), Yukaghir (Jochelson, Yukaghir, p. 94). 7 Ball, Things Chinese, p. 420 sq. 8 Genesis, xxiv. 65. 9 v. Schroeder, op. cit. p. 72 sqq. Samter, Familienfeste der Griechen und Romer, p. 47 sqq. Sartori, op. cit. i. 79. Piprek, op. cit. p. 190 (Slavs). Hamalainen, op. cit. p. 233 sqq. (Ugro-Finnic peoples) 528 THE HISTORY OF HUMAN MARRIAGE chap. particularly against the evil eye. The veil of the Esthonian bride is expressly said to serve this purpose ; and for fear of the same enemy the bridegroom, in addition, runs with her to the church as fast as he can.1 In Morocco, however, I have also found another idea connected with the veiling or covering of the bride : her own glance is considered dangerous to others. When she is painted with henna in her home, her eyes must be covered, because misfortune would befall any person or animal the bride looked at before she has seen her husband on her arrival at his house. The 'ammariya into which she is carried from her bed must be placed close to the door and covered with blankets in such a manner as to make it impossible for any outsider to see the bride, because anybody who should see her would become blind. And if she looked at anybody on her way to the bridegroom's place, there would be fighting and man- slaughter at the wedding that very day.2 Among the Touareg of the Ahaggar the bride is not allowed to look at the fantasia of the horsemen, because, " si, par malheur, elle regardait un des cavaliers, celui-ci tomberait et un accident se produirait."3 In ancient India the bridegroom had to protect himself against the evil eye of his bride.4 In Morocco the bridegroom also, to some extent, covers his face, pulling his hood over it and drawing his hayek over his mouth, and although bashfulness may from the beginning have had something to do with this custom, superstition has probably the greater share in it. Among the Oraons of Bengal a screen^is held round bride and bridegroom while the marriage rites are being performed, in order to protect them from the gaze of demons and of strangers who may have the evil eye.5 Among many peoples the bridegroom is not allowed to leave the house for a certain number of days, and a similar prohibition 1 Boecler-Kreutzwald, op. cit. p. 29. 2 Westermarck, op. cit. pp. 148, 163, 169, 172, 181, 189. Cf. ibid. p. 219. 3 Benhazera, op. cit. p. 15. 4 Oldenberg, Die Religion des Veda, p. 503. 5 Sarat Chandra Roy, Orcions of Chota-Nagpur, p. 363. XXV MARRIAGE RITES 529 applies still more frequently and for a longer period to the newly-wedded wife.1 At Fez she must remain inside the house for two months, or at least six weeks, not even being allowed to go on the roof. At Tangier she was formerly obliged to stay at home for a whole year, but this period has been reduced to three or four months. Among some of the North-East African Beduan she may not leave the house for three years.2 It seems that particular care is often taken to protect bride or bridegroom against dangers from above. In China, " when the bride ascends the bridal sedan she wears a hat of paper, and an old woman who has sons and grandsons holds an umbrella over her."3 In the towns of Palestine and Egypt she walks under a canopy, escorted on either side by a man with a drawn sword.4 The Jewish chuppah, or canopy, under which Jewish marriages are still celebrated, seems to have been derived from the canopied litter which, in ancient time, was occupied by the bride during the pro- cession.5 In the Scandinavian countries,6 England,7 and France8 a square piece of cloth (in French called carre, in English " care cloth," in Swedish pall, from the Latin pallium, or himmel) was held over the bride and bridegroom at the benediction. In Sweden and Swedish-speaking com- munities in Finland this practice has survived at the nuptial ceremony until recent times,9 and in some places a himmel 1 Crawley, op. cit. p. 333 sq. Westermarck, Marriage Ceremonies in Morocco, p. 323 n. 23. Sartori, op. cit. i. 120. 2 Munzinger, Ostafrikanische Studien, p. 148. 3 Doolittle, op. cit. i. 80. See also Stewart Lockhart, ' Chinese Folk-Lore,' in Folk-Lore, i. 365. 4 Van-Lennep, op. cit. p. 551. See also Lynch, op. cit. p. 448 (Jaffa). 5 Abrahams, op. cit. p. 193. 0 Grath, Svenska kyrkans brudvigsel, p. 30 sq. Troels-Lund, op. cit. xi. 49 sqq. 7 Brand, op. cit. p. 380 sq. Gloucestershire Notes and Queries, ii. 276. Calvor, Rituale ecclesiasticum, i. 107. 8 Martene, De antiquis ecclesiae ritibus, ii. 355 sq. 9 Troels-Lund, op. cit. xi. 50 sq. Hylfen-Cavallius, op. cit. ii. 437. Holmberg, Bohusldns historia och beskrifning, i. 243. Gaslander, loc. cit. p. 265. Lithberg, ' Brollopsseder pa Gottland,' in Fataburen, VOL II MM 530 THE HISTORY OF HUMAN MARRIAGE CHAP. was also held over the bride or the couple out-of-doors,1 or was fastened to the roof over their seats at the table,2 or arranged in the bridal chamber.3 Among the mining population of Fife, when the bridal company set out in procession for kirk, the bride and groom were sometimes " bowered," that is, an arch of green boughs was held over their heads.4 In some parts of Germany the bridegroom wears on the wedding day a tall hat, which he only removes in church.5 In some parts of SkAne, in Sweden, he kept his head covered while dancing,® and among various Slavonic peoples he does so at the table.7 In Bohemia8 and among some Ugro-Finnic peoples9 he wears a fur cap even though the marriage is celebrated in the summer; whilst a Syryenian bride has her head covered with a cap made of sheepskin until the nuptial ceremony, and does not remove it even at night.10 Bride and bridegroom must be protected against dangers not only from above but from below. In Morocco the bridegroom must avoid sitting on the ground. At the feast when he is painted with henna or, sometimes, has flour rubbed on his hand, he has underneath him a cafpet and a sack or a saddle, and I was told that the 1907, p. 173 sq. Wikman, ' Frieri, forlofning och brollop i Delsbo,' ibid. 1913, p. 79- Tegengren, ' Magi och vidskepelse, hanforande sig fill trolovning, brollop o.s.v. (Frin Osterbotten),' in Hembygden, iii. 41. Lindrcos och Andersson, ' Eft brollop i Pellinge,' ibid. 1910, p. 157. Nikander, manuscript notes relating to Aland. 1 Troels-Lund, op. cit. xi. 26. Nikander, manuscript notes relating to Eckero, Aland. 2 Holmberg, op. cit. i. 243. Djucklou, Ur Nerikes folks prdk och folklif, p. 50. Lithberg, loc. cit. p. 87 sq. Aina Wadstrom, * Frieri- och brollopsbruk fran Dagsmark i Lappfjard,' in Hembygden, ii. 84. 3 Lithberg, loc. cit. p. 88. 4 Rorie, in Simpkins, op. cit. p. 392. 5 Meyer, Badisches Volksleben, p. 290. Sartori, op. cit. i. 79. 0 Nicolovius (Loven), Folklifvet i Skytts Harad i Skdne, p. 140. 7 Piprek, op. cit. pp. 95 (Czechs), 106 (Slovaks). Rajacsich, Das Leben, die Sitten und Gebrauche, der im Kaiserthume Oestevreich lebenden Sudslaven, p. 181 (Serbs). 8 Ida v. Diiringsfeld and v. Reinsberg-Duringsfeld, op. cit. p. 180. 9 Hamalainen, op. cit. p. 264. For the bridegroom's keeping his head covered on various occasions see ibid. p. 263 sq. 10 Dobrotvorskij, quoted ibid. p. 264. xxv MARRIAGE RITES 531 object of this is to prevent his being affected by evil influ- ences. After the ceremony he is in some tribes carried away by his best-man or other bachelor friends; and at Fez, on the great occasion when his head is ceremonially shaved, he is likewise carried by his friends, from the house where he has been sitting with them to " the house of the wedding," is there put down on a large chair, and is after- wards again carried away by the friends. Throughout the wedding he has in many tribes the backs of his slippers pulled up so as to prevent their falling off and his feet coming into contact with the ground, although there is also some fear that the slippers might fall into the hands of an enemy, who would harm him by working magic with them.1 Similar and still greater precautions are taken with regard to the bride ; as the Shareefa of Wazan writes, " a bride would be unlucky to put her foot to earth at this period."2 She, too, has the backs of her slippers pulled up. Sometimes she is carried to and from the place in the house where she is painted with henna. In country places she is carried to her new home on the back of an animal, on to which she is lifted sometimes from her bed, sometimes from the door of the house or the entrance of the tent, and sometimes after she has first taken a few steps on a blanket or cloak spread in front of her in order to prevent her stepping on the ground or the threshold. And when she arrives at the bridegroom's place she is carried to the nuptial bed or across the threshold.3 It may be asked why bride and bridegroom must not come into contact with the ground. In the first instance, they have baraka, or " holiness," and persons or objects possessed of this delicate quality are in many cases not allowed to touch the ground.4 Moreover, the real native country of the ^nun is under the ground and they are therefore always liable to haunt its surface. And as for the custom of carrying the bride through the entrance of the dwelling or otherwise preventing her touching the 1 See Westermarck, op. cit. General Index, s.v. ' Bridegroom.' 2 Emily, the Shareefa of Wazan, op. cit. p. 136. 3 See Westermarck, op. cit. General Index, s.v. ' Bride.' 4 Idem, Moorish Conception of Holiness, p. 132 sqq. MM2 532 THE HISTORY OF HUMAN MARRIAGE chap. threshold, it is obviously connected with the idea that the threshold of a house and the entrance of a tent are much haunted by faun. The Moors say that " the masters of the house," that is, its jinn-owners, are walking out and in over the threshold. Nobody is allowed to sit down on the threshold of a house or at the entrance of a tent; should a person do so he would become ill himself or give bas to the house. And it would likewise be unlucky for the house and its inhabitants if anybody should pull up the backs of his slippers on the threshold.1 In Aglu in the south of Morocco, when the bridegroom enters the room where the bride is waiting for him, he puts his right foot twice over the threshold before he makes his entrance by a third step. Very similar marriage customs are found in other countries. The bride is frequently carried to her future home-on an animal or a litter or a man's back or in some other manner. This may, of course, be done for the sake of convenience, or it may be a ceremonial expression of the reluctance which a virgin pretends to feel against being given away in marriage; but there can be no doubt that the fear of her touching the ground also has something to do with it. In Sierra Leone, according to an old account, when the bridal party approaches the bridegroom's town, the bride is taken on the back of an old woman and carried covered with a fine cloth so as not to be seen by any man ; mats are spread on the ground, " that the feet of the person who carries her may not touch the earth," and in this manner she is transported to the house of her intended husband.2 Among the peasantry of Lor- raine women used to carry the bride seated on their arms crossed, from the threshold of the house to the church.3 In South Westphalia the bride is often taken to the church on horseback, even if it is close by, and she always rides when she leaves it.4 Among some Ugro-Finnic peoples she is carried to the carriage by which she is to be conveyed to the bridegroom's house, and on her arrival there she is carried 1 Cf. Idem, Marriage Ceremonies in Morocco, p. 220 n. 1. 2 Matthews, Voyage to the River Sierra Leone, p. 118. 3 Remigius, quoted by Dalyell, Darker Superstitions of Scotland, p. 292. 4 Woeste, loc. cit. p. 36 sq. xxv MARRIAGE RITES 533 into the house ; and among the Cheremiss this custom is expressly said to be connected with the idea that a bride must not put her foot on the bare ground.1 The bridegroom, too, may be carried. Among the Indians of Guatemala and San Salvador the bridal pair were carried by their friends, and shut in a room.2 Among the Mundas of Chota Nagpur, if the bridegroom is not rich enough to afford to engage a palanquin or other conveyance, he and the bride are carried on the arms of relatives up to the limits of the bride's village and again from the boundary of his own village up to his house.3 Among the Kurmis of Bengal the bridegroom is carried to the bride's house on men's shoulders.4 In Cyprus, if the couple live in the same village, the bridegroom is on the wedding morning carried to the house of the bride by his friends, who, crossing arms and joining hands, form a kind of chair, on which he sits ; but if he comes from a distance, he and all his retinue arrive on horseback.5 In Egypt, if the bridegroom be a youth or young man, one of his friends carries him a part of the way up to the harem-according to Lane, because "it is con- sidered proper that he, as well as the bride, should exhibit some degree of bashfulness.'' 6 Now the carrying of the bridegroom may no doubt, like that of the bride, be a cere- monial expression of modesty, real or pretended ; but my experience from Morocco makes me inclined to believe that elsewhere, also, this custom may be to some extent connected with the idea that at a certain stage of the wedding it is dangerous for him to come in contact with the ground. At Foochow in China, again, the floor of the reception-room in the bridegroom's house is covered with red carpeting from the place where the sedan stops to the door of the bride's 1 Hamalainen, op. cit. p. 243 sq. 2 Bancroft, Native Races of the Pacific States of North America, i. 703. 3 Sarat Chandra Roy, Mundas, p. 454. 4 Dalton, Descriptive Ethnology of Bengal, p. 319. 5 Rodd, Customs and Lore of Modern Egypt, p. 101. J Lane, Account of the Manners and Customs of the Modern Egyptians, p. 182. Cf. Jafiur Shurreef, Qanoon-e-islam, or the Customs of the Mussulmans of India, p. 84. 534 THE HISTORY OF HUMAN MARRIAGE chap. room, in order to prevent her feet from touching the floor ;1 and it is presumably for a similar reason that a Chinese bride on leaving her own home walks all the way from her room to the sedan chair in her father's shoes, which are then left behind before she steps into the chair.2 Among the Morlaks of Dalmatia a sheet is spread between the door of the bride- groom's house and the horse on which the bride arrives, and on this she walks to the threshold.3 Among some Slovaks she walks into his house upon a rug ;4 and a similar custom recently prevailed in some parts of Epirus.5 In England there was " a custom at marriages of strewing herbs and flowers, and also rushes, from the house or houses where persons betrothed resided to the church."6 In Sunderland the footpath of the street in which the bride lives, and along which she must pass in order to be married at the church, is sprinkled with sawdust. Formerly sea-sand was used ; and if the custom was to be fully carried out in its integrity, the sand or sawdust should stretch all the way from the bride's house to the church gates.7 In Newcastle-on- Tyne sand is strewn on the pavement before a bridal pair tread on it.8 At Cranbrook in Kent, when a newly-wedded pair leave the church, the path is strewed with emblems of the bridegroom's calling; thus carpenters walk on shavings, butchers on sheepskins, shoemakers on leather parings, and blacksmiths on scraps of old iron.9 The red carpet at weddings is familiar to all of us. That these customs are, at least in part, due to super- stitious fear of too close a contact with the ground is the more probable because there are other practices apparently intended to protect bride and bridegroom against supposed 1 Doolittle, op. cit. i. 83. 2 Chen, Patriarchal System in China, p. 3. 3 Piprek, op. cit. p. 133. 4 Ibid. p. 106. See also Rajacsich, op. cit. p. 159 (Serbs of Syrmia). 6 Sakellarios, op. cit. p. 22. 6 Brand, op. cit. p. 364 sq. 1 Coleman, ' Sawdust Wedding,' in Notes and Queries, ser. v. vol. v. 186. 8 Henderson, Notes on the Folk-Lore of the Northern Counties of England and the Borders, p. 40. 9 Crooke, ' Lifting of the Bride,' in Folk-Lore, xiii. 245 sq. XXV MARRIAGE RITES 535 danger from below. In some tribes in Morocoo a needle or some salt is put into the right slipper of the bride, or of the bridegroom as well, as a charm against jnun or other evil influences.1 In North Germany bride and bridegroom strew dill and salt into their shoes as a protection against witchcraft.2 In many places a coin or coins are put into the shoe or shoes of the bride or the bridegroom or both.3 This practice, which is particularly common among the Scandinavian peasantry,4 is often supposed to prevent poverty or to produce wealth, but not infrequently it is expressly said to be regarded as a safeguard against evil influences.5 In certain parts of Germany it is believed that " if bride and bridegroom on the wedding day put a three- headed bohemian (a coin) under the sole of their right foot, it will be a happy marriage."6 In some parts of Scotland, according to Dalyell, " the bridegroom has sought protec- tion by standing with the latchet of his shoe loose and a coin under his foot, probably for interception from the earth."7 In Montenegro, again, the dever places a knife in front of the feet of the bride, with the edge turned towards her, and then she puts her right foot on the knife.8 To carry a bride over the threshold is a very widespread 1 Westermarck, Marriage Ceremonies in Morocco, pp. 150, 237, 256, 290. 2 Kuhn and Schwartz, N orddeutsche Sagen, Marchen und Gebrauche, P- 434- 3 Westermarck, op. cit. p. 150 (Aith Ytisi in Morocco), v. Hahn, op. cit. p. 145 ; Sepp, op. cit. p. 21 (Albanians). Piprek, op. cit. pp. 64, 79, 86 (Slavonic peoples). 4 Feilberg, Bidrag til en ordbog over jyske almuesmdl, p. 64 (Den- mark). Raaf, op. cit. i. 115 ; Holmberg, op. cit. p. 243 ; Gaslander, loc. cit. p. 276 ; Lithberg, in Fataburen, 1907. P- 173 ; Wikman, ibid. 1913, P- 8if; Djurklou, op. cit. p. 48 ; Norlind, Gamla brollopsseder hos svenska allmogen, p. 68 (Sweden). Sjoberg, ' Brudstugugiendet i Replot,' in Hembygden, vii. 81 ; Tegengren, ' Brdllopsbruk i Vdr5,' ibid, viii.-ix. 139 ; Nikander, in manuscript notes relating to Kokar (Swedish-speaking communities in Finland). 5 Raaf, op. cit. i. 115 ; Wikman, loc. cit. p. 81 (Sweden). Tegen- gren, loc. cit. p. 139 (VorA, in Finland). 6 Grimm, op. cit. iv. 1823. 7 Dalyell, op. cit. p. 312 sq. 8 Piprek, op. cit. p. 124. 536 THE HISTORY OF HUMAX MARRIAGE chap. practice.1 It occurs, or has occurred, in China,2 Palestine,3 Cairo,4 ancient Rome,5 modern Greece,6 Germany7, France,8 French Switzerland,9 and Great Britain.10 Of the Palestine custom it is said that a bride is often carried over the threshold " that her feet may not touch it, to do so being con- sidered unlucky." 11 Among the Slovenes, on the arrival at the bride's house after the ceremony in the church, the bride- groom carries her over the threshold " in order that nobody shall be able to do her harm."12 In Wales, on her return from the marriage ceremony, she was always carefully lifted over the threshold because " it was considered very unlucky for a bride to place her feet on or near the threshold," andtrouble was in store for the maiden who preferred walking into the house."13 In some parts of Scotland, in the beginning of the last century, when the wedding party arrived at the bridegroom's house, " the young wife was lifted over the threshold, or first step of the door, lest any witchcraft or ill e'e should be cast upon and influence her."14 We may assume that in other cases also the custom of carrying 1 v. Schroeder, op. cit. p. 88 sqq. Trumbull, Threshold Covenant, p. 36 sqq. Samter, Geburt, Hochzeit und Tod, p. 136 sqq. Frazer, Folk-Lore in the Old Testament, iii. 6 sqq. 2 Davis, China, i. 329. 