THE HISTORY OF HUMAN 
MARRIAGE 
VOL. I THE HISTORY OF 
HUMAN MARRIAGE 
BY 
EDWARD WESTERMARCK. 
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. 1 
FIFTH EDITION REWRITTEN 
THE ALLER 1'ON BOOK COMPANY 
NEW YORK 
1922 PREFACE 
During the thirty years which have passed since the 
publication of the first edition of the present work the study 
of marriage and matters connected with it, especially among 
the lower races, has made such progress that I have found 
it necessary to reconsider the whole subject. Many new 
facts have been incorporated, and some old ones have been 
omitted. Various aspects of marriage, which were previously 
dealt with very inadequately or hardly touched upon, have 
been discussed at length. Objections raised by critics have 
been carefully considered. The old theories have been in 
some cases strengthened but in other cases modified. New 
theories set forth by other writers have been scrutinised. 
The matter has in many points been rearranged ; and the 
book has been rewritten throughout to such an extent that 
very few sentences of the earlier editions have remained 
unchanged. 
Only some of the changes can be here briefly indicated. 
A new introductory chapter on method, largely dealing 
with problems of recent growth, has taken the place of the 
old one. The statements quoted by certain writers as 
evidence of peoples living in a state of promiscuity have 
been more carefully examined, and the customs which have 
. been represented as survivals of such a state in the past have 
been more fully discussed. Thus the subjects of the jus 
^rimae noctis, religious prostitution, and the lending or 
exchange of wives now occupv sixtv-nine. pages instead of 
V VI 
PREFACE 
nine. In the treatment of the classificatory system of 
relationship the recent contributions to the subject, which 
largely tend to confirm my old views, have been taken notice 
of. In the discussion of the marriage age and certain other 
matters more attention has been paid to the laws of civilised 
countries. Religious celibacy and sexual modesty have each 
got a special chapter. The origin of female coyness has 
been discussed. With reference to the secondary sexual 
characters a suggestion has been made which, if correct, 
brings the sexual colours, odours, and sounds of animals 
into the closest possible analogy with the colours and odours 
of the flowers of plants. In the chapters on primitive means 
of attraction the older theories, though in some measure 
supported by new evidence, have in certain points been 
modified in accordance with the results of later research. 
A more thorough investigation of the exogamous rules has 
confirmed my belief in the substantial accuracy of my earlier 
theory as to their origin; and I hope that the restatement 
of it, in which the objections of critics have been taken into 
consideration, has made it more acceptable. The chapters 
on marriage by capture and marriage by consideration, to- 
gether with kindred subjects, contain copious additions and 
changes. 
The extremely defective treatment of marriage rites, 
which covered thirteen pages only, has been replaced by 
three chapters of more than twelve times that length ; but 
for a study of the marriage ritual as a sequence I must refer 
the reader to my book Marriage Ceremonies in Morocco. 
In the course of my six years' research in Morocco, and 
through books like The Mystic Rose and The Golden Bough, 
I have become aware of the enormous influence of magical 
beliefs on marriage rites ; whereas the value of these rites 
for the study of earlier forms of marriage now seems to me 
to be even less than I thought before. I am grateful to PREFACE 
VII 
the distinguished reviewer of the first edition of this work 
who expressed the belief that if its author had been a 
student of folk-lore he would in various cases have arrived 
at different conclusions. It drew my attention to a defect 
which I have since then endeavoured to remedy; but my 
conceptions of the earlier history of marriage have not 
been essentially changed thereby. 
In my discussion of monogamy and polygyny, and in 
many other sections of the book as well, I have considered 
the influence which economic conditions have exercised 
upon marriage, a point which was also greatly neglected 
in the earlier editions. Polyandry, which was dealt with 
on a few pages, now forms the subject-matter of two whole 
chapters ; and the question of group-marriage, which of 
late has much occupied the minds of sociologists, has been 
discussed in a chapter by itself. The treatment of divorce 
is much more detailed, both as regards the history of the 
subject and the present legislation on it. The list of 
authorities quoted has increased from thirty pages to over 
a hundred ; and the work as a whole has been expanded 
from one volume into three. In short, it is a new work 
much rather than a new edition. 
At the same time, amidst all the changes, the general 
character, as well as the structure, of the book has remained 
unchanged. The criticism passed on it has not essentially 
affected either its method or its fundamental ideas. This 
may perhaps be due to the fact that, although I opposed 
many theories in vogue at the time when the book was first 
written and the method which had led to them, my decision 
to write it did not spring from a desire for opposition. On 
the contrary, I commenced my work as a faithful adherent 
of the theory of primitive promiscuity and tried to discover 
fresh evidence for it in customs which I thought might be 
interpreted as survivals from a time when individual marriage PREFACE 
did not exist. I had not proceeded far, however, when I 
found that I was on the wrong track. I perceived that 
marriage must primarily be studied in its connection with 
biological conditions, and that the tendency to interpret all 
sorts of customs as social survivals, without a careful 
examination into their existing environment, is apt to lead 
to the most arbitrary conclusions. Later treatises on the 
subject have only confirmed this conviction ; and the present 
revival of the old method is not, in my opinion, likely to 
yield lasting results. 
I am indebted to the Press and public both in this country 
and elsewhere for the encouraging interest they have taken 
during all these years in the work of my youth ; to the earlier 
translations of it have been added subsequent ones into 
Spanish and Japanese. I am again under obligation to 
friends and correspondents for valuable advice and informa- 
tion. The largest part of the material has been collected 
in the Reading Room of the British Museum, and I take this 
opportunity to thank its officials for their unfailing courtesy. 
E. W. 
Woodman's Cottage, 
Boxhill, Surrey, 
May, 1921. INTRODUCTORY NOTE TO THE 
FIRST EDITION 
ALFRED R. WALLACE 
Having read the proofs of Mr. Westermarck's book I am 
asked by the publishers to say a few words by way of intro- 
ducing the work to English readers. This I have great 
pleasure in doing, because I have seldom read a more 
thorough or a more philosophic discussion of some of the 
most difficult, and at the same time interesting, problems of 
anthropology. 
The origin and development of human marriage have been 
discussed by such eminent writers as Darwin, Spencer, 
Morgan, Lubbock, and many others. On some of the more 
important questions involved in it all these writers are in 
general accord, and this agreement has led to their opinions 
being widely accepted as if they were well-established 
conclusions of science. But on several of these points 
Mr. Westermarck has arrived at different, and sometimes 
diametrically opposite, conclusions, and he has done so 
after a most complete and painstaking investigation of all 
the available facts. 
With such an array of authority on the one side and a 
hitherto unknown student on the other, it will certainly be 
thought that all the probabilities are against the latter. 
Yet I venture to anticipate that the verdict of independent 
thinkers will, on most of these disputed points, be in favour 
IX X 
INTRODUCTORY NOTE TO THE FIRST EDITION 
of the new comer who has so boldly challenged the con- 
clusions of some of our most esteemed writers. Even those 
whose views are here opposed, will, I think, acknowledge 
that Mr. Wcstermarck is a careful investigator and an acute 
reasoner, and that his arguments as well as his conclusions 
are worthy of the most careful consideration. 
I would also call attention to his ingenious and philo- 
sophical explanation of the repugnance to marriage between 
near relatives which is so very general both among savage 
and civilised man, and as to the causes of which there has 
been great diversity of opinion; and to his valuable 
suggestions on the general question of sexual selection, in 
which he furnishes an original argument against Darwin's 
views on the point, differing somewhat from my own though 
in general harmony with it. 
Every reader » the work will admire its clearness of 
style, and the wonderful command of what is to the author 
a foreign language. PREFACE TO THE FIRST EDITION 
I need scarcely say how fully I appreciate the honour 
of being introduced to English readers by Mr. Alfred R 
Wallace. I am also greatly obliged for his kindness in 
reading the proofs, and in giving me the benefit of his 
advice with regard to various parts of the subject. 
It is difficult for me to acknowledge sufficiently my 
obligations to Mr. James Sime for his assistance in preparing 
this book for the press. The work, as originally written, 
naturally contained a good many foreign modes of expres- 
sion. Mr. Sime has been indefatigable in helping me to 
improve the form of the text; and, in our discussions on 
the main lines of the argument, he has made several important 
suggestions. I am sincerely obliged for the invaluable aid 
he has given me. 
My cordial thanks are due to Mr. Charles J. Cooke, 
British Vice-Consul at Helsingfors, who most kindly aided 
me in writing the first part of the book in a tongue which 
is not rny own. I am indebted also to Dr. E. B. Tylor, 
Professor G. Croom Robertson, Mr. James Sully, and 
Dr. W. C. Coupland for much encouraging interest; to 
Mr. Joseph Jacobs for the readiness with which he has 
placed at my disposal some results of his own researches ; 
and to several gentlemen in different parts of the world who 
have been so good as to respond to my inquiries as to their 
personal observation of various classes of phenomena 
XI XII 
PREFACE TO THE FIRST EDITION 
connected with marriage among savage tribes. The informa- 
tion I have received from them is acknowledged in the 
passages in which it is used. 
A list of authorities is given at the end of the book- 
between the text and the index,-and it may be well to add 
that the references in the notes have been carefully verified. 
E. W. 
London, May, 1891. 
PREFACE TO THE SECOND EDITION 
In this new edition of my book I have made no essential 
changes, but here and there the argument has been 
strengthened by the addition of facts which have come 
to my knowledge since the appearance of the first edition. 
The most important of these new facts will be found in the 
second chapter. 
I take this opportunity of expressing my warm apprecia- 
tion of the thorough way in which the ideas set forth in this 
book have been discussed by many critics in England and 
elsewhere. Translations of the work have appeared, or 
are about to appear, in German, Swedish, French, Italian, 
and Russian. 
E. W. 
London, January, 1894. PREFACE TO THE THIRD EDITION 
I much regret that the demand for a new edition of this 
book should come at a time when circumstances prevent me 
from undertaking such a revision of the work as I feel to be 
required. Since the appearance of the Second Edition many 
important facts bearing upon the subject have been brought 
to light, new theories have been advanced, and old theories, 
supported by fresh arguments, have been revived. To all 
this, however, I can do no justice, as I am at present being 
engaged in anthropological research in Morocco. This 
edition is, in consequence, a mere reprint of the second. 
But I purpose, after my return to Europe, to issue an 
Appendix, in which the book will be brought more up to 
date and some criticism will be replied to. 
E. W. 
Mogador (Morocco), 
August, 1901. 
XIII CONTENTS 
INTRODUCTION 
ON THE METHOD OF INVESTIGATION 
The comparative method in sociology, p. i sq.-The resemblances of 
culture-phenomena due partly to mental similarity and partly to 
transmission, p. 2 sq.-The difficulty in deciding whether they are 
due to the one cause or to the other, p. 3 sq.-The ethnological method, 
so unfriendly to the idea of independent origins, has itself inde- 
pendently originated in two different countries, p. 4 sq.-The customs 
of savages subject to spontaneous changes, which may lead to similar 
results in different cases, p. 5 sq.-The investigations of the evolu- 
tionary school, which chiefly endeavours to find the psychological 
and sociological origin of culture-phenomena, and those of the ethno- 
logical school, which is concerned with their wanderings, supplement 
each other, p. 6.-How the comparative method helps the sociologist 
to find the origin of customs, p. 7 sq.-Dr. Rivers' criticism of the 
endeavour to discover the psychical causes of social phenomena, 
p. 8 sq.-Social survivals, p. 9 sq.-Dr. Rivers' distrust in our capacity 
of learning the motives by which social conduct is determined, 
p. 10 sq.-The field-ethnologist's inquiry into motives, p. n sq.- 
The hypothetical character of many explanations of social phenomena, 
P- 12--The complaint that the use of the comparative method is 
hardly compatible with a sufficiently careful scrutiny of authorities 
and sources, p. 12 sq.-The trustworthiness of ethnographical evidence, 
p. 13 sq.-The complaint that the comparative method detaches the 
cultural phenomenon from the organic whole of which it forms a 
part and thereby easily represents it in a wrong light, pp. 14-17.- 
The study of a cultural phenomenon as it is distributed among 
different races and the study of it which is restricted to a particular 
ethnic group complement each other, p. 17 sq.-The homogeneous 
elements of the human mind underrated and the homogeneity of 
the group-mind overrated by the school of Durkheim, pp. 17-19.- 
An error of method prevalent among the evolutionary school, p. 19 sq. 
-The only condition on which the universal prevalence of a social 
phenomenon in the past may be assumed, p. 20 sq.-Criticism of the 
method of Durkheim, p. 21.-Biological facts underlying marriage 
and many particular customs and rules relating to it, p. 22.-The 
influence which people's ideas and beliefs have exercised upon their 
marriage customs, ibid.-The influence of industrial culture, p. 22 sq.- 
The particular subjects discussed in the present work, p. 23.-The 
method of dealing with them, pp. 23-25. 
XV XVI 
CONTENTS 
CHAPTER I 
THE ORIGIN OF MARRIAGE 
Definition of marriage as a social institution, p. 26 sq.-The institution 
of marriage has probably developed out of a primeval habit, p. 27 sq.- 
The relations between the sexes and parental care among the Inverte- 
brata, p. 28.-Among Fishes, p. 29.-Among Reptiles, ibid.-Among 
Birds, p. 29 sq.-Among the lower Mammals, p. 30 sq.-Among the 
Quadrumana, pp. 31-35.-Why in certain species male and female 
remain together till after the birth of the offspring, p. 35 sq.-The 
need of marital and paternal protection among the man-like apes, 
p. 36 sq.-The family consisting of parents and children, with the 
father as its protector, in mankind, p. 37 sq.-A few cases in which 
the husband is said not to live with his wife, p. 38 sq.-Cases in which 
the mother's brother is said to have greater rights over a child than 
the father, p. 40 sq.-The elementary paternal duties seem to be 
universally recognised while the children live in the father's house, 
pp. 41-43.-Paternal rights among matrilineal peoples, p. 43 sq.- 
Dr. Hartland's hypothesis that the authority of the father and even 
that of the maternal uncle are the results of later development, 
pp. 44-46.-That the functions of the husband and father in the 
family involve the duty of protecting and supporting the wife and 
children, testified by an array of facts, pp. 46-49.-A man often pro- 
hibited from marrying until he has given som.e proof of his ability to 
fulfil these duties, pp. 49-53.-A man sometimes obliged to support 
his fiancee, p. 53.-To support his divorced wife, ibid.-The duty of 
marrying the widow of a deceased brother, ibid.-The more or less 
durable union between man and woman and marital and paternal 
care probably due to instincts once necessary for the preservation 
of the species, p. 53 sq.-The social conditions of savages who know 
neither cattle-rearing nor agriculture, unless perhaps of the most 
primitive kind, pp. 54-68.-Unreasonable to suppose that primitive 
men were more permanently gregarious than many modern savages, 
p. 68 sq.-The family in mankind presumably an inheritance from the 
parent species out of which the Anthropoids and the Hominides 
gradually developed, p. 69.-The tendency of habits to become rules 
of conduct, p. 69 sq.-The marital and parental instincts, p. 70 sq.- 
They give rise not only to habits but to rules of custom, or institu- 
tions, p. 71.-An alternative definition of the term "marriage," 
p. 71 sq.-Marriage rooted in the family rather than the family in 
marriage, p. 72.-Cases in which true married life does not begin, or 
a marriage does not become definite, until a child is born or there are 
signs of pregnancy, pp. 72-75.-Cases in which the occurrence of 
pregnancy or child-birth is a preliminary to, or as a general rule 
leads to, marriage, pp. 75-77. 
• 
CHAPTER II 
A HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
Hypotheses as to the periodicity in the sexual life of animals, p. 78.- 
The sexual season adapted to the requirements of each species, 
p. 79 sq.-Wild species without a definite pairing season, p. 80 sq.- 
A sexual season among the man-like apes, p. 81.-The pairing of our 
earliest human or half-human ancestors probably likewise restricted 
to a certain season of the year, ibid.-Peoples who are said to have CONTENTS 
XVII 
an annual pairing time or whose sexual desire seems or is supposed to 
undergo a periodical increase at a certain time of the year, pp. 81-91. 
-The doubtful testimony of certain feasts at which sexual licence 
prevails, pp. 90-92.-The custom of contracting or celebrating 
marriages at some special time of the year, pp. 92-94.-The conclusion 
that an annual increase of the sexual desire or of the reproductive 
power, generally in spring, is of frequent occurrence in mankind, 
supported by statistical data relating to the distribution of births 
over the different months of the year, pp. 94-96.-This increase 
probably a survival of an ancient pairing season, p. 97.-The winter 
maximum of conceptions, pp. 97-100.-Why man is no longer 
restricted to any particular season for the procreation, pp. 100-102.- 
Variations in the sexual life of domesticated animals and of wild 
animals kept in captivity, p. 101.-The bearing which the hypothesis 
of a pairing season in primitive times has upon the question of the 
origin of marriage, p. 102. 
CHAPTER III 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : ALLEGED INSTANCES 
OF PEOPLES LIVING IN PROMISCUITY 
The hypothesis that the human race originally lived in a state of 
promiscuity, p. 103 sq.-The general nature of the evidence adduced 
in support of it, p. 104 sq.-Traditions of a time when marriage did 
not exist, p. 105 sq.-Greek and Roman statements about tribes 
living in promiscuity, pp. 106-110.-Promiscuity ascribed by 
mediaeval writers to certain European peoples, pp. 110-112.- 
Examination of alleged instances of promiscuity mentioned by Lord 
Avebury, pp. 112-116.-By Bernhoft, p. 116 sq.-By Bastian, 
p. nq sq.-By Garcilasso de la Vega and Admiral Fitzroy, p. 118 sq.- 
By Wilken, pp. 119-123.-By Pleyte, R. Schmidt, and van der Lith, 
p. 122 sq.-By Wilutzky and Giraud-Teulon, p. 123 sq.-No known 
savage people living in promiscuity, p. 124.-The hypothesis of a 
general stage of promiscuity in any case entirely groundless, ibid.- 
Sexual relations most nearly approaching to promiscuity not found 
among the very lowest races, but among more advanced peoples, 
p. 124 sq. 
CHAPTER IV 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY I PRE-NUPTIAL 
UNCHASTITY 
The sexual freedom often enjoyed by both sexes among uncivilised peoples 
previous to marriage regarded as evidence of primitive promiscuity, 
p. 126.-Many of the facts could not, in any circumstances, be relics 
of such a state, because they are known to be of later growth, 
pp. 126-131.-Or because they do not represent promiscuity at all, 
sexual connections between a boy and a girl very frequently being 
a preliminary to their marriage, pp. 131-136.-Marriage upon trial, 
p. 135.-Instances of promiscuity outside marriage, p. 136.- 
Prostitution among uncivilised peoples, pp. 136-138.-Though of 
frequent occurrence, pre-nuptial unchastity not anything like a 
universal characteristic of the lower races, pp. 138-157.-Promiscuity 
among the unmarried an exception in the customs of unadulterated 
savages, p. 157.-The standard of pre-nuptial chastity in a tribe not 
VOL. I b XVIII 
CONTENTS 
proportionate to its degree of culture, p. 158 sq.-In the lowest tribes 
chastity apparently more respected than in the higher ones, p. 159 sq.- 
Causes of pre-nuptial unchastity, pp. 159-164.-Why promiscuous 
pre-nuptial unchastity cannot be regarded as a survival of earlier 
general promiscuity, pp. 162-164.-Preference given to virgin brides, 
p. 163.-The taste for variety does not justify the hypothesis of 
original promiscuity, p. 163 sq.-Courtesans sometimes held in greater 
estimation than married women, p. 164 sq.-Cannot be regarded as 
representatives of " communal wives " in primitive times, p. 165. 
CHAPTER V 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY I THE JuS Primae 
Noctis 
The jus primae noctis accorded to a priest, king, chief, or nobleman inter- 
preted as expiation for individual marriage, p. 166.-Instances of 
such a right, or of customary defloration by such a person, pp. 166-174. 
-The supposed existence of a droit du seigneur in feudal Europe and 
of a similar right accorded to the clergy, pp. 174-180.-A recent 
instance of the jus primae noctis reported from Albania, p. 180.- 
How the right in question may be explained, pp. 180-196.- 
Reluctance on the part of the bridegroom to deflower the bride, or 
to do so in the manner indicated by nature, pp. 180-191.-Artificial 
defloration by the bridegroom, p. 180 sq.-By somebody else, 
p. 181 sq.-Defloration by extra-matrimonial intercourse an opera- 
tion which the husband is anxious to avoid, pp. 182-189.-The 
tali kettu ceremony of the Nayars presumably a relic of pre-nuptial 
defloration, pp. 184-187.-Defloration by a foreigner, p. 187 sq.- 
By the father of the girl, p. 188 sq.-Why a bridegroom is reluctant 
to have intercourse with a virgin bride, pp. 189-191.-The fear of 
defloration closely connected with the fear of hymeneal blood, p. 190 sq. 
-Why defloration is performed by a holy man, pp. 191-193.-The 
defloration by the king regarded as a duty incumbent on him, p. 193 sq. 
-The jus primae noctis as a consequence of might, pp. 194-196.- 
Why it cannot in any case be regarded as the relic of an ancient 
communal right, p. 196.-Sexual intercourse with a bride or an 
unmarried girl, involving defloration, accorded to wedding guests or 
friends, pp. 196-199.-Girls earning a dowry by prostitution before 
they marry, p. 200.-Certain defloration customs in Australia, 
pp. 200-204. In East Africa, pp. 204-206. 
CHAPTER VI 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : RELIGIOUS PROSTITU- 
TION THE LENDING AND EXCHANGE OF WIVES FEASTS 
The temporary prostitution of a girl before marriage or of a matron at a 
temple or in the service of religion, pp. 207-210.-Theories adduced 
to explain it, pp. 210-213.-The author's theory, pp. 213-218.- 
The more or less permanent prostitution of women attached to a 
temple, pp. 219-223.-How it may be explained, p. 223 sq.-The 
custom of lending wives to visitors, pp. 225-230.-The temporary 
exchange of wives, pp. 230-234.-Promiscuous intercourse indulged 
in at certain feasts, p. 234 sq. CONTENTS 
XIX 
CHAPTER VII 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : THE CLASSIFICATORY 
SYSTEM OF RELATIONSHIP 
" Descriptive " and " classificatory systems of relationship," pp. 236-238.- 
The classificatory systems, p. 238 sq.-Conclusions drawn from the 
classificatory terms of relationship with reference to earlier marriage 
customs or relations between the sexes, pp. 239-242.-Terms of re- 
lationship borrowed from the children's lips, pp. 242-245.-No classi- 
ficatory term known to refer directly to the act of begetting or to the 
fact of having been begotten, p. 245 sq.-'The classificatory terms 
influenced by sex, p. 246.-By age, pp. 246-249.-By social relations, 
probably from the beginning combined with vague ideas of consan- 
guinity, pp. 249-257.-The classificatory terms used as terms of 
address, p. 251 sq.-The social factor also presumably felt in the 
distinction between persons of different sex and of different age, 
pp. 253-255.-The influence of the habit of living together, p. 255.- 
The importance of the social factor evident in terms used in addressing 
strangers, pp. 255-257.-The correlation between the presence of a 
term of relationship and special social relations or functions associated 
with it by no means complete, pp. 257-259.-To infer the earlier 
existence of certain social conditions from certain terms of relationship 
can in most cases be nothing more than a guess, p. 259 sq.-Dr. Rivers' 
conclusion that certain classificatory terms indicate the earlier existence 
of cross-cousin marriage, pp. 260-263.-A common nomenclature for 
persons representing two or more relationships no indication of a 
certain form of marriage in the past, pp. 263-267.-The classificatory 
system gives no evidence of the earlier existence of group-marriage 
or sexual communism, pp. 267-274. 
CHAPTER VIII 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY I MOTHER-RIGHT 
Mother-right supposed to be due to uncertain paternity resulting from 
early promiscuity, p. 275 sq.-What the term " mother-right " implies, 
p. 276 sq.-The prevalence of mother-right and of father-right, 
p. 277 sq.-The supposition that mother-right has everywhere preceded 
father-right, pp. 278-280.-No evidence that mother-right is more 
primitive than father-right, pp. 280-283.-The uncertainty of father- 
hood, pp. 283-285.-No general coincidence of what we consider moral 
and immoral habits with the prevalence of father- and mother-right 
among existing savages, p. 284 sq.-Ideas regarding the cause of 
pregnancy or child-birth, pp. 285-294.-The question whether ideas 
of this sort have influenced the method of counting descent, p. 294 sq.- 
Mother-right as well as father-right depending on social conditions, 
pp. 295-297.-The tie between mother and child stronger than that 
which binds the child to its father, p. 295.-The influence of polygyny, 
p. 295 sq.-Of matrilocal marriage, p. 296 sq.-The various facts 
adduced as evidence of the hypothesis of promiscuity nO evidence at 
all, p. 297 sq. XX 
CONTENTS 
CHAPTER IX 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : MASCULINE JEALOUSY 
Masculine jealousy a bar to promiscuity, p. 299 sq.-The supposition that 
savages are little addicted to jealousy, p. 300.-The connection between 
jealousy and the sense of ownership, p. 300 sq.-The nature of jealousy, 
p. 301 sq.-The prevalence of male jealousy among the lower races, 
pp. 302-310.-Among civilised nations, p. 310 sq.--Customs and laws 
concerning adultery, pp. 311-316.-Jealousy with reference to the 
past, p. 316 sq.- The demand that a wife shall belong to her husband 
not only during his lifetime but after his death, pp. 317-328.-Widows 
killed, pp. 317-320.-Certain duties of widows, p. 320 sq.-Widows 
forbidden to marry again or to remarry for a certain period, 
pp. 321-325.-The causes of these prohibitions, pp. 325-329.- 
Certain formalities required before the remarriage of a widow, 
PP- 327-329--Widowers compelled to live single for a certain length 
of time or sometimes for the rest of their lives, p. 329 sq.-The causes 
of these rules, p. 330 sq.-Criticism of the inference that certain savage 
customs prove the lack of jealousy, p. 331 sq.-The instances of 
peoples little addicted to jealousy give no support to the theory of 
promiscuity, pp. 332-334.-Criticism of the suggestion that man's 
gregarious way of living made promiscuity a necessity, p. 334.-The 
peculiar violence of masculine jealousy possibly due to infertility 
produced by promiscuity, pp. 334-336.-Certain statements relating 
to polyandrous peoples apparently support the view that intercourse 
between one woman and several men is unfavourable to reproduction, 
p. 335 -The hypothesis of promiscuity essentially unscientific, 
P- 336. 