3 Wilson, Peasant Life in the Holy Land, p. 114. 4 Burckhardt, Arabic Proverbs, p. 116. 5 Plutarch, Quaestiones Romanae, 29. Rossbach, op. cit. pp. 351, 359 S<H- Marquardt, op. cit. p. 55. 6 Wachsmuth, op. cit. p. 97. 7 Weinhold, Die deutschen Frauen in dem Mittelalter, i. 407. Drechsler, op. cit. i. 264 (Silesia). Temme, Die Volkssagen. der Altmark, p. 73. * 8 Ida v. Diiringsfeld and v. Reinsberg-Diiringsfeld, op. cit. pp. 251, 258. Laisnel de la Salle, op. cit. ii. 47. 9 Ida v. Diiringsfeld and v. Reinsberg-Diiringsfeld, op. cit. p. 106. 10 Mrs. Gutch and Mabel Peacock, County Folk-Lore, vol. v. Lincoln- shire, p. 2335^. Marie Balfour, op. cit. p. 93 (Northumberland). Marie Trevelyan, Folk-Lore and Folk-Stories of Wales, p. 273. Napier, Folk-Lore, or Superstitious Beliefs in the West of Scotland, p. 51, Dalyell, The Darker Superstitions of Scotland, p. 291. 11 Wilson, op. cit. p. 114. * 12 Piprek, op. cit. p. 113. 13 Marie Trevelyan, op. cit. p. 273. 14 Napier, op. cit. p. 51. XXV MARRIAGE RITES 537 the bride over the threshold is due to fear lest her feet should touch it.1 For there is a widespread belief that the threshold is haunted by spirits or otherwise fraught with danger; and side by side with the carrying or lifting of the bride over the threshold there is the custom which simply requires 1 Other explanations of this custom have been suggested. With reference to the ancient Romans, Plutarch (Quaestiones Romanae, 29) writes :-" Why do they not permit the new-married woman herself to step over the threshold of the house, but the bridesmen lift her over ? What if the reason be that thqy, taking their first wives by force, brought them thus into their houses, when they went not in of their own accord ? Or is it that they will have them seem to enter into that place as by force, not willingly, where they are about to lose their virginity ? " Many modern writers have likewise tried to explain the custom as a survival of marriage by capture, e.g., Rossbach (op. cit. p. 360), Lord Avebury (Origin of Civilisation, p. 102), v. Schroeder (op. cit. p. 92), and Jevons (in his edition of Plutarch's Romans Questions, p. xcv. sq.} ; but without success (cf. Samter, Geburt, Hochzeit und Tod, p. 137 sq. ; Zachariae, ' Zum altindischen Hochzeitsritual,' in Vienna Oriental Journal, xvii. 140 sqq.}. In many cases the man who carries the bride over the threshold is neither the bridegroom nor any of his friends, but a man of her own family (cf. Crooke, * Lifting of the Bride,' in Folk-Lore, xiii. 239) ; and, as noticed in the text, the custom which simply requires the bride to avoid stepping upon the threshold alternates with that of lifting her. Equally unsatisfactory is Zacha- riae's suggestion (loc. cit. p. 142 sq.} that the chief feature of the cere- mony is the lifting-up of the bride, not her avoidance of touching the threshold with her feet. " Das Heben an sich," he says, " war augenscheinlich eine feierliche, bedeutsame Handlung. Personen, die man feiern oder ehren wollte, wurden in die Hohe gehoben : so bei der Hochzeit die Braut und . . . auch der Brautigam." This may apply to some other ceremonies, but does not explain why the bride is carried over the threshold (cf. Samter, Geburt, Hochzeit und Tod, p. 138 sqq.}. The Roman custom of lifting a bride over the threshold has been said to serve the object of preventing the bad omen which would be caused by her stumbling on it (Preller, Rbmische Mythologie, p, 585). According to Mr. Crooke (' Lifting of the Bride,' in Folk- Lore, xiii. 242), "the lifting over the threshold may ... be in some cases a fertility charm ; in others it may have been intended to pro- tect the bride from some contamination, or to avoid ill-luck." Sir James Frazer (in his recent work, Folk-Lore in the Old Testament, iii. 11) favours the theory expressed in the text, which is a general application of the explanation of the Moorish custom given in my Marriage Ceremonies in Morocco, p. 219 sqq. 538 THE HISTORY OF HUMAN MARRIAGE ch^. her to avoid stepping on it.1 This rule prevailed in ancient India, where the bridegroom, on reaching the house, should instruct the bride to place the right foot first and avoid treading upon the threshold.2 The same rule is still followed by some of the Southern Slavs.3 Among the White Russians a fur is placed on the threshold of the bridegroom's house for the bridal pair to walk on when they arrive there.4 The threshold may be considered dangerous to the bride- groom also. Thus in Salsette, an island near Bombay, the bridegroom is carried into the house by his maternal uncle, and afterwards he himself lifts his bride .over the threshold.5 As for the fear of the threshold, I venture to believe that it is chiefly due to that uncanny feeling which superstitious people are apt to experience when they first enter a dwelling, passing through the doorway from daylight into dimness. This feeling easily gives rise to the idea that the threshold is haunted by mysterious beings, whether the souls of dead people or spirits like the jinn, or is, generally, a seat of supernatural danger.6 1 See Frazer Folk-Lore in the Old Testament, iii. i sqq., particularly p. 6 sqq.. Various superstitions relating to the threshold are found in Trumbull's book The Threshold Covenant. 2 Grihya-SHtras, ii. 193, 263. Winternitz, Das altindische Hochzeitsrituell, pp. 23, 72. 3 Krauss, op. cit. p. 430 sq. 4 Piprek, op. cit. p. 65. 5 d'Penha, ' Superstitions and Customs in Salsette,' in Indian Antiquary, xxviii. 117. 6 Dr. Samter (Geburt, Hochzeit und Tod, p. 140 sqq.), following a suggestion made by Dr. Winternitz (Das altindische Hochzeitsrituell, p. 72), maintains that the threshold is feared because it is believed to be haunted by the souls of dead people and thinks that this belief may have originated in an earlier custom of burying the dead underneath the threshold, although he can adduce no evidence for this supposition. Sir James G. Frazer (' Folk-Lore in the Old Testament,' in Anthropological Essays presented to E. B. Tylor, p. 172. sq. ", Folk-Lore in the Old Testament, iii. 13 sqq.), however, has given some instances of burying still-born children under the threshold in Russia and India and of burying abortive calves under the threshold of the cowhouse in England. But these instances are so few and of so exceptional a character that in my opinion no general theory as regards the world-wide fear of the threshold can possibly be based on them. I even strongly doubt that Frazer is right in his suggestion (ibid. p. 15 sq.) that " the glamour which surrounds the XXV MARRIAGE RITES 539 Fear of dangers threatening bride and bridegroom from below may also be the origin of the familiar custom of throwing an old shoe or old shoes after them, which is found not only in England and Scotland,1 but in Denmark,2 on the Rhine,3 and among the Gypsies of Transylvania,4 and evidently occurred in ancient Greece, as appears from the representation of a wedding on a vase in the museum of Athens.6 At a Turkish wedding the bridegroom " has to run for his life to the harem under a shower of old shoes ; for, according to the Turks, an old slipper thrown after a man is an infallible charm against the evil eye."6 In most cases the shoe is thrown after the bridal pair, when they leave for church or return from church or after the wedding breakfast ; but a shoe may also be thrown after each separately. Train writes in his Account of the Isle of Man :-" On the bride- groom leaving his house, it was customary to throw an old shoe after him, and in like manner an old shoe after the bride on leaving her home to proceed to church, in order to ensure good luck to each respectively ; and if by stratagem either of the bride's shoes could be taken off by any spectator on her way from church, it had to be ransomed by the bride- groom."7 threshold in popular fancy may be in part due to an ancient custom of burying dead infants or dead animals under the doorway." He adds, however, that this custom cannot completely account for the superstition, since the superstition attaches to the thresholds of tents as well as of houses, and, so far as he is aware, there is no evi- dence or probability of a custom of burying the dead in the doorway of a tent; besides, " in Morocco it is not the spirits of the dead, but the jinn, who are supposed to haunt the threshold." On the other hand, the suggestion made in the text, that the superstitions relating to the threshold are due to the uncanny feeling experienced by him who enters a dwelling, applies to tents as well as to houses, and is in no way affected by the particular nature of the evil influences which in the different cases are attached to the threshold ; for they all spring from the fear of the uncanny. 1 Gregor, op, cit. p. 91. 2 Feilberg, op. cit. iii. 642 ; Tillaeg og rettelser, p. 64. 3 Samter, Geburt, Hochzeit und Tod, p. 196. 4 v. Wlislocki, Vom wandernden Zigeunervolke, p. 189. 5 Samter, op. cit. p. 196. 8 Cox, Introduction to Folk-Lore, p. 18. » Train, Account of the Isle of Man, ii. 129. 540 THE HISTORY OF HUMAN MARRIAGE chap. Various explanations have been given of this custom. McLennan suggested that the English practice of hurling an old shoe after the bridegroom may be a relic of marriage by capture, being " a sham assault on the person carrying off the lady " j1 but whatever else may be said against this explanation, it is disproved by the fact that an old shoe was also thrown after the bride. Some writers believe that the main object of the custom was to ensure fecundity.2 In support of this explanation we are told that the wander- ing Gypsies of Transylvania throw old shoes or sandals on a newly-married pair when they enter their tent, expressly to enhance the fertility of the union. But this idea can hardly be primitive ; Dr. Aigremont's suggestion why an old shoe is a symbol of fecundity seems somewhat fantastic.3 Nor can the English belief that the throwing of an old shoe brings good luck be primitive. Zachariae maintains that it was in the first place meant to avert evil influences, such as malevolent spirits or the evil eye, and that its connection with the idea of good luck is secondary ; but to the question why it was looked upon as a safeguard against evil influences he can find no better answer than that spirits perhaps are afraid of leather.4 Dr. Ros6n has more recently expressed a similar opinion, with reference to the ceremonial use of shoes on various occasions ;5 but this theory is damaged by the fact that in some cases the shoes to be thrown are expressly stated to be wooden ones.6 Dr. Samter, again, 1 McLennan, Studies in Ancient History, p. 14 sq. n. 3. This suggestion was accepted by Lord Avebury (Origin of Civilisation, p. 103). 2 Sartori, ' Der Schuh im Volksglawben,' in Zeitschr. d. Vereins f. Volkskunde, iv. 153. Aigremont, Fuss- und Schuh-Symbolik und -Erotik, p. 55. Hartland, Primitive Paternity, i. 109. Heikel, Sandalion, p. 36 sq. 3 Aigremont, op. cit. p. 55 : " Der alte Schuh dient mit deutlicher Anspielung auf die Vulva als Symbol des ehelichen Glucks, der ehelichen Fruchtbarkeit (die vielbenutzte vulva)." 4 Zachariae, loc. cit. p. 138. See also Crooke, Popular Religion and Folk-Lore of Northern India, ii. 33 sqq. 5 Rosen, Om dodsrike och dodsbruk i fornnordisk religion, p. 152 sqq. 6 Ibid. pp. 141, 158. Strackerjan, Aberglaube und Sagen aus Oldenburg, i. 94. Cf. Samter, op. cit. p. 200. XXV MARRIAGE RITES 541 suggests that the hurling of a shoe after bride and bridegroom was originally intended as an offering to dangerous spirits but his attempt to show why an old shoe was chosen as a' particularly acceptable gift to the spirits is anything but convincing. It is very probable that the various rites in which shoes are used owe their origin to more than one idea. When the shoe is a charm against disease or evil spirits or the evil eye, or is used in funeral rites with a view to preventing a dead person from coming back to trouble t living,2 the reason for it seems to be that the shoe is closely associated with walking, and is therefore supposed to cause the evil influences or the dead person to pass by or to go away. But the throwing of a shoe after bride and bride- groom suggests the idea of protection rather than expulsion. It is thrown at them when they go somewhere-either when they start from their respective homes for the church, or from the church for the house where the marriage is celebrated, or from the latter place after the wedding break- fast. The throwing of it occurs side by side with practices apparently intended to protect them against evil influences from below. And it is also a custom found in England, Denmark,3 Germany,4 and elsewhere5 to throw a shoe or a slipper after a person who goes on a journey or to do business or a shooting. Brand says that in England it is accounted lucky by the vulgar to throw an old shoe after a person when they wish him to succeed in what he is going about.6 These facts suggest that the old shoe was meant to serve the persons in question as an extra magical pro- tection on their way, in addition to the shoes or boots they wore. They remind us of the extreme, care taken by 1 Samter, in Neue Jahrbucher filr das klassische Altertum, xix. 134 sqq. Idem, Geburt, Hochzeit rind Tod, p. 201 sqq. 2 Rosen, op. cit. p. 143 sqq. 3 Kristensen, op. cit. Tillaegsbind, iii. 130. 4 Sartori, in Zeitschr. d. Vereins f. Volkskunde, iv. 153. Stracker- jan, Aberglaube und Sagen aus Oldenburg, i. 94. 5 Rosen, op. cit. p. 141 (Sweden). Rochholz, Schweizersagen aus dem Aargau, i. 79. 6 Brand, op. cit. p. 672. 542 THE HISTORY OF HUMAN MARRIAGE ch. xxv Moorish brides and bridegrooms to prevent their slippers from falling off their feet and of the Chinese bride wearing her father's shoes for fear of contact with the ground. In Scotland it was the custom to wish brides and bridegrooms " a happy foot."1 1 Brand, op. cit. p. 673. CHAPTER XXVI MARRIAGE RITES (Concluded) Evils are averted from bride and bridegroom not only by positive rites but by abstinences of various kinds.1 In Morocco, when the bride is taken to her future home, she must not turn her head lest her husband should die,2 and when the bridegroom enters the room where the bride is waiting, he too must take care not to turn his head back, as otherwise evil spirits may enter with him.3 In the north-east of Scotland, when the bride set out, at the head of the party going to church, she was likewise on no account to look back, such an action entailing disaster of the worst kind during the married life.4 So also in many parts of Germany5 and Sweden6 the bride or the bridegroom as well must not look round when on the way to church or (in Sweden) at the nuptial ceremony ; if one of them does so, she or he is looking for another partner, that is, will become a widow 1 Cf. Crawley, op. cit. p. 342 sqq. 2 Westermarck, Marriage Ceremonies in Morocco, p. 172. 3 Ibid. p. 231. 4 Gregor, op. cit. p. 91. 5 Wuttke, op. cit. pp. 220, 221, 371. Schonwerth, Aus der Oberpfalz, i. 79 sq. Samter, Geburt, Hochzeit und Tod, p. 147. Grimm, Teutonic Mythology, iv. 1791, 1799. Tetzner, Die Slawen in Deutsch- land, p. 372. 6 Gaslander, ' Beskrifning, om Allmogens Sinnelag, etc. i Jon- kbpings Lahn och Wassbo Harad,' in Nyare Bidrag till kdnnedom om de svenska landsmdlen ock svenskt folklif, Bihang i. p. 276. Lith- berg, ' Brollopsseder pa Gottland,' in Fataburen, 1907. p. 173. 544 THE HISTORY OF HUMAN MARRIAGE chap. or widower, or some other misfortune will be the consequence. Nor must they stand so far apart that anyone can see between them ; otherwise they will pull two ways, or " evil people " or the devil or the evil eye will come between them and make the marriage unhappy.1 It is easy to understand the associations of ideas which have led to these beliefs ; but the real origin of the taboos undoubtedly lies in the feeling that persons like a bride and bridegroom, who are in a dangerous state or a source of dangers to others, must observe the utmost caution in all their doings. They should do as little as possible, and any mishap is an evil foreboding. In Italy neither of them dares to ex- tinguish the candle on the wedding night, because the one who does so will die first.2 In the Upper Palatinate a bride or bridegroom who drops anything on the way to church will soon become widowed and never get another partner.3 In Esthonia, if the bride falls on the way from her home, her first three or four children will die.4 There are taboos prohibiting bride and bridegroom from eating or drinking in public, from eating much, from eating certain victuals, or from eating anything at all-evidently with the object of preventing evil influences from entering the system by means of food.5 Such taboos are found in Europe,6 Morocco,7 and elsewhere.8 The old custom in 1 Grimm, op. cit. iv. 1804, 1817. Wuttke, op. cit. p. 371 sq. Birlinger, Voiksthumliches aus Schwaben, i. 479. Kuhn and Schwartz, Norddeutsche Sagen, Mdrchen und Gebrauche, p. 434. Temme, Die Volkssagen der Altmark, p. 74. Nicolovius (Loven), Folklifvet i Skytts H trad i Skine, p. 130. Lithberg, loc. cit. p. 173. Fernow, op. cit. p. 254. Raaf, op. cit. i. in. Norlind, op. cit. p. 90. 2 Placucci, Usi e pregiudizj dei contadini della Romagna, p. 60. Bacci, Usanze nuziali del contado della Valdelsa, p. 14. 3 Schonwerth, Aus der Oberpfalz, i. 81. See also Lithberg, in Fataburen, 1907, p. 173 (Gotland). 4 Grimm, op. cit. iv. 1842. 5 Cf. Crawley, op. cit. p. 343. h Schonwerth, op. cit. i. 97 ; Sartori, op. cit. i. 94 (Germany). Boecler-Kreutzwald, Der Ehsten abergldubische Gebrauche, p. 35. Piprek, Slawische Brautwerbungs- und Hochzeitsgebrauche, pp. 87, 95, 106, 128. Rajacsich, Das Leben, die Sitten und Gebrauche, der im Kaiserthume Oesterreich lebenden Sudslaven, pp. 157, 161, 181. 7 See Westermarck, op. cit. p. 323 n. 24 sq. 8 Dale, in Jour. Anthr. Inst. xxv. 199 sq. (Wabondei). XXVI MARRIAGE RITES 545 Russia is for the betrothed pair to eat nothing on the day of their marriage until after the ceremony.1 Among the Peruvian Indians bride and bridegroom fasted for two days before marriage, eating no salt, no pepper, and no flesh, and drinking none of the native wine.2 The young Macusi bridegroom-elect abstains from meat for some time before marriage.3 Among the Tlingit the couple are required to fast for two days, " in order to ensure domestic concord and happiness " ; at the expiration of that time they are allowed to partake of a little food, when a second fast of two days is added.4 The Masai believe that if the bride or bridegroom partakes of food at the wedding the taboo-breaker will have eruptions round the mouth.5 Among the Basoga in the Uganda Protectorate, again, the bride is present at the wedding meal in her new home, but is fed by a sister ; and from this time on to the end of her period of seclusion she is not allowed to touch food with her hands, but is fed like a child, her sister smearing the food over her mouth exactly as a mother does with her baby.6 The greatest possible inactivity is frequently required of a bride, and this rule may have to be observed for a considerable time after the marriage. Among the Nandi the bridal pair are for a whole month waited on by the bridegroom's mother, as it is unlawful for the bride during this period to work.7 Among the Wataveta of British East Africa, according to Mr. New, " brides are set apart for the first year as something almost too good for earth. . . . They are screened from vulgar sight, exempted from all household duties, and prohibited from all social intercourse with all of the other sex except their husbands. They are never left alone, are accompanied by some one wherever they may wish to go, and are not permitted to exert them- selves in the least; even in their short walks they creep 1 Romanoff, op. cit. p. 192. 2 de Herrera, General History of the Vast Continent and Islands of America, iv. 342. 3 Im Thurn, Among the Indians of Guiana, p. 222. 4 Bancroft, op. cit. i. 111. 5 Merker, op. cit. p. 48. « Roscoe, Northern Bantu, p. 210. See also ibid. p. 264 (Bateso). 7 Hollis, Nandi, p. 64. VOL. II N N 546 THE HISTORY OF HUMAN MARRIAGE chap. at a snail's pace, lest they should overstrain their muscles." They are not allowed to speak to outsiders, nor to be spoken to by them; and in this manner they are treated until they bear a child or the hope of such a desired event has passed away.1 'Silence is frequently imposed on brides. A Moorish bride must not speak at the wedding, or if she has something to say to any of the women attending her she must speak in a whisper; nor should she be spoken to in the presence of others. In Persia the bride " must not speak to any one " ;a and in Corea she is expected to keep absolute silence on the wedding day and in the nuptial chamber.3 Among various peoples women are forbidden to speak with anybody but their husbands for some time after marriage or, as among the Wataveta, until they have borne a child ; and in certain African tribes young wives are even forbidden to speak with their husbands.4 These taboos probably spring from different motives; Sir James G. Frazer is inclined to believe that the silence of the wife till her first child is bom rests on some superstitious belief touching her first pregnancy which as yet we do not under- stand.5 But it seems obvious that the silence of brides, at least, is connected with that fear of evil influences which is at the bottom of other abstinences. In Morocco the bridegroom, too, must not speak aloud ;6 and I was told that in one tribe, when he on entering the nuptial chamber bums some gum-benjamin to please the spirits of the place, the couple refrain from speaking for fear of the spirits who are then supposed to be about.7 At the same time, however, there can be no doubt that the silence of bride and bride- groom, and their extremely reserved behaviour in general, express or symbolise sexual bashfulness-a feeling which may easily be combined with superstitious fear. Some peoples consider it necessary for bride and bride- groom to keep awake.8 In Java they must not sleep 1 New, Life, Wanderings, &c. in Eastern Africa, p. 360 sq. 8 Customs and Manners of the Women of Persia, trans, by Atkinson, p. 70. 3 Griffis, Corea, p. 247 sq. 4 Frazer, Totemism and Exogamy, iv. 234 sqq. 5 Ibid. iv. 236 sq. 6 Westermarck, op. cit. p. 129. "^Ibid. p. 244. 8 Crawley, op. cit. p. 342 sq. Samter, Geburt, Hochzeit und Tod, p. 131. XXVI MARRIAGE RITES 547 during the night preceding the wedding, lest some grave misfortune should befall them.1 Among some Sumatrans the pair sit up all night in state.2 Among some Dyaks of Borneo they may not go to sleep, " else evil spirits would make them ill."3 Among the Nufors of Geelvink Bay, in Dutch New Guinea, bride and bridegroom are obliged to pass the first night or, according to another account, the first four nights after marriage sitting with their backs turned to each other without falling asleep, and if they grow drowsy, they are waked by their friends ; they believe that to keep awake on the wedding night is a means of ensuring a long and happy life.4 So also among the Khyoungtha of Chittagong " the bride and bridegroom are expected to sit up all night."6 At a Brahman wedding in India bride and bridegroom, when entering the nuptial chamber, find there a number of young women whose object is to prevent their sleeping during the rest of the night.6 In the Hi^ina, in Morocco, I was told that the bridegroom must not let the bride wait for him so long that she dozes and is then awakened and frightened by the noise he makes, because if this happens she may easily be struck by jnun and get a distorted face or lose her senses. Very frequently continence has to be observed for a shorter or longer time after marriage.7 Instances of this may be quoted from all parts of the world. Among the Tlingit bride and bridegroom were not only obliged to fast for some days but had to defer the consummation of the 1 Schmidt, Liebe und Ehe im alien und modernen Indien, p. 422. 8 Marsden, History of Sumatra, p. 269. 8 Perelaer, Ethnographische beschrijving der Dajaks, p. 53. * van Hasselt, ' Die Noeforezen,' in Zeitschr. f. Ethnol. viii. 181 sq. Guillemard, Cruise of the Marchesa, p. 389. Finsch, Neu-Guinea, p. 103. Frazer, Folk-Lore in the Old Testament, i. 511 sq. 6 Lewin, op. cit. p. 130. 6 Schmidt, Liebe und Ehe im alten und modernen Indien, p. 370. 7 Various instances of this have been given by Crawley, op. cit. p. 344 sqq. ; by v. Reitzenstein, ' DerKausalzusammenhang zwischen Geschlechtsverkehr und Empfangnis in Glaube und Brauch der Natur- und Kulturvolker,' in Zeitschr. f. Ethnol. xli. 677 sq. ; and especially by Frazer, Folk-Lore in the Old Testament, i. 497 sqq. 548 THE HISTORY OF HUMAN MARRIAGE chap. marriage for four weeks.1 Among the Thompson Indians of British Columbia " a newly-married couple, although sleeping under the same robe, were not supposed to have connubial connection until from two to seven nights- generally four nights-after coming together. The young wife slept with her husband, but still wore her maiden's breech-cloth. At last, having had connection with her husband, she arose before daybreak and repaired to the water, where she washed herself and spent the day in seclusion."2 Among the Nootka a newly-married pair used to refrain from intercourse for ten days after marriage.3 Among other North American Indians there were some who after being married lived in perfect continence for months or even a year. The object of this, they said, was to show that they had married " not because of lust, but purely through affection " ; and a young woman would even be pointed at if she were found with child within a year of marriage.4 Among the ancient Mexicans the married pair passed " four days in prayer and fasting, dressed in new habits, and adorned with certain ensigns of the gods of their devotion, without proceeding to any act of less decency, fearing that otherwise the punishment of heaven would fall upon them. . . . Until the fourth night the marriage was not consummated ; they believed it would have proved unlucky, if they had anticipated the period of consumma- tion."5 The Mazatek bridegroom had no sexual inter- course with his wife during the first fifteen days of his wedded life, both of them spending the time in fasting and penance.6 1 Holmberg, ' Ethnographische Skizzen uber die Volker des russischen Amerika,' in Acta Societatis Scientiarum Fennicae, iv. 314 sq. Bancroft, op. cit. i. in. 2 Teit, ' Thompson Indians of British Columbia,' in Publications of the Jesup North Pacific Expedition, i. 326. 3 Grant, quoted by Bancroft, op. cit. i. 198 n. 75. 4 Perrot, ' Memoir on the Manners, Customs, and Religion of the Savages of North America,' in Blair, Indian Tribes of the Upper Mississippi Valley and Region of the Great Lakes, i. 69. de Charlevoix, Voyage to North-America, ii. 38. Jesuit Relations, xix. 68 sq. ; xxxvii. 154 sq. Lafitau, Maeurs des sauvages ameriquains, i. 574 sq. 5 Clavigero, History of Mexico, i. 320 sq. " Bancroft, op. cit. ii. 261. XXVI MARRIAGE RITES 549 Among the Musos and Colimas of New Granada, if the matrimony happened to be consummated during the first three days, the woman was looked upon as lewd and wicked ; and the correct thing for the man was to lie by her " all that moon without consummation."1 A Mundrucu bridegroom keeps away from his bride on the wedding night.2 Among the Karaya on the Rio Araguaya in Brazil bride and bridegroom sleep for the first four nights on the same mat but with a wide interval between them, not being allowed to approach each other; on the fifth day the bride returns to her father's house, and the marriage is consummated only when the bridegroom after a few days' hunt comes back with his game.3 Among the Guaycurus the bride- groom spends the wedding night together with the bride in her father's house but without touching her.4 Among the Herero bride and bridegroom have no con- jugal intercourse on the wedding night, although they meet in secret.5 So also among the Bohindu, on the Congo, " les relations entre mari et femme ne sont pas permises imme- diatement apres le mariage."6 Among the Pangwe it is the custom for the bride, even though she sleeps with the bridegroom, not to allow him to have connection with her on the first night, because she is ashamed of being talked about.7 Among the Banyoro marriage was consummated in the evening of the second day after the wedding ceremony, when the guests had departed ;8 among the Baganda9 and Banyankole,10 on the third night. Of the Fors we are told that bride and bridegroom are not present at the wedding and do not see one another during the following week ; 1 de Herrera, op. cit. vi. 184. 2 v. Martius, Beitrdge zur Ethnographic Amerika's, i. 113. 3 Ehrenreich, Beitrdge zur Volkerkunde Brasiliens, p. 29. 4 do Prado,' Historia dos Indios Cavalleiros, da Nagao Guaycuru,' in O Patriota, 1814, no. 4, p. 20. 5 Bensen, quoted by Kohler, ' Das Recht der Herero,' in Zeitschr. f. vergl. Rechtswiss. xiv. 303. « Torday and Joyce, Les Bushongo, p. 271. 7 Tessmann, Die Pangwe, ii. 255 sq. 8 Roscoe, Northern Bantu, p. 40. 9 Idem, Baganda, p. 91. 10 Idem, Northern Bantu, p. 120. 550 THE HISTORY OF HUMAN MARRIAGE chap. only in the afternoon of the seventh day after the wedding they meet, but even then marriage is not consummated for some nights.1 In Usambara the young people meet at the house of a friend, where two bedsteads are placed one on either side of the room, with a big fire between ; and on these the bride and bridegroom recline in the sight of each other for four or five days without food.2 Among the Wataveta a bride sleeps with four little bridesmaids for five nights after marriage, and it is not till about a week afterwards that the bridegroom is allowed to take possession of her.3 Among the Nandi marriage may not be consummated the first night,4 and according to one account it is not so until the couple take possession of their own house, which usually happens on the third day.5 Among the Akamba the young wife sleeps in the husband's bed during the first night, but they may not have any intercourse; early next morning, while the others are still asleep, she gets up, sweeps out the hut, and makes up the fire for cooking, and then goes to bed again, because she is shy of her mother-in-law and wants to show herself to her as little as possible.6 In some Berber tribes of Morocco the bridegroom has his first intercourse with the bride only on the day after her arrival when all the guests have gone away, and if he is shy of his parents he waits till the evening before he goes to see her. But in one of the tribes with whose marriage customs I am acquainted, the Ait TamSldu on the southern slopes of the Great Atlas 1 Felkin, ' Notes on the For Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xiii. 228 sq. An Arab traveller, Muhammad ibn 'Umar, says that in Darfur " the marriage is seldom considered as completely celebrated until the seventh day, and never until after the third. A husband always shuns the insulting epithet of the impatient man. Each day of temperance is dedicated to some particular person : the first to the father of the bride, the second to the mother, and so on " {Travels of an Arab Merchant in Soudan, abridged from the French by Bayle St. John, p. 107). • Farler, ' Usambara Country in East Africa,' in Proceed. Roy. Geograph. Soc. N.S. i. 92. Baumann, op. cit. p. 133 sq. 8 Hollis, ' Notes on the History and Customs of the People of Taveta,' in Jour. African Soc. i. 115 sqq. 4 Idem, Nandi, p. 63. 6 Johnston, Uganda Protectorate, p. 878. 8 Lindblom Akamba, p. 73. XXVI MARRIAGE RITES 551 range, a bridegroom has not conjugal intercourse with his bride until the third night after she has been brought to his house, or, if his parents are alive, not until all the guests have gone away.1 In Egypt it is customary for husbands to deny themselves their conjugal rights duiing the first week after marriage with a virgin bride.2 Similar customs have been found among the- Australian aborigines. Strehlow tells us that among the Arunta and Loritja the young wife spends the first night with her mother and only the second night with her husband.3 Of the Narrinyeri we are told that "it is a point of decency for the couple not to sleep close to each other for the first two or three nights ; on the third or fourth night the man and his wife sleep together under the same rug. This arrangement is for the sake of decency. At the marriage many persons are present, sleeping in the same camp; so the newly married couple wait till they have moved off, and only a few relatives are left with them. They then often make a little hut for themselves."4 In the tribes of Western Victoria, where a marriage used to be observed with some- what elaborate ceremonies and attended by a large number of the friends and relatives of the couple, bride and bride- groom had for two months to sleep on opposite sides of the fire in the new hut which had been erected for them by the bridegroom's friends. During all this time they were not allowed to speak to or look at each other; and the observ- ance of these rules was ensured by a bachelor friend of the bridegroom who slept with him on one side of the fire, and the nearest unmarried female relative of the bridegroom who slept with the bride on the other side of it.5 Among the Euahlayi tribe bride and bridegroom camp for a moon he on one side of the fire and she on the other; till her grand- 1 Westermarck, Marriage Ceremonies in Morocco, p. 249 sqq. a Lane, Modern Egyptians, p. 508. 8 Strehlow, Die Aranda- und Loritja-Stdmme in Zentral-Australien, vol. iv. pt. i. 90, 100. 4 Taplin, Folklore, Manners, Customs, and Languages of the South Australian Aborigines, p. 35. B Dawson, Australian Aborigines, p. 31 sq. 552 THE HISTORY OF HUMAN MARRIAGE chap. mother tells her that she must camp on the same side of the fire as her betrothed, and as long as she lives be his faithful and obedient wife.1 Among the natives of Fraser Island, Queensland, bride and bridegroom do not come together for nearly two months after marriage.2 In Tikopia the consummation of marriage is postponed for several days.3 At Saa in Malanta, one of the Solomon Islands, although the girl goes to take up her abode in her new family, it is sometimes a long time before the marriage is consummated, " through the shyness of the bridegroom, though the parents encourage the young couple to be friendly."4 Among the Roro-speaking tribes of British New Guinea cohabitation is not supposed to begin for a few weeks after the wedding, the husband sleeping in the men's clubhouse and the wife in her father-in-law's house ; but it appears that in most cases intercourse takes place in or near the gardens soon after marriage. " It was stated that formerly it was not customary for a woman to have children until her garden was bearing well, that is to say, until she had been married from one to two years."5 We have previously seen how a newly-wedded pair spend the first night or the first four nights among the Nufors of Geelvink Bay.6 Continence for some time after the celebration of the marriage ceremony is the rule in many parts of the Indian Archipelago. Among some of the Dyaks of Dutch Borneo the couple may not come together on the first night of marriage, which is spent by the bride in the house of her mother or of some other female relative.7 In the tribes of the Barito valley, in the same island, a bridegroom usually observes continence during the first three nights, which he passes in the company of his friends, although he visits his wife from time to time to eat and drink with her and over- 1 Mrs. Langloh Parker, Euahlayi Tribe, p. 58. * Brough Smyth, Aborigines of Victoria, i. 84 n.* 8 Rivers, History of Melanesian Society, i. 309 sq. 4 Codrington, Melanesians, p. 239. 5 Seligman, Melanesians of British New Guinea, p. 270. 6 Supra, ii. 547. 7 Perelaer, op. cit. p. 53. XXVI MARRIAGE RITES 553 come her shyness.1 Among the Madurese and in some parts of Eastern Java the marriage is not consummated until the third night after the wedding;2 among the Sundanese of the same island not until the fourth day, the bride sitting during the first three nights beside the bride- groom silent and with downcast eyes;3 among the inhabi- tants of the Teng'ger Mountains in Java not until the fifth day ;4 and among the princes and the wealthy of that island sometimes not before three months have passed.5 In many villages of Central Sumatra some old women of the family keep watch over'the young couple for three nights, preventing them from having intercourse.6 The Achehnese, says Snouck Hurgronje, are averse to great intimacy in the beginning of married life ; during the first seven nights the couple always sleep under the surveillance of a Jeunganjo, or bride's duenna.7 In families of high rank in Southern Celebes, where the marriage ceremonies sometimes last for a month, the bride is all this time attended by eight old women, who sleep at night with the couple and prevent all intimacy between them.8 So also in Endeh, in the island of Flores, eight women sleep with them, though only for the first four nights after marriage, two of the women always keeping awake to prevent the bride and bride- groom from getting too near each other.9 In the Babar Islands the pair may after the performance of the marriage ceremony sleep in the same room, but for the first few nights the bride sleeps with some female relatives and the bride- groom with some male relatives ; should they, however, contrive to come together in the dark, the relatives go away 1 Schwaner, Borneo. Beschrijving van het stroomgebied van den Barito, i. 197. 2 Veth, Java, iv. 396. 3 Hardouin and Ritter, Java, p. 29. 4 Raffles, History of Java, i. 369. 5 Barrington d'Almeida, quoted by Frazer, Folk-Lore in the Old Testament, i. 510. G van Hasselt, Volksbeschrijving van Midden-Sumatra, p. 280. 7 Snouck Hurgronje, Achehnese, i. 325. See also Kruijt, Atjeh en de Atjehers, p. 193. 8 Matthes, op. cit. pp. 29, 35. 9 Roos, ' lets over Endeh,' in Tijdschrift voor indische taal-, land- en volkenkunde, xxiv. 525. 