CHAPTER X 
THE FREQUENCY OF MARRIAGE AND THE MARRIAGE AGE 
Among the uncivilised races, as a general rule, nearly every man endeavours 
to marry when he has reached the age of puberty and practically 
every woman gets married, pp. 337-343.--Unmarried persons looked 
upon as unnatural beings or objects of contempt or ridicule, p. 343 sq.- 
Among savages girls married at an earlier age than among the peoples 
of Western civilisation and the same probably in most cases true of 
the men, pp. 344-361.-Instances from America, pp. 345-349.- 
From Asia, pp. 349-352.-From the Indian Archipelago and the 
islands of the Pacific Ocean, pp. 352-356.-From Australia, pp. 356- 
358.-From Africa, pp. 358-361.-To most savages marriage seems 
absolutely indispensable, pp. 361-363.-Little difficulty experienced 
by savages in supporting a family, p. 363.-Circumstances in savage 
life which may compel persons to live unmarried for a longer or shorter 
. time, pp. 363-370.--The influence of polygyny on the frequency of 
marriage, pp. 366-368.-A military life a hindrance to marriage, 
pp. 368-370.-The causes of infant- or child-betrothal, p. 370 sq.- 
Younger sons or daughters must not marry before the elder ones, 
PP- 372-374--Among peoples of archaic culture also, celibacy a 
great exception and marriage regarded as a duty, pp. 374-387.- 
Infant-macriage in India, pp. 380-383.-The lowest ages at which 
marriage is allowed in Christian countries, p. 387 sq.-Modern civilisa- 
tion has proved unfavourable to the frequency of marriage and has 
raised the average age at which marriages are concluded, pp. 388-390. 
-The causes of the decline of the marriage-rate and of the rise of the 
age for marriage in countries of Western civilisation, pp. 390-394. CONTENTS 
XXI 
CHAPTER XI 
CELIBACY 
Celibacy in extraordinary cases commanding more respect than marriage 
or even looked upon as a stringent duty, pp. 395-400.-Christian 
views and rules regarding celibacy,, pp. 400-403.-The causes of 
religious celibacy, pp. 403-417.-The' priestess regarded as married 
to the god whom she is serving, pp. 403-405.-Religious celibacy a 
means of self-mortification supposed to appease an angry god or 
intended to raise the spiritual nature of man, p. 405 sq.-The idea 
that sexual intercourse is defiling and in certain circumstances a 
mysterious cause of evil, pp. 406-413.-The idea of sexual defilement 
particularly conspicuous in connection with religious observances, 
pp. 409-413.-The effect of pollution upon holiness, pp. 410-413.- 
Sexual pollution to be particularly avoided by priests, p. 413 sq.- 
Admission into the priesthood preceded by a period of continence, 
p. 413.-The notion that celibacy is more pleasing to God than 
marriage, p. 414 sq.-Why sexual intercourse is looked upon as defiling, 
PP- 
CHAPTER XII 
SEXUAL MODESTY 
The nature of sexual modesty, p. 418 sq.-Some degree'of sexual modesty 
probably universal in mankind, pp. 419-424.-It shows itself in 
various ways in the preliminaries to, or the conclusion of, a marriage, 
pp. 424-429.-Various hypotheses as to the origin of sexual modesty, 
pp. 429-432.-The coyness of the female when courted by the male 
one root of it in mankind, p. 432 sq.-Sexual modesty also closely 
connected with the aversion to incest, as appears from the fact that 
it is especially strongly developed in relations between members of 
the same family circle or kin, pp. 433-439.-The taboos imposed 
on the intercourse between a married man and his parents-in-law or 
other members of his wife's family, pp. 439-442.-Criticism of various 
theories on the origin of such taboos, pp. 442-452.-The tendency 
in persons to feel sexual shame particularly in the presence of members 
of the family circles of their mates and in the presence of members 
of their own family circles an offshoot of the aversion to sexual inter- 
course between persons who have grown up in the same home, p. 453 sq. 
CHAPTER XIII 
COURTSHIP 
The male generally plays a more active and the female a more passive 
part in courtship, p. 455 sq.-The comparative strength of the sexual 
impulse in males and in females, p. 456.-Females making the first 
advances among some birds, ibid.-Among not a few uncivilised 
peoples the women said to be the courters or to make proposals of 
marriage, pp. 456-462.-The males fighting for females among the 
lower animals, p. 462.-Among men, pp. 462-466.-The rivalry of 
the suitors taking less violent and more playful forms, pp. 466-469.- 
Peaceful methods of courtship among men, pp. 469-471.-Among the XXII 
CONTENTS 
lower animals, p. 471 sq.-Where the male is the courter the female 
does not remain completely passive, pp. 472-475.-The coyness of 
the female, pp. 473-475.-The choice exercised by the female, p. 475.- 
The fighting of the males acts as a sexual stimulant, ibid. 
CHAPTER XIV 
SECONDARY SEXUAL CHARACTERS IN ANIMALS-^FEMALE COYNESS 
Darwin's theory of sexual selection, p. 477 sq.-The contradiction between 
the theories of natural and sexual selection, p. 478 sq.--The colours 
of flowers, p. 479.-Wallace's theory of the sexual colours of animals, 
p. 479 sq.-The sexual colours of animals seem to have been developed 
because they make it easier for the sexes to find each other during the 
pairing time, p. 481.-"They occur exactly in those species whose 
habits make these colours most visible, pp. 481-483.-The odours 
of flowers, p. 483.-Sexual odours and sounds among animals, 
p. 483 sq.-The sexual colours, odours, and sounds of animals comple- 
mentary to each other in the way that is best suited to make the 
animals easily discoverable, pp. 484-486.-The untenableness of 
Darwin's theory, p. 486.-The secondary sexual characters due to 
natural selection, p. 486 sq.-Wallace's views, p. 487 sq.-Sexual 
colours, sounds, and odours useful to the species not only because 
they facilitate reproduction by making it easier for the sexes to find 
each other, but probably also because they tend to prevent in-breeding 
by attracting individuals from a distance, p. 488 sq.-Animal " orna- 
ments," p. 489.-The great variability of the secondary sexual 
characters in different species, p. 489 sq.-Secondary sexual characters 
as sexual stimulants, p. 490 sq.-The part played by secondary sexual 
characters in courtship, p. 491.-The origin of feminine coyness, 
pp. 492-496. 
CHAPTER XV 
PRIMITIVE MEANS OF ATTRACTION 
The desire for self-decoration, p. 497 sq.-1The things which people attach 
to their bodies not all ornaments, p. 498 sq.-The ornanlentation of 
savages largely a sexual allurement, pp. 499-502.-Operations for 
the purpose of attaching objects to the body, pp. 502-505.-Ill- 
treatment of the teeth, pp. 505-507.-The hair-dress, pp. 507-510.- 
Savage delight in showy colours, p. 510.-The painting of the body, 
pp. 510-514.-The practice of tattooing, pp. 514-525.-The practice 
of scarification, pp. 525-527.-The motives for these various practices, 
pp. 527-529.-Savage women less decorated than savage men, 
pp. 529-532.-The sexual instinct of women more discriminating than 
that of men, p. 532 sq. 
CHAPTER XVI 
PRIMITIVE MEANS OF ATTRACTION 
(Concluded) 
Various motives for clothing, p. 534 sq.-The covering of the sexual organs 
partly due to the desire of protection, pp. 535-538.-Among various 
peoples only the married women cover themselves, p. 537 sq.-The 
fear of causing disgust as a motive for covering the sexual centre, CONTENTS 
XXIII 
p. 538 sq.-The theory that-clothing owes its origin to shame, p. 539 sq. 
-Savages who go naked without showing a trace of shame or who, 
when they dress themselves, are utterly neglectful of what we consider 
the first requirements of decency, pp. 540-546.-Sexual modesty 
among such savages, p. 546.-Nakedness not alluring, pp. 546-548.- 
Naked savages who are ashamed to cover themselves, p. 548.-The 
sort of covering savages often use, pp. 548-551.-Covering used as 
a means of embellishment, pp. 551-553.-Every reason to believe 
that in many cases at least the pubic covering was originally adopted 
as a sexual lure, pp. 553-558.-Naked savages covering themselves 
only at dances or festivals, p. 555 sq.-Peoples among whom only the 
marriageable girls cover their nakedness, p. 556 sq.-Covering resorted 
to at the age of puberty, p. 557.-Used only by harlots, p. 558.- 
Said to be more common among savage men than women, ibid.- 
The tattooing of the pubic region, p. 558 sq.-The sexual organs 
subject to mutilations or other practices, the purpose of which may 
have been to make the person more attractive to the opposite sex, 
PP- 559-56i.-Circumcision, pp. 561-564.-The tying up of the 
prepuce, p. 564.-Mutilations and other transformations of the sexual 
organs of girls, ibid.-Certain attitudes exhibited by women among 
naked tribes, p. 565.-The connection between covering and the 
feeling of shame, pp. 565-569.-The tendency to look for superstitious 
origins of savage customs, p. 569 sq.-Consideration of an argument 
brought against the theory here advocated, p. 570 sq. THE HISTORY OF HUMAN 
MARRIAGE 
INTRODUCTION 
ON THE METHOD OF INVESTIGATION 
The method followed in the present work is the com- 
parative method, which for half a century has been dominant 
among British students of social anthropology. When ap- 
plied to the study of human civilisation, this method starts 
from the fact that there are great similarities between the 
products of culture-such as implements, weapons, objects 
of art, customs, institutions, and beliefs-among different 
peoples in different countries. Weapons, for example, are 
classed under spear, club, sling, bow and arrow, and so 
forth. Myths are divided under such headings as myths of 
sunrise and sunset, eclipse-myths, earthquake-myths, and 
local myths which account for the names of places by some 
fanciful tale. Under religious beliefs and practices there 
are animism, totemism, ancestor-worship, polytheism, mono- 
theism. Under institutions occur, for instance, marriage, 
clanship, chieftainship, slavery; and under each heading there 
are sub-headings, like marriage by consideration, monogamy, 
polygyny, polyandry, group-marriage. These classifications 
of the various details of culture, as Tylor remarks, may be 
compared with the species of plants and animals as studied 2 
THE HISTORY OF HUMAN MARRIAGE 
by the naturalist. " To the ethnographer, the bow and arrow 
is a species, the habit of flattening children's skulls is a species, 
the practice of reckoning numbers by tens is a species. The 
geographical distribution of these things, and their trans- 
mission from region to region, have to be studied as the 
naturalist studies the geography of his botanical and zoo- 
logical species." 1 And the same is true of social institutions 
and their various aspects. 
But the task of comparative sociology is not restricted 
to that of classifying the different phenomena of culture 
with a view to making out their distribution in geography 
and history. Its ultimate object is, of course, the same as 
that of every other science, namely, to explain the facts with 
which it is concerned, to give an answer to the question, 
Why ? Hence, when similar customs, beliefs, legends, or 
arts, are found among different peoples, the question arises 
how the similarity is to be accounted for. In answer to this 
question Tylor made the following general statement:- 
" Sometimes it may be ascribed to the like working of men's 
minds under like conditions, and sometimes it is a proof of 
blood relationship or of intercourse, direct or indirect, 
between the races among whom it is found."2 Sir James 
G. Frazer likewise speaks of " the essential similarity in the 
working of the less developed human mind among all races, 
which corresponds to the essential similarity in their bodily 
frame revealed by comparative anatomy. But," he adds, 
" while this general mental similarity may, I believe, be 
taken as established, we must always be on our guard against 
tracing to it a multitude of particular resemblances which 
may be and often are due to simple diffusion, since nothing 
is more certain than that the various races of men have 
borrowed from each other many of their arts and crafts, 
their ideas, customs, and institutions." 3 I quote these state- 
ments in reply to the charge made in a Presidential Address 
to the Anthropological Section of the British Association a 
few years ago, that where similarities are found in different 
1 Tylor, Primitive Culture, i. 7 sq. 
2 Idem, Researches into the Early History of Mankind, p. 5. 
3 Frazer, Balder the Beautiful, vol. i. p. vi. sq. INTRODUCTION 
3 
parts of the world it is assumed by the leading school of 
British anthropologists, " almost as an axiom, that they are 
due to independent origin and development."1 So little is 
this charge in accordance with facts that Tylor, in his 
pioneer work ' Researches into the Early History of Man- 
kind,' found it hard to account for the occurrence in so many 
distant times and places of customs like the cure by sucking 
and the couvade, and of superstitions like those connected 
with sneezing, " on any other hypothesis than that of deep- 
lying connections by blood or intercourse, among races 
which history, and even philology, only know as isolated 
sections of the population of the world."2 
It is, no doubt, true that the question whether a certain 
custom or institution has sprung up spontaneously among the 
people or tribe practising it, or whether it has been introduced 
from some other people or tribe, is seldom discussed in 
comparative treatises. But this by no means implies the 
assumption of independent origins ; on the contrary, when 
the custom or institution occurs among related or neighbour- 
ing peoples, there is, at least in many cases, a tendency 
almost to take for granted that it has been derived from a 
common source in the different cases-that its prevalence 
is due either to a common descent or to social intercourse. 
One reason why the question of transmission is not more 
frequently discussed is the lack of evidence in the case of 
peoples whose history is unknown to us. Tylor justly 
spoke of " the constant difficulty in deciding whether any 
particular development is due to independent invention, or 
to transmission from some other people to those among 
whom it is found " ; 3 and this difficulty has certainly not 
been removed by later investigations. Dr. Graebnerlays down 
two main criteria which, he thinks, enable us to trace similar 
culture-phenomena to a common source : first, the criterion 
of form, as he calls it, that is, correspondence of qualities not 
inherent to the nature of the object, and secondly, that of 
1 Rivers, British Association far the Advancement of Science. 
Portsmouth, 1911. Address to the Anthropological Section, p. 2. 
2 Tylor, Researches, etc., p. 378 sq. 
• Ibid. p. 373. 4 
THE HISTORY OF HUMAN MARRIAGE 
quantitative correspondence, that is, the coincidence of 
several phenomena between which there is no necessary and 
intrinsic connection.1 But when customs are concerned, 
it may be doubted whether either criterion is, generally, 
sufficient to save the conclusion from being much more than 
a mere guess, unless the peoples in question belong to the 
same race or may on linguistic, historical, or geographical 
grounds be supposed to have had contact with each other. 
And even in such cases it may often be difficult or impossible 
to decide with certainty whether similar customs have a 
common origin or not. Dr. Graebner himself admits that 
it is possible, although not proved, that identical customs 
grow up independently among peoples in different parts of 
the world; if so, it is obviously also possible that identical 
customs grow up independently among peoples who are of 
the same stock or have come into contact with one another. 
Indeed, the more similar two peoples are, the greater is the 
probability that also new details in their culture should 
resemble each other; from seeds of the same kind very 
similar plants spring up. If the custom of providing a 
bride with a marriage portion is found among different 
Indo-European peoples, we are not therefore entitled to 
assume that this custom is either an inheritance from the 
primitive Indo-European period or has in historical times 
been adopted by one people from the other. And if 
marriage by purchase is found among two unrelated 
neighbouring tribes, it is by no means certain that one of 
them has borrowed this custom from the other, however 
many coincidences there may be in their culture. 
It is strange that the method of the modern German 
school, which is so unfriendly to the idea of independent 
origins and " in every case where British anthropologists 
see evolution, either in the forms of material objects or in 
social and religious institutions, . . . sees only the evidence 
of mixture of cultures "-that this very method should 
itself have independently originated in two countries the 
peoples of which are partly of the same race and have had 
culture-contact also in later days. Dr. Rivers says, " I 
1 Graebner, Methode der Ethnologie, p. 108 sqq. INTRODUCTION 
5 
have been led quite independently to much the same general 
position as that of the German school by the results of my 
own work in Oceania."1 If customs and institutions 
and ideas could speak, they might also perhaps be justified 
in defending themselves .against the suspicion of being 
mere borrowings. To this Dr. Graebner would say, as he 
has indeed said in a general way, that in cases of parallelism 
we must not apply European evidence to savages, who 
almost entirely lack " the conscious endeavour after further 
development." 2 It seems as though he regarded the 
customs of savages as almost unchangeable, unless subject 
to influences from without. But there is sufficient proof 
that they are not so. 
Among the Central Australians, for instance, changes 
in aboriginal custom take place from time to time, and 
Messrs. Spencer and Gillen are of opinion that these changes 
are in part due to the influence of individuals of superior 
ability. They write, " After carefully watching the natives 
during the performance of their ceremonies and endeavouring 
as best as we could to enter into their feelings, to think as 
they did, and to become for the time being one of themselves, 
we came to the conclusion that if one or two of the most 
powerful men settled upon the advisability of introducing 
some change, even an important one, it would be quite 
possible for,this to be agreed upon and carried out." 3 Dr. 
Landtman tells me that during his stay among the Kiwai- 
speaking people of New Guinea he was struck by the fluctua- 
tions of their habits and customs independently of all 
foreign influence. The funeral customs, for instance, have 
greatly changed within the memory of the present generation 
in a manner which, partly at least, excludes the possibility 
of influence either from neighbouring tribes or from the 
whites. Indeed, do not the frequent variations of custom 
in neighbouring related tribes, or within the same tribe, 
by themselves show that the customs of a people are subject 
to spontaneous changes also at the lower stages of culture ? 
1 Rivers, op. cit. p. 3. 
2 Graebner, op. cit. p. 114. 
3 Spencer and Gillen, Native Tribes of Central Australia, p. 12 sqq. 6 
THE HISTORY OF HUMAN MARRIAGE 
And if this is the case, it is only natural that the changes 
often should lead to similar results in different instances. 
For the possibilities in cultural development are always 
limited, and often limited in a very high degree. For ex- 
ample, it is necessary that the bodies of the dead should be 
disposed of in some way or other, and there are not many 
methods to choose between; hence the same method must- 
be in use among different peoples, quite independently of 
any culture-contact. 
For my own part, I cannot find that there is any reasonable 
ground for quarrelling between the ethnological school, 
which particularly studies the influence that one people 
has exercised upon another owing to contact of their cultures, 
and the evolutionary school; their subjects of investigation 
differ, and therefore also their methods. Both schools 
deal with resemblances of culture-phenomena, but whilst 
the evolutionary school chiefly endeavours to find the 
psychological and sociological origin of these phenomena, 
the ethnological school is concerned with their wanderings. 
The two kinds of investigation supplement each other, 
and their results should exercise a wholesome influence on 
each other; but they cannot replace each other. To 
regard the ethnological analysis of culture-relations as the 
chief task of the history of civilisation-as some German 
scholars do1 -is to deprive this science of its loftiest aims 
and also to disregard many of its most important achieve- 
ments. It should be remembered that even when the 
historical connection between customs found among 
different peoples has been well-established, the real origin 
of the customs has not been explained thereby. It is not 
a sufficient explanation of a custom to say that it has been 
derived from ancestors or borrowed from neighbours. 
This only raises the question how it originated among those 
who first practised it; for a custom must have had a be- 
ginning. It is with questions of this sort that the evolu- 
tionary school of sociologists have pre-eminently occupied 
1 Graebner, op. cit. p. 107 : " So bleibt denn als erstes und 
Grundproblem der Ethnologic wie der ganzen Kulturgeschichte die 
Herausarbeitung der Kulturbeziehungen." INTRODUCTION 
7 
themselves. And their comparative method has greatly 
helped them in their task. 
The simultaneous occurrence of certain social phenomena 
in many different groups of people may prove that there is 
a causal connection between them, though no such connection 
is proved by their simultaneous occurrence in a single 
group. It was this fact that led Tylor to his statistical 
method of investigating the development of institutions.1 
Moreover, a comparison of the circumstances in which a 
custom is practised by different peoples may lead to the 
discovery of the motive underlying it. For example a 
comparative study of the practice of human sacrifice shows 
that human victims are frequently offered in war, before a 
battle, or during a siege ; for the purpose of stopping or 
preventing epidemics ; in order to put an end to a de- 
vastating famine ; when the earth fails to supply the people 
with water ; with a view to averting perils arising from the 
sea or from rivers ; and for the purpose of preventing the 
death of some particular individual, especially a chief or 
a king. And from these facts we are justified in drawing 
the conclusion that human sacrifice is, largely at least, a 
method of life-insurance, based upon the idea of substitu- 
tion ; whilst the famine-sacrifice and the principle under- 
lying it lead to the supposition that the frequent custom of 
securing good crops by means of such a sacrifice, even when 
there is no famine, may also be traced to the same principle, 
especially as there are obvious links between this custom and 
the actual famine-sacrifice.2 Very frequently the know- 
ledge of the cause of a certain custom found among one people 
helps us to understand the meaning of the same, or more 
or 'less similar, customs among other peoples. Little 
details which by themselves would hardly attract our 
attention may, when viewed in the light of the comparative 
method, become conclusive evidence or, in other cases, 
lead to valuable suggestions, some of which may be within 
1 Tylor, ' On a Method of investigating the Development of 
Institutions,' in Jour. Anthr. Inst, xviii. 245 sqq. 
* See Westermarck, Origin and Development of the Moral Ideas, 
i. 440 sqq. 8 
THE HISTORY OF HUMAN MARRIAGE 
the range of future confirmation. In this way the customs 
and institutions of savages have thrown rays of light on 
the early history of civilised nations. 
The endeavour of the evolutionary school to discover 
the psychical causes of social phenomena, however, has of 
late been subject to criticism. Dr. Rivers writes, " The 
proper task of the sociologist is the study of the correlation 
of social phenomena with other social phenomena, and the 
reference of the facts of social life to social antecedents, 
and only when this has been done, or at any rate when this 
process has made far greater advances than at present, 
will it be profitable to endeavour to explain the course of 
social life by psychological processes."1 At present sociology 
and social psychology should, so far as possible, be treated 
as if they were independent disciplines, because each of 
them is liable to make assumptions, belonging to the other 
science, which are readily mistaken for explanations. But 
through " pure " sociology we may ultimately hope to 
attain knowledge of social psychology, though this channel 
must be long and tortuous. The question why such a course 
should be necessary, and why we cannot follow the more 
obvious way of inquiring directly into the motives which 
actuate the conduct of men as members of society, is 
answered by Dr. Rivers as follows :-" Among the people 
whose social conduct has been the special object of my own 
investigations, there is no more difficult task than that of 
discovering the motives which lead them to perform social 
actions. There is no more depressing and apparently 
hopeless task than that of trying to discover why people 
perform rites and ceremonies and conform to the social 
customs of their community." Moreover, it has been 
gradually recognised " that social conduct is not directed 
by intellectual motives, but, predominantly, often it would 
seem exclusively, by sentiments or even instincts," and 
" no mental states are more difficult to introspect than 
emotions and sentiments, to say nothing of instincts." 2 
1 Rivers, ' Survival in Sociology,' in Sociological Review, vi. 304. 
8 Idem, * Sociology and Psychology,' in Sociological Review, 
ix, 3, 10 sq, INTRODUCTION 
9 
For the present, then, we should, on this principle, care- 
fully refrain from assuming, for example, that courtship 
and marriage have anything to do with the sexual instinct, 
that the retaliation of adultery springs from jealousy and 
revenge,1 that the secrecy observed in the performance of 
the sexual function is connected with sexual modesty. 
We should refrain from trying to find any motives for the 
practice of polygyny, the prohibition of incest, the various 
marriage rites, and so forth. We should only correlate 
these phenomena with other social phenomena or refer them 
to social antecedents. Dr. Rivers says, in fact, that it 
would be possible " to write volumes on that group of 
social processes which we sum up under the term ' marriage,' 
without the use of a single psychological term referring to 
instincts, emotions, sentiments, ideas or beliefs," and that 
such, a treatment of marriage would nevertheless be " capable 
of producing valuable contributions to our knowledge." 2 
After those volumes had been written we might perhaps 
be allowed to consider that people not only marry but fall 
in love, and that the marriage customs are not merely 
muscular movements standing in relations to other muscular 
movements, but that they are actuated by intentions and 
motives. I should not be surprised, however, if social 
psychology, when at last permitted to speak, should raise 
violent objections to many of the classifications and con- 
clusions made by " pure " sociology. For it seems to me 
that " pure " sociology is liable to commit the most fatal 
mistakes by detaching social phenomena from their motive 
powers and treating them as mechanical processes, just as 
if men as members of society were a sort of automata. 
From Dr. Rivers' own constructive works and his article, 
on ' Survival in Sociology ' it appears that he thinks of the 
study of organised social processes in their relations in time 
largely as a study of social survivals. But what knowledge 
of the past can be expected from the interpretation of a 
social process as a survival, if no notice is taken of the mental 
1 Dr. Rivers actually blames me for assuming that the blood-feud 
is due to the feeling of revenge {ibid. p. 4 sqq.). 