554 THE HISTORY OF HUMAN MARRIAGE chap. on the fourth or fifth day or even earlier.1 In the Kei Islands an old woman, or sometimes a child who is too young to run about, sleeps between the couple for three nights.2 Among the Tinguianes of Luzon the bride and bridegroom sleep on the wedding night with a space of two ells between them, in which lies a boy, six or eight years old ; and they are not even allowed to speak to each other.3 In Riukiu, an island between Formosa and Japan, they spend the wedding night in different rooms and live together only from the following day.4 In the Patani States of Malacca a bride and bridegroom are supposed to lie together for at least three nights after the wedding without having intercourse ; but this restric- tion is not necessary in the case of old men who are not married for the first time.5 Among the Kachins or Ching- paw of Upper Burma, " as a rule, cohabitation does not take place for some days after marriage, the only reason given being that the parties are ashamed."6 But according to Father Gilhodes, most young couples among them whom he has known have had no child before the third or fourth year after their marriage ; and when he has asked their relatives to explain, this, the answer has been that " les jeunes 6poux par honte n'ont pas de bonne heure des relations maritales."7 Among the Khyougtha, one of the hill tribes of Chittagong, a bridegroom refrains from consummating his marriage until he and his wife (sleeping apart) have for seven days eaten together seven times a day.8 In some of the wild hill tribes of Assam the young couple are forbidden to come together until they have slept under the same roof at least three nights without intercourse; but the prohibition is 1 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, p. 351. 2 Ibid. p. 236. 3 Blumentritt, Versuch einer Ethnographic der Philippinen, p. 38. 4 Iguchi, ' Wenig bekannte japanische Hochzeitsbrauche,' in Globus, Ixviii. 272. 8 Annandale and Robinson, Fasciculi Malayenses, ii. 75. 0 Scott and Hardiman, Gazetteer of Upper Burma, pt, i. vol. i. 407. 7 Gilhodes, ' Manage et Condition de la Femme chez les Katchins,' in Anthropos, viii. 372. 8 Lewin, op. cit. p. 130. XXVI MARRIAGE RITES 555 relaxed in the case of the marriage of widows.1 Among the Kacharis the interval of time between the bride's entering her husband's house and the consummation of the marriage sometimes amounts to five days.2 Among the Hindu Chutiyas of Assam it is said that " matrimonial etiquette requires postponement of consummation of the marriage for a week or so after the completion of the wedding cere- monial."3 Among the Mongsen, an Ao tribe of the Nagas, six men and six women sleep for six nights in the house of the newly-wedded pair, the men with the bridegroom and the women with the bride.4 In the Angami tribe of Nagas two women and one man sleep with the bride on the wedding night in the house of the bridegroom's parents while the bridegroom sleeps in the bachelors' hall; and it is only after nine or ten more days have passed and the high priest of the clan has completed the ceremony of marriage by sacrificing a chicken that bride and bridegroom are allowed to cohabit.5 In the Naga tribes of Manipur " marital intercourse within the dwelling house is prohibited for the initial nights of the married life. This prohibition extends in some cases over a period of a month, but is always less in cases of remarriage." 6 Among the Mikirs, when the bridegroom comes with his party to the bride's house, the bride prepares the bed for him ; " but if the lad is ashamed, he sends one of his garments to take his place in the bed."7 Among the Rajjhars, a caste of farm servants in the north of the Central Provinces, the couple sleep on their wedding night with a woman lying between them.8 Among various castes and tribes in Cochin the consummation of marriage is deferred until the night of the fourth day,9 or even till a week or two have passed after the wedding.10 In the Agasa 1 Hodson, 'The "Genna" amongst the Tribes of Assam,' in Jour. Anthr. Inst, xxxvi. 97. 2 Endle, Kachdris, p. 46. 3 Ibid. p. 95. 4 Gait, Census of India, 1891, Assam, i. 245. 8 Ibid. p. 239. 6 Hodson, Naga Tribes of Manipur, p. 87. ' Stack, Mikirs, p. 18. 8 Russell, Tribes and Castes of the Central Provinces of India, iv. 407. 9 Anantha Krishna Iyer, Cochin Tribes and Castes, ii. 139 (Vari- yars), 143 (Pisharotis), 192 (Nambuthiris). 10 Ibid. i. 61 (Ulladans). 556 THE HISTORY OF HUMAN MARRIAGE chap. caste in Mysore, if a girl is married when she has already attained womanhood, there must be an interval of fifteen days between marriage and consummation ; whilst the Kurubas insist on an interval of three months, as they consider it inauspicious that a child should be born within a year of the marriage.1 So also among the Kammas, a Telugu caste of Southern India, consummation does not take place till three months after the marriage ceremony, because it is held to be unlucky to have three heads of a family in a house- hold during the first year of marriage.2 The Wazirs of Bannu, in the Punjab, say that it was formerly their practice not to consummate the marriage for a long time after the wedding ceremony.3 In Baluchistan the bride often con- tinues to share her bed with a kinswoman for three nights after the celebration of the marriage ; " and when her husband eventually joins her, he is expected in some tribes to defer consummation for a considerable period."4 Continence for some time after marriage was prescribed for the Vedic householder by ancient Indian law. In the Grihyasutras he is frequently enjoined to abstain from conjugal intercourse during the first three nights.5 In one of these codes it is said that through three nights he and his wife shall sleep on the ground, be chaste, and avoid salt and pungent food ; and " between their sleeping-places a staff is interposed, which is anointed with perfumes and wrapped round with a garment or a thread."6 In other codes the period of continence imposed on the couple or recommended to them is extended to six or twelve nights, or four or six months, or even a year.7 One of the lawgivers tells us that an abstinence of three nights would be rewarded with the birth of a Vedic scholar of the ordinary type, an abstinence 1 Thyagaraja Aiyar, Census of India, 1911, vol. xxi. (Mysore) Report, p. 100. 2 Thurston, Castes and Tribes of Southern India, iii. 103. 3 Rose, Glossary of the Tribes and Castes of the Punjab and North- West Frontier Province, iii. 507. 4 Bray, Census of India, 1911, vol. iv. (Baluchistan) Report, p. 113. 5 Grihya-Sutras, i. 43, 384; ii. 48. 0 Ibid. ii. 267. Winternitz, Das altindische Hochzeitsrituell, p. 25. 7 Grihya-Sutras, i. 171,286. Haas, loc. cit. p. 330 sq. Winternitz, op. cit. p. 86. XXVI MARRIAGE RITES 557 of twelve nights with that of a really first class Vedic scholar, an abstinence of four months with that of a Brahman of a still more exalted rank, an abstinence of six months with that of a Rishi, or saint, and an abstinence of a whole year with that of a god.1 Professor v. Schroeder believes that the custom of practis- ing continence for some time after marriage may be traced back to the primitive period of the Indo-European race.2 It certainly exists, or has in recent times existed, in many parts of Europe.3 In Herzegovina and Montenegro, accord- ing to Bogisic, the bride sleeps on the marriage night fully dressed with the bridesman, generally one of the husband's brothers, and passes the following nights with her sisters- in-law ; and this separation of husband and wife may last a long time, since the husband's mother alone has the right to decide when it shall end.4 Miss Durham was assured by an old-style Montenegrin from the mountains that the longer the consummation of the marriage was put off the more honourable it was, and that he knew brides who had lived for many months in a separate house with their brothers-in-law.5 At Risano, in Dalmatia, it was formerly the custom for the bride to sleep the first three nights with the two bridesmen.6 Among the Serbs of the Banat she sleeps the first night with the bridesman, who is generally a small child.7 Among Roman Catholic towns-people in Upper Albania a well-educated bride is supposed to resist the bridegroom's advances for the first three nights, but custom does not allow her to do so for a longer period.8 Von Hahn says that in Albania the bride sleeps with the women and the bridegroom with his friends.9 1 Winternitz, op. cit. p. 86. 2 v. Schroeder, op. cit. p. 192 sqq. 3 Ibid. p. 193 sqq. 4 Demelic, Le droit coutumier des slaves meridionaux d'apr&s les recherches de M. V. Bogisit, quoted by Frazer, Folk-Lore in the Old Testament, i. 504. See also Krauss, op. cit. p. 456. 5 According to manuscript notes which Miss Durham has kindly placed at the author's disposal. 6 Krauss, op. cit. p. 456. 7 Piprek, op. cit. p. 135. 8 Gopcevic, Oberalbanien und seine Liga, p. 456. 8 v. Hahn, Albanesische Studien, p. 146. 558 THE HISTORY OF HUMAN MARRIAGE chap. In the year 1738 the parson of Daxlanden, near Karlsruhe, complained of the custom according to which the brides- man and bridesmaid slept together with the bride and bridegroom during the first night in order to prevent their having conjugal intercourse.1 In various parts of Germany and Switzerland continence is observed for three nights after marriage, which are frequently known as the " Tobias nights " ; it is believed that otherwise the wedded life of the couple would be unlucky, whereas if they abstain from intercourse the devil will not be able to do any harm.2 In some parts of France also continence is or recently was practised for three3 or two4 nights after marriage or on the first night,5 and in several places in Brittany the bride is during this night entrusted to the supervision of the best- man and the bridesmaid.6 The first night is dedicated sometimes to God and sometimes to the Holy Virgin, the second to the Holy Virgin or to Saint Joseph, the third to the husband's patron saint. In Lower Brittany the con- summation of marriage is sometimes put off for a fortnight or even longer.7 Among the peasants of Romagna bride and bridegroom do not sleep together on the marriage night because some of the guests remain overnight and the house is small.8 In the latter part of the eighteenth century Lord Hailes was informed that abstinence on the wedding night was " still observed by the vulgar in some parts of Scot- 1 Meyer, Badisches Volksleben im neunzehnten Jahrhundert, p. 272. 2 Wuttke, op. cit. p. 375. Simrock, Handbuch der Deutschen Mythologie, p. 600 sq. Schonwerth, Aus der Oberpfalz, i. 112. Birlinger, Volksthumliches aus Schwaben,i. 479; ii. 334, 354. Meyer, Badisches Volksleben, p. 319. Reiser, Sagen, Gebrduche und Sprich- wbrter des Allgdus, ii. 276. 3 Hanauer, Les paysans d'Alsace au Moyen-Age, p. 137 n. 2. Ogee, Dictionnaire historique et geographique de la province de Bretagne, ii. 890 (Scaer), 932 (Tr6meoc). 4 S6billot, Coutumes populaires de la Haute-Bretagne, p. 132 (Matignon). 6 Ibid. p. 137. Ida v. Diiringsfeld and v. Reinsberg-Diiringsfeld, op. cit. p. 250 (Sachelay in the department of Seine-et-C)ise). c de Nore, op. cit. p. 195. Sebillot, op. cit. p. 137. 7 Sebillot, op. cit. p. 132. 8 Placucci, op. cit. p. 59 sq. XXVI MARRIAGE RITES 559 land."1 Similar customs have also been found among some non-Aryan peoples belonging to the former Russian empire. Among the Esthonians it was formerly the rule that a husband might not unloose his wife's girdle nor take any other liberty with her on the wedding night.2 Pallas was told that among the Samoyed a bride remained un- touched for a whole month, although the couple slept on the same bed and not, as was the case among the Ostyak, on separate furs.3 Now it may be argued that the continence observed after marriage in so many parts of Europe is not a survival of an ancient pagan custom, but is due to the teaching of the Christian Church. A decree of the alleged fourth Council of Carthage, said to have been held in the year 398/ enacted that when the bridegroom and bride have received the benediction, they shall remain that same night in a state of virginity out of reverence for that benediction.5 This enactment was received into the canon law; and by sub- sequent enactments the period of chastity which bride and bridegroom were required or recommended to observe after marriage was extended from one to two or three nights, often with special reference to the example set by Tobias, who by advice of the archangel Raphael abstained from carnal intercourse with his wife Sarah for three nights.6 It is conceivable that the same horror of sexual defile- 1 Lord Hailes, Annals of Scotland, iii. 15 n. * 8 Boeder-Kreutzwald, op. cit. p. 25. 8 Pallas, Reise durch verschiedene Provinzen des Russischen Reichs, iii. 79 sq. According to Lepechin (quoted by v. Stenin, ' Das Gewohnheitsrecht der Samojeden,' in Globus, lx. 172), a Samoyed bridegroom spends the first night vzith the bride without touching her. 4 See v. Hefele, Conciliengeschichte, ii. 68 sq. 6 Concilium Carthaginense quartum, 13 (Migne, Patrologiae cursus, Ixxxiv. 201). 0 Lord Hailes, op. cit. iii. 15 sq. Veuillot, Le droit du seigneur, p. 191 sq. Schmidt, Jus primae noctis, p. 153. Frazer, Folk-Lore in the Old Testament, i. 497 sq. Benedictus, Capitularium collectio, 463 (Migne, op. cit. xcvii. 859) : " Et biduo vel triduo oratiqnibus vacent et castitatem custodiant, ut bonae soboles generentur et Domino suis in actibus placeant." The archangel Raphael's advice and the continence of Tobias are related in the Vulgate, Liber Tobiae, chs. 6, 8. 560 THE HISTORY OF HUMAN MARRIAGE chap. ment as induced the Church to prescribe continence in connection with various other religious acts also might independently have led to the decree imposing conti- nence in connection with the sacrament of marriage ; but it seems more probable that this decree and the subsequent appeal to the archangel's advice to Tobias only gave religious sanction and scriptural support to an old pagan custom which was highly congenial to the ascetic tendencies of the Church. A similar view has recently been advocated by Sir James Frazer with much fulness of detail.1 This view derives support, first, from the fact that the rule of continence after marriage is not only found among pagan peoples in all parts of the world but existed among the Vedic Aryans; and, secondly, from its persistence in European folk-custom, which suggests a deeper foundation than ecclesiastical injunctions alone. At the same time it must be admitted that the custom of deferring the consummation of the marriage for a time may have a different origin in different cases. Sometimes it is attributed to resistance on the part of the bride, and there may be some truth in this. In many tribes in Morocco it is usual, and seems even to be the proper thing, for a bride to make resistance ; and that this may be of a serious character is apparent from the fact that the bridegroom's best-man often remains outside the door of the room or tent ready to step in, if the bridegroom needs his assistance, and tie up the bride.2 More frequently, however, the custom of observing continence is ascribed to sexual bashfulness in the bridegroom or in both parties ; and when intercourse is said to be postponed till the guests have gone away, and in some other cases quoted above as well, this seems a very natural explanation of the postponement. The objection to having a child within a year of the marriage or at a still later period may also perhaps have something to do with sexual modesty, though there are probably other reasons for it. In his description of the Greenlanders in the eigh- 1 Frazer, op. cit. i. 505. 2 Westermarck, Marriage Ceremonies in Morocco, pp. 235, 239, 242, 244, 248, 250, 251, 253. XXVI MARRIAGE RITES 561 teenth century Egede says that if married couples among them had children before a year was past, or if they had large families, they were blamed and compared to dogs.1 Yet it can hardly be doubted that the rule of continence to which bride and bridegroom are subject, like other taboos imposed upon them, is mainly the outcome of superstitious fear.2 If it is considered dangerous for them to speak or eat or sleep, it is not surprising if sexual intercourse between them is supposed to be fraught with danger. In most cases of compulsory continence we are not told of any such reason for it; but sometimes it is said that the observance of it is essential for a happy wedded life, or that it will make the offspring good, or that it will prevent the devil from doing harm,3 whilst we read in the ' Book of Tobit ' that a wicked demon, named Asmodaeus, out of spite and jealousy slew the woman's seven bridegrooms as soon as they had gone in to her on the wedding night.4 With reference to the Vedic practice Oldenberg says that its original meaning, though it was obviously no longer under- stood by the people, must be sought in the fear of spirits who, in the act of copulation, might slip into the woman and en- danger her offspring or might even themselves impregnate her, but were supposed to be misled by a pretence of omitting the consummation of marriage.5 The idea that evil spirits may slip into women when they have sexual intercourse is familiar to Muhammadans even at the present day. I was told in Morocco that it is always necessary for the husband before having intercourse with his wife to say bismilldh, " In the name of God," lest the devil should enter the woman and make the child a villain; and this belief has the support of the Muhammadan tradition.6 Sir James G. Frazer, again, 1 Egede, Description of Greenland, p. 143 note. 2 Cf. Crawley, op. cit. p. 343; Sartori, op. cit. i. no; Frazer, Folk-Lore in the Old Testament, i. 520. 3 See supra, ii. 548, 556 sqq. 4 Liber Tobice, chs. 3, 6. Book of Tobit, chs. 3, 6. 6 Oldenberg, op. cit. p. 271. c El-Bukhari, Sahih, Ixvii. 66, French translation by Houdas and Mamais, iii. 578 :-" D'aprds Ibn-'Abbas, le Prophete a dit: ' Ehbien! si 1'un de vous, voulant avoir commerce avec sa femme, disait: " Au nom de Dieu. 0 mon Dieu, dcarte de moi le ddmon ; vol. n 00 562 THE HISTORY OF HUMAN MARRIAGE chaf. observes that " the intention of the custom is perhaps not so much to deceive the demons by pretending that the marriage is not to be consummated, as to leave them free scope for making love to the bride in the absence of the bridegroom " j1 and a similar explanation has previously, and in more positive terms, been suggested by Baron von Reitzenstein.2 Yet I fail to find any evidence for it in existing facts. Frazer appeals to the story of Tobias and Sarah, as it is related in the Vulgate ; in the Greek text of the Septuagint and in the English translation, which follows this text, no mention is made of the nights of continence. He observes that in that story the practice of continence is enjoined for the purpose of defeating the jealous devil, who had already massacred Sarah's seven earlier husbands, and who would have killed her eighth, if that bold man had not received timely warning and prudently abstained from exercising his conjugal rights for three nights after marriage. " The inference suggested by the narrative," Frazer adds, " is that by this abstinence Tobias left the field open to his spiritual rival, who, after enjoying the bride undisturbed for three nights, was content to pass her on to her lawful husband for the term of his natural life."3 But the story tells us that the archangel Raphael advised Tobias not only to observe continence and give himself up to prayer, but, on the first night, to bum the heart and liver of a fish by which he had been attacked on his way to Ecbatana and which he had caught and killed by the advice of the angel, and to make a smoke with it, in order to put the demon to flight; and we are also told that the demon no sooner smelled the ill-savour than he fled away into the utmost parts of Egypt, where the angel bound him fast. This incident is found in the Greek text also,4 and, as Frazer himself points out,5 we may conclude that it also belonged to the original dcarte le d6mon du fruit de notre union," et qu'ensuite le Destin ou la Predestination fit naitre un enfant de ces relations, le demon ne pourrait jamais nuire a cet enfant.' " 1 Frazer, op. cit. i. 520. 2 Reitzenstein, in Zeitschr.f. Ethnol. xli. 656, 661, 676. 3 Frazer, op. cit. i. 519 sq. 4 Book of Tobit, chs. 6, 8. 6 Frazer, op. cit. i. 518. XXVI MARRIAGE RITES 563 version of the story. But if Tobias thus drove the demon away on the first night, how can it be inferred that he let him enjoy the bride undisturbed for three nights ? We may suppose that the practice of continence was another pre- cautionary measure calculated to safeguard Tobias-several different methods of averting evil influences are commonly resorted to at weddings-but I can find no indication that it was meant to serve as a sort of invitation to the demon to take possession of the bride. I am of opinion that anthropologists are often apt to look for too much reasoning at the bottom of primitive customs. Many of these are based on vague feelings rather than on definite ideas. Sexual intercourse, which is in many cases regarded as a mysterious cause of evil, is considered,dangerous to bride and bridegroom, who are particularly exposed or sensitive to evil influences, and is therefore abstained from while the danger lasts. But to speculate on the specific nature of the danger and of the evil influences causing it, or on the way in which continence is supposed to avert that danger, is, in the absence of direct evidence, the more- precarious, as it is very doubtful whether the people them- selves have any clear theory on the subject.1 The prophylactic observances which play such an impor- tant part at marriages in all parts of the world raise the interesting question, Why are bride and bridegroom supposed to be in a dangerous condition, and why is the bride con- sidered dangerous to others ? In order to answer these questions, at least so far as Moorish ideas are concerned, 1 Since the above was already in type Dr. Karsten has, in his Contributions to the Sociology of the Indian Tribes of Ecuador (Acta Academiae Aboensis, Humaniora, i. no. 3), p. 72, published a fact which is in full agreement with the theory of Reitzenstein and Frazer. Among the Indians of Canelos the young couple do not spend the first night after the nuptial feast together because they believe that, if they did, the husband would die. The supai, a most dangerous demon, claims the right to spend that night with the bride, and the right is voluntarily'ceded to him by the husband. Even the following night is critical, since the demon wants to continue to have the woman for himself and is jealous of her husband ; and sometimes the danger is not supposed to be wholly over until 0 0 2 564 THE HISTORY OF HUMAN MARRIAGE chap. I have compared the rites practised in Morocco at weddings where the bridegroom is a bachelor and the bride a maiden with those practised in cases where either bride or bride- groom or both have been married before. This comparison showed that the rites of a purificatory or protective kind to which the bride or bridegroom is subject depend on the circumstance whether she or he, but not both parties, have been married before or not. A bridegroom who is a bachelor is subject to the same ceremonies whether the bride be a maiden, a widow, or a divorced wife, whereas these cere- monies are omitted in the case of a bridegroom who has or has had another wife, quite independently of the state of the bride ; and a bride who has not been married before is subject to the same ceremonies whether the bridegroom be a bachelor, a widower, or a polygamist, whereas these ceremonies are, if not altogether done away with, at all events much reduced in the case of a bride who is a widow or a divorced wife, quite independently of the state of the bridegroom. From all this I conclude that, even though some of the purificatory and protective marriage rites have sprung from fear of hymeneal blood or from the idea that the bride may carry evil with her both as a newcomer into the bridegroom's household and in her capacity of being a woman, the bulk of these rites are due to the fact that the person who is subject to them is bride or bridegroom for the first time. She or he enters into a new state, the wedding is, to use a phrase coined by M. van Gennep, a rite de passage ; and to pass into a new condition or to do a thing for the two or three children have been born in the marriage (ibid. p. 74). Although this statement is undoubtedly a valuable contribution to the psychology of the subject, we must not assume that the same idea as that held by the Indians of Canelos underlies all the in- numerable cases of continence required of newly-married people, and I find no reason to alter the general views expressed in the text. It seems strange that the Canelos Indian should so readily allow the supai to have intercourse with his wife, as he believes that " the consequence of intercourse with the demon will be that the woman will either fall ill and die or give birth to a monstrous child " (ibid. p. 69). But in a private discussion on the subject Dr. Karsten affirmed that he was certain of the accuracy of his statement. XXVI MARRIAGE RITES 565 first time is not only in this, but in many other cases, con- sidered to be attended with danger. But it must in addition be noticed that in the present instance the nature of the act itself which is sanctioned by the wedding is apt to increase the supposed peril, sexual intercourse, as we have seen, being looked upon as defiling and under certain circumstances as a mysterious cause of evil.1 I venture to believe that all this in the main holds true not only of the Moors but of other peoples as well. Evidence from a detailed comparison between the prophylactic and cathartic rites practised at first and those practised at second marriages is unfortunately wanting. But we often hear that widows or divorced wives are married with less formality than girls.2 In our classification. of marriage rites we have hitherto made a distinction between protective or purificatory rites and such as are held to result in more positive benefits. But we have seen that in many cases it is difficult or im- possible to draw any such definite distinction. The same substance or action may sometimes be looked upon as a means of averting or expelling evil, sometimes as a source of good, and sometimes as both at once. Even when a certain rite is originally performed for a prophylactic or cathartic purpose it may easily come to be regarded as a means of securing positive benefits, and, on the other hand, 1 See Westermarck, Marriage Ceremonies in Morocco, p. 329 sqq. 2 de Piedrahita, Historia general de las conquistas del nuevo reyno de Granada, p. 23 (Mozcos). Henry, Les Bambara, p. 199 n. 1. Hurel, ' Religion et Vie domestique des Bakerewe,' in Anthropos, vi. 292. Munzinger, Ueber die Sitten und das Recht der Bogos, p. 62. Ranga Rao, ' Yanddis of the Nellore District,' in the Madras Govern- ment Museum's Bulletin, iv. 100. Sturrock, Madras District Manuals : South Canara, i. 161 (Bants). Hislop, Aboriginal Tribes of the Central Provinces, p. 3 (hill tribes). Dehon, ' Religion and Customs of the Uraons,' in Memoirs Asiatic Soc. Bengal, i. 165. Shaw, ' Inhabitants of the Hills near Rdjamahall,' in Asiatick Researches, iv. 71. Endle, Kachdris, pp. 89 (Morans), 96 (Chutiyas). Logan, ' Orang Binua of Johore,' in Jour. Indian Archipelago, i. 270. Plutarch, Questiones^ Romanae, 105. Placucci, op. cit. p. 62 (Romagna). 566 THE HISTORY OF HUMAN MARRIAGE chap. a rite which is in the hrst place meant to bring such benefits may come to be looked upon as a rite of purification. A mixture of the two kinds of motives is found in certain classes of practices relating to the celebration of a marriage which still have to be considered. It is noticeable in the selection of the time when a mar- riage is to be celebrated. Many peoples consider it a matter of the utmost importance to find out the right day for the wedding, by consulting astrologers or otherwise.1 Very frequently, also, a certain period or day is chosen, or avoided, because it is regarded as lucky, or unlucky, quite indepen- dently of any individual circumstances. In Ireland it was an ancient idea that people should not marry in the autumn in " binding " time, as they were sure to be unbound afterwards.2 The Chinese and Hindus have not only their fortunate months, mentioned above in another connection,3 but also their unlucky months. In China no marriages, except in cases of extreme urgency, take place in the ninth month, which is regarded as very unpropitious.4 Sir W. H. Sleeman writes, " Certain it is that no Hindoo will have a marriage in his family during the four months of the rainy season; for among eighty millions of souls, not one doubts that the Great Preserver of the universe is, during these 1 Davy, Account of the Interior of Ceylon, p. 285 ; Sirr, Ceylon and the Cingalese, ii. 169 ; Forbes, Eleven Years in Ceylon, i. 328 (Sinhalese). Winternitz, Das altindische Hochzeitsrituell, p. 29. Schmidt, Liebe und Ehe im alten und modernen Indien, p. 360 sqq. Anantha Krishna Iyer, op. cit. ii. 185 (Brahmans of Malabar, Cochin, and Travancore). Gopal Panikkar, op. cit. p. 26 (Nayars). Moegling, Coorg Memoirs, p. 37. Kearns, Marriage Ceremonies of the Hindus of South India, p. 54 (Naickers). Forsyth, Highlands of Central India, p. 149 (Gonds and Korkiis). Lewin, op. cit. p. 126 sq. (Khyoungtha). Shway Yoe, The Burman, p. 64. Young, Kingdom of the Yellow Robe, p. 94 (Siam). Taupin, reviewed in L' Anthropologic, ii. 488 (Laos). Wells Williams, Middle Kingdom, i. 785 (China). Kiichler, ' Marriage in Japan,' in Trans. Asiatic Soc. Japan, xiii. 121.' Das, ' Marriage Customs of Tibet,' in Jour. Asiatic Soc. Bengal, vol. Ixii. pt. iii. 16. Bergmann, Nomadische Streifereien unter den Kalmiiken, iii. 147 sq. Waitz, Anthropologic der Naturvblker, iv. 132 (ancient Mexicans). 2 Wood-Martin, Traces of the Elder Faiths of Ireland, ii. 39. 3 Supra, i. 94. 4 Gray, China, i. 187 sq. XXVI MARRIAGE RITES 567 four months, down on a visit to Rajah Bull, and, conse- quently, unable to bless the contract with his presence."1 The Romans considered May and the first half of June unlucky for marriages ;2 and marriages in May are still avoided,or have been so up to recent times, in Italy,3 Greece,4 France,6 Scotland,6 Germany,7 and Bohemia.8 To marry in May is said to be a cause of death9 or madness.10 In Sicily11 and France12 August, also, is an unlucky month; it is supposed that those who marry then will be always jealous. The Sicilians say, " La spusa majulina Nun si godi la curtina " ; and " La spusa agustina Si la porta la lavina." In Sardinia marriages are avoided in July.13 No Jewish marriage takes place between Passover and Pentecost, and Mr. Abrahams thinks " there can be little doubt that we are here in presence of a variant of the Roman super- stition which forbade marriages in May."14 In Morocco marriages are avoided in Moharram, the first month of the Muhammadan year, at least by shereefs ; and the Egyptians likewise consider it unlucky to make a marriage contract in that month.16 1 Sleeman, Rambles and Recollections of an Indian Official, i. 50. 2 Ovidius, Fasti, v. 490 ; vi. 225. 8 Bacci, Usanze nuziali del contado della Valdelsa, p. 12. Frescura, ' Fra i Cimbri dei Sette Comuni Vicentini,' in Archivio per lo studio delle tradizioni popolari, xvii. 46. Pitrd, Usi e costumi credenze e pregiudizi del popolo siciliano, ii. 48. 4 Rodd, Customs and Lore of Modern Greece, p. 90. 6 Thiers, op. cit. iv. 429. de Nore, op. cit. p. 242. S6billot, . Coutumes populaires de la Haute-Bretagne, p. 112 sq. 0 Brand, op. cit. p. 387. Dalyell, op. cit. p. 287. Rogers, Scotland Social and Domestic, p. 112. Gregor, Notes on the Folk-Lore of the North-East of Scotland, p. 88. 7 Sartori, op. cit. i. 60. Meyer, Deutsche Volkskunde, p. 174. 8 Wuttke, op. cit. p. 368. 9 Ibid. p. 368. " O' marriages in May Bairns die in decay " (Auchterderran; Simpkins, County Folk-Lore, vol. vii. Fife, p. 152). 10 Bacci, op. cit. p. 12. de Nore, op. cit. p. 242. 11 Pitrd, op. cit. ii. 48. 12 de Nore, op. cit. p. 242. S6billot, op. cit. p. 112 sq. 13 Faggiani, in Provenzal, Usanze e feste del popolo italiano, p. 232. 14 Abrahams, Jewish Life in the Middle Ages, p. 184. 16 Lane, Account of the Manners and Customs of the Modern Egyptians, p. 432. 568 THE HISTORY OF HUMAN MARRIAGE chap. The choice of the time when marriages are celebrated is often influenced by the moon. The Esthonians marry when the moon is new, because they believe that it will make the marriage happier and prevent the women from aging.1 In some parts of Sweden marrying at the new moon was supposed to make the couple wealthy.2 In Shetland all marriages must commence with the new moon, since otherwise the marriage will be an unlucky one.8 In the Orkney Islands the people never marry but when the moon is growing, and some even wish for a flowing tide ; to marry when the moon is waning is believed to make the marriage-bed barren.4 In the southern districts of Scotland the inhabitants likewise preferred the increase of the moon for their marriages, whilst in other parts auspicious conse- quences were anticipated from their celebration at full moon.5 In various parts of Germany marriages are cele- brated during the increase of the moon, so that there shall be no want, or when it is at full, so that everything shall be in full.6 The ancient Greeks7 and Hindus,8 also, married at the time of the crescent moon. The Kachins of Burma never marry when the moon is on the wane, for fear of shortening their lives thereby.9 The Saorias of the Raj- mahal Hills believe that if marriages were consummated during the dark o' the moon, they might not be fruitful, and that there might be general bad luck and maladies as well.10 1 Boecler-Kreutzwald, op. cit. p. 24. 2 Gaslander, loo. cit. p. 276. Raaf, op. cit. i. no. 3 Black, County Folk-Lore, vol. iii. Orkney and Shetland Islands, p. 207 sq. * Ibid. p. 214. 5 Dalyell, op. cit. p. 285. See also Gregor, op. cit. p. 88. 6 Wuttke, op. cit. p. 368. Sartori, op. cit. i. 60 n. 4. Simrock, op. cit. p. 600. Tetzner, Die Slawen in Deutschland, p. 373. Schon- werth, Aus der Oberpfalz, i. 91. Kuhn and Schwartz, Norddeutsche Sagen, Marchen und Gebrauche, p. 434 sq. The custom of marrying when the moon is growing is also found in Denmark (Kristensen, op. cit. Tillaegsbind, iv. 59). 7 Becker-Goll, op. cit. iii. 360. 8 Grihy a-Sutras, i. 164, 277. 8 Gilhodes, in Anthropos, viii. 367. 10 Bainbridge, ' Saorias of the Rajmahal Hills,' in Memoirs Asiatic Soc. Bengal, ii. 50. XXVI MARRIAGE RITES 569 Certain days of the week are chosen for the celebration of marriages because regarded as auspicious, whereas others are avoided as ill-omened; but these " lucky " and " un- lucky " days vary in different countries and sometimes even within a comparatively small area.1 Among Teutonic peoples Tuesday and Thursday used to be, or still are, favourite days for weddings, according to E. H. Meyer probably because they were dedicated to the marriage gods Tiu or Ziu and Donar.2 In some parts of Germany Friday is considered a lucky day for celebrating marriages,3 whereas in other parts, under the influence of Christian ideas, it is regarded as inauspicious.4 Sunday is also chosen,5 and in various parts of Sweden it has become the regular day for weddings.6 In the north-east of Scotland the marriage day was either Tuesday or Thursday, more rarely Saturday.7 In the Orkneys and Shetland Thursday is the day generally 1 Cf. Tetzner, op. cit. p. 82 (Lithuania). 2 Meyer, Deutsche Volkskunde, p. 174. Sartori, op. cit. i. 60 sq. Wuttke, op. cit. pp. 60, 61, 368. Simrock, op. cit. p. 600. Sepp, Volkerbrauch bei Hochzeit, Geburt und Tod, p. 56. Woeste, ' Aber- glaube und Gebrauche in Siidwestfalen,' in Jahrbuch d. Vereins f. niederdeutsche Sprachforschung, 1877, p. 135. Lithberg, ' Brollops- seder pci Gottland,' in Fataburen, 1906, p. 79 sq. Ullberg, ' Brol- lopsseder i Sodra Sibbo,' in Hembygden, viii.-ix. 144 ; Lindroos och Andersson, ' Ett brollop i Pellinge,' ibid. 1910, p. 156 (Tuesday) ; Nikander, according to manuscript notes relating to Parna (Thurs- day) ; these three statements refer to Swedish-speaking communities in Finland. In some parts of Germany, especially where Slavonic influence has made itself felt, no marriages are celebrated on Thursday (Wuttke, op. cit. p. 60I. 3 Wuttke, op. cit. pp. 61, 368. Sartori, op. cit. i. 61. Sepp, op. cit. p. 56. Bartsch, Sagen, Mdrchen und Gebrduche aus Mecklen- burg, ii. 59. Schell, ' Bergische Hochzeitsgebrauche,' in Zeitschr. d. Vereins f. Volkskunde, x. 163. Woeste, loc. cit. p. 135. Tetzner, op. cit. p. 373. 4 Wuttke, op. cit. pp. 61, 368. Sartori, op. cit. i. 61. Simrock, op. cit. p. 600. Schonwerth, op. cit. i. 92. According to a Nether- landish belief, it is not good to be married on a Friday (Thorpe, Northern Mythology, iii. 330). 6 Wuttke, op. cit. pp. 58, 368. Sartori, op. cit. i. 61. 8 Lithberg, loc. cit. p. 79 sq. Wikman, ' Frieri, fbrlofning och brollop i Delsbo,' in Fataburen, 1913, p. 76. Djurklou, Ur Nerikes folksprak och folklif, p. 47. 7 Gregor, op. cit. p. 88. 570 THE HISTORY OF HUMAN MARRIAGE chaf. fixed for marriage ; but in the Orkneys people are also inclined to marry on Fridays,1 and the same has been the case in Scotland 2-indeed, at St. Monans it is very rare to see a wedding on any other day, although Friday is generally regarded as ominous of evil.3 According to an English tradition, it is unlucky to marry on a Thursday, perhaps because Thor is partly identified with the devil, and a couple married on a Friday are doomed to lead a cat-and-dog life.4 The brides of the Elizabethan dramas are usually represented as married on Sundays.5 Speaking of this custom, Jeaffreson remarks :-•" In our feudal times and long after the Reformation, Sunday was of all days of the week the favourite one for marriages. . . . Long after the theatres of London had been closed on Sundays, the day of rest was the chief day for weddings with Londoners of every social class."6 In Wales Saturday is considered a lucky day for a wedding.7 In France in the seventeenth century a man avoided marrying on a Wednesday for fear of becoming a cuckold.8 There is still the saying, "On ne se marie pas le mercredi, De peur d'avoir nom Jean-Jeudi."9 But in Berry the man who marries on a Thursday is said sooner or later to become Jean-Jeudi, that is, a cornard.10 In Upper Brittany no marriage is celebrated on a Thursday, because that was the day when the devil married his mother.11 As regards Friday opinion differs in France, as elsewhere : in some places weddings are frequently held on that day, whereas in other places it is regarded as an unfortunate day for marrying ;12 Thiers states that it was considered so in the 1 Black, op. cit. p. 214. 