2 Ibid, p, 2. 10 
THE HISTORY OF HUMAN MARRIAGE 
aspect of social conduct ? Dr. Rivers regards a custom as 
a survival " if its nature cannot be explained by its present 
utility but only becomes intelligible through its past 
history." 1 I presume that utility here means supposed 
utility, since a custom can be quite intelligible through 
existing conditions if it is merely regarded as useful by those 
who practise it. Before a custom is pronounced a survival 
it is thus necessary to examine whether it can be explained 
by present conditions or not, and to these conditions ob- 
viously belong the feelings and ideas of those who practise 
it. It is just the neglect of making such an examination that 
is responsible for many of those arbitrary and unscientific 
conclusions in which the study of early history abounds. 
Customs have been interpreted as surviving traces of other, 
hypothetical, customs in the past, simply because there is 
some external resemblance, often of the most superficial 
character, between them, or because they seem to be possible 
ingredients or consequences of such customs. The reader 
will find numerous instances of this in the present work, 
for example in the chapters on the theory of promiscuity, 
sexual modesty, and marriage by capture. I had hoped 
that such purely sociological interpretations were, on the 
whole, things of the past, and find it truly alarming to hear 
from one who may almost be regarded as the leader of a new 
school in sociology in this country that social phenomena 
should be referred to social antecedents before any attempt 
is made to examine the psychological processes underlying 
them. 
It is strange that this extraordinary faith in sociological 
explanations should be coupled with an equally extreme dis- 
trust in our capacity of learning the motives by which social 
conduct is determined. The mental constitution of men 
is, in spite of all racial and individual differences, essentially 
similar everywhere. This is implied in the fact that they 
are members of the human species and is confirmed by 
their external behaviour. It is true that the more different 
people are from ourselves, and the less we know them, the 
more difficult it is for us to know the motives for their 
actions ; and to understand them in every detail is beyond 
1 Rivers, in Sociological Review, vi. 295. INTRODUCTION 
11 
our power. But the mental facts that lead to the customs 
of peoples are not of a very subtle character. They are 
general instincts, sentiments, or emotions, or particular 
ideas, which, if still prevailing, ought to be accessible to a 
penetrating inquiry. Dr. Rivers maintains that the appa- 
rently hopeless task of trying to discover why people perform 
rites and ceremonies is partly due to the abstract nature of 
such inquiries : " directly one approaches the underlying 
meaning of rite or custom . . . one meets only with un- 
certainty and vagueness unless, as is most frequently the 
case, the people are wholly satisfied with the position that 
they are acting as their fathers have done before them."1 
So far as my own experience goes, this is true of some cases 
but not of others, in which most valuable information has 
been obtained from -the natives themselves.2 Their ex- 
planations are not always alike, and the reason for- this is 
probably that the real origin of the rite has been partly or 
wholly forgotten and a new interpretation substituted for 
the idea from which it rose. This, however, should not 
make the field-ethnologist less eager to find out the present 
meaning attached to the facts he records; for whether or 
no it be the original meaning, it gives us in any case some 
insight into the ideas of existing people, and these are by 
themselves important subjects of inquiry. But the direct 
inquiry into motives is not the only way in which they may 
be ascertained : excellent information may be obtained 
from the words by which ceremonies are accompanied,3 
1 Idem, in Sociological Review, ix. io. 
2 See my Marriage Ceremonies in Morocco; Ceremonies and 
Beliefs connected with Agriculture, certain Dates of the Solar Year, 
and the Weather in Morocco (Cfversigt af Finska Vetenskaps- 
Societetens Fdrhandlingar. Bd. LIV. 1911-1912. Afd. B. N:o 1) ; 
The Moorish Conception of Holiness (Baraka) (Ofversigt, etc. Bd. 
LVIII. 1915-1916. Afd. B. N:o 1) ; The Belief in Spirits in 
Morocco (Acta Academiae Aboensis. Humaniora, I:/); 'The Popular 
Ritual of the Great Feast in Morocco,' in Folk-Lore, xxii. 
3 It was, for example, partly in this way that I found the meaning 
attached to the fire-ceremonies at Midsummer or on other occasions 
in Morocco (' Midsummer Customs in Morocco,' in Folk-Lore, xvi; 
Ceremonies and Beliefs connected with Agriculture, etc.), which led 
Frazer to revise his views on the European fire-festivals (see his 
Balder the Beautiful, vol. i. p. vit sq.). 12 
THE HISTORY OF HUMAN MARRIAGE 
and from comparisons between the circumstances in which 
more or less similar rites or practices are performed, com- 
bined with a solid knowledge of native ways of thinking. 
Such a knowledge, however, presupposes a protracted stay 
among the people whose customs are investigated and 
familiarity with their language. To acknowledge and 
emphasise the need of field-ethnologists who fulfil these 
requirements would, in my opinion, be of greater service to 
sociology than to give up as hopeless the endeavour to 
discover the sources of social action. Even where the 
meaning of a custom is obscure or lost, the field-ethnologist's 
knowledge of the native mind and its modes of thinking 
and feeling ought to enable him to make valuable conjectures. 
Hence I thoroughly disagree with the principle which I once 
heard expressed by the reader of a paper on some savage 
tribes at a meeting of the Royal Anthropological Institute, 
that the field-anthropologist should only aim at collecting 
facts and leave it to the anthropologists at home to explain 
them. 
Many of the psychological explanations of social 
phenomena must of course be more or less hypothetical; 
but this is no reason, why they should not be sought for. 
Hypotheses are no more foreign to the ethnological school, 
which is concerned with the mixture of cultures, than to 
the evolutionary school; I know of no work which is more 
profuse of conjectures than Dr. Rivers' ' History of 
Melanesian Society.' It is in innumerable instances much 
easier to find the psychological origin of a custom or rite 
than to decide if and how-it has come into existence as the 
result of the contact or blending of peoples. Who would 
doubt that water-ceremonies performed for the purpose of 
producing rain are due to the law of association of ideas by 
similarity? But who would dare to trace their wanderings ? 
An objection frequently offered to the comparative method 
is that the use of this method is hardly compatible with a 
sufficiently careful scrutiny of authorities and sources. 
There is undoubtedly some truth in this. Every sociologist 
who has made use of it on a large scale has good reason to 
cry peccavi, and even he who merely deals with some special INTRODUCTION 
13 
group of kindred phenomena has rarely the same oppor- 
tunity as the writer of a monograph on a certain people to 
subject his facts to a searching criticism. In the earlier 
editions of the present work I emphasised that, " as the 
sociologist is in many cases unable to distinguish falsehood 
from truth, he must be prepared to admit the inaccuracy of 
some of the statements he quotes " ; and I cannot possibly 
conceive how Dr. Graebner has been able to construe this 
admission into an attempt on my part to make a virtue of 
necessity.1 In fact, I think that a similar admission might 
be reasonably expected also from sociologists of other 
schools. It is often simply impossible for the most scruti- 
nising critic to decide whether a certain statement is 
accurate or not, and it may even be difficult to form a just 
idea of the general trustworthiness of an ethnographical 
author. Dr. Graebner, for example, considers Curr's 
' The Australian Race ' to be so worthless a book that he 
blames me for quoting it; 2 whereas Dr. Malinowski, who 
in his monograph ' The Family among the Australian 
Aborigines ' has sifted his material with the greatest possible 
care, maintains that Curr had especially good opportunities 
for observation,3 and often refers to him. 
Generally speaking, I must confess that I have become 
more distrustful of ethnographical evidence the longer I 
have myself been in the field. In my own field-work I made 
it long ago a stringent rule never to accept information 
given by anybody but a native of the country, because I 
found that the statements of European residents are very 
frequently lacking in accuracy. I further made it a rule 
never to use information given me about a tribe by members 
of other tribes, without specially mentioning the more or 
less unauthoritative character of the statement. I am also 
somewhat suspicious of that, fortunately insignificant, 
portion of my material which I collected before I could 
freely converse with the natives without the aid of an 
1 Graebner, op. cit. p. 38 n. 2. 
2 Ibid. p. 38 n. 2. 
3 Malinowski, The Family among the Australian Aborigines, 
p. 22 sq. 14 
THE HISTORY OF HUMAN MARRIAGE 
interpreter, although my interpreter was a very intelligent 
and absolutely trustworthy native, with a remarkable 
command of English, who accompanied me on all my jour- 
neys in Morocco. For I have come to the conclusion that 
even the best interpreter is apt to omit details which, though 
apparently trivial, may be of the greatest importance for a 
right understanding of the custom or belief in question, or 
to let his attention slacken for a moment, or to give an 
inaccurate meaning to expressions which baffle all direct 
translation. I, moreover, made it an invariable habit to 
repeat to my informants in full their statements so as to 
avoid all misunderstanding, and occasionally tested their 
accuracy and attention by deliberately misrepresenting 
their statements; and all this could hardly be equally well 
done through the medium of an interpreter. Now, I cannot 
say that my standard of trustworthiness is exactly the same 
when I am using other people's materials as when I am col- 
lecting my own. It could not be the same, considering how 
extremely seldom a field-anthropologist lets his readers 
know minutely how he has obtained his information. But 
nobody can deny that there is in the comparative method 
itself a test which, if carefully applied, gives the investigator 
some confidence in his facts, namely, the test of recurrence. 
As Tylor puts it, "if two independent visitors to different 
countries, say a mediaeval Mohammedan in Tartary and a 
modem Englishman in Dahome, or a Jesuit missionary in 
Brazil and a Wesleyan in the Fiji Islands, agree in describing 
some analogous art or rite or myth among the people they 
have visited, it becomes difficult, or impossible to set down 
such correspondence to accident or wilful fraud."1 
A common complaint against the comparative method is 
that it detaches the cultural phenomenon from the organic 
whole of which it forms a part and thereby easily represents 
it in a wrong light. Customs and beliefs are not the property 
of individuals but belong to the whole social group among 
which they are found: they express its whole corporate 
soul-life. Hence, it is said, they cannot be explained by the 
psychology of the individual, but must, in order to be under- 
1 Tylor, Primitive Culture, i. 9. INTRODUCTION 
15 
stood, be viewed in the light of the culture and social struc- 
ture of the group concerned, instead of being abstracted 
from their social context and classed together with customs 
or beliefs of other groups. In this argument, also, there is, 
I think, a great deal of truth, as well as exaggeration. I 
have myself 1 expressed the opinion that, so far as the lower 
stages of civilisation are concerned, there are, next to socio- 
logical field-work, no other investigations so urgently 
needed as monographs on some definite class of social 
phenomena or institutions among a certain group of related 
tribes-just because social phenomena are not isolated 
phenomena but largely influenced by local conditions, by 
the physical environment, by the circumstances in which 
the people in question live, by their habits and mental 
characteristics. All these factors can much more easily be 
taken into account when the investigation is confined to a 
single people or one ethnic unit than when it embraces a 
social institution as it exists throughout the whole uncivil- 
ised world. 
I presume that most books in which the comparative 
method is followed contain facts which have been classified 
under wrong headings on account of external resemblances 
with other facts. There is a tendency to assume that similar 
customs and rites, practised by different peoples, have their 
roots in similar ideas, and, although this tendency is easy 
to explain and very often results in accurate classifications, 
it is also apt to lead to ill-founded or erroneous conclusions. 
So far as ceremonies are concerned, it should be remembered 
that, especially among simple peoples, the means of express- 
ing ideas in actions are so limited that the same kind of 
activity, or the making use of similar objects, may frequently 
have a different meaning in different cases. To take an 
instance from rites practised at weddings : the eggs so 
commonly used on these occasions are sometimes said to 
promote fecundity (on account of physiological connections), 
sometimes to give good luck or make the weather fine 
(on account of their white colour), sometimes to secure the 
1 In my ' Prefatory Note ' to The Tribe, and Inter-tribal Relations 
in Australia, by G. C. Wheeler, p. v. sq. 16 
THE HISTORY OF HUMAN MARRIAGE 
consummation of the marriage (on account of the fragility 
of their shells), sometimes to facilitate delivery; and it is 
quite possible that all these various interpretations represent 
original motives or parts of a mixed motive. Instances of 
mistaken classification might no doubt be quoted from my 
own writings (and will probably be found in the present 
work also) ; but as it is more agreeable to find fault with 
others than with one's self, I shall choose an example from 
the investigations of an esteemed colleague. 
In his book ' Primitive Paternity,' Dr. Hartland shows that 
in various countries bathing is practised as a method of 
obtaining children, and he traces this practice to an ancient 
belief " that pregnancy was caused otherwise than by sexual 
intercourse." In support of this view he quotes, besides 
many other facts, a statement of mine referring to a tribe 
in Southern Morocco.1 It is there the custom for a married 
woman who is anxious to know if she will be blessed with a 
child or not to go to the sea-shore on Midsummer Day, and 
on the two following days as well, and let seven waves go 
over her body ; then she knows that if she does not get a child 
soon she will have none at all. In this case magic has 
dwindled into divination, as is obvious from a similar 
custom practised in another tribe in Morocco, where the 
young wife goes to the sea on the fortieth day after her arrival 
at her new home, and, while the seven waves are going over 
her body, says to the sea, " O my uncle the Sea, I am 
troubled with spirits, give me children and health." Now 
these facts can by no means serve as evidence for a theory 
of primitive paternity. In Morocco, at least, the effect 
which water is held to have upon fecundity is only indirect, 
that is, it is supposed to remove the evil influences which 
cause sterility, as appears from the idea that an infertile 
woman or animal is troubled with evil spirits and from the 
fact that the very same procedure as is adopted as a cure 
for barrenness is also supposed to remove or prevent sickness 
or misfortune in general.,, But in extenuation of Dr. Hart- 
land's guilt it should be added that, in the article of mine 
from which he made his quotation, I had not expressly 
1 Hartland, Primitive Paternity, i. 80. INTRODUCTION 
17 
mentioned the Moorish view of sterility. 'This case may be 
considered typical. The mistakes made by sociologists of 
the school following the comparative method, when they 
detach facts from their environment and interpret them in 
their own ways, are largely due to the incompleteness of 
their sources. Hence one of the chief defects of their method 
may be considerably reduced by the strenuous efforts of 
field-workers to collect not only external facts but to enter 
into the thoughts and feelings of the people they investi- 
gate, as also by monographs of the kind already 
mentioned. 
There is no real opposition between the study of a cultural 
phenomenon as it is distributed among different races and 
the study of it which is restricted to a particular ethnic 
group. Here again the methods differ simply because the 
subjects differ. But the two kinds of investigation com- 
plement each other. Whilst the student of a custom or 
institution in its generality must be grateful to the special- 
ist who provides him with the results of his detailed research, 
the comparative treatment, which in the first place bears 
out general resemblances, often helps the specialist to explain 
facts which he could hardly understand in full if his know- 
ledge were restricted to a limited area. It is easy to criticise 
the comparative method in the point we are now con- 
sidering, but it is impossible for any modern student of 
human civilisation to ignore its results. The writings of 
Professor Durkheim and his disciples are thoroughly pervaded 
by the teachings of the very school whose method they 
have so severely criticised. Does not this show that there 
must be exaggeration in their criticism ? They have not 
sufficiently considered an extremely simple but extremely 
important fact, namely, that all the different ethnic groups 
belong to the same animal species and therefore must 
present resemblances which have a deeper foundation than 
all differences which are the effects of the social environment. 
How could we disclose these resemblances in any other way 
than by comparison ? How could we otherwise distinguish 
that which is local from that which is general ? Nay, how 
could we fully explain the social environment itself without 18 
THE HISTORY OF HUMAN MARRIAGE 
taking into account the mental characteristics of the human 
species ? I think there is sufficient evidence to show that 
innumerable customs and beliefs are not so closely inter- 
woven with the social tissue that they cannot, with due 
precaution, be abstracted from it for the sake of com- 
parison. And in any case we may expect to find a 
specifically human element mingled with local peculiari- 
ties. 
But if the French sociologists-I except of course M. 
van Gennep, who does not belong to the school of Durkheim 
■-have underrated the homogeneous elements of the human 
mind, I think they have, on the other hand, somewhat 
overrated the homogeneity of the group-mind. That the 
minds of the people are profoundly influenced by the fact 
that they live and act together is a truth which nobody 
doubts. We implicitly recognise this when we speak of the 
customs, beliefs, or religion of a people-expressions which 
are much older than Bastian's Volkergedanke. But we 
must not forget that the homogeneity of thoughts and 
actions inside a society is not absolute. This is true not 
only of civilised men but, in some degree, also of savages. 
Dr. K. Donner points out that among the Samoyed the 
religious beliefs and the ideas relating to the soul's fate 
after death vary considerably in different individuals p 
and, with special reference to ideas of the latter kind, he 
tells me that this is the case even in a small tribe consisting 
of some five or six hundred individuals. One man gave 
him a minute description of the other world which was based 
upon his own experience of it in a dream. Dr. Landtman, 
again, informs me that the Kiwai-speaking Papuans often 
drew his attention to the differences of habits, not only 
within different groups of the same tribe or village, but 
even in the case of different individuals. A native said to 
him, for instance, " One man has one method of catching 
the dugong, another man has another method." Among 
their ceremonies there are such as are common to the whole 
tribe, but there are many others, referring to agriculture, 
hunting, or other occupations, that are practised only by 
1 Donner, Bland Samojeder i Sibirien, p. 119 sq. INTRODUCTION 
19 
members of the same family or by single individuals who have 
directly or indirectly learnt them from some spirit or ghost 
appearing in a dream. Dr. Landtman maintains that it 
is hardly possible to distinguish in every case between 
practices and beliefs which are general and such as are indi- 
vidual. Considering that distinctions of this sort are not 
generally found in anthropological books, the ethnologist 
of the study must be warned against making too liberal a 
use of the term " collective ideas," or that favourite expres- 
sion of the French sociologists, " representations collec- 
tives." 
An error of method which was very prevalent among the 
evolutionary school in those days when the first edition of 
this work was written, and which is still committed by 
certain writers, is the practice of inferring, without sufficient 
reason, from the prevalence of a custom or institution 
among some savage peoples, or from facts interpreted as 
survivals of it, that this custom or institution is a relic from 
a stage of development which the whole human race once 
went through. Thus the assumption that primitive men 
lived in tribes or hordes all the men of which had promiscuous 
intercourse with all the women, where no individual 
marriage existed, and the children were the common pro- 
perty of the tribe, is founded, in the first place, on the 
statements of some travellers and ancient writers as to 
peoples among whom this custom is said actually to prevail, 
or to have prevailed, and, in the second place, on certain 
supposed survivals of that custom. Dr. Post went still 
further in his book ' Die Geschlechtsgenossenschaft der 
Urzeit und die Entstehung der Ehe.' Without adducing 
any satisfactory reason for his opinion, he considered it 
probable that " monogamous marriage originally emerged 
everywhere from pure communism in women, through the 
intermediate stages of limited communism in women, 
polyandry, and polygyny " ; 1 but he subsequently revised 
his views.2 Lewis H. Morgan, in his ' Systems of Con- 
1 Post, Die Geschlechtsgenossenschaft der Urzeit, p. 17. 
2 Idem, Studien zur Entwickelungsgeschichte des Familienrechts, 
p. 58. 20 
THE HISTORY OF HUMAN MARRIAGE 
sanguinity and Affinity of the Human Family,' suggested 
no fewer than fifteen normal stages in the evolution of 
marriage and the family, assuming the existence and 
general prevalence of a series of customs and institutions 
" which must of necessity have preceded a knowledge of 
marriage between single pairs, and of the family itself, in 
the modern sense of the term." 1 Marriage by capture has 
been regarded as the primitive and once universal method 
of acquiring a wife; and to this day we hear of an early 
stage of group-marriage and of a stage of mother-right 
preceding father-right. 
Now, it seems obvious that we have no right to assume 
the universal prevalence of a social phenomenon in the past 
unless we may assume that the cause or causes to which it 
is due have been universally operating. If speculating on 
such problems, we have therefore first to find out the causes 
of the social phenomena ; then, from the prevalence of the 
causes, we may infer the prevalence of the phenomena 
themselves, if the former may be assumed to have operated 
without being checked by other causes. In this way we 
may conclude with absolute certainty that there were always 
intimate relations between a mother and her young child, 
since the causes of these relations must always have operated 
in a mammalian species, like man. A similar way of reason- 
ing has also led me to the hypothesis that the family, 
consisting of father, mother, and children, existed already 
in primeval times and probably among our pre-human 
ancestors, owing to the offspring's need of care and protec- 
tion and to the economic obstacles in the way of a permanent 
living in hordes ; but this conclusion has a less solid founda- 
tion than the former one, because the necessity of paternal 
care is not so certain as that of maternal care. On the other 
hand, the causes, or hypothetical causes, to which marriage 
by capture, group-marriage, and mother-right may be 
traced are not such as to justify the belief in the universal 
prevalence of any of these customs at any stage of human 
civilisation-to say nothing of promiscuity, the existence 
of which as the exclusive form of the relations between 
1 Morgan, Systems of Consanguinity and Affinity, p. 479. INTRODUCTION 
21 
the sexes even among a single people is extremely 
doubtful. 
The comparative method and the manner in which it has 
been applied are thus by no means above criticism. But 
the defects from which it suffers, and the errors to which it 
has led, hardly justify the air of superiority with which it 
has not infrequently been treated by the advocates of other 
methods. Its weaknesses and pitfalls are easy to detect, 
because it has been applied on a large scale in the course of 
half a century; but I also think that the many important 
results achieved through it bear ample testimony to its 
merits. In order to form a just appreciation of a method 
it must be sufficiently put to the test, and this cannot be 
said to have been the case either with the sociological method 
of the Trench school or the ethnological method, which as 
yet have been in a greater degree subjects of theoretical 
discussion than of practical application. As for the French 
method, I cannot help saying that there are disquieting 
signs of a tendency to expansion beyond its legitimate 
limits. Its followers have not always been satisfied with 
restricting their conclusions to social phenomena belonging 
to the same area, but have regarded their method as a 
direct means of arriving at results of a much wider scope. 
Thus Professor Durkheim, in his book on the totemic system 
in Australia with the significant title * Les formes el£men- 
taires de la vie religieuse,' confidently asserts that this 
system contains " all the great ideas and all the principal 
ritual attitudes which are at the bottom even of the most 
advanced religions " ; and he then proceeds to a discussion 
of religion in general, in the belief that if you have carefully 
studied the religion of one people only, you are better able 
to lay down the main principles of the religious life than if 
you follow the comparative method of a Tylor or Frazer.1 
It almost seems as though some kind of sociological intuition 
were to take the place of comparative induction. When 
properly applied, however, the different methods have their 
special problems to solve; and this should exclude every 
rivalry between them. All of them are beset with difficulties. 
1 Durkheim, Les formes elementaires de la vie religieuse, p. 593 sqq. 22 
THE HISTORY OF HUMAN MARRIAGE 
For the problems are often complicated, and the material 
at our disposal is defective. 
With reference to the particular subject-matter of the 
present work, it should be added that the investigator often 
must go beyond the psychological causes of the phenomena 
with which he deals and try to find their biological founda- 
tion. Marriage, so far as I can see, is rooted in instincts 
which can only be explained by biological facts-either 
peculiar to mankind and its nearest relatives, or of a more 
general character-and so are many particular customs and 
rules relating to marriage. Such instincts have been formed 
in accordance with the need of the species, which again 
depends upon anatomical and physiological factors of various 
kinds. This is true both of the sexual instinct with its 
peculiarities and of other instincts by which marriage is 
determined. It was a great defect of the earlier treatises 
on marriage that the biological aspect of the problem was 
entirely ignored, and even now it is not sufficiently recog- 
nised. In a review of the first edition of this work, Tylor 
pointed out that its distinguishing character was its " effort 
to work the biology-side and the culture-side of anthropology 
into one connected system " ; and he added that " there 
can be no doubt of the value of the resulting discussions, 
which will develop further as the inquiry goes on in this 
direction."1 In the present edition discussions of this kind 
occupy an even more prominent place than before. There 
is no question here of biological analogies applied to the 
explanation of social evolution-as has been mistakenly 
said-but we are concerned with biological facts underlying 
psychical and social phenomena. At the same time careful 
attention must also be given to the influence which people's 
ideas and beliefs have exercised upon their marriage customs, 
a subject on which much light has been thrown in the course 
of the last twenty years.2 And another factor of importance 
1 Tylor, in Academy, October 3, 1891. Cf. Steinmetz, ' Die 
neueren Forschungen zur Geschichte der menschlichen Familie,' 
in Zeitschrift fur Socialwissenschaft, ii. 811. 
2 See, especially, the works of Frazer and The Mystic Rose by 
Crawley. INTRODUCTION 
23 
is the influence of industrial culture.1 Marriage is not a 
mere sex-relation, but an economic institution as well, 
and is consequently more or less influenced by the sources 
of subsistence. 
I do not propose to discuss marriage in all its aspects, but 
shall, in the main, restrict myself to the following subjects :- 
the origin of marriage and questions connected with it, such 
as sexual periodicity, the various groups of facts which have 
been regarded as evidence of primitive promiscuity (alleged 
instances of peoples living in promiscuity, pre-nuptial 
unchastity, the jus frimae noctis, religious prostitution, 
the lending and exchange of wives, feasts at which pro- 
miscuous intercourse is indulged in, the classificatory system 
of relationship, mother-right), and masculine jealousy; 
the frequency of marriage and the marriage age ; celibacy; 
sexual modesty, which has a bearing both on celibacy and 
on some marriage customs; courtship and its various 
features; primitive means of attraction; sexual selection 
as influenced both by preferences and aversions, and 
the endogamous and exogamous rules which are rooted in 
the latter; the methods of contracting a marriage, such 
as capture, consent, and the giving of a consideration 
for the wife either in the form of the exchange of bride for 
bride, or of service, or of true purchase, or of gifts, or of 
the exchange of presents ; the marriage portion ; marriage 
rites ; monogamy, polygyny, polyandry, and group- 
marriage and other group-relations; the duration of 
marriage and the rules relating to its dissolution. 