2 Dalyell, op. cit. p. 285. 3 Simpkins, op. cit. p. 163. 4 Henderson, op. cit. p. 33. For the English idea that it is un- lucky to marry on a Friday, see also Mrs. Gutch, County Folk-Lore, vol. ii. North Riding, etc. p. 290 ; Mrs. Gutch and Mabel Peacock, County Folk-Lore, vol. v. Lincolnshire, p. 146. 6 Thiselton Dyer, Folk Lore of Shakespeare, p. 336. 8 Jeaffreson, Brides and Bridals, i. 288. 7 Marie Trevelyan, Folk-Lore and Folk-Stories of Wales, p. 270. 8 Thiers, op. cit. iv. 429. 9 S6billot, op. cit. p. 113, 10 Laisnel de la Salle, op. cit. ii. 22. 11 S6billot, op. cit. p. 113. 13 Ibid. p. T13 sq. XXVI MARRIAGE RITES 571 seventeenth century.1 All over Italy weddings are univer- sally avoided on Fridays and Tuesdays-days when curses are supposed to possess extraordinary efficacy ; a proverb says, " Ne di Venere ne di Marte, non si sposa ne si parte."2 On the other hand, Sunday is a favourite marriage day both in Sicily3 and Piedmont.4 In modern Greece mar- riages are generally celebrated on Sundays,5 and the same is the case elsewhere in South-Eastern Europe.6 Among the Jews, on the other hand, a marriage cannot take place on a Sabbath or a holiday, although it can do so on any other day of the week.7 Among Muhammadans the bridegroom frequently receives his bride on the eve of Friday, which is regarded by them as a blessed night, Friday being their Sabbath-day.8 In some parts of Morocco Sunday is also considered a suitable day for the beginning of married life as being the first day of the week, and in other parts Monday ; 9 in the same country Sunday is held to be the most favourable day for the beginning of the autumn ploughing, and in some tribes Thursday and Monday are likewise regarded as suitable for this purpose.10 In the 1 Thiers, op. cit. iv. 429. 2 Bustico, ' Il matrimonio nel Bellunese,' in Provenzal, op. cit. p. 21. Bacci, op. cit. p. 12. Frescura, loc. cit. p. 46 (Sette Comuni ; Thursdays are also avoided). Estella Canziani, ' Courtship, Mar- riage and Folk-Belief in Vai d'Ossola (Piedmont),' in Folk-Lore, xxiii. 459. Faggiani, loc. cit. p. 232 (Sardinia ; Mondays are also avoided). Pitre, op. cit. ii. 50 (Sicily). 3 Pitre, op. cit. ii. 50. 4 Estella Canziani, loc. cit. p. 459. 6 Rodd, op. cit. p. 90. b Heuzey, op. cit. p. 276 (Valakhs). Stern, op. cit. ii. 105 (Rumanians in Macedonia), 107 (Bulgars). Kaindl, Ruthenische Hochzeitsgebrauche in der Bukowina,' in Zeitschr. d. Vereins f. Volkskunde, xi. 284 (also on Thursdays). Piprek, op. cit. p. 34 (Little Russians). 7 Rosenau, op. cit. p. 157. Abrahams, op. cit. p. 186. 8 Lobel, op. cit. p. 25 n. 1 (Turks). Lane, Modern Egyptians, p. I75- Mrs. Todd, Tripoli the Mysterious, p. 94. Westermarck, Marriage Ceremonies in Morocco, p. 86 sq. 9 Westermarck, op. cit. p. 87. In Cairo the eve of Monday is also a fortunate period for receiving the bride (Lane, op. cit. p. 175 sq.}. 10 Westermarck, Ceremonies and Beliefs connected with Agriculture, certain Dates of the Solar Year, and the Weather in Morocco, p. 8. 572 THE HISTORY OF HUMAN MARRIAGE chap. Kwantan district in Central Sumatra the same three days are by preference chosen for the celebration of marriages.1 In Dahomey they are usually celebrated on a Sunday.2 There are also certain dates on which marriages are avoided and certain dates which are regarded as particularly lucky for marrying. In ancient Rome it was necessary to refrain from marrying on the so-called dies parentales (February 13th-21st) and the days of Kalends, Ides, and Nones.3 Brand says that in England people never marry on Childermas Day, which, for whatever reason, " is a black day in the calendar of impatient lovers."4 At Helmsley, in Yorkshire, there is the belief that to be wed on St. Thomas's Day makes a wife a widow ere long;5 but in Lincolnshire this day was, on the contrary, favourably regarded as a wedding day, because it afforded " less time " for repentance .than any other.6 At Dunfermline the last day of the year-Hogmanay.-was considered especially lucky as a day for marrying.7 It should be added, however, that there may be other than superstitious reasons for celebrating marriages by preference at a certain period or on a certain day. In Dukkala, in Morocco, I was told that the weddings are held in the full of the moon in order that the people shall be safe from robbers ; and I have no reason to doubt the accuracy of this explanation, as I never heard in Morocco of any superstitious connection between the celebration of a marriage and the moon, whereas the fear of robbers is always present to the minds of the people. Among the 1 Maass, ' Durch Zentral-Sumatra,' in Zeitschr. f. Ethnol. xli. 160. 2 Ellis, Ewe-speaking Peoples of the Slave Coast of West Africa, P- 155 sq. 3 Smith, Wayte, and Marindin, Dictionary of Greek and Roman Antiquities, ii. 142. Rossbach, op. cit. p. 272. Marquardt, op. cit. P- 43- 4 Brand, op. cit. pp. 290, 397 sq. 5 Mrs. Gutch, County Folk-Lore, vol. ii. North Riding of Yorkshire, York and the Ainsty, p. 290. 0 Mrs. Gutch and Mabel Peacock, County Folk-Lore, vol. v. Lincolnshire, p. 235. 7 Simpkins, op. cit. p. 163. XXVI MARRIAGE RITES 573 Jews the favourite wedding day in the Middle Ages was Friday, although the selection of this day was entirely against the Talmudic prescriptions on the subject, Wednes- day and Thursday being the marriage days of the Mishna. But, as Mr. Abrahams points out, the convenience of marrying on Friday was so obvious that mediaeval authorities, while deploring the custom, did not seriously attempt to effect a change ; for the proximity in time to the Sabbath as a day of rest-on which no marriage could be celebrated -and the opportunity given for associating the wedding with the synagogue service of the following day, gave to Friday a peculiar appropriateness.1 In Gotland, in Sweden, the wedding day was in the last century changed from Thursday to Friday, probably, as Dr. Lithberg suggests, in order that there should be no interruption of the festivities, which lasted during three days including a Sunday, when the guests had to accompany the bridal pair to church. But subsequently the time for feasting was reduced, and as a result of this Saturday took the place of Friday as the most usual marriage day in Gotland.2 The popularity of Sunday weddings is also, no doubt, partly due to the custom of performing the nuptial ceremony in church, as well as to the leisure of the weekly holiday. In some cases the very idea that a certain period or day is " lucky " or'" un- lucky " for the celebration of marriages may have originated in the fact that it was regularly chosen, or avoided, not for any superstitious reason, but purely for the sake of convenience. Besides marriage rites which are purely magical, there are others of a religious or semi-religious nature, often per- formed by a priest. In some cases these rites are intended to serve some specific purpose, such as the securing of off- spring,3 but most frequently their object is to promote the welfare of the couple in general either by bestowing on 1 Abrahams, op. cit. p. 186. 8 Lithberg, in Fataburen, 1906, p. 79 sq. • Percy Smith, ' Futuna,' in Jour. Polynesian Soc. i. 38. John- ston, George Grenfell and the Congo, p. 680 (Bakongo). 574 THE HISTORY OF HUMAN MARRIAGE chap. them positive benefits or by protecting them against evil. They arc found both among savage and civilised peoples. In some African tribes sacrifices are offered to ancestral spirits1 or to some fetish.2 Among the Mpfumo clan of the Thonga, in South-Eastern Africa, the father of the bride performs a religious ceremony standing behind the bride and groom. He speaks to the gods, that is, the spirits of his ancestors, asking them to look at her and accompany her where she will live. " May she also found a village," he says, " may she have many children, may she be happy, good and just. May she be on good terms with those with whom she will be."3 All over Madagascar it is the custom for the bride, when she leaves her home, to be blessed by her parents, who ask God and their ancestors to give her a long life, happiness, wealth, and especially a numerous progeny.4 Among the Maori an aristocratic marriage was accompanied by a great feast (umu kotore) at which a priest recited certain prayers or invocations over the couple, ' in order to preserve them in health and prosperity, to ward off from them all evil, and to cause the woman to be fruitful and to cleave to her husband.5 In Tahiti, after the pre- liminaries had been adjusted, the parties repaired to the temple, where the priest addressed the bridegroom usually in the following terms :-" Will you not cast away your wife ? " to which the bridegroom answered, " No." Turn- ing to the bride, he proposed to her a like question, and received a similar answer. He then addressed them both, saying, " Happy will it be if thus with you two." And finally he offered a prayer to the gods on their behalf, imploring that they might live in affection, and realise 1 Thomas, Anthropological Report on Ibo-speaking Peoples of Nigeria, i. 65. Johnston, Uganda Protectorate, p. 553 (Lendu). 2 Hartter, ' Sitten und Gebrauche der Angloer (Ober-Guinea),' in Zeitschr. f. Ethnol. xxxviii. 44. 3 Junod, Life of a South African Tribe, i. in sq. 4 Grandidier, Ethnographic de Madagascar, ii. 184 sq. See also ibid. ii. 182., 5 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New Zealand Institute, xxxvi. 44 sq. Idem, ' Lore of the Whare-Kohanga, ' in Jour. Polynesian Soc. xv. 5. Tregear, The Maori Race, p. 295. XXVI MARRIAGE RITES 575 the happiness which marriage was designed to securer.1 Among the Igorot of Luzon a priestess performs the mar- riage ceremony, praying to the spirits of the deceased in the presence of all the kinsfolk of the couple.2 Among the Bagobo of Mindanao a. mbalian, orfemale shaman, " spreads a mat on the floor, places on it many valuable articles, and then offers all to the spirits, in order that they may be pleased to give the couple a long and prosperous life together. Finally, she puts a dish of rice on the mat and, after offering it to the spirits, places it between the boy and the girl," who feed each other with it.3 At a Khasi marriage there is an elaborate religious ceremony at which God the creator, the god or goddess of the State, and, what is probably more important, the ancestress or ancestor of the clan are in- voked.4 Among the Khyoungtha5 and Garos6 a priest beseeches the gods to bless the union. In ancient India various deities were invoked at the weddings, but the magical element by far predominated in the marriage ritual.7 Modern Hindus, except the very lowest, consider it essential for the validity of a marriage that a Brahman, acting as priest, should be present at its celebration.8 Among the Parsees the wedding service is conducted by two priests, the elder of whom blesses the couple, praying that Ahura Mazda may grant them " pro- geny of sons and grandsons, abundant means, strong friendship, bodily strength, long life, and an existence of 150 years."9 Religious marriage rites occurred in ancient Greece ; thus the bride dedicated to various deities that 1 Ellis, Polynesian Researches, i. 271. 2 Meyer, ' Die Igorrotes von Luzon,' in Verhandl. Berliner Gesellsch. Anthr. 1883, p. 385. 3 Cole, 'Wild Tribes of Davao District, Mindanao,' in Field Museum of Natural History, Anthr. Ser. xii. 101 sq. 4 Gurdon, Khasis, pp. 127, 129 sq. 8 Lewin, op. cit. p. 129. 0 Dalton, op. cit. p. 64. 7 Haas, in Weber, Indische Studien, v. 312 sqq. Oldenberg, Die Religion des Veda, p. 462 sqq. 8 Schmidt, Liebe und Ehe im alten und modernen Indien, p. 370. * Mary Billington, op. cit. p. 85. Jivanji Jamshedji Modi, ' Marriage (Iranian).-1. Zoroastrian,' in Hastings,.op. cit. viii. 456. 576 THE HISTORY OF HUMAN MARRIAGE chap. superintended the union of the sexes her girlish toys and other gifts, and more especially her maiden tresses, now shorn.1 At the Roman confarreatio, as noticed above, a cake of far was offered to Jupiter Farreus,2 and in the historic period an animal sacrifice was made at a wedding in Rome ;3 but we do not know to what deity it was offered, or, indeed, if it was offered to any deity at all. Generally speaking, the religious side of ancient Indo-European mar- riage rituals has been exaggerated by earlier writers, who have put a religious interpretation upon many purely magical ceremonies by associating them with the worship of divine beings. The founder of Christianity did not prescribe any par- ticular ceremonies in connection with marriage, but it has been assumed that the celebration of it among Christians was from the very fir$t accompanied with suitable acts of religious worship. The testimony of the Fathers, from the middle of the third century onwards, shows that marriages contracted without any formal benediction did occur, but they were discountenanced by the Church.4 Yet, though the dogma that marriage is a sacrament gradually developed from St. Paul's words, to java-rppiov tovto pfya earriv6 -in the Vulgate translated, " Sacramentum hoc magnum est "-and was fully recognised in the twelfth century,6 marriage without benediction was nevertheless regarded as valid in the Church till the year 1563, when the Council of Trent decreed that thenceforth no marriage should be considered valid unless celebrated by a priest in the presence of two or three witnesses.7 Luther's opinion that all matrimonial affairs belong not 1 Pausanias, Descripiio Graeciae, ii. 32. 1. Frazer, Pausanias's Description of Greece, iii. 279 sq. 2 Supra, ii. 450. 3 Rossbach, op. cit. p. 309 sqq. Idem, Romische Hochzeits- und Ehedenkmdl^r, p. 105 sqq. 4 Tertullian, Ad uxorem, ii. 9 (Migne, Patrologies cursus, i. 1415 sqq.). Idem, De pudicitia, ch. 4 (Migne, ii. 1038 sq.). 5 Ephesians, v. 32. 6 v. Scheurl, Das gemeine deutsche Eherecht, p. 15. 7 Roguin, Traitd de droit civil compart. Le mariage, pp. 103, 104, 128 sqq. XXVI MARRIAGE RITES 577 to the Church, but to the jurists, was not accepted by the legislators of the Protestant countries. Marriage certainly ceased to be thought of as a sacrament, but continued to be regarded as a divine institution. And sacerdotal nuptials became no less obligatory on Protestants than on Roman Catholics. It was the French Revolution that first gave rise to an alteration in this respect. The Constitution of the 3rd Sep- tember, 1791, declares in its seventh article, title ii.:- " La loi ne considere le mariage que comme contrat civil. Le pouvoir legislatif etablira pour tous les habitants, sans distinction, le mode par lequel les naissances, manages et deces seront constates et il designera les officiers publics qui en recevront les actes."1 To this obligatory civil act a sacerdotal benediction may be added, if the parties think proper. Since then civil marriage has gradually obtained a footing in the legislation of most European countries, although in some of them, as in England, the parties may choose the religious or the civil rite, just as they like, both making the marriage equally valid by law.2 The legal importance which has been attached to the religious ceremony in Christian countries has no counterpart either in Jewish or Muhammadan law. Although the former regards marriage as a divine institution, the omission of the benediction would not invalidate a marriage.3 The priestly benediction is mentioned neither in the Bible nor the Tal- mud ; and the regular presence of a Rabbi at a wedding is not earlier than the fourteenth century.4 Nor does Mu- hammadan law require religious rites for the contraction of a valid marriage. In all cases the religious ceremony is left entirely to the discretion of the qazi or person who performs the ceremony, and consequently there is no uni- formity of ritual.5 1 Glasson, Le mariage civil et le divorce, p. 253. Roguin, op. cit. p. 140 sq. 2 See Roguin, op. cit. p. 141 sqq. ; Glasson, op. cit. p. 282. 3 Rosenau, op. cit. p. 158. 4 Abrahams, op. cit. pp. 199, 200 n. 1. Grunwald, ' Marriage Ceremonies,' in Jewish Encyclopedia, viii. 341. 6 Hughes, Dictionary of Islam, p. 318. VOL. II p p 578 THE HISTORY OF HUMAN MARRIAGE chap. The office of the priest at a Christian wedding, however, has not been restricted to the performance of the nuptial ceremony: according to Roman Catholic rituals he has also had to bless the bridal bed, and this was considered one of the most important of the marriage rites. Thus in England, in the Papal times, no marriage could be consum- mated until the bed had been blessed. " On the evening of the wedding-day," says Jeaffreson, " when the married couple sat in state in the bridal-bed, before the exclusion of the guests who assembled to commend them yet again to Heaven's keeping, one or more priests, attended by acolytes swinging to and fro lighted censers, appeared in the crowded chamber to bless the couch, its occupants, and the truckle-bed, and fumigate the room with hallowing incense " ;x and the parties were also sprinkled with holy water.2 That the clergy knew how to make profit by this custom appears from an old manuscript in which it is said that " new married Couples were made to wait till Midnight after the Marriage Day, before they would pro- nounce a Benediction, unless handsomely paid for it; and they durst not undress without it, on pain of excommunica- tion."3 The object of the ceremony was partly to bestow upon the couple a long life and progeny and other good things, but partly also to protect them against evil influences ; as appears from the formula given in the manual for the use of Salisbury, where it is said:-" Benedic, Domine, thalamum istvm et omnes habitantes in eo; ut in tua pace consistant, et in tua voluntate permaneant: et in" amore tuo vivant et senescant et multiplicentur in longitudine dierum. . . Qui custodis Israel, custodi famulos tuos in hoc lecto quiescentes ab omnibus fantasmaticis demonum illusionibus."^ 1 Jeaffreson, op. cit. i. 98. 3 Douce, Illustrations of Shakspeare, p. 123. 3 Brand, op. cit. p. 403. 