I shall discuss each of these questions separately, and 
in doing so I shall consider the customs or laws of peoples 
at all stages of civilisation. I shall not, however, enter into 
all the details of the marriage laws of modern civilised peoples, 
which are of comparatively little theoretical interest and, 
1 The importance of this factor was emphasised by Grosse, in 
Die Formen der Familie und die Formen der Wirthschaft, and by 
Hildebrand, in Recht und Sitte auf den verschiedenen wirthschaftlichen 
Kulturstufen. But the most valuable contribution to the subject is, 
on account of the wealth of its material and its statistical method, 
The Material Culture and Social Institutions of the Simpler Peoples, 
by Hobhouse, Wheeler, and Ginsberg. 24 
THE HISTORY OF HUMAN MARRIAGE 
besides, can easily be found in legal treatises. Whenever it 
is possible, I shall group together peoples of the same stock 
or living in the same area, but in other cases I shall not 
hesitate to mention even in the same paragraph peoples 
belonging to different races, living far away from each other, 
and standing on different levels of culture, but whose cus- 
toms or beliefs invite comparison. This will in certain 
quarters be called bad method ; but I beg to refer to Darwin 
and others who, in dealing with some particular biological 
phenomenon, speak in the same breath even of different 
species of animals, and nevertheless have succeeded in reach- 
ing conclusions of some importance. 
When the subject allows it, I shall make it a rule first to 
give a descriptive account of the phenomenon I am dis- 
cussing, as it occurs among different peoples, and afterwards 
try to find its cause or causes-whether biological, psycho- 
logical, or sociological. I am, of course, fully aware that 
the explanation may be incomplete, that it may refer to 
some cases but not to all, and that it may be partly or 
wholly conjectural. If I do not discuss the question of a 
common origin, either through the peoples' descent from the 
same ancestors or through culture-contact, that must by 
no means be interpreted as an assumption on my part that 
there is no common origin. When I speak of phenomena 
which I regard as specific characters-such as the sexual 
instinct, masculine jealousy, female coyness, or the family- 
I assume, on the contrary, eo ipso that they have a common 
source ; and when I speak, for example, of identical marriage 
rites among the peoples of the Indo-European group, I never 
think of independent origins in each case. But to seek for 
the homestead of the various marriage customs would in 
most cases be a most unprofitable task. 
The method I am following often necessitates long 
enumerations of facts. These are not meant merely to 
illustrate some particular theory of the author-as has been 
alleged by certain critics-but they form the basis on which 
the theory is built, and cannot, therefore, be avoided, 
however much they may tax the patience of the reader. 
The general public will also, perhaps, object to being told INTRODUCTION 
25 
certain facts which hurt its sense of modesty. But the 
concealment of truth is the only indecorum known to 
science; and to keep anything secret within its cold and 
passionless expanses would be as prudish as to throw a cloth 
round a naked statue. CHAPTER I 
THE ORIGIN OF MARRIAGE 
Marriage is generally used as a term for a social institu- 
tion. As such it may be defined as a relation of one or more 
men to one or more women which is recognised by custom 
or law and involves certain rights and duties both in the 
case of the parties entering the union and in the case of the 
children bom of it. These rights and duties vary among 
different peoples, and cannot therefore all be included 
in a general definition ; but there must, of course, be some- 
thing which they have in common. Marriage always implies 
the right of sexual intercourse : society holds such inter- 
course allowable in the case of husband and wife, and, 
generally speaking, even regards it as their duty to gratify 
in some measure the other partner's desire.1 But the 
right to sexual intercourse is not necessarily exclusive. 
It can hardly be said to be so, from the legal point of view, 
unless adultery is regarded as an offence which entitles the 
other partner to dissolve the marriage union, and this, as 
we know, is by no means always the case. 
At the same time, marriage is something more than a 
regulated sexual relation. It is an economic institution, 
which may in various ways affect the proprietary rights of 
the parties. It is the husband's duty, so far as it is 
possible and necessary, to support his wife and children, but 
it may also be their duty to work for him. As a general 
1 For the conception of a " right," see my Origin and Development 
of the Moral Ideas, i. 139 sqq. 
26 ch. i THE ORIGIN OF MARRIAGE 
27 
rule he has some power over them, although his power over 
the children is generally of limited duration. Very often 
marriage determines the place which a newly-born indi- 
vidual is to take in the social structure of the community 
to which he or she belongs ; but this cannot, as has been 
maintained,1 be regarded as the chief and primary function 
of marriage, considering how frequently illegitimate chil- 
dren are treated exactly like legitimate ones with regard 
to descent, inheritance, and succession.2 It is, finally, 
necessary that the union, to be recognised as a marriage, 
should be concluded in accordance with the rules laid 
down by custom or law, whatever these rules may be. 
They may require the consent of the parties themselves 
or of their parents, or of the parties as well as of their 
parents. They may compel the man to pay a price for 
his bride, or the parents of the latter to provide her with 
a dowry. They may prescribe the performance of a parti- 
cular marriage ceremony of one kind or other. And no 
man and woman are regarded as husband and wife unless 
the conditions stipulated by custom or law are complied 
with. 
As for the origin of the institution of marriage, I con- 
sider it probable that it has developed out of a primeval 
habit. It was, I believe, even in primitive times, the habit 
for a man and a woman (or several women) to live together, 
1 Rivers, History of Melanesian Society, ii. 145. 
2 See, e.g., Solberg, ' Gebrauche der Mittelmesa-Hopi (Moqui) 
bei Namengebung, Heirat und Tod,' in Zeitschr. f. Ethnol. xxxvii. 
629; Bogoras, Chukchee, p. 572 ; Gurdon, Khasis, p. 77 (War 
country) ; Kohler, ' Das Recht der Marschallinsulaner, ' in Zeitschr. 
vergl. Rechtswiss. xiv. 423 ; Torday and Joyce, * Notes on the 
Ethnography of the Ba-Huana,' in Jour. Anthr. Inst, xxxvi. 285 ; 
Ploss-Renz, Das Kind, ii. 689 ; Hartland, Primitive Paternity, 
i. 277, 319, 320, and ii. 105, 178, 226. In China all sons born in 
the household have an equal share in the inheritance, whether 
the mother be the principal wife or a concubine or a domestic slave ; 
and according to Muhammadan law no distinction in point of 
inheritance is made between the child of a wife and that borne 
by a slave to her master, if the master acknowledge the child to 
be his own (Westermarck, Origin and Development of the Moral 
Ideas, ii. 48). See also Benzinger, Hebrdische Archdologie, p. 114. 28 
THE HISTORY OF HUMAN MARRIAGE chap. 
to have sexual relations with one another, and to rear their 
offspring in common, the man being the protector and 
supporter of his family and the woman being his helpmate 
and the nurse of their children. This habit was sanctioned 
by custom, and afterwards by law, and was thus trans- 
formed into a social institution. In order to trace marriage 
in its legal sense to its ultimate source, we must therefore 
try to' find out the origin of the habit from which it 
sprang. 
Our task is much facilitated by the fact that similar 
habits are found among many other species of the animal 
kingdom. Not, however, among the very lowest. In 
the great sub-kingdom of the Invertebrata the relations 
between the sexes are generally of the most fugitive nature, 
and even the mothers are exempted from nearly all anxiety 
as regards their offspring. In the highest order, the Insects, 
the eggs are hatched by the heat of the sun, and the mother, 
in most cases, does not even see her young, her care being 
generally limited to seeking out an appropriate place for 
laying the eggs, and to fastening them to some proper 
object and covering them, if this be necessary for their 
preservation.1 Yet there are several species in which 
the mother prepares board and lodging for her offspring,2 
and among some beetles the male also takes part in the work, 
and, together with the female, even guards the eggs.3 
The Spanish Copris, for instance, digs a burrow under a heap 
of sheep's dung and gathers there victuals from the heap ; 
and Fabre also suspects " the husband of lending a hand 
to his partner with the harvesting and the storing. . . . 
But, once the house is well supplied, he retires discreetly, 
returns to the surface and goes and settles down elsewhere, 
leaving the mother to her delicate functions. His part in 
the family-mansion is ended."4 
1 Brehm, Thierlebsn, ix. 16. Fabre, Life and Love of the Insect, 
p. 2. 
2 Reuter, Lebensgewohnheiten und Instinkte dev Insekten, pp. 204, 
205, 208 sqq. Fabre, op. cit. p. 1 sq. 
3 Reuter, op. cit. p. 227 sqq. 
4 Fabre, op. cit. p. 67 sq. I THE ORIGIN OF MARRIAGE 
29 
In the lowest classes of the Vertebrata parental care is 
likewise almost unheard of. In the immense majority of 
species, young fishes are hatched without the assistance of 
their parents, and have, from the outset, to help thenjselyes. 
Many Teleostei, however, form an exception ; and in these 
cases it is the male on which the parental duty generally 
devolves. In some instances he constructs a nest, and 
jealously guards the ova deposited in it by the female; 
while the male of certain species of Arius carries the ova 
about with him in his capacious pharynx.1 'Most of the 
Reptiles place their eggs ;n a convenient and sunny spot 
between moss and leaves, and take no further trouble about 
them. But several of the larger serpents have a curious 
fashion of laying them in a heap, and then coiling themselves 
around them in a great hollow cone.2 And female crocodiles, 
as also certain aquatic snakes of Cochin China, observed by 
Dr. Morice, carry with them even their young.3 
Among the lower Vertebrata, it rarely happens that both 
parents jointly take care of their progeny. Milne Edwards 
states that in the Pipa, or Toad of Surinam, the male helps 
the female to disburthen herself of her eggs ; 4 and concerning 
some Chelonia M. Espinas observes, " La femelle vient sur 
les plages sablonneuses au moment de la ponte, accom- 
pagn£e du male, et .construit un nid en forme de four ou 
la chaleur du soleil fait Delore les oeufs."5 But it may 
be regarded as an almost universal rule that the relations 
of the sexes are utterly fickle. The male and female come 
together in the pairing time ; but having satisfied their sexual 
instincts they part again, and have nothing more to do with 
one another. 
The case is very different with the large majority of 
Birds. Among them male and female keep together not 
only during the breeding season, but also after it, and the 
1 Guenther, Introduction to the Study of Fishes, p. 163. 
2 Wood, Illustrated Natural History, iii. 3. 
3 Espinas, Des societes animates, p. 416. 
4 Milne Edwards, Lemons sur la -physiologic et I'anatomie comparee, 
viii. 496. 
6 Espinas, op. cit. p. 417. 30 
THE HISTORY OF HUMAN MARRIAGE chap. 
parental instinct has reached a high degree of intensity 
on the father's side as well as on the mother's. The two 
birds help each other to build the nest, the male generally 
bringing the materials and the female doing the work. 
In fulfilling the numberless duties of the breeding season 
both birds take a share. Incubation rests principally with 
the mother,1 but the father, as a rule, helps his companion, 
taking her place when she wants to leave the nest for 
a moment, or providing her with food and protecting her 
from every danger. During the first few days after hatching 
most birds rarely leave their young for long, and then only 
to procure food for themselves and their family ; and in 
cases of danger both parents bravely defend their offspring. 
As soon as the first period of helplessness is over and the 
young have grown somewhat, they are carefully taught 
to shift for themselves ; and it is only when they are per- 
fectly capable of so doing that they leave the nest and the 
parents. 
There are, it is true, a few birds that from the first day 
of their ultra-oval existence lack all parental care; and 
in some species, as the ducks, it frequently happens that 
the male leaves family duties wholly to the female. But 
as a general rule both share prosperity and adversity. 
Brehm believes that most birds, with the exception of those 
belonging to the Gallinaceous family and some other species, 
when pairing, do so once for all till either one or the other 
dies. He is so filled with admiration for their exemplary 
family life, that he enthusiastically declares that " real 
genuine marriage can only be found among birds."2 
The same cannot be said of most of the Mammals. The 
mother is, no doubt, very ardently concerned for the welfare 
of her young, generally nursing them with the utmost 
affection, but in the majority of species the relations 
between the sexes are restricted to the pairing season, 
1 The ostrich, however, forms a curious exception, the male- 
not the female-sitting on the eggs and bringing up the young 
birds (Brehm, Bird-Life, p, 324). 
2 Ibid. p. 285. The statements concerning birds are taken from 
Brehm's Thierleben, vol. iv., the same author's Bird-Life, and Her- 
mann Muller's Am Neste. I THE ORIGIN OF MARRIAGE 
31 
and sometimes the male even acts as an enemy to his own 
progeny. Yet there are various species in which the union 
between male and female is of a more durable character. 
This is the case with whales,1 seals,2 the hippopotamus,3 
the Cervus campestrisA gazelles,6 the Neotragus Hemprichii 
and other small antelopes,6 rein-deer,7 the Myopotamus 
bonariensis (a South American aquatic rodent),8 squirrels,9 
moles,10 the ichneumon,11 and some carnivorous animals, 
as a few cats and martens,12 the yaguarundi in South 
America,13 the Canis Azarce,1* and possibly also the wolf.16 
Among all these animals the sexes are said to remain together 
even after the birth of the young, the male being the 
protector of the family. 
This is frequently the case among the Quadrumana. The 
natives of Madagascar assert that in some species of the 
Prosimii male and female nurse their young in common16- 
a statement, however, the accuracy of which remains to be 
proved. The mirikina (Nyctipithecus trivirgatus} seems, 
according to Rengger, to live in pairs throughout the whole 
year, for, whatever the season, a male and a female have 
always been found together.17 Of the Mycetes Car ay a, 
Cebus Azarch,18 and Ateles paniscus,™ single individuals 
are very seldom or never seen, whole families being generally 
met with. Among the Arctopitheci, the male parent is 
expressly said to assist the female in taking care of the 
young ones.20 
1 Brehm, Thierleben, iii. 679. 2 Ibid. iii. 593, 594, 599. 
3 Ibid. iii. 578. 
4 Rengger, Naturgeschichte dev Sdugethiere von Paraguay, p. 354. 
5 Brehm, Thierleben, iii. 206. 
• Ibid. iii. 256. Espinas, op. cit. p. 447. 
7 Brehm, Thierleben, iii. 124. 
8 Rengger, op. cit. p. 240. 9 Brehm, Thierleben, ii. 270. 
10 Ibid. ii. 263. 11 ibid. ii. 39. 
12 Ibid. i. 347. 13 Ibid j 387 
14 Rengger, op. cit. p. 147 sq. 15 Brehm, Thierleben, i. 535. 
16 Ibid. i. 244. 17 Rengger, op. cit. p. 62. 
18 Ibid. pp. 20, 38. 
19 Schomburgk, Reisen in Britisch-Guiana, iii. 767. 
20 Brehm, Thierleben, i. 228. 32 
THE HISTORY OF HUMAN MARRIAGE chap. 
In a recent work on his travels in Sumatra, Moszkowski 
states that the higher monkeys or apes usually live in 
families consisting of father, mother, and one or two young ; 
he says that he has often witnessed this himself, but, unfor- 
tunately, does not mention the name of any particular 
species.1 Diard was told by the Malays, and found it 
afterwards to be true, that the young siamangs, when in 
their helpless state, are carried about by their parents, 
the males by the father, the females by the mother.2 C. de 
Crespigny, who was wandering in the northern part of 
Borneo in 1870, gives the following description of the orang- 
utan :-"They live in families-the male, female, and a 
young one. On one occasion I found a family in which 
were two young ones, one of them much larger than the 
other, and I took this as a proof that the family tie had 
existed for at least two seasons. They build commodious 
nests in the trees which form their feeding-ground, and, so 
far as I could observe, the nests, which are well lined with 
dry leaves, are only occupied by the female and young, 
the male passing the night in the fork of the same or another 
tree in the vicinity. The nests are very numerous all over 
the forest, for they are not occupied above a few nights, 
the mias (or orang-utan) leading a roving life."3 According 
to Rajah Brooke, these apes " are never found in numbers 
together, in general only single, though occasionally the 
male and female are in company."4 Mohnike says that 
the old males generally live with the females during the 
rutting season only.5 Hornaday writes " The orang is 
quite solitary in his habits, the old males always being found 
alone; nor are two adult females ever found together. 
On two occasions I found three individuals together, but 
1 Moszkowski, Auf neuen Wegen durch Sumatra, p. 246. 
2 Brehm, Thierleben, i. 97. 
3 de Crespigny, ' On Northern Borneo,' in Proceed. Roy. Geo. Soc. 
xvi. 177. 
4 Brooke, Narrative of Events in Borneo and Celebes from the 
Journals of James Brooke, i. 227. 
6 Mohnike, ' Die Afton auf den indischen Inseln,' in Das Ausland, 
xlv. 850. See also Hartmann, Die menschenahnlichen Affen, p. 230. I THE ORIGIN OF MARRIAGE 
33 
one was an old female with a nursing infant, and the third 
was her next oldest offspring, apparently about a year and 
a half old, who had not yet left his mother's side to shift 
for himself." From this our informant draws the conclusion 
that the young one does not leave its mother until nearly 
two years of age 1-an assumption which well agrees with 
Rajah Brooke's statement referring to an adult female with 
a young one at her breast and " a second a year or two 
old (probably her former offspring) in company."2 Nor 
did Wallace ever see two full-grown animals together; 
but he sometimes found females and at other times males 
accompanied by half-grown young ones.3 It thus seems that 
the offspring of the orang-utan are not devoid of all paternal 
care. 
More unanimous in this respect are the statements we 
have regarding the gorilla. According to Savage, these 
apes live in bands, but all his informants agreed in the 
assertion that but one adult male is found in every band. 
" It is said that when the male is first seen he gives a terrific 
yell that resounds far and wide through the forest. . . . 
The females and young at the first cry quickly disappear; 
he then approaches the enemy in great fury, pouring out 
his horrid cries in quick succession."4 Schweinfurth says 
that the gorillas are not found in herds, but either in pairs 
or even quite alone, and that it is only the young that 
occasionally may be seen in groups.6 Du Chaillu found 
" almost always one male with one female, though sometimes 
the old male wanders companionless " ;6 and Winwood 
Reade states likewise that the gorilla goes " sometimes alone, 
sometimes accompanied by his female and young one."7 
The same traveller was told that when a family of these 
apes ascend a tree and eat a certain fruit, the old father 
1 Hornaday, Two Years in the Jungle, p. 402 sq. 
2 Brooke, op. cit. i. 221. 
3 Wallace, Malay Archipelago, i. 93. 
4 Savage, Description of Troglodytes Gorilla, p. 9 sq. 
5 Schweinfurth, Heart of Africa, i. 522. 
e Du Chaillu, Explorations and Adventures in Equatorial Africa, 
P- 349- 
7 Reade, Savage Africa, p. 214. 34 
THE HISTORY OF HUMAN MARRIAGE chap. 
remains seated at the foot of the tree. And when the 
female is pregnant he builds a rude nest, usually about 
fifteen or twenty feet from the ground ; here she is delivered, 
and the nest is then abandoned.1 According to von 
Koppenfels, the male spends the night crouching at the foot 
of the tree, against which he places his back, and thus 
protects the female and their young, which are in the nest 
above, from the nocturnal attacks of leopards. Once he 
observed a male and a female with two young ones of different 
ages, the elder being perhaps about six years old, the younger 
about one year.2 
A somewhat different account of the habits of the gorilla 
was given to Mr. Guthrie in Kamerun by natives of the 
Bulu tribe. According to this account, " the gorillas of 
Kamerun live in small companies, scarcely to be called 
families, except in the younger days of the band, when 
only two, three, or four individuals are found together. A 
company seldom comprises more than twelve members, 
and is said never to exceed fifteen or sixteen. The smaller 
companies consist of one male with his one, two, or three 
wives, and some small children. A company of six or 
seven members would probably have two adult males. 
As the younger members grow up they take, or rather 
keep, their places in the company. When the old male 
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." When a gorilla's 
family is threatened by natives, he will attack them. Mr. 
Guthrie relates :-" In one instance a band of gorillas was 
attacked by two Bulu men. The old gorilla of the band 
first got his family out of danger, and then returned to 
the encounter."3 
From these accounts it appears that the gorilla lives in 
family groups, consisting of one adult male (according to 
1 Ibid. pp. 218, 214. 
2 v. Koppenfels, ' Meine Jagden auf Gorillas,' in Die Gavtenlaube, 
1877, p. 418 sq. 
3 Jenks, ' Bulu knowledge of the Gorilla and Chimpanzee, 
in American Anthropologist, N.S. xiii. 56, 58. 1 THE ORIGIN OF MARRIAGE 
35 
one account sometimes two), one or more females, and one 
or more young ones of different ages, and that the adult 
male, or father, guards, warns, and protects his family, 
and, apparently, builds a nest for them. The habits of 
the chimpanzee are said to be very similar. Schweinfurth 
states that this ape also is found either in pairs or even quite 
alone, only the young occasionally being seen in groups.1 
According to Savage, "it is seldom that more than one or 
two nests are seen upon the same tree or in the same neigh- 
bourhood ; five have been found, but it was an unusual 
circumstance. They do not live in ' villages.' . . . They 
are more often seen in pairs than in gangs.'' He also 
tells us that " it is not unusual to see ' the old folks ' sitting 
under a tree regaling themselves with fruit and friendly 
chat, while ' their children ' are leaping around them and 
swinging from branch to branch in boisterous merriment."2 
Von Koppenfels says that the chimpanzee, like the gorilla, 
builds a nest for the young and female on a forked branch, 
and that the male himself spends the night lower down in 
the tree.3 Mr. Guthrie was told by his native informants 
that the chimpanzee in Kamerun lives in bands, as does 
the gorilla ; but " the old males eventually become solitary, 
though the young on maturing are believed to remain in 
the kinship group."4 
If we ask why in certain animal species male and female 
remain together not only during the pairing season but 
till after the birth of the offspring, I think that there can 
be no doubt as regards the true answer. They are induced 
to do so by an instinct which has been acquired through the 
process of natural selection because it has a tendency to 
preserve the next generation and thereby the species. 
This is shown by the fact that in such cases the male not 
only stays with the female and young but also takes care 
of them. Marital and paternal instincts, like maternal 
1 Schweinfurth, op. cit. i. 522. 
2 Savage, ' On Troglodytes niger,' in Boston Journal of Natural 
History, iv. 384 sq. See also Du ChaiJlu, op. cit. p. 358 ; Hartmann, 
op. cit. p. 221. 
3 v. Koppenfels, loo. cit. p. 418. 4 Jenks, loc. cit. p. 61. 36 
THE HISTORY OF HUMAN MARRIAGE chap. 
affection, are necessary for the existence of certain species, 
although there are many other means by which a species 
may be enabled to subsist. Where parental care is lacking, 
we may be sure to find compensation for it in some other 
way. Among the Invertebrata, Fishes, and Reptiles, both 
parents are generally quite indifferent as to their progeny. 
An immense proportion of the latter therefore succumb before 
reaching maturity ; but the number of eggs laid is propor- 
tionate to the number of those lost, and the species is 
preserved nevertheless. If every grain of roe spawned 
by the female fishes were fecundated and hatched, the 
sea would not be large enough to hold all the creatures 
resulting from them. The eggs of Reptiles need no maternal 
care, the embryo being developed by the heat of the sun ; and 
their young are from the outset able to help themselves, 
leading the same life as the adults. Among Birds, on the 
other hand, parental care is an absolute necessity. Equal 
and continual warmth is the first requirement for the develop- 
ment of the embryo and the preservation of the young ones. 
For this the mother almost always wants the assistance of 
the father, who provides her with necessaries, and sometimes 
relieves her of the brooding. Among Mammals, the young 
can never do without the mother at the tenderest age, but the 
father's aid is generally by no means indispensable. In 
some species, as the walrus,1 the elephant,2 the Bos ameri- 
canus,3 and the bat,4 there seems to be a rather curious 
substitute for paternal protection, the females, together 
with their young ones, collecting in large herds or flocks 
apart from the males. 
In the case of the man-like apes there are some obvious 
facts which might account for the need of marital and 
paternal protection. One is the small number of young; 
the female brings forth but one at a birth.5 Another is 
the long period of infancy ; the orang-utan is said to be full- 
grown only at the age of fifteen.6 If the family life of this 
1 Brehm, Thievleben, iii. 649. 2 Ibid. iii. 479. 
3 Ibid. iii. 400. 4 Ibid. i. 299. 
6 Jenks, loo. cit. pp. 57, 61 (gorilla, chimpanzee). Hornaday, 
op. cit. p. 403 (orang-utan). 0 Mohnike, loc. cit. p. 850. I THE ORIGIN OF MARRIAGE 
37 
ape, nevertheless, is more defective than that of the gorilla 
and chimpanzee, the reason may perhaps be that it is 
exposed to fewer dangers. " Except man," says Mohnike, 
" the orang-utan in Borneo has no enemy of equal strength ' j1 
and on account of his great strength, says Rajah Brooke, 
" an old male might attack a single man if provoked."2 
Finally, as we have seen, none of these apes can be called 
gregarious animals. In this respect they differ from the 
smaller monkeys; and the reason for this is probably 
just their larger size, which, on the one hand, makes the 
protection afforded by gregariousness less necessary, and, 
on the other hand, makes it more difficult to live in larger 
herds owing to the greater quantities of food required. 