4 Douce, op. cit. p. 123. In Norway the custom of the clergy blessing the bridal bed still persisted in the beginning of the seven- teenth century, although formally abolished by the Reformation (Troels-Lund, op. cit. xi. 66 sq.). Among German Catholics it is found to this day (Meyer, Badisches Volksleben, p. 306 ; Reiser, op. cit. ii. 250 sq.). XXVI MARRIAGE RITES 579 In this connection may also be mentioned the old Scotch custom according to which " the parson who presided over the marriage ceremony uniformly claimed it as his alienable privilege to have a smack at the lips of the bride immediately after the performance of his official duties " ; for it was sturdily believed that the happiness of every bride lay involved in the pastoral kiss.1 The rites which we have hitherto considered have in the first place reference to the welfare of the two individuals who enter into the married state and therefore form the centre of the ritual, and there can be no doubt that the large majority of marriage rites belong to this class.2 But at the same time many of them have reference to other persons as well. The future offspring are concerned in fertility and other rites ; indeed, when a certain ceremony is necessary for the validity of the union, its performance may materially affect the rights and social status of the children. Moreover, the conclusion of a marriage implies that either party enters into new relations to the other party's family or to the larger social group of which he or she is a member, and this also influences the marriage ritual. There are rites which spring from the intimate contact into which the bride comes with the bridegroom's family, especially his mother, when she goes to live in his place. She is often, as we have seen, ceremonially received by her mother-in-law. In Romagna, on her arrival at her new home, the mother-in-law presents to her the key of the house, addresses her as the " mistress of the house," and gives her a kiss.3 Among the Slovenes of Carinthia the bride sits down at the table in the bridegroom's house, a woman places on it two glasses, from .which the bride and her mother-in-law drink together, and the bride drops some money into her glass as a present for the mother-in-law.4 In Bulgaria, when the bride has arrived at the door of the 1 Simpkins, op. cit. p. 164. 2 Cf. Westermarck, Marriage Ceremonies in Morocco, p. 365. 3 Placucci, op. cit. p. 58. 1 Piprek, op. cit. p. 114. 580 THE HISTORY OF HUMAN MARRIAGE chap. bridegroom's house, she is presented with honey, which she smears on all the thresholds she has to pass, " in order that she shall love her new relatives."1 At Fez, a few days after the proposal has been accepted, some eight or ten women of the young man's family or kin, including his mother, go to visit the girl's mother, who entertains them with tea, food, and honey; and I was told that the object of the honey is to make the daughter " sweet " to the family of her future husband so that there shall be no quarrel between them. Among the Tsui, an Arabic-speaking mountain tribe in the interior of Morocco, the bride's mother removes the right eye of the sheep which her husband kills on the occasion when the bride is painted with henna, and the dried eye is afterwards made into powder and, mixed with various spices, put into the food which the mother gives to the bridegroom's family to eat, " so that they shall look upon her daughter with affectionate eyes." Among the Berbers of the Aith Yusi a date or raisin is put into the right slipper of the bride " that she shall become dear to her husband's family." In the same tribe the bridegroom's mother offers her some seksu on her uncovered right thigh, and the bride snaps up a little of it three times, slightly biting the mother-in-law ; this, I was told, is supposed to make the two women friendly to each other. In China, on the third day after marriage, there is a ceremony called " washing of the feet," which consists in the bride washing the feet of her mother-in-law and is supposed to be a sign tKat the mother-in-law will be no longer troubled with domestic affairs.2 Among the pastoral clans of the Banyoro " a bride was taken to live with her husband's parents, who received her as a daughter. She sat first in the lap of her mother-in-law and afterwards in the lap of her father-in-law." When the term of her seclusion ended, the bridegroom took her to visit her parents, and he was admitted into .their family as a son by first sitting in the lap of his mother-in-law and afterwards in that of his father-in-law.3 Among the common 1 Ibid. p. 145. 2 Stewart Lockhart, in Folk-Lore, i. 366. 3 Roscoe, Northern Bantu, p. 40; XXVI MARRIAGE RITES 581 people in the Stlatlumh tribes of British Columbia, when a youth who wishes to marry a certain girl is informed that he is expected by her family, he visits the house of his future father-in-law. " When he enters he is made welcome and invited to sit down with the family alongside of his bride. It is this formal inclusion in the family circle of the bride that constitutes the marriage." He stays with his father-in-law for at least four days, but sometimes continues to live in the family of the latter. " This in- clusion of the son-in-law within the family circle gives him all the rights of sonship and his offspring are regarded as belonging to his wife's family just as much as to his own."1 Among various peoples the wedding is sooner or later followed by a ceremonial visit paid by the newly-married couple to the parents of the bride. This is the case in some tribes in Morocco, the husband presenting them food and kissing their heads. A marriage, however, establishes new relations not only between the bride and the bridegroom's family and between the bridegroom and the bride's family, but between the other members of the two families as well. As we have noticed above, the feasting with which it is celebrated not only serves the purpose of giving publicity to the event, but also brings the families together and strengthens the ties which unite them,2 Of the distribution of bread and salt at a betrothal feast in the High Vosges it is said, " Quand on a partage le pain et le sei, en pareille circonstance, on ne fait plus qu'une famille."3 The following ceremony takes place at a Yakut wedding :-•" The father of the groom, rising with a choice bit of meat in his hand, made an ap- propriate speech and gave the meat to the father of the bride. This is repeated a little later with the mother of the bride, then with her other relatives, and then with the most important members of her sib. Then the other companions of the groom complimented the parents and relatives of the bride in the same manner. The point of 1 Hill Tout, ' Report on the Ethnology of the Stlatlumh of British Columbia,' in Jour. Anthr. Inst. xxxv. 132 sq. 8 Supra, ii. 438. 3 Sauvd, Le Folk-Lore des H antes-Vosges, p. 83. 582 THE HISTORY OF HUMAN MARRIAGE chap. all the speeches was, ' We are now related to each other; we will hereafter live in friendship and concord.' " 1 By being attached to her husband's family and com- munity the bride also enters into relations to the spiritual agents connected with them. In China the introduction ceremonies begin with a worship of the ancestors of the husband's family. After the performance of this ceremony and the worship of Heaven and Earth " she will present herself in the company of her husband before the latter's parents, and they will both pay their obeisance by kneeling down and kowtowing to them. . . . On the following day the bride will be formally introduced with such pomp as is usual on such occasions to all the relatives and intimate friends of the husband, and when this is done the bridegroom will call with the bride on her parents, and, with the excep- tion of the worship of heaven and earth, all the other ceremonies will be repeated in the same way as has been done in the husband's house."2 The Roman bride brought with her three coins (asses'), one of which she gave to her husband, one she laid on the hearth, and the third she threw down at the nearest compitum (crossways). " Here," Mr. Fowler observes, " she seems to be making an offering to the genius of her husband, to the spirit of the hearth-fire, and to the Lar of the family's land allotment, who dwelt in a sacellum at the compitum."3 Among the Southern Slavs the bride -still frequently offers a coin to the hearth in the husband's house.4 In the Aland Islands in Finland, inhabited by Swedish-speaking people, the bride, on her arrival at the bridegroom's place, " buys " the spirit of it (tomten) by throwing a coin into the fire-place or putting a coin underneath the front-door steps or between two boards in the floor.5 In Esthonia, the moment the bride enters the bridegroom's house, she is led through every part of it 1 Sieroshevski, ' Yakuts,' in Jour. Anthr. Inst. xxxi. 82. 2 Chen, Patriarchal System in China, p. 4. Cf. Leong and Tao, Village and Town Life in China, p. 107 sq. ; Gray, China, i. 205. 3 Rossbach, Untersuchungen uber die romische Ehe, p. 373 sqq. Fowler, ' Marriage (Roman),' in Hastings, op. cit. viii. 465. Cf. Idem, Religious Experience of the Roman People, p. 77. 4 Piprek, op. cit. p. 170. 5 Dr. Nikander, in manuscript notes. XXVI MARRIAGE RITES 583 and through the stables and gardens, and is bound to drop ribbons or money at each place and into the well and the fire.1 In modem Greece, on the third day after the wedding, the young wife is taken by other married women to the spring or well which supplies the household with water ; she drinks out of it, throws a few coins and some food into it, and dances round it together with the other women.2 In Andjra, in Morocco, the wife, on the evening of the seventh day after her arrival at her new home, takes a loaf of bread which has been previously presented to her by her mother-in-law and goes in secret to the spring of the house, drops some pieces of the loaf on her way, puts some round the spring, and throws others into the water, saying, " I am one of the guests of God and of your guests, O owners of the land." Being a stranger, she thus places herself under the protection of the spirits and saints of the district. Next morning before sunrise she goes secretly, accompanied only by a young brother of her husband, to the sanctuary of the patron saint of the village or some other saintly place (siyid) hear the house, taking with her a cock, a coin, bread, and incense. She addresses the saint with the words, " I am one of the guests of God and of your guests, O my lord the saint " ; as a stranger she does not yet know, or is not supposed to know, the name of the saint. The cock is killed by the boy, and the knife with which it is done, and also the bread and coin, are left at the siyid. In the same tribe, on the evening of the seventh day, the young wife also gives seven pieces of bread to the dog of the house, putting them one after the other on the top of her foot and letting the dog take them from there so as to make it friendly. Among the Berbers of the Ait Warain, again, the water with which the bridegroom's mother has washed the right foot and hand of the bride over one of the fire-stones3 is then sprinkled on the cattle, so as to make the bride fond of them. In Delsbo in Sweden, where the wedding was held in the bridegroom's home if the 1 Boecler-Kreutzwald, op. cit. p. 33. 2 Sakellarios, op. cit. p. '27. 3 See supra, ii. 513. 584 THE HISTORY OF HUMAN MARRIAGE chap. couple were going to live there, the bride, on her return from church, went into the cow-house and gave to each animal a piece of bread or something else to eat, in order that she in the future should have good luck with cows and calves. Then she went into the stable and did the same to the horses ; and she also had a look-in at the cook-house and larder so that there should be plenty of food.1 In some parts of Germany, too, the bride goes to the cattle and throws food to them, wishing them good luck.2 A marriage implies not only that the parties enter into new relations to each other's people, but, very frequently, that one of them, through the change of domicile, is actually transferred to the other one's family group. And it implies other changes in the social grouping of people : either party passes from one social class into another, the bridegroom from the class of the bachelors to that of the married men, and the bride from the class of the girls to that of the married women. This re-grouping also finds expression in the marriage ritual, as when the hair of the bride is arranged in the fashion of married women,3 or she ceremonially assumes the head-dress worn by them,4 or when the bride dances first with the unmarried girls and then with the married women and the bridegroom first with the bachelors and then with the married men.5 Sometimes it even leads to sham fights, just as does the removal of the bride to the bridegroom's house. A very common feature of peasant weddings in Denmark,6 Sweden,7 and the Swedish-speaking 1 Wikman, ' Frieri, forlofning och brollop i Delsbo,' in Fataburen, 1913, p. 82. 2 Wuttke, op. cit. p. 373 sq. Sartori, op. cit. i. 117. 8 See, e.g., Westermarck, Marriage Ceremonies in Morocco, pp. 252, 276 sq. 4 See, e.g., v. Schroeder, op. cit. p. 144 sqq. ; Sartori, op. cit. i. 100. 6 Lithberg, in Fataburen, 1911, p. 161 sqq. (Gotland) ; Wikman, ibid. 1913, p. 84 (Delsbo) ; Nicolovius (Loven), Folklifvet i Skytts Hdrad i Skane, p. 140 sq. (Sweden). Allardt, Nyldndska folkseder och bruk, pp. 27, 29. 6 Kristensen, op. cit. iv. 76. 7 Nicolovius (Loven), op. cit. p. 140 sq. Djurklou, Ur Nerikes folksprah och folklif, p. 52. Hofberg, Nerikes gamla minnen, p. 201 sq. Gaslander, loc. cit. p. 267 (Wassbo). Hammarstedt, in Fest- skrift til H. F. Fetlberg, p. 490 sq. XXVI MARRIAGE RITES 585 communities in Finland1 is a struggle between the bachelors and the married men about the bridegroom, and between the girls and the married women about the bride ; and a similar custom is found among some Slavs.2 In Morocco the married men make violent attempts to catch the bride- groom or rob him of some of his belongings, and the bachelors of the village, who are surrounding him all the time, defend him ; whilst the bride is surrounded by the unmarried girls, who must never leave her alone, lest the married women should interfere with her property.3 But sometimes the bachelors also beat the bridegroom, who is defended by his best-man,4 and although this is said to rid him of evil influences it may at the same time be a ceremonial punish- ment inflicted on him by his bachelor friends because he is deserting their class. Though the marriage ritual centres round the bride and bridegroom, there are rites which are supposed to influence the welfare of other persons even independently of their relations to the principals. In Morocco the baraka, or holiness, attributed to bride and bridegroom makes a wedding an occasion from which persons who take part in it, or otherwise have anything to do with it, expect to derive certain benefits.5 Thus, when milk is offered to the bride on her way to the bridegroom's place, she dips her finger into it or drinks a few drops and blows on the rest, so as to impart to it a little of her holiness ; and the milk is then mixed with other milk to serve as a charm against witch- craft, or poured into the churn to make the butter plentiful.6 1 Allardt, op. cit. p. 27 sq. (Nyland). Lindroos and Andersson, ' Ett brollop i Pellinge,' in Hembygden, i. 158. Aina Wadstrom, ' Frieri- och brollopsbruk fran Lappfjard,' ibid. ii. 88 sq. Nordman, Brollop i Houtskar,' ibid. vi. 82 sq. Mattans, ' Brdllopsseder i Korsnas,' ibid. vi. 139. Tegengren, ' Brollopsbruk i Vora,' ibid. viii.-ix. 138. * Piprek, op. cit. p. 87. 3 See Westermarck, Marriage Ceremonies in Morocco, p. 346 and the references in the foot-notes. 4 See supra, ii. 517. a Westermarck, op. cit. p. 360 sqq. 8 Ibid. pp. 170, 171, 183 190. 586 THE HISTORY OF HUMAN MARRIAGE chap. The bread or dried fruit which is thrown over the bridal box and falls on the ground is picked up by people who want to benefit their com by putting it underneath the heap on the threshing-floor;1 or the bride throws the barley which is offered her on the people, who catch of it what they can and mix it with their own barley.2 Roasted barley brought from the bride's home is distributed among the wedding guests, owing to the holiness with which it is saturated.3 And when, after the consummation of the marriage, the people come to look at the blood-stained garment of the bride, they mb their eyes with the stains, which are sup- posed to contain bar aka and be wholesome for the eyes.4 Among the benefits expected from a wedding there are such as are closely connected with the event it celebrates. Owing to a natural association of ideas, a wedding is looked upon as a potential cause of other weddings. In Morocco, before the bride is painted with henna, seven girls pour water over her at a spring and wash her body, hoping that by doing so they will get married themselves.5 The egg which is put into the henna bowl is subsequently eaten by one of the bride's girl friends who wants to get a husband soon.6 If any unmarried woman or girl is living in the house of the bride's parents, the bride is told to " drag her foot " when she leaves it, so as to help the unmarried one to a husband.7 When the bridal box is taken to the bride's village on the back of a mule, an unmarried youth sits inside it in order to get married soon ;8 or when the bride, on her arrival at the bridegroom's house, has been lifted down from the mare which carried her thither, a bachelor for the same purpose mounts the animal and has a ride on it.9 In Brittany the bride, on her arrival at the bridegroom's house, distributes the bread and butter then presented to her among the young people escorting her, who eat it with eagerness because they hope that they thereby will get married during the year;10 or the pins which were fixed to 1 Ibid. p. 195. 2 Ibid. p. 211. 3 Ibid. p. 246. 4 Ibid. p. 159. 5 Ibid. p. 153. (i Ibid. p. 145. 7 Ibid. p. 165. 8 Ibid. p. 168. 9 Ibid. p. 196. 10 Og6e, op. 'it. ii. 357. XXVI MARRIAGE RITES 587 the crown of the bride are distributed among the unmarried girls, or boys as well, and this is supposed to have a similar effect.1 In Worcestershire it is said that a girl who has a pin out of a bride's veil will soon be a bride herself.2 In Northumberland the nuptial ring was in the evening dropped into a posset, which was instantly attacked by all the unmarried laddies and lasses, as the one who discovered it would be the first to get married. With the same object the bride, or a female attendant, threw her left stocking over her shoulder among those in the room, the one on whom b fell or who secured it being the person who would be married next.3 This custom, with small variations in details, occurred, or still occurs, in Scotland and Shetland, as well as in England.4 At Gardenston the bridegroom drew off his stocking and threw it among the bystanders to scramble for.5 But in some places " the bride's stockings were taken by the young men, and the bridegroom's by the girls ; each of whom, sitting at the foot of the bed, threw the stockings over their own heads, endeavouring to make them fall upon those of the bride, or of her spouse. If the bridegroom's stockings, thrown by the girls, fell upon the bridegroom's head, it was a sign of their own speedy mar- riage ; and a similar prognostic was derived from the falling of the bride's stockings, as thrown by the young men."6 In Aberdeenshire7 and Shetland,8 when the bridegroom had his feet washed, a ring was thrown into the tub, and after the ceremony was completed there was a contest for it, 1 Sebillot, Coutumes populaires de la Haute-Bretagne, pp. 133, 135. 2 Ladbury, ' Scraps of English Folklore, III. Worcestershire,' in Folk-Lore, xx. 345. 