It is said that the gorilla hardly ever spends two nights in 
the same place, each family roaming about in the bush from 
place to place in search of food. 3 Savage tells us that the 
chimpanzees are more numerous in the season when the 
greatest number of fruits come to maturity,4 which seems to 
indicate that the solitary life in separate family groups 
generally led by this ape is due chiefly to the difficulty it 
experiences in getting food at other times of the year. And 
the comparatively greater sociability attributed to it by 
the natives of Kamerun may perhaps have something 
to do with the fact that " the immense forests furnish 
an abundance of varied food, so the chimpanzee usually 
experiences little trouble in satisfying its hunger."6 
When we from the highest monkeys pass to man, we meet 
with the same phenomenon. Among the lowest savages, as 
well as the most civilised races of men, we find the family 
consisting of parents and children, and the father as its 
protector and supporter. There are, it is true, statements 
according to which certain peoples five or have lived in a 
state of promiscuity without any family ties; there are 
various customs which have been interpreted as survivals of 
1 Ibid. p. 894. 8 Brooke, op. cit. i. 225. 
3 Jenks, loc. cit. p. 57. 
4 Savage, in Boston Journal of Natural History, iv. 384. Cf. von 
Koppenfels, loc. cit. p. 419. 
5 Jenks, loc. cit. p. 6r. 38 
THE HISTORY OF HUMAN MARRIAGE chap 
such a state in the past; and the hypothesis has been set 
forth that promiscuity prevailed universally among primitive 
men. I shall, in subsequent chapters, try to show that most 
of the statements in question are obviously inaccurate, and 
none of them can be proved to be, or is even likely to be, 
true ; that the so-called survivals of earlier promiscuity may 
be interpreted in a much more satisfactory manner other- 
wise ; and that the hypothesis of a primitive stage of pro- 
miscuity not only lacks all foundation in fact, but is utterly 
opposed to the most probable inference we are able to make 
as regards the early condition of man. On the other hand, 
there are undoubtedly peoples among whom a child stands 
in a more intimate relation to its maternal uncle than to its 
father. 
In a few exceptional cases it is said to be the custom for 
the husband not to live with his wife at all, but merely to 
pay her visits in the place where she dwells with her maternal 
relatives ; and the children she bears then remain with her. 
Of the Orang Mamaq in Sumatra, who are divided into a 
number of exogamous matrilineal clans, we are told that 
man and wife generally continue to live each with his or 
her own clan, though it sometimes happens that the husband 
moves to his wife's home. The head of the family is the 
eldest brother of the wife, whereas the husband and father 
is not considered to belong to it at all. Yet their marriages 
are said to be not only monogamous but indissoluble save 
by death.1 Among the kindred Malays of the Padang 
Highlands in the same island there is a similar institution. 
Married life, we are told, reveals itself merely in the form of 
visits which the husband pays to his wife ; in the beginning, 
at least, he comes by day, helps her in her work in the rice- 
fields and takes his midday meal with her, but later he 
generally comes privately in the evening to his wife's house 
and stays there, if he be a faithful husband, until the fol- 
lowing morning. As a rule, he does not sufficiently provide 
1 Graafland, ' De verbreiding van bet matriarchaat in het 
landschap Indragiri,' in Bijdragen tot de taal-, land- en volkenkunde 
van Nederlandsch-Indie, ser. v. vol. v. (vol. xxxix.) p. 43 sq. Hagen, 
Die Orang Kubu auf Sumatra, p. 164. I THE ORIGIN OF MARRIAGE 
39 
for his wife and children ; here, too, it is the maternal uncle 
(mamaq) of the latter that is their father, so far as duties 
and rights are concerned, and the head of the family in the 
narrower sense of the term, the so-called sa-mandei.1 Among 
the Syntengs of the Jaintia Hills in Assam, the husband 
likewise only visits his wife at her mother's house ; "in 
Jowai," says Major Gurdon, " some people admitted to me 
that the husband came to his mother-in-law's house only 
after dark, and that he did not eat, smoke, or even partake 
of betel-nut there, the idea being that because none of his 
earnings go to support this house, therefore it is not etiquette 
for him to partake of food or other refreshment there.''2 
Among the Nayars of Malabar, who practised a sort of 
polyandry, the woman lived apart from her husbands, or 
lovers, who cohabited with her by agreement amoiig them- 
selves. They contributed to' maintain her, and according 
to several authorities the children as well; but in some 
accounts all paternal duties are said to be ignored, the child- 
ren being brought up by their maternal uncle. Even to 
this day, when polyandry has almost entirely ceased to exist 
among the Nayars, it frequently happens that the wife 
remains in her own tarwad (the common residence of the 
children of the same maternal ancestor), and that the hus- 
band only visits her in her house in the night and goes 
home the next morning ;3 nay, from the strictly legal point 
of view, wife and children possess no privilege of claiming 
maintenance from the husband and father.4 These cases, 
however, are certainly rare exceptions to a well-nigh universal 
rule, and must by no means be regarded as anything like 
representative of the relations between man and woman 
among peoples with matrilineal descent. As for the Nayars, 
it seems that their peculiar customs were closely connected 
with their military organisation.5 
1 Pistorius, Studien over de inlandsche huishouding in de Padangsche 
Bovenlanden, p. 42 sqq. van Hasselt, Volksbeschrijving van Midden- 
Sumatra, p. 245 sq. 
2 Gurdon, Khasis, p. 76. 3 Infra, on Polyandry. 
4 Gopal Panikkar, Malabar and its Folk, p. 22. 
5 Infra, on Polyandry. 40 
THE HISTORY OF HUMAN MARRIAGE chap. 
More frequently it is said that the mother's brother has 
greater rights over a child than the father, or that the latter's 
authority is very slight or even nil, although the children 
live with their parents, at least in earlier years. This is 
reported of various African peoples mentioned by Dr. 
Hartland-the Alladians on the Ivory Coast, the people of 
Loango, the Igalwas of the Lower Congo, the Kimbunda, 
the Ewhe of Anglo in Upper Guinea, the Kunama of Northern 
Abyssinia, the Suahili, the Wanyika in the hinterland of 
Mombasa-and of the Melanesians of the Gazelle Peninsula 
of New Britain.1 Yet it is hard to believe that the 
father really is devoid of all power over his children while 
they remain in his house. Among the Bambala, according 
to Mr. Torday, the children of a " betrothed " wife belong 
to their maternal uncle, although those of a purchased wife 
belong to their father, and after the age of seven or eight 
the children of the former class may go to the uncle ; but 
we are expressly told that " as long as children remain with 
the father he is supreme."2 Perhaps something like might 
be said of various other Congo natives, as well as of the 
coast people of the Gazelle Peninsula, among whom the 
children or boys move to the village of their maternal uncle 
when they reach a certain age.3 On the Lower Congo, 
according to Mr. Weeks, when the lad is about fourteen or 
fifteen years old his maternal uncle brings a calabash of 
palm wine to the father and claims the lad. " The father 
has no power to withhold him from going with his uncle, 
but the lad himself can refuse to go, and thus elect to remain 
under the tutelage of his father as long as he likes."4 
Generally speaking, I think it is necessary to receive with 
some caution statements which attribute unqualified power 
1 Hartland, Primitive Paternity, i. 277, 281, 282, 284, 285, 287, 
288, 290. 
2 Torday, Camp and Tramp in African Wilds, p. 95. 
3 Ibid. p. 134 (Bayaka). Torday and Joyce, ' Notes on the 
Ethnography of the Ba-Huana,' in Jour. Anthr. Inst, xxxvi. 285 sq. 
Johnston, George Grenfell and the Congo, ii. 673 (natives of South 
Congoland). Burger, Die Kusten- und Bergvolker der Gazellehalb- 
insel, p. 30. 
4 Weeks, Among the Primitive Bakongo, p. 119. I THE ORIGIN OF MARRIAGE 
41 
to the maternal uncle to the exclusion of the father. Tra- 
vellers are naturally impressed by the difference between the 
European family system and that of the people they visit, 
and are therefore liable to emphasise this difference some- 
what more than is justified by the actual facts. In this way 
we may explain certain inconsistencies occasionally found in 
their accounts. Dr. Ruelle, for instance, states that among 
the Lobi in French West Africa the children belong to their 
uncle, although they live with their parents; but on the 
next page we read, " Il n'existe aucune autre autorite que 
celle du pere de famille qui 1'exerce reellement, soutenu, au 
besoin, par les membres qui la constituent." Nay, in case 
of divorce all children from six years of age upwards remain 
with the father.1 Concerning the Ewhe-speaking peoples on 
the Slave Coast, Major Ellis tells us in one place that " the 
eldest brother is the head of the family," but speaks in 
another place of the father as the " owner or master of the 
household," and says that when a man wants to marry a 
certain girl the negotiations are made with her parents ; 
whilst in the case of betrothals between children a present 
is made by the parents of the male child to those of the 
female child.2 Magyar informs us that among the Kimbunda 
the father has no power over his sons even during their 
minority; but, nevertheless, we hear that " the grown-up 
son leaves his parents' house as soon as he is marriageable, 
and establishes himself in an independent position," and 
that the daughters are given away in marriage by their 
parents.3 With reference to the same people, it is of interest 
to note that, although no rights over the sons are attributed 
to the father, they are said to be under his guardianship 
(Obhut).* 
The elementary paternal duties seem to be recognised 
universally while the children live in the father's house, 
1 Ruelle, ' Notes anthropologiques, ethnographiques et sociolo- 
giques sur quelques populations noires du 2e territoire militaire de 
l'Afrique occidentale fran^aise,' in L'Anthropologic, xv. 661 sq. 
2 Ellis, Ewe-speaking Peoples of the Slave Coast, pp. 207, 211, 155, 
201. 
x 3 Magyar, Reisen in Slid-Afrika, pp. 284, 281 sq. 4 Ibid. p. 284. 42 
THE HISTORY OF HUMAN MARRIAGE chap. 
however limited be his rights over them. The Ewhe of 
Anglo, among whom the maternal uncle is said to have 
greater power than the father, consider it to be incumbent 
upon the latter to take care of the boys when they grow 
older, whilst the education of the girls devolves on the 
mother; and when a boy gets old enough to marry, the 
father and the uncle jointly procure a wife for him.1 On the 
Lower Congo, " until the uncle comes with the palm wine 
the lad is under the protection of his father, who is responsible 
for him to the boy's family, but on the boy going with the 
uncle, the father's responsibility is ended."2 The Herero, 
among whom the father's power is said to consist merely in 
the right of chastising his child,3 compel him to pay com- 
pensation to his wife's kin if he neglects a child so that it 
dies, nay apparently if it dies without any fault on his part.4 
In Garenganze, to the south-west of Lake Moero, "if a 
freeborn child were lost or devoured by wild animals, the 
father would have to pay its value to his wife's relatives, 
as . . . freeborn children are in some parts supposed to 
belong entirely to their mother."5 Among the Kaupuis of 
Manipur, on the death of a child, munda, or " bone-money," 
is demanded by the wife's father.6 Among the Papuans of 
the Tami Islands, if a child dies the father makes presents 
to the mother's kin-obviously, as Kohler remarks, because 
the father is held responsible for the death and redeems his 
liability with a gift.7 Dr. Hartland mentions the last four 
cases as instances of " the alien position occupied among 
matrilineal peoples by the father in regard to his children."8 
1 Hartter, ' Sitten und Gebrauche der Angloer,' in Zeitschr. f. 
Ethnol. xxxviii. 43. 
2 Weeks, op. cit. p. 119. 
3 Dannert, Zum Rechte der Herero, p. 47. 
4 Ibid. p. 47. Kohler, ' Das Recht der Herero,' in Zeitschr. vergl. 
Rechtswiss. xiv. 307. 
5 Arnot, Garenganze, p. 242. 
G Watt, ' Aboriginal Tribes of Manipur,' in Jour. Anthr. Inst. 
355- 
7 Kohler, ' Das Recht der Papuas,' in Zeitschr. vergl. Rechtswiss. 
xiv. 351. 
8 Hartland, op. cit. i. 275-279, 281. I THE ORIGIN OF MARRIAGE 
43 
In this connection they are of interest as showing that even 
where the mother's kindred have certain rights over the 
children, the father is held responsible for their lives. 
But it must not be supposed that it is a general rule among 
matrilineal peoples that the maternal uncle or any other 
member of the mother's kin has more authority over the 
children than the father. In all Australian tribes, whether 
patrilineal or matrilineal, the father is most distinctly the 
head of the family.1 The same is the case in many parts 
of Melanesia, where descent is traced through the mother ; 
as Dr. Codrington puts it, " the house of the family is the 
father's, the garden is his, the rule and government are his."2 
Of the Khasis in Assam, Major Gurdon writes :-" It is 
true that the kni, or mother's elder brother, is the head of 
the house, but the father is the executive head of the new 
home, where, after children have been bom to him, his wife 
and children live with him. It is he who faces the dangers 
of the jungles, and risks his life for wife and children. . . . 
In his own family circle a father and husband is nearer to 
his children and his wife than u knif'3 In Madagascar the 
prevalence of matrilineal descent does not prevent the 
commands of a father or an ancestor from being " held as 
most sacredly binding upon his descendants."4 Among the 
Mpongwe, who reckon descent through the mother, the 
father has by law unrestricted power over his children.5 
Concerning the matrilineal Algonkin of North America, 
Charlevoix says that even though the father " is not regarded 
1 Curr, The Australian Race, i. 60, 62, 69. Eylmann, Die Einge- 
borenen der Kolonie Siidaustr alien, p. 168. Malinowski, The Family 
among the Australian Aborigines, p. 67 sqq. See also infra, i. 66 sq. 
2 Codrington, Melanesians, p. 34. See also Hagen, Unter den 
Papua's, p. 224 sq. (natives of Kaiser Wilhelm Land) ; Vetter, 
' Bericht fiber papuanische Rechtsverhaltnisse,' in Nachrichten uber 
Kaiser Wilhelms-Land, 1897, p. 86 (natives of Simbang) ; Jung, 
' Rechtsanschauungen der Eingeborenen von Nauru,' in Mittheil. 
Deutsch. Schutzgeb. x. 65 (natives of Nauru, one of the Marshall 
Islands). 
3 Gurdon, op. cit. p. 78 sq. 
4 Sibree, The Great African Island, p. 326. 
6 Hiibbe-Schleiden, Ethiopien, pp. 151, 153. 44 
THE HISTORY OF HUMAN MARRIAGE chap. 
as father, he is always respected as the master of the cabin."1 
Of the Iroquois-who, on the authority of Morgan,2 have 
been represented as one of the few instances of mother-right 
" in its most typical form," where the father has no authority 
in the household3-an earlier authority tells us that the 
mother superintends the children, but that the word of the 
father is law and must be obeyed by the whole household.4 
These are only a few instances of a very widespread right 
granted to the father among peoples who have the matri- 
lineal system of descent. 
Dr. Hartland-to mention only the latest exponent of 
" mother-right "-regards the authority of the father, nay 
even that of the maternal uncle, as the result of later develop- 
ment. The potestas, he says, vests in the first instance in 
the elders of the kin at large. " As the consciousness of kin 
becomes gradually more vivid and defined the elders of the 
inchoate family absorb the headship of their more immediate 
kin and administer its concerns. Gradually the headship 
becomes concentrated in the hands of one man, often chosen 
by the family from among a small number specially qualified 
by age experience wisdom or courage, or designated by 
propinquity of blood to the predecessor in office." But 
" when the family under motherright emerges the power is 
found to be wielded not by the husband but by the wife's 
brothers, or her maternal uncles, a circle constantly narrow- 
ing until the definition of these terms approximates to our 
own, one of whom takes ultimately the lead and appropriates 
the greater part or sometimes the whole of the potestas. 
Nor does the transition to the reckoning of descent through 
the father entirely and at once divest him of it. Enough 
survives in his hands to form very material evidence of the 
more archaic social organisation which preceded the estab- 
1 Charlevoix, Histoire de la Nouvelle France, v. 424. 
2 Morgan, League of the Iroquois, p. 325. 
8 Rivers, ' Mother-Right,' in Hastings, Encyclopedia of Religion 
and Ethics, viii. 851. 
4 Seaver, Narrative of the Life of Mrs. Mary femison, p. 165. 
Mrs. Jemison was a white woman who was captured by the Indians 
in 1755- t THE ORIGIN OF MARRIAGE 
45 
lishment of fatherright." And we are told that " the result 
of anthropological investigations during the past half- 
Century has been to show that motherright everywhere 
preceded fatherright and the reckoning of descent in the 
modem civilised fashion through both parents." 1 In short, 
the family consisting of father, mother, and children has 
everywhere been preceded i by a social organisation of 
mother-right where the father was a wholly subordinate 
personage. 
The main facts on which this theory is based have already 
been mentioned. Among a few peoples, not even half a 
dozen in number, it is said to be the custom for husband 
and wife to remain permanently in their own communities, 
apart from each other, and for the children to stay with the 
mother. Among various other peoples, the mother's kin, 
and particularly the maternal uncle, are said to have greater 
rights over the children than the father, if not exclusive 
rights over them ; whereas among many matrilineal peoples 
the father's potestas is paramount. I can find no reason 
whatever to assume that the latter peoples, also, had full- 
fledged mother-right in former times. If a certain institu- 
tion is highly developed among some peoples and much less 
developed among others, it does not follow that it was once 
highly developed among the latter as well. In the present 
case any such conclusion is particularly illegitimate, con- 
sidering that the fullest mother-right prevails among agri- 
cultural tribes, whereas the matrilineal system is nowhere 
feebler than among the Australian aborigines, who still live 
in the hunting and food-collecting stage. This is very 
significant on account of the close connection which exists 
between the family organisation and factors of an economic 
character. Nor can I accept the statement that " mother- 
right everywhere preceded fatherright " ; but this question 
will t>e most conveniently discussed together with the 
hypothesis which regards the matrilineal system as a relic 
of early promiscuity. Those who advocate a primitive stage 
of mother-right without paternal rights and paternal duties 
are faced by the formidable fact, which will be dealt with 
1 Hartland, op. cit. i. 299, 300, 256 sq. 46 
THE HISTORY OE HUMAN MARRIAGE chap. 
presently, that among the lowest savages, who chiefly or 
exclusively subsist on game and such products of nature as 
they can gather without cultivating the soil or breeding 
domestic animals, the family consisting of parents and 
children is a well-marked social unit, with the father as its 
head and protector. 
That the functions of the husband and father in the 
family are not merely of the sexual and procreative kind, 
but involve the duty of protecting the wife and children, 
is testified by an array of facts relating to peoples in all 
quarters of the world and in all stages of civilisation. The 
North American Indian considered it disgraceful for a man 
to have more wives than he was able to maintain.1 Powers 
says that among the Patwin, a rude hunting tribe of Cali- 
fornia, " the sentiment that the men are bound to support 
the women-that is, to furnish the supplies-is stronger even 
than among us."2 Among the Iroquois it was the office of 
the husband " to make a mat, to repair the cabin of his 
wife, or to construct a new one." The product of his 
hunting expeditions belonged during the first year of mar- 
riage to his wife, and afterwards he shared it equally with 
her, whether she remained in the village or accompanied 
him on the chase.3 Among the Tarahumafe of Mexico the 
bridegroom's father at the wedding makes a speech in which 
he tells the young man " that he has to kill deer and take 
care always to bring some animal home to his wife, even 
if it be only a chipmunk or a mouse. He also has to plough 
and to sow corn and to raise crops, that he and she may 
always have enough to eat and not go hungry."4 Among the 
Pawnee " a young man did not expect to marry until he 
had come to be an expert hunter, and so was able to support 
a wife."5 So also among the Guaraunos on the Orinoco 
"un gargon n'entre en menage que lorsqu'il pent nourrir une 
1 Waitz, Anthropologic der Naturvolker, iii. 109. Carver, Travels 
through the Interior Parts of North America, p. 367. 
2 Powers, Tribes of California, p. 222. 
3 Heriot, Travels through the Canadas, p. 338. 
4 Lumholtz, Unknown Mexico, i. 269. 
6 Grinnell, Story of the Indian, p. 42 sq. i THE ORIGIN OF MARRIAGE 
47 
femme avec les produits de sa chasse et sa peche."1 Of the 
Indians of the North-West Amazons we are told that " the 
husband, once he has obtained his wife, is entirely responsible 
for her maintenance."2 Among the Charruas of Uruguay, 
" du moment ou un homme se marie, il forme une famille a 
part, et travaille pour la nourrir."3 Among the Fuegians, 
" as soon as a youth is able to maintain a wife, by his exer- 
tions in fishing or bird-catching, he obtains the consent of 
her relations."4 
Concerning the Tonga Islanders, Mariner remarks that 
" a married woman is one who cohabits with a man, and 
lives under his roof and protection."5 In Samoa, it is said, 
" whatever intercourse may take place between the sexes, 
a woman does not become a man's wife unless the latter 
take her to his own house."6 Among the Maori " the mis- 
sion of woman was to increase and multiply ; that of man to 
defend his home."7 In Pentecost, of the New Hebrides, the 
father of the bride, or some friend of consequence at the 
wedding, formally exhorts the bridegroom " to feed his wife 
properly and treat her kindly."8 Among some of the Mar- 
shall Islanders even natural children were received by the 
father into his house as soon as they were able to walk.9 
.Among the.Bangerang tribe in Victoria "the ordinary 
business of the married men was to see to the safety of the 
family, procure meat or fish, and opossum skins enough to 
keep them clad " ; whilst the women provided the daily 
supply of wild roots and vegetables.10 With reference to the 
1 Chaffanjon, L'Orenoque et le Caura, p. n. 
2 Whiffen, North-West Amazons, p. 66. See also ibid. p. 164 sq. 
3 Azara, Voyages dans V Amerique meridionale, ii. 22. 
4 King and Fitzroy, Voyages of the Adventure and Beagle, ii. 182. 
For other instances from South America see Rivet, ' Les Indiens 
Jibaros,' in L'Anthropologic, xviii. 607 ; Cardus, Las Misiones 
Fransiscanas entre los inflates de Bolivia, p. 263 (Tobas). 
6 Mariner, Natives of the Tonga Islands, ii. 167. 
fi Pritchard, Polynesian Reminiscences, p. 134. 
7 Johnstone, Maoria, p. 28 sq. s Codrington, op. cit. p. 240. 
9 Kotzebue, Voyage of Discovery into the South Sea, iii. 173. 
Cf. Kohler, in Zeitschr. verN- Rechtswiss. xiv. 423. 
111 Curr, Recollections of Squatting in Victoria, pp. 255, 251. 48 
THE HISTORY OF HUMAN MARRIAGE chap 
Kumai in Gippsland, Dr. Howitt states that " the man has 
to provide for his family with the assistance of his wife. His 
share is to hunt for their support, and to fight for their 
protection." 1 Among the North-West-Central Queensland 
aborigines, according to Dr. Roth, "it is the husband's 
business in the main to supply the animal food for the 
family."2 In the Encounter Bay tribe in South Australia 
paternal care is considered so indispensable that if the 
father dies before the child is bom, it is put to death by the 
mother, as there is no longer any one to provide for it.3 
In his study of the South Australian natives Dr. Eylmann 
observes that it is also the business of the husband and 
father to erect the hut.4 
The Sea Dyak women " generally regard marriage as a 
means of obtaining a man to work for them. A woman 
will often separate from her husband simply because he is 
lazy, and will not do his fair share of the work."5 Among 
the Barito tribes in the south-east part of Borneo it is 
considered to be a husband's duty to provide his wife with 
food and clothing and other necessaries, as also to protect 
her from all dangers.6 The Nagas are not permitted to 
marry until they are able to set up house on their own 
account.7 Among the Eravallens, a jungle tribe in the 
Cochin State, " a young man is never allowed to marry 
unless he is able to support a wife."8 Among the Mal- 
divians, " although a man is allowed four wives at one time, 
it is only on condition of his being able to support them." 9 
1 Fison and Howitt, Kamilaroi and Kumai, p. 206. 
2 Roth, Ethnological Studies among the N orth-West-Central Queens- 
land Aborigines, p. 184. 
3 Meyer, ' Manners and Customs of the Encounter Bay Tribe,' 
in Woods, Native Tribes of South Australia, p. 186. 
4 Eylmann, op. cit. p. 168. 
6 Goines, Seventeen Years among the Sea Dyaks, p. 128. 
c Schwaner, Borneo, i. 199. 
7 Stewart, ' Notes on Northern Cachar,' in Jour. Asiatic Soc. 
Bengal, xxiv. 614. 
8 Anantha Krishna Iyer, Cochin Tribes and Castes, i. 44. 
u Rosset, ' On the Maldive Islands,' in Jour. Anthr. Inst. xvi. 
168 sq. 1 THE ORIGIN OF MARRIAGE 
49 
The Rock Veddas in Ceylon " acknowledge the marital 
obligation and the duty of supporting their own families."1 
In Nyasaland " native custom expects the husband to 
maintain his wife in clothes, to build the hut in which they 
are to live, to pay the Government tax on it, and generally 
to protect his spouse and treat her with reasonable con- 
fidence."2 The Wasania in British East Africa prohibit a 
man from having more than three wives, as they consider 
him incapable of supporting a larger number.3 Among the 
Shambaa in East Africa a wife is allowed to leave her 
husband if he neglects to provide her with food, clothing, 
and a dwelling;4 and the like is said of the Negroes of 
Angola.5 So also among the Bavili a wife may demand 
dissolution of the marriage if her husband is long absent 
from home and fails to support her.6 Among the Xosa- 
Kafirs the father must provide for the maintenance and 
welfare of the members of the family.7 
The husband and father being a supporter and pro- 
tector of his family, a man is often not permitted to marry 
until he has given some proof of his ability to fulfil these 
duties. 
Among the Macusis of British Guiana, before a young 
man is allowed to choose a wife, " he must prove that he 
is a man, and can do man's work. Without flinching, 
he suffers the infliction of wounds in his flesh ; or he allows 
himself to be sewn up in a hammock full of fire-ants ; or 
by some other similar tests he shows his courage. And he 
1 Tennent, Ceylon, ii. 441. 
2 Duff, Nyasaland under the Foreign Office, p.317. 
3 Barrett, ' Notes on the Customs and Beliefs of the Wa-Giriama, 
etc., British East Africa,' in Jour. Roy. Anthr. Inst. xli. 30. 
4 Dahlgriin, ' Heiratsgebrauche der Schambaa,' in Mittheil. 
Deutsch. Schutzgeb. xvi. 227. 
6 Monteiro, Angola and the River Congo, i. 264 sq. 
e Dennett, At the Rack of the Black Man's Mind, p. 40. 
7 Kropf, Das Volk der Xosa-Kajfern im bstlichen Sudafrika, p. 139. 
For other African instances see Poupon, * £tude ethnographique 
des Baya de la circonscription du M'Bimou,' in L'Anthropologic, 
xxvi. 124 ; Vergette, Certain Marriage Customs of some of the Tribes 
in the Protectorate of Sierra Leone, p. 21 sq. 
VOL. I K 50 
THE HISTORY OF HUMAN MARRIAGE chap 
clears a space in the forest to be planted with cassava, 
and brings in as much game and fish as possible, to show 
that he is able to support himself and others."1 Some similar 
exhibition of courage or the ability to support a family is 
required of a bridegroom in various other South American 
tribes.2 Thus, among the Bororo every young man who 
wishes to marry must have killed either five peccaries or one 
jaguar, whilst he who kills five jaguars has the right to have 
two wives.3 The Indians of Pennsylvania considered it a 
shame for a youth to think of a wife before he had given 
some proof of his manhood.4 Among the Kinipetu and 
some other Eskimo in the neighbourhood of Hudson's Bay, 
a young man is not permitted to marry until he, through 
his skill in hunting and otherwise, has shown that he can 
support not only his wife and children, but his parents-in- 
law as well.6 The Koyukun of Alaska believe that a man 
who marries before he has killed a deer will have no children.6 
Among the Koryak and Yukaghir, according to Dr. 
Jochelson, the custom of serving for a bride is intended 
to test the young man's ability to work, the bridegroom 
being required to be a good hunter and fisherman and 
capable of doing everything necessary in a household. 
Among the Yukaghir, besides serving, he had to pass through 
another test as well:-" The prospective father-in-law would 
go to the woods and fell as thick a tree as he could find. 
1 Im Thurn, Among the Indians of Guiana, p. 221. See also 
v. Martius, Beitrage zur Ethnographic Amerika's, i. 645 ; Appun, 
' Die Indianer von Britisch-Guayana, ' in Das Ausland, xliv. 446. 
2 Crevaux, Voyages dans l'Amerique du Sud, pp. 307 (Apalai, 
Roucouyennes), 612 (Guaraunos). Wallace, Narrative of Travels on 
the Amazon and Rio Negro, p. 498 (Uacarras). v. Martius, op. cit. i. 247 
(Guat6s), 688 (Arawaks). Ignace, ' Les Capiekrans,' in Anthropos, 
v. 477. Fernandez, Relacion de las missiones de los Indios, que 
Haman Chiquitos, p. 33. Schmidt, ' Uber das Recht der tropischen 
Naturvolker Sudamerikas, ' in Zeitschr. vergl. Rechtswiss. xiii. 307 sq. 
3 Erie and Radin, ' Contribution to the Study of the Bororo 
Indians,' in Jour. Anthr. Inst, xxxvi. 390. 
4 Buchanan, Sketches of the History, etc. of the North American 
Indians, p. 323. 
5 Klutschak, Ais Eskimo unter den Eskimos p. 233. 
6 Dall, Alaska and its Resources, p. 196. i THE ORIGIN OF MARRIAGE 
51 
The bridegroom had to drag the trunk of the tree to the 
house of his future father-in-law, and throw it upon the tent, 
so that it would fall. Then the father-in-law would say, 
' This is a good man ; he will be able to support us and to 
care for our safety.' "x Among the Koryak, again, on the 
evening of the wedding day the parents of the bride attacked 
the bridegroom with sticks, after which everybody else was 
allowed to do the same; and only on showing himself as 
a man by receiving the flogging with fortitude and without 
making resistance was he allowed to take away the bride.2 
So also among some Arabs of Upper Egypt the man must 
undergo an ordeal of whipping by the relations of the bride 
in order to test his courage ; and if he wishes to be con- 
sidered worth having, he must receive the chastisement, 
which is sometimes exceedingly severe, with an expression of 
enjoyment.3 We shall see, however, that customs of this 
kind may be something else than mere tests of courage.4 
Mr. Campbell of Selangor, in writing of the marriages 
of the Ulu Lang at Sakai, tells us that one custom was for 
the relations on both sides to sit on the ground round an 
ant-heap, and for the bride or her father to question the 
bridegroom as follows" Are you clever with the blow- 
pipe ? " " Can you fell trees cleverly ? " " Are you a good 
climber ? " and " Do you smoke cigarettes ? " If these 
questions were answered in the affirmative, the bridegroom 
gave a cigarette to the bride and lighted one himself. They 
then ran round the mound three times. If the man suc- 
ceeded in catching the woman the ceremony was completed, 
and they were declared married, whereas if he failed to catch 
her he tried again another day.5 In the Ladrone Islands the 
wedding took place only after the bridegroom through 
1 Joclielson, Koryak, p. 740. Idem, Yukaghir, p. 87 sq. 
2 v. Dittmar, ' Uber die Koraken,' in Melanges russes tires du 
bulletin hist.-philol. de I'Academic imper. des Sciences de St.-Petersbourg, 
iii. 25. Cf. Kennan, Tent Life in Siberia, p. 137 sq. 
3 Baker, Nile Tributaries of Abyssinia, p. 125. 
4 See infra, on Marriage Rites, 
5 Campbell, quoted by Skeat and Blagden, Pagan Races of the 
Malay Peninsula, ii. 67 sq. 52 
THE HISTORY OF HUMAN MARRIAGE chap. 
careful tests had proved capable of maintaining a wife.1 
And according to Don Luis de Torres, no Caroline Islander 
" is allowed to marry until he has given proofs of his dexterity 
in steering a proa."'2. 
Of the Atayals of Formosa,3 the Gaddanes of Luzon,4 
the Alfoors of Ceram,6 the Dyaks of Borneo,6 and the Nagas 
of Upper Assam,7 it is reported that no man can marry 
without having first procured at least one human head as 
a token of his valour ; but with reference to the Sea Dyaks, 
Mr. Gomes observes that this is true only of their chiefs.8 
As to the hill tribes of Assam, Mr. Hodson is inclined to 
believe " that success in head-hunting was at one time, if 
not essential to marriage, regarded at least as a token of 
having passed from adolescence to maturity." 9 The ancient 
Karmanians, according to Strabo, were considered marriage- 
able only after they had killed an enemy.10 The desire of a 
Galla warrior is to deprive the enemy of his genitals, the 
possession of such a trophy being said to be a necessary 
preliminary to marriage.11 Formerly, no Masai was able to 
marry until he had been on several raids.12 Among the 
Wapokomo of British East Africa early marriage is, in some 
districts of their country, prevented by the rule that no 
man may marry until he has killed a crocodile, and given a 
part of the flesh to the woman to eat.13 Among the Bechuana 
and Kafir tribes south of the Zambesi a youth is only 
1 Meinicke, Die Inseln des Stillen Oceans, ii. 407. Freycinet, 
Voyage autour du monde, ii. 277 sq. 
2 Arago, Voyage round the World, ii. 16. 
3 Davidson, Island of Formosa, p. 566. 
4 Worcester, Philippine Islands, p. 439. 
5 Bickmore, Travels in the East Indian Archipelago, p. 205. 
G Wilkes, United States Exploring Expedition, v. 363. Bock, 
Head-Hunters of Borneo, pp. 216, 221, &c. 
7 Dalton, Descriptive Ethnology of Bengal, p. 40. 
8 Gomes, op. cit. p. 74. Cf. ibid. pp. 73, 87. 
9 Hodson, ' Head-Hunting among the Hill Tribes of Assam,' 
in Folk-Lore, xx. 141. 
19 Strabo, Geographica, xv. 2. 14. 
11 Krapf, Reisen in Ost-AJrika, i. 274. 
12 Hollis, Masai, p. 302 n. 1. 
13 Gregory, Great Rift Valley, p. 343. T THE ORIGIN OF MARRIAGE 
53 
allowed to take a wife after lie has killed a 'rhinoceros,1 
and among the Auin Bushmen, after he has killed some big 
game.2 
Among some uncivilised peoples a man is even obliged 
to support the girl he is going to marry from the day of 
their betrothal.3 Among others, again, custom requires 
the former husband to support his divorced wife with her 
children.4 And on a man's death the obligation of main- 
taining his widow often devolves on his heir, the widespread 
custom of a man marrying the widow of his deceased 
brother being, as will be seen in a subsequent chapter, 
not only a privilege, but among several peoples a duty 
incumbent upon him. 
When we 'find in mankind a habit which it has in common 
with many other animal species, including those most 
nearly related to it, we naturally ask whether it may have 
a similar origin in all these cases. May we suppose that the 
more or less durable union between man and woman and 
the care which the man takes of the woman and their 
common offspring are due to instincts which were once 
necessary for the preservation of the human race ? We 
found reasons to believe that the marital and paternal 
relations among the man-like apes are the results of instincts 
which are needed for the subsistence of the offspring, 
because their number is small, the period of infancy is long, 
and the kind of food on which the species lives and the 
quantity of it required prevent a gregarious mode of life. 
Now there can be no doubt whatever that in mankind, 
also, the number of children has always been comparatively 
very small and the period of infancy comparatively very 
1 Livingstone, Missionary Travels, p. 147. 
2 Kaufmann, ' Die Auin,' in Mittheil.Deutsch. Schutzgeb. xxiii. 157. 
3 E.g., the Botocudos (v. Tschudi, Reisen durch Sildamerika, 
ii. 283), Ladrone Islanders (Meinicke, op. cit. ii. 407), Fanti (Sarbah, 
Fanti Customary Laws, p. 50). 
4 E.g., the Basuto (Endemann, ' Mittheilungen uber die Sotho- 
Neger,' in Zeitschr. f. Ethnol. vi. 40), Munda Kols of Chota Nagpur 
(Jellinghaus, ' Sagen, Sitten und Gebrauche der Munda-Kolbs in 
Chota Nagpore,' ibid. iii. 370). 54 
THE HISTORY OF HUMAN MARRIAGE chap. 
long. We have also good reasons to believe that our earliest 
human or half-human ancestors subsisted on essentially the 
same diet-chiefly but not exclusively vegetable1-and re- 
quired about the same quantities of food as the man-like apes.2 
Is it not likely, then, that the same causes have produced 
the same results in either case ? The objection will perhaps 
be raised that man, unlike the anthropoid apes, is now an 
extremely social animal and could therefore, like other 
social animals, easily have dispensed with those marital 
and paternal ties which for some reason or other exist, but 
could hardly have been needed for the subsistence of the 
species. But this objection loses its force when we consider 
the social conditions of savages who know neither cattle- 
rearing nor agriculture-unless perhaps of the most primitive 
kind-and exclusively or almost exclusively subsist on what 
nature directly gives them-game, fish, fruit, roots, and so 
forth. 
The natives of Tierra del Fuego are social in a very small 
degree. Snow states that " they reside in families."3 
According to Bishop Stirling, family life is exclusive among 
1 The chimpanzee has even a marked inclination for animal 
food (Cambridge Natural History, x. 576). 
2 Kollmann's view (see his ' Neue Gedanken uber das alte Problem 
von der Abstammung des Menschen,' in Globus, lxxxvii. 141 sqq.) 
that primitive man was a pygmy who had descended from a small 
anthropoid at most one metre high has not been accepted (see 
particularly Schwalbe, ' Zur Frage der Abstammung des Menschen,' 
ibid. Ixxxviii. 159 sqq., and in his Studien zur V orgeschichte des 
Menschen, p. n sqq.}. Professor Keith observes (Antiquity of Man, 
p. 498) that the great anthropoids-the gorilla, chimpanzee, and 
orang-" are so like man in structure of body that we must, to 
account for the degree of similarity, regard all of them as collateral 
descendants of a common stock." He adds (ibid. p. 498 sq) :- 
" We do not hesitate to think that the anthropoids retain, to a 
much greater degree than man, the structure and manner of living 
of the ancient stock from which all four have been evolved. If, 
therefore, we try to form a picture of the world of ancient and 
primitive humanity, we must base it on the conditions now existing 
among anthropoids, not on those which hold for the modern world 
of mankind." 
8 Snow, ' Remarks on the Wild Tribes of Tierra del Fuego,' 
in Trans. Ethn. Soc. London, N.S., i. 264. t THE ORIGIN OF MARRIAGE 
55 
them. " Get outside the family," he says, " and relation- 
ships are doubtful, if not hostile. The bond of a common 
language is no security for friendly offices."1 ' Wilkes 
observes that they " appear to live in families and not in 
tribes, and do not seem to acknowledge any chief."2 Other 
writers tell us that among the Yahgans (who live in the 
southern part of the archipelago, south of the Beagle Channel 
to the islands off Cape Horn) and the Alacalufs (who live 
on the western islands) each family is perfectly independent 
of all the others, and that only the necessity of common 
defence now and then induces a few families to form small 
gangs without a chief.3 With reference to the former, 
Hyades writes, " La famille est bien constitute, mais la 
tribu n'existe pas, a proprement parler."4 In a letter to me 
Mr. T. Bridges gave the following account of them :-" They 
live in clans, called by them ucuhr, which means a house. 
These ucuhr comprise many subdivisions ; and the members 
are necessarily related. But the Yahgans are a roving 
people, having their districts and moving about within 
these districts from bay to bay and island to island in canoes, 
without any order. The whole clan seldom travels together, 
and only occasionally, and then always incidentally, is it 
to be found collected. The smaller divisions keep more 
together. . . . Occasionally as many as five families are to 
be found living in a wigwam, but generally two families." 
In a printed article Mr. Bridges says that " family influence 
is the one great tie which binds these natives together, and 
the one great preventive of violence."5 Admiral Fitzroy 
observes that " scarcity of food, and the facility with which 
they move from one place to another in their canoes, are, 
no doubt, the reasons why the Fuegians are always so 
1 Stirling, ' Residence in Tierra del Fuego,' in South American 
Missionary Magazine, iv. n. 
2 Wilkes, op. cit. i. 124. 
3 Bove, Patagonia. Terra del Fuoco, p. 134. Lovisato, ' Appunti 
etnografici sulla Terra del Fuoco,' in Cosmos di Guida Cora, viii. 150. 
4 Hyades, ' Ethnographic des Fu6giens,' in Bull. Soc. d'Anthr. 
Paris, ser. iii. vol. x. 333. 
6 Bridges, ' Manners and Customs of the Firelanders,' in A Voice 
for South America, xiii. 204. 56 
THE HISTORY OF HUMAN MARRIAGE chap. 
dispersed among the islands in small family parties, why 
they never remain long in one place, and why a large 
number are not seen many days in society."1 Of 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,2 Senor Gallardo writes that they lead 
a nomadic life in bands the size of which depends on the 
supply of food, but that they do not join in groups sufficiently 
large to be called tribes. He adds that every man is the 
chief of his family and is obeyed by its members.3 Of the 
Patagonian Tehuelches, who are essentially hunters without 
having a fixed residence, Hutchinson states that they " are 
divided into a number of small tribes, dispersed into families." 4 
The natives of the Patagonian Channel Region between the 
Magellan Straits and the Gulf of Penas, according to Dr. 
Skottsberg, " live in families and have no idea of a com- 
munity. Now and then some families keep together, pro- 
bably those related to each other, as, for instance, two 
brothers with their wives and children." These Indians are 
still in the stone age, inhabiting a country devoid of available 
metals and with a climate inimical to agriculture of any 
kind.6 
Of the Guayaki, on the right side of the Upper Parana, 
opposite the Argentine territory Misiones, Father Vogt writes 
that they " live, as it seems, not in tribes but in families " ; 
they have no agriculture but subsist chiefly on game, fish, 
and fruits.8 Von Martius states that the Guachis of the 
Gran Chaco and the Guatos, a tribe near the sources of 
the Araguaya, are for the most part scattered in families.7 
1 King and Fitzroy, op. cit. ii. 177 sq. 
2 Cojazzi, Los indios del Archipielago Fueguino, p. 6. 
3 Gallardo, Tierra del Fuego-Los Onas, p. 207 sq. 
4 Hutchinson, ' Tehuelche Indians of Patagonia,' in Trans. Ethn. 
Soc. N.S. vii. 317. 
5 Skottsberg, Wilds of Patagonia, p. 97. Idem, ' Observations on 
the Natives of the Patagonian Channel Region,' in American 
Anthropologist, N.S. xv. 594, 596. 
G Vogt, ' Material zur Ethnographic und Sprache der Guayaki- 
Indianer,' in Zeitschr. f. Ethnol. xxxiv. 35, 37. 
7 y. Martius, op. cit. i. 244, 247. T THE ORIGIN OF MARRIAGE 
57 
Concerning the Guatos, Castelnau observes that they never 
unite in villages. Each family lives alone and builds for 
itself a hut in some inaccessible place. More than one man 
is never found in the same hut, for as soon as a son reaches 
the age of puberty he seeks a wife and makes an establish- 
ment of his own. At certain times, but only twice a year, 
the men assemble in places which have been fixed before- 
hand by their chiefs, but they remain there together for two 
days only.1 According to Bates, the social condition of the 
Caishanas, inhabiting the forests of the Tunantins, among 
whom each family has its solitary hut, " is of a low 
type, very little removed, indeed, from that of the brutes 
living in the same forests.''2 The Maraua Indians, on the 
Lower Jurua, are likewise dispersed in separate families or 
small hordes, and so were some of the other tribes visited 
by Bates.3 Ehrenreich states that the southern Karaya 
Indians, on the Araguaya, are compelled to scatter owing to 
lack of food, and that in the neighbourhood of S. Jose small 
bands consisting of eight to ten individuals are still met 
with during the dry season. According to the same authority, 
the forest Indians of Amazonas, on the Purus, and the 
nomadic hunters on the coast, live in small gangs comprising 
two or three families each.4 Concerning the Botocudos, a 
Brazilian tribe on the Tocantins, v. Tschudi writes that " the 
family is the only tie which joins these rude children of 
nature with each other."5 Von Martius observes that 
travellers in Brazil often meet with a language used only by 
a few individuals connected with each other by relationship, 
who are thus completely isolated, and can hold no com- 
munication with any of their other countrymen far or near.6 
The Togiagamiut-an Eskimo tribe of Alaska, never 
visited by white men in their own country until the year 
1 Castelnau, Expedition dans les parties centrales de V Amerique 
du Sud, iii. 12 sq. See also Max Schmidt, in Verhandl. Berliner 
Gesellsch. Anthr. 1902, p. 86. 
2 Bates, The Naturalist on the River Amazons, ii. 376. 
3 Ibid. ii. 381, 377 sq. ; i. 328. 
4 Ehrenreich, Beitrdge zur Volkerkunde Brasiliens, p. 12. 
5 v. Tschudi, op. cit. ii. 283. 6 v. Martins, op. cit. i. 46. 58 
THE HISTORY OF HUMAN MARRIAGE chap. 
1880-who lead a thoroughly nomadic life, wandering from 
place to place in search of game or fish, appear, according to 
Petroff, " to live in the most perfect state of independence of 
each other. Even the communities do not seem bound 
together in any way; families and groups of families con- 
stantly changing their abode, leaving one community and 
joining another, or perhaps forming one of their own. 
The youth, as soon as he is able to build a kaiak and to 
support himself, no longer observes any family ties, but goes 
where his fancy takes him, frequently roaming about with 
his kaiak for thousands of miles before another fancy calls 
him to take a wife, to excavate a miserable dwelling, and to 
settle down for a time."1 
Among the Reindeer Chukchee, in the north-eastern 
extremity of Asia, the family, says Dr. Bogoras, " forms the 
basis of the social relations between members of the tribe. 
Even family ties arc not absolutely binding, and single 
persons often break them and leave their family relations. 
Crown-up sons frequently leave their parents and go away 
to distant localities in search of a fortune. ... It may be 
said that a Ione man living by himself forms the real unit of 
Chukchee society." There are, however, camps, usually 
consisting of two or three families, with ten or fifteen persons 
together ; camps of four to six families form but a slight 
minority, and a camp with ten houses is almost impossible 
unless formed for special reasons, like the temporary camps 
in trading-places. In most cases the camp consists of related 
families-for instance, of brothers, cousins, etc., with their 
wives and children ; but each family has a house of its own. 
A group of kindred families, designated by the term va'rat 
(literally, " collection of those who are together ") " may 
perhaps be called an embryo of a clan ; it is unstable, how- 
ever, and the number of families that ' are together ' changes 
almost every year."2 The Reindeer Koryak usually wander 
in groups consisting of a few families, but " it sometimes 
happens that separate families wander far from their native 
1 Petroff, ' Report on the Population, Industries, and Resources 
of Alaska,' in Tenth Census of the United States, p. 135. 
2 Bogoras, Chukchee, pp. 537, 54 t, 612. I THE ORIGIN OF MARRIAGE 
59 
places, leaving the groups to which they originally belonged 
for one reason or another; as, for instance, on account of 
quarrels, lack of pastures in the old places for their reindeer, 
or the establishment through marriage of new family ties."1 
The Yukaghir, as a hunting-tribe, " frequently have to 
scatter in separate families, or groups of related families, 
in search of food."2 Among the reindeer-breeders of the 
Northern Yenisei valley, says Miss Czaplicka, " the family 
is practically the social unit, though they were probably 
originally organised in clans. The conditions of their life 
prevent these people from living together in sufficient 
numbers to form anything like villages. They are wanderers, 
roaming with their herds of reindeer wherever the prospect 
of good hunting or fishing, combined with that of finding an 
abundance of moss for the herds, may lead them. Occasion- 
ally several families are found living together in two or three 
chums (tents), but this is never a permanent arrangement, 
nor in accordance with any customary rules."3 
The forest Veddas of Ceylon only occasionally assemble 
in greater numbers.4 According to Bailey, the Nilgala 
Veddas, who are considered the wildest, " are distributed 
through their lovely country in small septs, or families, 
occupying generally caves in the rocks, though some have 
little bark huts. They depend almost solely on hunting for 
their support, and hold little communication even with 
each other."5 The brothers Sarasin tell us that during the 
dry season every family wanders about on its hunting- 
ground and only seldom comes into contact with its neigh- 
bours, unless induced to do so by disturbing influences from 
the outside. During the rainy season the various families 
living in the same district retire to their rock centre and 
stay there in their caves. The families who concentrate 
themselves in the same rock centre during this period form 
1 Jochelson, Koryak, p. 431. 2 Idem, Yukaghir, p. 86. 
3 Miss Czaplicka, My Siberian Year, p. 101. 
4 Pridham, Account of Ceylon, i. 454. Hartshorne, ' Weddas,' 
in Indian Antiquary, viii. 320. 
5 Bailey, ' Wild Tribes of the Veddahs of Ceylon,' in Trans. Ethn. 
Soc. N.S. ii. 281. 60 
THE HISTORY OF HUMAN MARRIAGE citap. 
a clan ; but the different clans have nothing to do with one 
another.1 M. Deschamps writes, " Chaque village est forme 
par une ou deux huttes pour une ou deux families, mais le 
plus generalement une famille s'etablit seule."2 To this 
day, says Mr. Nevill, " if one wishes to approach a Vaedda's 
home, one waits nearly a quarter mile off it, and shouts until 
the dogs bark."3 
In his history of Mindanao and Sulu, published in 1667, 
Father Francisco Combes says that the Subanu, a sub- 
Visayan mountain folk of Mindanao, lack human inter- 
course, " living, as they do, in high, wild country, with as 
little sociability as animals, and having their houses placed 
a league apart, wherever one of them may be pleased to 
make himself a settlement." Nor have they in later times 
congregated into villages. " The family," say Messrs. Finley 
and Churchill, " is the governmental unit. The father is the 
head of the family and its absolute ruler. ... As soon as 
the young men take to themselves wives they break away 
from the old family home and establish new family units 
at remote-points, where they can enjoy all the freedom of 
their peculiar nomadic life." They have associations or 
confederations of families forming communities, each under 
the leadership of a timuai, or chief. But " family rights are 
supreme and therefore the right of secession from the 
community inheres in the head of the family." The Subanu 
draws his sustenance from the earth by primitive agricultural 
methods ; "he seeks the isolated and wildest portions of the 
interior and relies upon his strength and native ingenuity 
to cope with nature and wring from it a means of living for 
himself and his family."4 The Negritos of Northern Luzon, 
most of whom do not practise agriculture at all, are said 
to be " usually scattered here and there through the forest, 
although occasionally a group of one or two dozen will be 
1 Sarasin, Ergebnisse naturwissenschaftlichcv Forschungen auf 
Ceylon, iii. 476, 477, 481, 485. 