3 Murray's Handbook for Travellers in Durham and Northumber- land, p. 140. Marie Balfour, County Folk-Lore, vol. iv. Northumber- land, pp. 94, 96 sqq. 1 Simpkins, op. cit. p. 393 (mining population of Fife). Gregor, Some Marriage Customs in Cairnbulg and Inverallochy, ' in Folk- Lore Journal, i. 120. Hibbert, Description of the Shetland Islands, p. 555. Brand, op. cit. p. 399 sqq. 5 Gregor, Notes on the Folk-Lore of the North-East of Scotland, p. 100. 0 Brand, op. cit. p. 400. 7 Gregor, in Folk-Lore Journal, i. 119. 8 Hibbert, op. cit p. 554. 588 THE HISTORY OF HUMAN MARRIAGE chap. the finder being the person who would be first married. In the northern counties of England it is deemed an augury of speedy marriage to rub shoulders with the bride1 or bridegroom ; and she who receives from the bride a piece of cheese, cut by her before leaving the table, will be the next bride among the company.2 Brand states that in the North, and perhaps all over England, " slices of the bride- cake are thrice, some say nine times, put through the wed- ding ring,, and are afterwards laid under the pillows of young folk when they go to bed, for the purpose of making them dream of their lovers ; or of exciting prophetic dreams of love and marriage."3 This custom still survives in some parts of the country.4 At a wedding in Holdemess in Yorkshire, " as soon as the bride and bridegroom had left the house, and had the usual number of old shoes thrown after them, the young folks rushed forward, each bearing a tea-kettle, of boiling water, which they poured down the front door-steps, that other marriages might soon follow, or, as one said, ' flow on.' "5 A common Swedish marriage custom is that of " dancing the coronet off the head of the bride." It presents varia- tions in details, but the following may be regarded as a fair description of it as practised in some places. With her eyes bandaged, and whilst the maidens present are dancing a ring-dance around her, the bride takes the crown from off her own head and places it, haphazard, on that of one or other of the damsels ; and she on whom the honour has been bestowed is believed to be the first to obtain a husband. This girl, in her turn, places it on the head of a second, 1 For a similar superstition among the mining population of Fife see Simpkins, op. cit. p. 393. 2 Henderson, op. cit. p. 35. 3 Brand, op. cit. p. 396. 4 Mrs. Gutch and Mabel Peacock, op. cit. p. 231 (Lincolnshire). Mrs. Gutch, County Folk-Lore, vol. ii. North Riding of Yorkshire, etc., p. 297. In Northumberland there is a belief that if the cake is broken by the bridegroom over the head of the bride and the pieces are then thrown up and scrambled for, they have much more prophetic virtue than when they' are merely put nine times through the ring (Marie Balfour, op. cit. p. 96). 5 Mrs. Gutch, County Folk-Lore, vol. vi. East Riding of Yorkshire, p. 129 sq. XXVI MARRIAGE RITES 589 and so on with the whole of the party. The bride is then lifted on a chair, above the heads of her companions, an^ amidst tremendous cheering she drinks a toast, the implica- tion of which is " a hope that all maidens may soon change their condition."1 It seems that at European weddings not only speedy marriage but good luck in general is expected from contact with the bride or bridegroom, or something worn by them. In his description of Scottish superstitions Dalyell states that " an auspicious fortune was anticipated from gaining possession of certain parts of the apparel of the wedded pair," hence a struggle sometimes ensued, even in church, for the bridegroom's gloves.2 In Yorkshire, as soon as the bride re-enters her father's house, after the ceremony is concluded, there is a general scrambling on the part of the guests to get the first kiss of the bride, and it is believed that the person who does so will be extremely lucky.3 In some parts of Scotland, after the marriage ceremony, the bride was expected to proceed round the apartment, attended by her maidens, and to kiss every male in the company, and a dish was then handed round, in which every one placed a sum of money.4 At Bourges it was the custom for brides on coming out of church to embrace indifferently all whom they met in the street, and in the province of the Marche they were said to do so before the marriage service.5 At Swedish country weddings it is the custom for the bride to dance with all the men,6 and frequently also for the bridegroom to dance with all the women.7 The former custom is found elsewhere in Europe,8 1 Lloyd, op. cit. p. 27. See also Djurklou, op. cit. p. 51 ; Norlind, op. cit. pp. in, 114, 116 sq. 2 Dalyell, op. cit. p. 292. 3 ?drs. Gutch, County Folk-Love, vol. vi. East Riding, p. 129. 1 Rogers, op. cit. p. 112. 5 Laisnel de la Salle, op. cit. ii. 66. 6 GaSlander, loc. cit. p. 266. Craelius, Forsok till ett Landskaps Beskvifning, p. 423. Wiknian, in Fatabuven, 1913, p. 84. 7 Nicolovius (Loven), op. cit. p. 139 sq. Eva Vigstrorn, ' Folkseder i Ostra Goinge harad i Skane,' in Hazelius, Bidrag till var odlings hafder, vol. ii. pt. i. 63. Djurklou, op. cit. p. 51. Hofberg, op. cit. p. 201. Lithberg, in Fatabuven, 1911, p. 157. 8 Hartland, Legend of Perseus, ii. 356. 590 THE HISTORY OF HUMAN MARRIAGE chap. as in various parts of Germany 1 and among the Slovaks, where every guest who dances with the bride has to pay her some money.2 Dr. Hartland connects the bride-dance and the kiss which the bride bestows upon the masculine guests with the custom of the Nasamonians and some other ancient peoples, according to which she was on the wedding night considered as common property, and concludes that they hint at the former prevalence of group-marriage among the ancestors of the European nations.3 But if the right of kissing, or dancing with, the bride is to be interpreted as the survival of an earlier right of having sexual intercourse with her, what conclusions are to be drawn from the Swedish4 and Danish5 custom which prescribes that the officiating clergyman shall have the first dance with the bride, and from the Scottish parson's right to the first kiss ? It should also be remembered that the bride dances with all the women and the bridegroom with all the men, and sometimes the latter even gives a farewell kiss to each of his bachelor friends.6 The ceremonial dancing and kissing at European weddings may have a different meaning in different cases ; but to see in it traces of a primitive marriage institution which did not exist among any Indo-European people in the historic age, and the existence of which in prehistoric times is a sheer guess, is to my mind one of those errors of method which have more than anything else led astray many students of early society. Those who join in a wedding may not only try to derive positive benefits from it but may also, like the principals, have to take precautionary measures against evil influences. In Morocco the same kinds of purifying or protective matters as are applied to the bride or the bridegroom are also in some places applied to their friends : the bridegroom's best- man or the bachelors of the village who surround him smear 1 Sepp, op. cit. p. 67 sq. Sartori, op. cit. i. 106. 2 Piprek, op. cit. p. 108. 3 Hartland, op. cit. ii. 358 sqq. 4 Craelius, op. cit. p. 423. Nicolovius (Loven), op. cit. p. 139. Gaslander, loc. cit. p. 266. Djurklou, op. cit. p. 51. Hofberg, op. cit. p. 201. Dr. Nikander has found the same custom in Aland. 8 Kristensen, op. cit. iv. 66. 6 Nicolovius (Loven), op. cit. p. 140. XXVI MARRIAGE RITES 591 henna on their hands or clothes1 and paint their eyes with antimony and their lips with walnut-root ;2 and the same things, as well as saffron, are used by the bride's girl friends who surround her, and even by all the women who are present at the wedding.3 In cases like these it may be difficult to distinguish between imitation and self-protection ; but there seems to be an idea that persons who come in close contact with the bride or the bridegroom are also exposed to some danger. In one case I was expressly told that before the bride is carried into the bridegroom's tent guns are fired off close to her in order to prevent her evil influences from affecting the bridegroom's bachelor friends.4 So also the ceremonial wedding fights may serve a prophy- lactic or cathartic object for all who engage in them.5 Sometimes the dung of animals is used on these occasions,6 or the wedding guests smear porridge on each other's faces ;7 and purifying qualities are attributed to these substances.8 Among some peoples it is not the principals but the assist- ants at a marriage who appear disguised in the costume of the other sex ;9 and self-protection may be the object of this practice also. From this survey of the marriage rites of many different peoples it appears that they are not empty formalities, but practices which are supposed materially to influence the welfare of individuals, families, or whole communities.10 Some of them may no doubt be regarded as survivals of earlier, either occasional or regular, methods of concluding a marriage. Certain rites, as we have noticed before, may have been suggested by genuine bride-capture,11 and in other cases marriage by consideration may have left traces in 1 Westermarck, Marriage Ceremonies in Morocco, pp. 98, 113, 120, 327. 2 Ibid. pp. 105, I2o, 202, 282, 327. 3 'Ibid. pp. 153, 156, 157, 161, 283, 327 sq. 4 Ibid. pp. 210, 218, 327. 6 Ibid. pp. 128, 224, 245, 261, 268, 327. 6 Ibid. pp. 245, 327. 7 Ibid. pp. 242, 327. 8 Ibid. pp. 267, 269, 327. 9 Frazer, Totemism and Exogamy, iv. 256 sq. 10 Dr. Karsten says (Contributions to the Sociology of the Indian Tribes of Ecuador, p. 72) that among the Canelos Indians the nuptial feast " is considered absolutely necessary, because without it the young husband would soon die owing to the machinations of the supai," or demon. 11 Supra, ii. 261. 592 THE HISTORY OF HUMAN MARRIAGE chap. the wedding ritual after it has ceased to exist as a reality ; but, generally speaking, the importance of marriage rites as means of studying earlier forms of marriage or relations between the sexes has been greatly exaggerated. Various rites are partly or exclusively fossilised expressions of such emotional states as sexual bashfulness, sorrow, or anger, whilst others are expressions of joy or erotic feelings. To the latter class belong dancing, which forms a regular feature of wedding feasts in many parts of the world,1 and the sexual licence in which the guests are often allowed to indulge. But dancing as a marriage rite may also, in particular cases, have a symbolic or magical significance,2 and, gene- rally, be a method of attaining tumescence.3 And the sexual indulgence of the wedding guests may, on the principle of homoeopathic magic, be a means of assisting bride and bride- groom in achieving the reproductive aims of their union.4 1 Taplin, Folklore, &c. of the Australian Aborigines, p. 35 (Narrin- yeri). Moszkowski, ' Die Volkerstamme am Mamberamo in Hol- landisch-Neuguinea,' in Zeitschr. f. Ethnol. xliii. 322. Man, ' Aboriginal Inhabitants of the Andaman Islands,' in Jour. Anthr. Inst. xii. 138. d' Penha, ' Superstitions and Customs in Salsette,' in Indian Antiquary, xxviii. 116. Dennett, Notes on the Folklore of the Fjort (French Congo), p. 21. Trenk, ' Die Buschleute der Namib,' in Mittheil. Deutsch. Schzitzgeb. xxiii. 168. Theal, Yellow and Dark-skinned People of Africa south of the Zambesi, p. 215 sqq. Decle, op. cit. pp. 159 (Matabele), 347 (Wanyamwezi). Miss Werner, Natives of British Central Africa, p. 131 (natives of the Western Upper Shire district). Dale, ' Natives inhabiting the Bondei Country,' in Jour. Anthr. Inst. xxv. 199 sq. Felkin, ' Notes on the Madi or Moru Tribe of Central Africa,' in Proceed. Roy. Soc. Edinburgh, xii. 321. Weiss, Die Volkerstamme im Norden Deutsch- Ostafrikas, p. 383 (Masai). Hobley, Eastern Uganda, pp. 24 (Bantu Kavirondo), 37 (Nandi). Munzinger, Ueber die Sitten und das Recht der Bogos, p. 63. Westermarck, Marriage Ceremonies in Morocco, see Index s.v. ' Dancing.' de Gubernatis, op. cit. p. 189 sqq. ; Sartori, op. cit. i. 103 sqq. ; Jeaffreson, op. cit. i. 240 sqq. (Europe) ; &c. 2 Cf. Sartori, op. cit. i. 103 sqq. ; Norlind, op. cit. p. 145 sq. 3 Cf. Ellis, Studies in the Psychology of Sex, (vol. iii.) Analysis of the Sexual Impulse, p. 40 n. 1. Among the Xosa Kafirs "the dance at a marriage is considered of more importance than any of the others except the war dance '' (Theal, op. cit. p. 215). 4 Cf. Norlind, op. cit. p. 144 sq., with special reference to Swedish marriage customs. XXVI MARRIAGE RITES 593 In spite of the great importance which is so frequently attached to marriage rites there are many peoples who are said to have no such rites. This is particularly the case with American1 and Australian2 tribes, but also with various 1 Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 133 ; Hyades and Deniker, Mission scientifique du Cap Horn, vii. 378 (Yahgans). Gallardo, Tierra del Fuego-Los Onas, p. 219; Cojazzi, Los indios del Archipielago Fueguino, p. 17 (Onas). Falkner, Description of Patagonia, p. 124. Grubb, An Unknown People in an Unknown Land, p. 214 (Lengua Indians of the Paraguayan Chaco). Church, Aborigines of South America, p. 237 (Chiriguanos). SAnchez Labrador, El Paraguay Catolico, ii. 25 sq. (Guarani). Cardus, Las Misiones Fransiscanas entre los infieles de Bolivia, pp. 71 (Guarayos), 254 (Matacos). Herrmann, ' Die ethnographischen Ergebnisse der Deutschen Pilcomayo-Expedition,' in Zeitschr. f. Ethnol. xl. 129 (Sotegaraik). ' Extracts out of the Historic of John Lerius,' in Purchas, His Pilgrimes, xvi. 562 (Tupis). von den Steinen, Unier den Naturvblkern Zentral-Brasiliens, p. 331 (Bakairi)v Wallace, Travels on the Amazon, p. 512 (Muras). Can- delier, Rio-Hacha et les Indiens Goajires, p. 209. Brett, Indian Tribes of Guiana, p. 101 (Arawaks). Letherman, ' Navajos,' in Smithsonian Report, j 1855, p. 294. Schoolcraft, Indian Tribes of the United States (Archives of Aboriginal Knowledge), ii. 132 (Comanche) ; iv. 223 (Bonaks of California) ; v. 268 (Creeks). Dodge, Our Wild Indians, p. 198 (Plains Indians). Marston, ' Letter to Jedidiah Morse,' in Emma Helen Blair's Indian Tribes of the Upper Mississippi Valley,, ii. 167 ; Forsyth, ' Sauk and Fox Nations of Indians Tradition,' ibid. ii. 214 (Sauk and Foxes). James, Expedition from Pittsburgh to the Rocky Mountains, i. 230 (Omaha). Keating, Expedition to the Source of St. Peter's River, ii. 157 (Ojibway). Heriot, Travels through the Canadas, p. 332 (Iroquois). Richardson, Arctic Searching Expedition, ii. 24 (Chipewyan). Coxe, Russian Discoveries between Asia and America, p. 230 ; v. Langsdorf, Voyages and Travels, ii. 47 ; Sarytschew, ' Voyage of Discovery to the North- East of Siberia, &c.,' in Collection of Modern and Contemporary Voyages and Travels, vi. 76 (Aleut). Lisiansky, Voyage round the World, p. 198 sq. (Kaniagmiut). Murdoch, ' Ethnological Results of the Point Barrow Expedition,' in Ann. Rep. Bur. Ethn. ix. 411 ; Gilder, Schwatka's Search, p. 249 ; Hall, Arctic Researches, p. 567 ; Holm, * Konebaads-Expeditionen til Gronlands Ostkyst,' in Geo- grafisk Tidskrift, viii. 91 ; Hodge, Handbook of Indians of Canada, p. 275 (Eskimo). 2 Curr, The Australian Race, i. 107. Idem, Recollections of Squat- ting in Victoria, p. 248 (Bangerang). Beveridge, Aborigines of Victoria and Riverina, p. 22. Wilhelmi, ' Manners and Customs VOL. II Q Q 594 THE HISTORY OF HUMAN MARRIAGE chap. South Sea Islanders1 and several Asiatic 2 and African 3 peoples. Statements to this effect, however, need not indicate the complete absence of marriage rites. It will often be found, as Mr. Crawley points out, that " there is of the Australian Natives, in particular of the Port Lincoln District, ' in Trans. Roy. Soc. Victoria, v. 179. Withnell, Customs and Tra- ditions of the Aboriginal Natives of North Western Australia, p. 16. Brown, ' Three Tribes of Western Australia,' in Jour. Roy. Anthr. Inst, xliii. 158 (Kariera tribe). Browne, ' Die Eingebomen Aus- traliens,' in Mittheil. aus Justus Perthes' Geographischer Anstalt, 1856, p. 450 (natives of King George's Sound). 1 Best, in Trans, and Proceed. New Zealand Inst, xxxvi. 35 (common people among the Maori). Elton, ' Natives of the Solomon Islands,' in Jour. Anthr. Inst. xvii. 94 (natives of Ugi and San Cristoval). Parkinson, Dreissig Jahre in der Sudsee, p. 160 (Baining of New Britain), de Rochas, La Nouvelle Caledonie, p. 231. Miklucho-Maclay, ' Ethnographische Bemerkungen uber die Papuas der Maclay-Kiiste in Neu-Guinea,' in Natuur kundig Tijdschrift voor Nederlandsch Indie, xxxvi. 299. Rawling, Land of the New Guinea Pygmies, p. 131 (natives of the Mimika district of Dutch New Guinea). Finsch, Neu-Guinea, p. 62 (Outanatas). Kubary, ' Die Bewohner der Mortlock Inseln,' in Mittheil. d. Geograph. Gesellsch. in Hamburg, 1878-79, p. 260. 2 Knocker, ' Aborigines of Sungei Ujong,' in Jour. Roy. Anthr. Inst, xxxvii. 293 (Orang Bukit). Dalton, op. cit. p. 19 (Chalikata Mishmis). Rivers, Todas, p. 504. Dahmen, ' Paliyans, a Hill- Tribe of the Palni Hills (South India),' in Anthropos, iii. 27. Lowis, Census of India, 1911, vol. ii. (Andaman and Nicobar Islands) p. 100 (Nicobarese). Sarasin, Ergebnisse naturwissenschaftlicher Forschungen auf Ceylon, iii. 460 sq. ; Nevill, * Vaeddas of Ceylon,' in Taprobanian, i. 178; Deschamps, 'Les Veddas de Ceylan,' in L' Anthropologic, ii. 313 (some of the Veddas). Jochelson, Yukaghir, p. 89 (some of the Yukaghir). Labbe, ' L'ile de Sakhaline,' in Bull. Soc. geograph, commerciale Paris, xxiii. 124. 3 Campbell, Travels in South Africa, p. 439 ; Idem, Travels in South Africa, Second Journey, i. 29 sq. (Bushmen). Magyar, Reisen in Sud-Afrika, p. 282 (Kimbunda). Decle, op. cit. p. 78 (Barotse). Duff, Nyasaland under the Foreign Office, p. 316. Baumann, Durch Massailand zur Nilquelle, p. 172 (Wataturu). Hinde, The Last of the Masai, p. 71. Delhaise, Les Warega, pp. 169, 177. Delafosse, ' Les Agni,' in L' Anthropologic, iv. 429. Ruelle, ' Notes anthro- pologiques sur quelques populations noires du 2« territoire militaire de 1'Afrique occidentale fran^aise,' ibid. xv. 661 (Lobi). Pout on, ' Istude ethnographique des Baya,' ibid. xxvi. 124. Clozel and Villamur, Les coutumes indigenes de la C6te d'Ivoire, p. 101 sq. (Baoule). de Espinosa, Guanches of Tenerife, p. 35. XXVI MARRIAGE RITES 595 some act performed which is too slight or too practical to be marked by an observer as a ' ceremony,' but which when analysed turns out to be a real marriage rite " / and even when positive rites are wanting, there may be abstinences of some kind or other connected with the conclusion of a marriage. The marriage ritual is particularly profuse among peoples who have reached a higher degree of culture and among tribes which have been in close contact with such peoples. It is nowhere richer than among the peasantry of Indo-European nations and among peoples of Semitic culture ; but among the latter it seems largely to be of comparatively recent origin and distinctly suggests that Indo-European influence has been at work.2 On the other hand, modern civilisation has proved destructive to the old rituals and has had practically nothing new to add in- stead. This is a natural consequence of the fact that the large bulk of marriage rites have originated in magical ideas which have vanished along with the progress in intellectual culture. 1 Crawley, op. cit. p. 318. 2 See my Marriage Ceremonies in Morocco, passim. END OF VOL. II.