2 Deschamps, * Les Veddas de Ceylan,' in L' Anthropologic, ii. 305, 
3 Nevill, ' Vaeddas of Ceylon,' in Taprobanian, i. 186. 
4 Finley and Churchill, Stibanu. Studies of a Sub-Visayan Mountain 
Folk of Mindanao, pp. 12, 15, 24 sq. i THE ORIGIN OF MARRIAGE 
61 
found together " ;1 or to live in small hordes of twenty to 
thirty persons, who recognise no other ties but those* of 
the family.2 The Kubus of Southern Sumatra, who chiefly 
subsist on game and fish and the products of the forest 
though they also practise a kind of primitive agriculture, 
are said by Volz to live, as a rule, in settlements of one or 
two families each, which are spread over a very wide area. 
They seldom come together in small hordes.3 
Among the Orang Muka Kuning, of the Orang Laut in 
Malacca, who neither cultivate any plants nor breed any 
animals save dogs, the families " live scattered in the forest 
in small huts beneath the trees."4 Of the Orang Semang 
Vaughan Stevens says that they are true nomads who" never 
remain long at the same place, and that even the families 
among them dissolve, to reunite again afterwards. Dr. 
Martin quotes this with approval,5 and makes the general 
statement that the pure tribes of the Malay Peninsula 
" always live in small groups- only, consisting of one to at 
most six families. . . . These family groups form the 
separate hordes, which are in the habit of nomadising 
together but also on occasion separate from one another." 
Every family erects a shelter or hut for itself ; at their 
halting-places more than two or three huts are seldom seen 
together, each inhabited by two to seven individuals. 
Thus " the separate family (Sonderfamilie) forms every- 
where the fundamental element, which grows into a family 
group (Grossfamilie) " with the eldest man as its regular 
chief. The connection between the different family groups 
is exceedingly loose, and there is no such organisation as 
a phratry or a tribe.6 
1 Worcester, ' Non-Christiau Tribes of Northern Luzon,' in 
Philippine Journal of Science, i. 808 sq. 
2 Blumentritt, Versuch einer Ethnographic der Philippinen, p. 8. 
Meyer} Die Philippinen. II. Negritos, p. 15. 
3 Volz, ' Zur Kenntniss der Kubus in Siidsumatra,' in Archiv J. 
Anthrop. N.S. vii, 98, 100, 107. 
4 Logan, ' Orang Muka Kuning,' in Jour. Indian Archipelago, 
i. 337 n. * 
5 Martin, Die Inlandstdmme der Malayischen lialbinscl, p. 861. 
0 Ibid. p. 859 sqq. 62 
THE HISTORY OF HUMAN MARRIAGE chap. 
The Bushmen of South Africa, according to Mr. McCall 
Theal, " lived in little communities, often consisting of only 
a few families. It was impossible for a large number, such 
as would constitute an important tribe, to gain a subsistence 
solely from the chase and the natural products of the earth 
in any part of South Africa at any time, and more especially 
after the Hottentots and the Bantu had taken possession of 
the choicest sections."1 Fritsch observes that they are almost 
entirely devoid of a tribal organisation, and that even when 
a number of families occasionally unite in a larger horde the 
association is more or less accidental and not regulated by 
any law.2 The families that thus associate are now and 
then obliged to disperse, as the same spot will not afford 
sufficient sustenance for all; " the smaller the number, the 
easier is a supply of food procured."3 Indeed, a horde fre- 
quently consists of the different members of one family 
only, at least if the children are old and strong enough to 
help their parents to find food.4 There may be a chief in 
the horde, but his power is hardly greater than that of any 
other man who rules over his own wife and children-and 
over the latter only before they are grown-up.5 Of the Auin, 
who belong to the Kalahari Bushmen, Kaufmann states 
that the size of the settlement depends on the circumstances : 
during the dry season it is mostly formed by one or two 
families only, whereas towards the end of and shortly after 
the rainy period, when food is more plentiful, even as many 
as thirty families may join into one " village." The dwellings 
are of the most primitive character.6 Among the Tati 
Bushmen-who inhabit the Bechuanaland Protectorate, 
the Kalahari, and the portions of Southern Rhodesia 
1 Theal, History of the Boers in South Africa, p. 17.- 
2 Fritsch, Die Eivgeborenen Slid-Afrika's, p. 443 sq. 
3 Lichtenstein, Travels in Southern Africa, ii. 49, 194. 
4 Ibid. i. 48. Thulie, ' Instructions sur les Bochimans,' in Bull. 
Soo. d'Anthr. Paris, ser. iii. vol. iv. 409 sq. 
Schinz, Deutsch-Sud-West-Afrika, p. 396. v. Francois, Nama 
und Damara Deutsch-Sud-West-Afrika, p. 234. Passarge, ' Die 
Buschmanner der Kalahari,' in Mittheil. Deutsch. Schutzgeb. xviii. 
213, 271 sqq. Kaufmann, ' Die Auin,' ibid, xxiii. 154 sq. 
0 Kaufmann, loc. cit. pp. 136, 138. 1 THE ORIGIN OF MARRIAGE 
63 
adjoining these territories-a few " clans or families may 
combine in the face of danger, but the combination soon 
conies to an end as soon as the danger is past. They nkver 
seem to feel the need of unity. Each family goes its own way, 
and the father is a despot as long as he can maintain his 
position." Our informant has " seldom seen more than 
four families numbering twenty-two individuals together, 
and this was a camp at permanent water."1 From all that 
has been said it is obvious that among the Bushmen, also, 
the family consisting of parents and children is the funda- 
mental and, so far as living together is concerned, among 
many of them at least, the only permanent social unit.2 
The Central African Pygmies live in communities of 
variable size. Hutereau speaks of groups of two to twenty 
families, subject to the rule of a family chief ; but " parfois 
plusieurs groupes de families se reunissent pour se deplacer 
et pour chasser ensemble."3 Stuhlmann found " villages," 
consisting of two to four huts and others consisting of one to 
two hundred.4 Each hut seems to be occupied by one family 
only, and the people forming a community are probably, 
as a rule, branches of the same parent family.5 David says 
that the Wambutti of Ituri live in patriarchal family groups.6 
According to Casati, the huts serving as family dwellings 
" are usually scattered in the forests, or over the hills," 
and seldom form a village ; but he also found a good many 
families living without any shelter at all on the side of a 
stream or in the thickets of the forest.7 If the communities 
1 Dornan, ' Tati Bushmen (Masarwas) and their Language,' 
in Jour. Roy. Anthr. Inst, xlvii. 47, 53. 
2 See also Passarge, loc. cit. p. 213 sq. ; Trenk, ' Die Buschleute 
der Namib, ihre Rechts- und Familienverhaltnisse,' in Mittheil. 
Deutsch. Schutzgeb. xxiii. 166, 168 ; Moffat, Missionary Labours 
and Scenes in Southern Africa, p. 56. 
3 Hutereau, Notes sur la Vie familiale et juridique de quelqucs 
populations du Congo Beige, p. 1. 
4 Stuhlmann, Mit Emin Pascha ins Herz von Afrika, p. 449. 
6 Schmidt, Die Stellung der Pygmaenvblker in der Entwicklungs- 
geschichte des Menschen, pp. 61, 62, 187, 192. 
6 David, ' Notizen fiber die Pygmaen des Ituriwaldes,' in Globus, 
Ixxxvi. 196. 
7 Casati, Ten Years in Equatoria, i. 157 sq. 64 
THE HISTORY OF HUMAN MARRIAGE chap. 
of the Central African Pygmies arc more permanent than 
those of many other peoples who have now come under 
our notice, the reason for it is no doubt their richer food 
supply. They are skilful bowmen and exceptionally expert 
as hunters j1 and while'they do not themselves cultivate the 
soil, they get from their neighbours com, bulbs and other 
vegetable food in exchange for products of the chase, or, 
more simply still, make raids on their fields and carry off 
what they want.2 Junker says of the Akkas that " they are 
feared by all their neighbours, and despite their thievish 
habits, permitted to frequent the cultivated grounds." 3 
Among the Australian aborigines, who also are a race of 
hunters find food-collectors without agriculture and cattle- 
rearing, we find a much more definite social organisation 
than among any other people in the same state of economic 
culture. But what interests us in this connection is not 
their social organisation in general, but their actual manner 
of living. They live in hordes ; find here again we find that 
the size of the horde is regulated by the food supply. It 
has been remarked that where the country is sterile and un- 
productive the natives congregate in small numbers, whereas 
in fertile districts they are comparatively numerous.4 The 
horde or tribe has always a tendency to break up for a 
time in search of food. In his book on the aborigines of 
Victoria, Mr. Brough Smyth remarks that " in any large 
area occupied by a tribe, where there was not much forest 
land, and where kangaroos were not numerous, it is highly 
probable that the several families composing the tribe would 
withdraw from their companions for short periods, at certain 
seasons, and betake themselves to separate portions of the 
area, . . . and it is more than probable-it is almost 
certain- that each head of a family would betake himself, 
1 Junker, Travels in Africa during the Years 1882-1886, p. 85. 
Emin Pasha in Central Africa, p. 316. 
2 v. Wissmann, Wolf, v. Francois, and Mueller, Im Innern Afrikas, 
p. 261. Emin Pasha in Central Africa, p. 316. Casati, op. cit. i. 159. 
3 Junker, op. cit. p. 85. 
* Wilhelmi, ' Manners and Customs of the Australian Natives,' 
111 Trans. Roy. Soc. Victoria, v. 165. i THE ORIGIN OF MARRIAGE 
65 
if practicable, to that portion which his father had fre- 
quented."1 Among the natives of Victoria and Riverina, 
according to Beveridge, " a family, or perhaps several 
families, as the case may be, select a site for their camp, 
where abundance of game and other sources of food exist, 
and are procurable with the least expenditure of time and 
trouble."2 Concerning the Encounter Bay tribe in South 
Australia, the Rev. A. Meyer tells us that " the whole tribe 
does not always move in a body from one place to another, 
unless there should be abundance of food to be obtained at 
some particular spot ; but generally they are scattered in 
search of food."3 The natives of Port Jackson, in New South 
Wales, when visited by Captain Hunter at the end of the 
eighteenth century, were associated in tribes of many families 
living together, apparently with one fixed residence. Yet, 
he says, " you may often visit the place where the tribe 
resides, without finding the whole society there ; their time 
is so much occupied in search of food, that the different 
families take different routes ; but, in case of any dispute 
with a neighbouring tribe, they can soon be assembled."4 
Speaking of the six tribes living on the immediate banks 
of the MacLeay River near Port Macquarie in New South 
Wales, Hodgkinson observes that the whole body of a tribe, 
containing on an average from eighty to a hundred men and 
women, exclusive of children, " is never united on the same 
spot, unless on some important occasion. . . . They are 
more generally divided into small parties of eight or ten men. 
with their women and children, for the convenience of hunting, 
&c., and these detached companies roam over any part 
of the country within the prescribed limits of the main 
tribe to which they belong."5 
1 Brough Smyth, Aborigines of Victoria, i. 146 sq. 
2 Beveridge, Aborigines of Victoria and Riverina, p. 32. 
3 Meyer, ' Manners and Customs of the Aborigines of the En- 
counter Bay Tribe,' in Woods, Native Tribes of South Australia, 
p. 191. 
4 Hunter, Historical Journal of the Transactions at Port Jackson 
and Norfolk Island, p. 62. 
5 Hodgkinson, Australia, from Port Macquarie to Moreton Bay, 
p. 222. 66 
THE HISTORY OF HUMAN MARRIAGE chap. 
Whether assembled or scattered, the families belonging 
to the same community form distinct and very marked social 
units among the Australian natives.1 Mr. Stanbridge, who 
spent eighteen years in the wilds of Victoria, states that the 
land of each tribe is parcelled out amongst families and 
carefully transmitted by direct descent; and so sacredly 
are these boundaries maintained that the member of no 
single family will venture on the lands of a neighbouring 
one without invitation.2 Mr. Curr remarks that among the 
Bangerang in Victoria, and all other tribes he has known, 
each married couple had their own nia-nia or hut, although 
unmarried men and boys of eight or ten years and upwards 
lived together, and separate from their parents and sisters, 
in the bachelors' camp.3 Concerning the Gournditch-mara, 
Dr. Howitt says that " each family camped by itself." 4 
Among the Central Australian Arunta, who are distributed 
in a large number of small local groups each of which 
occupies a given area of country and has its own headman, 
every family, consisting of a man and one or more wives and 
children, has a separate lean-to of shrubs.6 There are 
similar statements 1 elating to many other tribes.6 Of the 
Kabi and Wakka tribes in Queensland, Mr. Mathew states 
that " the family, consisting of husband and wife, or wives, 
with their children, constituted a distinct social unit. They 
occupied the same gunyah (dwelling), they ate together, they 
travelled together."7 Bishop Salvado writes of West 
Australian natives, among whom he spent most part of bis 
life, that they, " au lieu de se gouvemer par tribus, paraissent 
1 See Malinowski, The Family among the Australian Aborigines, 
passim. 
2 Stanbridge, ' Tribes in the Central Part of-Victoria,' in Trans. 
Ethn. Soc. N.S. i. 286 sq. 
3 Curr, Recollections of Squatting in Victoria, pp. 259, 277. 
4 Fison and Howitt, Kamilaroi and Kumai, p. 278. 
6 Spencer and Gillen, Native Tribes of Central Australia, pp. 8,10,18. 
G Westgarth, Australia Felix, p. 87 (natives of New South Wales). 
Henderson, Excursions and Adventures in New South Wales, ii. 109. 
Mathews, Ethnological Notes on the Aboriginal Tribes of N. S. 
Wales, p. 99. 
7 Mathew, Two Representative Tribes of Queensland, p. 153. i THE ORIGIN OF MARRIAGE 
67 
se gouvcrner a la maniere patriarcale : chaque famille, qui 
generalement ne compte pas plus de six a neuf individus, 
forme comme une petite societe, sous la seule dependance 
de son propre chef. . . . Chaque famille s'approprie une 
espece de district, dont cependant les families voisines 
iouissent en commun si 1'on vit en bonne harmonie."1 
It seems extremely probable that Australian blacks com- 
monly are so much more sociable than most other hunting 
and food-collecting peoples because the food-supply of their 
country is naturally more plentiful, or, partly thanks to 
their boomerangs, more easily attainable. Speaking of 
West Australian natives, Calvert observes that they, as a 
rule, have an abundance of food, " although they may run 
a little short in the height of the rainy season, or when they 
are overcome with laziness in very hot weather."2 A Central 
Australian native is, generally speaking, well nourished ; 
" kangaroo, rock-wallabies, emus, and other forms of game 
are not scarce, and often fall a prey to his spear and boomer- 
ang, while smaller animals, such as rats and lizards, are 
constantly caught without any difficulty by the women."3 
Yet, as we have seen, separate families are often compelled 
to walk about in search of food, and the fathci has on these 
occasions an essential function to fulfil both as protector 
and maintainer of his family. 
Our knowledge of the Tasmanians, now an extinct race 
which was neither agricultural nor pastoral, is very defective. 
Captain Furneaux, who accompanied Cook on his second 
voyage as commander of the Adventure, wrote that they 
wandered about in small parties from place to place in search 
of food, and that more than three or four huts, only capable 
of containing three or four persons each, were never found 
in the same place.4 O'Connor says that they travelled in 
parties of ten to thirty,6 and Mrs. Prinsep that they moved 
1 Salvado, Memoires historiques sur V Australis, p. 265 sq. Idem, 
Voyage en Australie, p. 178. 
8 Calvert, Aborigines of Western Australia, p. 24. 
3 Spencer and Gillen, op. cit. pp. 7, 44. 
4 Cook, Voyage towards the South Pole, i. 114. 
6 O'Connor, quoted by Ling Roth, Aborigines of Tasmania, p. 105, 68 
THE HISTORY OF HUMAN MARRIAGE chap. 
" in large bodies, with incredible swiftness."1 Each tribe 
was divided into several families, consisting of a few in- 
dividuals, and wherever they located themselves each 
family kindled its separate fire " at fourteen to twenty yards 
apart." 2 It also " hunted separately, and erected a hut for 
its own accommodation."3 
From this survey of facts it appears that among modern 
savages living in the hunting and food-collecting stage, 
or at most acquainted with some primitive mode of agri- 
culture, the family consisting of parents and children is a 
very well-marked social unit. Yet it is certainly not the only 
association among them. If travellers speak of the family 
tie as the only one which joins individuals with one another, 
they without doubt use the term " family " in a wider sense. 
Closely related families not only hold friendly relations with 
each other, but live together in smaller or largei groups ; 
and there may be social organisations of a more comprehen- 
sive character, as among the Australian natives. At the 
same time, it is repeatedly stated that the families belonging 
to the same group do not always keep together, but often 
disperse in search of food and may remain separated even 
for a considerable time ; and this is the case not only in 
desolate regions where the supply of food is unusually scarce, 
but even in countries highly favoured by nature. Now I ask : 
is it reasonable to suppose that primitive men were more 
permanently gregarious than many modern savages ? The 
answer must be : they were undoubtedly less. Let us 
remember that all savages now existing stand high above 
our earliest human ancestors. Man, as we know him, has- 
to quote Darwin-" invented and is able to use various 
weapons, tools, traps, &c., with which he defends himself, 
kills or catches prey, and otherwise obtains food. He has 
made rafts or canoes for fishing or crossing over to neigh- 
1 Mrs. Prinsep, Journal of a Voyage from Calcutta to Van Diemen's 
Land, p. 78. 
2 Backhouse, Narrative of a Visit to the Australian Colonies, 
p. 104. Lloyd, quoted by Ling Roth, op. cit. p. 107. 
3 West, History of Tasmania, ii. 82. I THE ORIGIN OF MARRIAGE 
69 
bouring fertile islands. He has discovered the art of making 
fire, by which hard and stringy roots can be rendered 
digestible, and poisonous roots or herbs innocuous."1 In 
short, he has gradually found out many new ways of earning 
his living which his ruder ancestors had no idea of, and has 
thus more and more emancipated himself from direct 
dependence on surrounding nature. If this notwithstanding 
there are still so many savage peoples among whom the 
separate families often are compelled to give up the pro- 
tection afforded them by living together, in order to find the 
food necessary for their subsistence, I. think we have 
reason to believe that the family, implying marital and 
paternal care, was haidly less indispensable for primitive 
man than it is for the gorilla and chimpanzee. If this was 
the case, the family may have been an inheritance from the 
parent species out of which the Anthropoids and the 
Hominides-the Pithecanthropus, Homo primigenius, and 
Homo sapiens, according to Schwalbe2-gradually developed. 
This may be supposed to have been the case if that hypo- 
thetical species lived on the same diet as the man-like apes, 
or even on a diet somewhat more animal than that of the 
chimpanzee, and also, being of a comparatively large size, 
required about the same quantities of food as they ; if, 
further, it gave birth to the same small number of young ; 
and if its offspring were in need of parental care for a com- 
paratively long period. I want to emphasise that it is on 
such factors, and not merely on the habits of the gorilla and 
the chimpanzee as they are, that I base my supposition that 
the family consisting of parents and children existed among 
primitive man. 
I have so far spoken of habits, not of institutions. But 
there is an intimate connection between them. Social 
habits have a strong tendency to become true customs, that 
is, rules of conduct in addition to their being habits. A 
habit may develop into a genuine custom simply because 
1 Darwin, Descent of Man, i. 72. 
2 Schwalbe, Studien zur Vorgeschichte des Menschen, p. 5. 70 
THE HISTORX OF HUMAN MARRIAGE chap. 
men are inclined to disapprove of anything which is unusual.1 
But in the present case the transition from habit to custom 
has undoubtedly a deeper foundation. It is an instinct 
that induces the male to remain with the female and to 
take care of her even after the sexual relations have ceased. 
We may assume that the tendency to feel some attachment 
to a being which has been the cause of pleasure, in the present 
case sexual pleasure, is at the bottom of this instinct. Such 
a feeling may originally have caused the sexes to remain 
united and the male to protect the female though the sexual 
desire was gratified ; and if procuring great advantage to 
the species in the struggle for existence, conjugal attachment 
would naturally develop into a specific characteristic. 
An instinct must also be the cause of the care which the 
father takes of his offspring ; the paternal feeling seems to be 
hardly less prevalent among savages than among civilised 
men.2 This feeling, however, and also the kindred feeling 
of maternal love, are not quite adequately defined as the 
affection which attaches a creature to its young. Though 
most frequently and most strongly displayed in this relation, 
the so-called parental feeling is really excitable apart from 
parenthood, as Spencer justly observed. According to him. 
the common trait of the objects which arouse it is always 
relative weakness or helplessness.3 But this explanation 
contains only part of the truth ; even in a gregarious species 
mothers make a distinction between their own offspring and 
other young. To account for the maternal sentiment we 
must therefore assume the existence of some other stimulus 
besides the signs of helplessness, which produces, or at least 
strengthens, the instinctive motor response in the mother. 
This stimulus, so far as I can see, is rooted in the external 
relationship in which the helpless offspring from the begin- 
ning stand to the mother, being in close proximity to her 
from their tenderest age. And the stimuli to which the 
1 See my Origin and Development of the Moral Ideas, i. 159. 
2 Ibid. i. 529 sqq. 
3 Spencer, Principles of Psychology, ii. 623 sq. See also Hartley, 
Observations on Man, i. 497. THE ORIGIN OF MARRIAGE 
71 
paternal instinct responds are apparently derived from the 
same circumstances as those which call into activity the 
maternal instinct, that is, the helplessness and the nearness 
of the young. Wherever this instinct exists the father is 
near his offspring from the beginning, living together with 
the mother. Of course I here speak of the parental feelings 
only in their original simplicity ; later on they become more 
complex, through the association of other feelings, as those 
of property and pride, and tend to extend themselves 
beyond the limits of infancy and childhood.1 
In mankind these instincts give rise not only to habits 
but to rules of custom, or institutions. Social beings 
endowed with such instincts, as also with a sufficiently 
developed intellect, would feel moral resentment against a 
man who forsakes the woman with whom he has conjugal 
intercourse and the offspring resulting from it. And, as I 
have pointed out in another work, public or moral resentment 
or disapproval is at the bottom of the rules of custom and 
of all duties and rights.2 Thus the institutions of marriage 
and the family have the same root as the habits with which 
I have been dealing in this chapter ; indeed these institutions 
and habits are practically identical, except that in the one 
case there is social sanction or regulation and in the other 
not. Now as the word " family " is not merely used to 
denote a certain institution, I think we may be allowed to 
apply the term " marriage " also in a broader sense than that 
given to it above. We may alternatively define it as a more 
or less durable connection between male and female lasting 
beyond the mere act of propagation till after the birth of the 
offspring. This definition lays stress on the highly important 
fact, only too often overlooked by sociologists, that there is 
a vital difference between marriage and merely sexual 
relations, even though these be sanctioned by custom. 
It also implies living together-in agreement with the 
mediaeval saying, " Boire, manger, coucher ensemble est 
1 See Westermarck, op. cit. ii. 188-191, 193. 
* See ibid. i. 118-122, 135-137, 139 sqq. For the characteristics 
and origin of moral disapproval see ibid. vol. i. ch. ii. p. 21 sqq. 72 
THE HISTORY OF HUMAN MARRIAGE chap. 
mariage, ce me semble."1 And though rather vague, which 
is a matter of course, it has the advantage of comprehending 
in one notion facts which are essentially similar in nature 
and have a similar origin. The marriage of mankind, as 
we have seen, is not an isolated phenomenon, but has its 
counterpart in many other animal species and is probably 
an inheritance from some pre-human ancestor. It is in 
order to emphasise this that I have called my book " the 
history of human marriage," although I shall throughout 
deal with unions which are, or may be supposed to be, 
sanctioned by custom or law. 
From wTiat has been said above it appears that marriage 
and the family are most intimately connected with one 
another : it is originally for the benefit of the young that 
male and female continue to live together. We may there- 
fore say that marriage is rooted in the family rather than the 
family in marriage. Indeed, among many peoples true 
married life does not begin for persons who are formally 
married or betrothed, or a marriage does not become definite, 
until a child is bom or there are signs of pregnancy ; whilst 
in other cases sexual relations which happen to lead to 
pregnancy or the birth of a child are, as a rule, followed by 
marriage or make marriage compulsory. 
Thus among the Fuegians2 and the Eastern Greenlanders3 
marriage is not regarded as complete until the woman has 
become a mother. Among the Lengua Indians of the 
Paraguayan Chaco " there is only a marriage on approval, 
corresponding to our engagement, although the couple live 
together. No marriage is considered binding by native law 
until a child is born, and if this does not take place within a 
1 Schaffner, Geschichte der Rechtsverfasswng Frankreichs, iii. 186. 
Among the Maori, according to Mr. Tregear (The Maori Race, 
p. 293), marriage was merely alluded to as a " dwelling-together," 
or a " sleeping-together." The Timagami (Ojibway) term for 
husband and wife, which is used reciprocally, means " the one who 
lives with me " (Speck, Family Hunting Territories and Social Life 
of Various Algonkian Bands of the Ottawa Valley, p. 24). 
2 Hyades and Deniker, Mission scientifique du Cap Horn, vii. 
377 sq. 3 * East Greenland Eskimo/ in Science, vii. 172. i THE ORIGIN OF MARRIAGE 
73 
reasonable time they are justified in separating. But when 
once a child is bom to them, even should the child die or be 
put to death, they are considered to be bound to each other 
for life."1 Among the Bororo, of Central Brazil, a man after 
his marriage stays in the house of the bride until he has a 
family of his own, when he builds a house for himself.2 
In some Canadian tribes a married man was obliged to go 
to his father-in-law's house to find his wife when he had a 
mind to her company, until she brought forth a child ; then 
only she went to live with her husband.3 Among the Aleut 
the wife stays at her father's home for a certain time or 
until the birth of a child ; the husband is at liberty to visit 
her, but not to remove her to his own village until the ex- 
piration of the customary period, unless a child be born 
meanwhile.4 Among the Atkha Aleut a husband does not 
pay the purchase sum before he has become a father.5 
Of the Tsalisens of Formosa we are told that, " after 
matters have been definitely arranged, a month is allowed 
to intervene, and then on an appointed day the suitor visits 
the house of his intended and a simple ceremony sanctions 
the right of the couple to come together. The woman remains 
at the home of her mother until a child is bom, when she 
removes to the home of her husband, and the marriage is 
then considered to have been effected. Should she be 
without issue, however, her suitor ceases to call, and all 
familiarity between the couple comes to an end. Both 
parties are now free to seek a mate elsewhere."6 Among the 
Sea Dyaks of Borneo " intercourse often takes place between 
those who have been betrothed, but not formally married, 
simply to ascertain if the marriage will be fruitful. At the first 
signs of the desired result themarriage ceremony takes place."7 
1 Grubb, An Unknown People in an Unknown Land, p. 214. 
2 Eric and Radin, in Jour. Anthr. Inst, xxxvi. 390. 
3 Lahontan, New Voyages to North-America, p. 457. 
4 Veniaminov, trans.by Golder, in Jour. American Folk-Lore, xx. 134. 
5 Erman, * Ethnographische Wahrnehmungen an den Kusten 
des Berings-Meeres,' in Zeitschr. f. Ethnol. iii. 162. 
6 Davidson, op. cit. p. 573. 
7 Gomes, op. cit. p. 127. Brooke Low, quoted by Ling Roth, 
Natives of Sarawak, i. 115. 74 
THE HISTORY OF HUMAN MARRIAGE 
CHAP. 
Among the aborigines of North-West-Central Queensland 
a wife who happens to be but of kanari or kati-kati maro 
rank at the time of marriage neither prepares nor cooks her 
husband's food, and it is only after the birth of her first child 
that she remains at her husband's apartment permanently.1 
Tn some other Australian tribes a man may, at least in 
certain circumstances, retain as his wife a girl with whom he 
has eloped, if they stay away until a child is born.2 
Among the Badagas of the Nilgiris in Southern India the 
marriage bond is not really sealed until the fifth month of 
the first pregnancy, when the relatives are invited to be 
present at the ceremony of tying the marriage emblem round 
the neck of the woman.3 Among the Mezeyne tribe of 
Bedouins within the limits of the Sinai peninsula the bride 
runs away and is caught by the bridegroom, after which she 
flees back to her home and does not enter her husband's tent 
until she becomes far advanced in pregnancy.4 
Concerning the Wolofs in Senegambia, Berenger-Feraud 
writes, " Ce n'est que lorsque les signes de la grossesse sont 
irrecusables chez la fiancee, quelquefois meme ce n'est 
qu'apres la naissance d'un ou plusieurs enfants, que la 
ceremonie du mariage proprement dit s'accomplit."6 Among 
the Siena in the French Sudan " la famille n'est jugee 
reellement existante que lorsqu'elle comprend des enfants."6 
Among the Southern Bambala in Congo "the marriage seems 
to become definite only when a child is expected, for then 
conjugal fidelity becomes obligatory on both parties, as the 
child is otherwise supposed to die, and deaths of infants are 
generally attributed to this cause."7 Dr. Felkin states that 
1 Roth, op. cit. p. 180 sq. 
2 Howitt, Native Tribes of South-East Australia, pp. 259, 263. 
3 Thurston, ' Badagas of the Nilgiris,' in the Madras Government 
Museum's Bulletin, ii. 3 sq. See also Harkness, Neilgherry Hills, 
p. 116. 
4 Burckhardt, Notes on the Bedouins and Wahdbys, p. 153. 
5 Berenger-Fdraud, ' Le mariage chez les negres S6n6gambiens, ' 
in Revue d'anthropologic, 1883, p. 286 sq. 
0 Delafosse, ' Le peuple Si6na ou Senoufo,' in Revue des etudes 
ethnographiques et sociologiques, i. 483. 
7 Torday, Camp and Tramp in African Wilds, p. 203. i THE ORIGIN OF MARRIAGE 
75 
among the For tribe of Central Africa the husband lives 
with his wife at his father-in-law's place until his first child 
is born, when he is permitted to take her away and set up 
housekeeping on his own account ; and during the whole 
of this time the father-in-law has to pay all housekeeping 
expenses for the young couple, the husband being entitled 
to three meals during each night.1 Among various other 
peoples the husband takes up his abode with his father-in- 
law and never removes his wife till aftei the birth of a child.2 
Among the Khasis in Assam the husband goes to live in his 
mother-in-law's house, but after one or two children are 
born and if the married couple get on well together, he 
frequently takes away his wife and family to a house of his 
own.3 Among some of the Old Kuki clans a young man has 
to serve his future w'ife's father for three years ; during 
this time he has free access to the girl, and should she become 
enceinte the marriage ceremony must be performed and the 
price paid.4 
In many countries, including various parts of Europe, the 
occurrence of pregnancy or child-birth is a usual preliminary 
to, or as a general rule leads to, marriage.5 We are often 
1 Felkin, ' Notes on the For Tribe of Central Africa,' in Proceed. 
Roy. Soc. Edinburgh, xiii. 229. 
2 E.g., the Omaha (James, Account of an Expedition from Pittsburgh 
to the Rocky Mountains, i. 242 sq.), the Ainu of Yesso (v. Siebold, 
Die Aino auf Yesso, p. 31), one of the aboriginal tribes of China 
(Gray, China, ii. 304), the Khyens (Rowney, Wild Tribes of India, 
p. 203 sqf, the Wasania of British East Africa (Barrett, in four, 
Roy. Anthr. Inst. xli. 30 sqf, some of the Dinka (Cummins, ' Sub- 
tribes of the Bahr-el-Ghazal Dinkas,' in four. Anthr. Inst, xxxiv. 
151)- 
3 Gurdon, op. cit. p. 76. 
4 Shakespear, Lushei Kuki Clans, p. 154. 
5 Cardus, op. cit. p. 71 (Guarayos of Bolivia). Gait, Census of 
India, 1911, vol i. (India) Report, p. 243 (aboriginal tribes). 
Hutchinson, Account of the Chittagong Hill Tracts, p. 23. Kloss, 
In the Andamans and Nicobars, p. 188 (Andamanese). Strzoda, 
' Die Li auf Hainan,' in Zeitschr. f. Ethnol. xliii. 203 (Chinese 
tribes). Jenks, Bontoc Igorot, p. 66. Martin, Reisen in den Molukken, 
p 63 (natives of Amboyna and Uliase). Pfeil, Studien und Beobach- 
tungen aus der Sudsee, p. 31 (natives of the Bismarck Archipelago). 76 
THE HISTORY OF HUMAN MARRIAGE chap. 
told that a seducer or lover is compelled to marry the girl 
if she becomes with child ;x but he may also have the alter- 
native of paying a fine.2 In many of the wild tribes of 
Borneo, for instance, there is almost unrestricted inter- 
course between the young people of both sexes, but if 
pregnancy ensues marriage is considered necessary.3 Dr. 
A. Bunker told me the same of certain Karen tribes in 
Burma. Among the Wanyamwezi, between Zanzibar and 
Tanganyika, a man who renders a girl enceinte and does 
not marry her before the child is bom " is bound to pay for 
the woman and also for the child about three times the 
value of the ordinary dowry."4 In Tahiti, according to 
Cook, the father might kill his natural child, but if he 
suffered it to live the parties were considered to be in the 
married state.6 Dr. Rivers states that in Tikopia, " when 
the illicit intercourse of a youth and girl of the ordinary 
people results in offspring, the pair usually marry and if 
1 Lumholtz, Unknown Mexico, i. 466 (Tepehuane). Endle, 
Kacharis, p. 31. Thurston, Castes and Tribes of Southern India, 
i. 248 (Billavas of the South Canara district); vii. 348 (Velans of the 
Cochin State). Anantha Krishna Iyer, op. cit. i. 60 sq. (Ulladans 
of the Cochin State). Emin Pasha in Central Africa, p. 103 (Madi 
and some other Central African peoples). Cunningham, Uganda 
and its Peoples, p. 140 (Bavuma). Tremearne, ' Notes on the 
Kagoro and other Nigerian Head-Hunters,' in Jour. Roy. Anthr. 
Inst. xlii. 169. Among the Basoga (Roscoe, Northern Bantu, p. 233) 
and Thonga (Junod, Life of a South African Tribe, ii. 97) the man is 
asked to marry the girl. 
2 Geis, quoted by Rose and Brown, ' Lisu (Yawyin) Tribes of the 
Burma-China Frontier,' in Memoirs Asiatic Soc. Bengal, iii. 263. 
Taupin, reviewed in L' Anthropologic, ii. 488 (Laosians). Hutchinson, 
op. cit. p. 23 (Chittagong Hill tribes). Soppitt, Kuki-Lushai Tribes 
on the North-East Frontier, p. 15 (Nagas and Kukis). Baumann, 
Durch Massailand, p. 161. Felkin, loc. cit. p. 208 (Fors). 
3 St. John, ' Wild Tribes of the North-West Coast of Borneo,' in 
Trans. Ethn. Soc. N.S. ii. 237. Low, Sarawak, p. 195. 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. iv. 442 
Hose and McDougall, Pagan Tribes of Borneo, ii. 170 (Kayans), 
183 (Punans). 
4 Decle, Three Years in Savage Africa, p. 348. 
s Cook, Voyage to the Pacific Ocean, ii. 157. I THE ORIGIN OF MARRIAGE 
77 
they do so there would be no slur on the child. Should the 
man refuse to marry in such a case, the child would be 
killed as soon as born."1 
To the hypothesis set forth in this chapter as regards the 
origin of human marriage the objection will perhaps be 
raised that I have overlooked one most important difference 
which exists among man and the lower animals. " That 
which distinguishes man from the beast," Beaumarchais 
says, " is drinking without being thirsty and making love 
at all seasons." Where love-making is restricted to a certain 
season only, it cannot, of course, be the sexual impulse that 
keeps male and female together till after the birth of the 
offspring ; but the case is different with a species that pairs 
throughout the year. It may be argued, therefore, that in 
mankind the prolonged union between the sexes has 
originated, not in any specific instinct of the kind suggested 
above, but simply in a protracted tendency to procreation. 
Although this argument would still leave the father's 
relation to the offspring unexplained, it cannot be ignored. 
The permanency or periodicity of the sexual life must 
certainly affect the relations between the sexes. But when 
the question is whether this factor has played a part in the 
origin of human marriage, we must first consider the legiti- 
macy of the assumption that our early human or semi- 
human ancestors, like ourselves, made love at all seasons. 
The next chapter will be devoted to this problem. 
1 Rivers, History of Melanesian Society, i. 310. CHAPTER II 
A HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
It has been assumed by some physiologists that the 
periodicity in the sexual life of animals depends upon 
economic conditions, the reproductive matter being a surplus 
of the individual economy. Hence, it is said, their sexual 
season occurs when the proportion between receipts and 
expenditure is most favourable.1 
According to Mr. Heape, on the other hand, the sexual 
season is governed by a variety of influences. It may be 
influenced by the climate of the region in which the animal 
lives, by the seasons of the year when these are of marked 
variation, and by the supply of food, or possibly by the 
nature of the food, obtainable ; by special nervous, vascular, 
and secretory peculiarities of the individual and its habits 
of life ; and by the length of gestation, the claims of the 
newly-born offspring on the mother, and her powers of 
recuperation.2 
There can, of course, be no doubt that the periodicity 
in question is closely connected with certain conditions 
1 Reuckart, ' Zeugung,' in Wagner, Handworterbuch der Physi- 
ologic, iv. 862. Gruenhagen, Lehrbuch der Physiologic, iii. 528. Cf. 
Haycraft, ' Some Physiological Results of Temperature Variations,' 
in Trans. Roy. Soc. Edinburgh, xxix. 130. 
2 Heape, ' The " Sexual Season " of Mammals and the Relation 
of the " Pro-oestrum " to Menstruation,' in Quarterly Jour Micro- 
scot>. Science, N.S. vol. xliv. pt. i. 16 sqq. 
78 ch. 11 HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
79 
prevailing in the particular season when the pairing takes 
place ; but it seems to me equally obvious that the §exual 
functions are, at least to some extent, affected by different 
conditions in different species. This is shown by the fact 
that every month or season of the year is the pairing time 
of one or another species of Mammals.1 The sexual season 
is adapted to the requirements of each species. It is funda- 
mentally governed by the law that the young shall be bom 
at the time which is most favourable for their survival; 
and the influence of seasonal conditions upon the sexual 
functions, and the length of the period of gestation, are 
subordinate to this law. The periodicity in the sexual life 
of animals is the result of natural selection. 
This explains why Reptiles, Birds, and many Mammals 
bring forth their young early in spring, or, in tropical 
countries, at the beginning of the rainy season ; the period 
then commences when life is more easily sustained, when 
prey is most abundant, when there is enough water and 
vegetable food, and when the climate becomes more suitable. 
In highlands animals, generally speaking, pair later than 
in lower regions, and in the polar and temperate zones later 
than in the tropics ; species living in different latitudes have 
their pairing season earlier or later according to the differences 
1 Thus the'bat pairs in January and February (Brehm, Thierleben, 
i. 299) ; the Canis Azarce (Rengger, Naturgeschichte der Sdugethiere 
von Paraguay, p. 147) and the Indian bison (Forsyth, Highlands of 
Central India, p. 108), in winter; the wild-cat and the fox, in 
February (Brehm, op. cit. i. 453, 662) ; the weasel, in March (ibid. 
ii. 84) ; the kulan, from May to July (ibid. iii. 19) ; the musk-ox, 
at the end of August (ibid. iii. 377 ) ; the elk in the Baltic Provinces, 
at the end of August, and in Asiatic Russia, in September or October 
(ibid. iii. in) ; the wild yak in Tibet, in September (Prejevalsky, 
Mongolia, ii. 192) ; the reindeer in Norway, at the end of September 
(Brehm, op. cit. iii. 123) ; the badger, in October (ibid. ii. 149) ; 
the Capra pyrenaica, in November (ibid. iii. 311) ; the chamois, 
musk-deer (ibid. iii. 274, 95) and Antilope Hodgsoni (Prejevalsky, 
op. cit. ii. 205), in November and December ; the wild camel of 
Central Asia, in December, January, and February (Hedin, Central 
Asia and Tibet, i. 357 ; Prejevalsky, From Kulja to Lob-nor, p. 91) ; 
the wolf, from the end of December to the middle of February 
(Brehm, op. cit. i. 534). 80 
THE HISTORY OF HUMAN MARRIAGE chap. 
in climate.1 The dormouse (Muscardinus avellanarius), 
which feeds upon hazel-nuts, pairs in July and brings forth 
its young in August, when nuts begin to ripen ; then the 
young grow very quickly, so that they are able to bear the 
autumn and winter cold.2 While the Adelie penguin chooses 
the warmest and lightest months of the year for the rearing 
of its young, the Emperor penguin-by far the largest of all 
penguins-performs this duty in the darkest, coldest, and 
most tempestuous time. " The only reason," says Dr. 
Murray Levick, " that has been suggested for this custom is 
that many months must pass before the chicks are fully 
fledged. Were they hatched in December (midsummer), as 
are Adelies, autumn would find them still unfledged, and 
probably they would perish in consequence, whereas being 
hatched in the early spring, they are fostered by their parents 
until the warmer weather begins, and then have the entire 
summer in which to accomplish their change of plumage."3 
This shows not only that the sexual season is controlled by 
the needs of the next generation, but also how very different 
the conditions may be under which the reproductive func- 
tions are called into play even in related species. 
Although most of the higher animals breed only once or 
twice a year, there are certain species-as some whales,4 
the elephant,6 many rodents,6 and several of the lower 
monkeys7-that are said to have no definite pairing season. 
As to them, it is perhaps sufficient to quote Brehm's 
statement with regard to elephants, that " the richness of 
their woods is so great that they really never suffer want."8 
But although some monkeys in tropical countries possibly 
1 Brehm, op. cit. i. 370, 404, 431 ; ii. 6, 325, 420 ; iii. in, 158, 
159, 275> 3O2> 578, 599- Prejevalsky, Mongolia, ii. 193, 206. 
2 Brehm, op. cit. ii. 313. 3 Levick, Antarctic Penguins, p. 134 sq. 
4 Brehm, op. cit. iii. 699, 723. 
5 Ibid. iii. 482. 6 Ibid. ii. 440. 
7 Ibid. i. 119, 147, 182, 228. Schomburgk, Reisen in Britisch- 
Guiana, ii. 767. 
8 Brehm, op. cit. iii. 480. The birds on the Galapagos Islands, 
which are situated almost on the equator, likewise seem to have 
no definite sexual season (Markham, ' Visit to the Galapagos Islands,' 
in Proceed. Roy. Geograph. Soc. N.S. ii. 753). ii HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
81 
are in a condition to become pregnant at all times of the 
year, others are certainly not so j1 and to these belong the 
man-like apes. Dr. Mohnike and other writers mention 
the occurrence of a sexual season with the orang-utan,2 
although they do not inform us when it takes place. Mr. 
Wallace, however, told me that he " found the young sucking 
orang-utan in May ; that was about the second or third 
month of the dry season, in which fruits began to be plenti- 
ful." According to Winwood Reade, the male gorillas fight 
at the rutting season for their females.3 Sir Richard Burton 
says, " The gorilla breeds about December, a cool and dry 
month : according to my bushmen, the period of gestation 
is between five and six months."4 
Considering, then, that the sexual season largely depends 
on the kind of food on which the species fives, together 
with other circumstances connected with anatomical and 
physiological peculiarities, and considering further the close 
biological resemblance between man and the man-like apes, 
we have reason to believe that the pairing of our earliest 
human or half-human ancestors also was restricted to a 
certain season of the year. This belief derives additional 
probability from the fact that there are even now some rude 
peoples who are actually stated to have an annual pairing 
time, and other peoples whose sexual desire most decidedly 
seems to undergo a periodical increase at a certain time of 
the year. 
According to Mr. Johnston, the wild Indians of California, 
belonging to the lowest races on earth, " have their rutting 
seasons as regularly as have the deer, the elk, the antelope, 
or any other animals."6 With reference to some of these 
1 Heape, loc. cit. p. 32. Idem, ' Menstruation of Semnopithecus 
entellus,' in Philosophical Transactions, ser. B. vol. clxxxv. pt. i. 412 
sqq. Idem, ' Menstruation and Ovulation of Macacus Rhesus,' ibid. 
ser. B. vol. clxxxviii. 137 sq. Marshall, Physiology of Reproduction, 
p. 63. 
2 Mohnike, ' Die Affen auf den indischen Inseln,' in Das Ausland, 
xlv. 850. Hartmann, Die menschendhnlichen Affen, p. 230. Huxley, 
Man's Place in Nature, p. 33. 3 Reade, Savage Africa, p. 214. 
4 Burton, Trips to Gorilla Land, i. 248. 
5 Schoolcraft, Archives of Aboriginal Knowledge, iv. 224. 
VOL. I G 82 
THE HISTORY OF HUMAN MARRIAGE chap. 
Indians, Mr. Powers says that spring " is a literal Saint 
Valentine's Day with them, as with the natural beasts and 
birds of the forest."1 Of the Californian Hupa we are told 
that their marriages took place at the beginning of the 
summer season and were preceded by a courtship which 
extended through a summer and a winter.2 
Dr. Cook records that the Eskimo inhabiting the country 
lying between the seventy-sixth and seventy-ninth parallels 
exhibit a distinct sexual season, which recurs with great 
intensity at the first appearance of the sun, that little else 
is thought of for some time afterwards, and that the majority 
of the children are in consequence born nine months later.3 
This account is in agreement with statements made by 
Bosquet regarding other Eskimo.4 Among the reindeer- 
breeders of the Northern Yenisei valley the midwinter- 
the only time when they live for any length of time in one 
spot, since hunting and fishing are then impossible-is 
" the season of courtship and matchmaking. With the 
reappearance of the sun comes the wedding."5 
Friedrich M filler says that among the Australian aborigines 
marriage and conception mostly take place during the warm 
season, when there is the greatest abundance of food, and 
that the latter is even to some extent confined to this 
period.6 This statement is, partly at least, based on the 
following remark made by Oldfield in his description of the 
West Australian Watchandies :-" Like the beasts of the 
field, the savage has but one time for copulation in the year. 
About the middle of spring . . . the Watch-an-dies begin 
1 Powers, Tribes of California, p. 206. 
2 Goddard, Life and Culture of the Hupa, p. 55. 
3 Cook, 'Medical Observations among Esquimaux,' in New 
York fournal of Gynecology and Obstetrics, vol. iv., quoted by Heape, 
in Quarterly four. Microscop. Science, N.S. vol. xliv. pt. i. 37 
and Ellis, Studies in the Psychology of Sex, vol. i. Evolution of 
Modesty, &c., p. 126. 
4 Bosquet, ' Note on a Case of Absence of the Uterus,' in Obstetrical 
Transactions, vol. xxvii., quoted by Heape, loc. cit. p. 37. 
5 Miss Czaplicka, My Siberian Year, p. 102 sq. See also Idem, 
Aboriginal Siberia, p. 117. 
0 Muller, Allgemeine Ethnographic, p. 212 sq. ii HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
83 
to think of holding their grand semi-religious festival of 
Caa-ro, preparatory to the performance of the important 
duty of procreation."1 If this were literally correct the 
females would bring forth their offspring at a certain season 
only, but Curr emphatically denies that this is the case 
among the Australian aborigines ;2 and Sir W. B. Spencer 
informs me that Oldfield's statement holds true of none of 
the tribes known to him. On the other hand, there is Mr. 
Caldwell's communication to Mr. Heape, according to which 
those Queensland natives with whom he has been brought in 
contact have a distinct sexual season in September-that 
is, spring-and in consequence cannot be prevailed upon 
to do any work for some weeks at that time of the year.3 
Concerning the Papuans inhabiting the Maclay Coast of 
New Guinea, the Russian traveller Miklucho-Maclay writes:- 
" In the months of July and August I noticed in the Papua 
villages many women in a late stage of pregnancy and most 
of the births took place in September. It seems therefore 
that the procreation is mostly performed in a certain season 
of the year, and it is also during this time-December and 
January-that the Papuas living here are least occupied 
with field-work and celebrate most of their nocturnal feasts, 
which last for several nights in succession."4 Among the 
tribes inhabiting the mouth of the Wanigela River it is 
during the celebration of their most important feast, the 
annual kapa, that marriages usually take place ;5 and of some 
other New Guinea savages, the natives of Mailu, we are 
likewise told, by Dr. Malinowski, that their big annual feast 
is connected with marriage and sexual life. " The dancing, 
which takes place with increased intensity during the few 
1 Oldfield, ' Aborigines of Australia,' in Trans. Ethn. Soc. N.S. 
iii. 230? 
2 Curr, The Australian Race, i. 310 sq. 
3 Heape, in Quarterly Jour. Microscop. Science, N.S. vol. xliv. 
Pt. i. 37. 
4 Miklucho-Maclay, ' Anthropologische Beinerkungen fiber die 
Papuas der Maclay-Kfiste,' in Natuurkundig Tijdschrift voor Neder- 
landsch Indie, xxxiii. 245. / 
5 Guise, ' Tribes inhabiting the Mouth of the Wanigela River,' 
in Jour. Anthr. Inst, xxviii. 214. 
G 2 84 
THE HISTORY OF HUMAN MARRIAGE chap. 
days of the feast, seems to be associated with opportunities 
for short-lived intrigues, and occasionally there even seem 
to be features of licentiousness, groups absconding to- 
gether."1 Mr. J. H. Hadfield wrote to me from Lifu, near 
New Caledonia, that marriages there formerly took place' at 
various times, when suitable, but that " November used to 
be the time at which engagements were made." As the 
seasons in this island are the reverse of those in England, 
November falls at the end of spring and the beginning of 
summer. 
Among the Gaddanes of Luzon " it is-the custom of the 
young men about to marry, to vie with each other in present- 
ing to the sires of their future bride all the scalps they are able 
to take from their enemies, as proof of their manliness and 
courage. This practice prevails at the season of the year, 
when the tree-popularly called by the Spaniards ' the 
fire-tree '-is in bloom."2 Of the savages of Northern 
Formosa it is said in a Chinese work that " when vege- 
tation bursts forth women array themselves in their best 
clothing, and pay visits to their friends in the neighbouring 
tribes."3 In the translation of a Chinese chronicle Bowra 
tells us that the Li of Hainan celebrate a spring festival at 
which the men and women of neighbouring settlements leap 
about hand in hand and sing. This is the best occasion for 
match-making, as the parents have no power to oppose the 
choice made by their children and no regard is paid to family 
names. According to the same authority, it is the custom 
among the wild, or Sheng-, Li that the boys and girls of 
sixteen or seventeen play and sing together on beautiful 
spring evenings without being in any way interfered with by 
their parents ; and there are similar customs among various 
tribes of the continent.4 In his essay on the Peninsula of 
Lei-chou, which forms the southern point of the Chinese 
1 Malinowski, 'Natives of Mailu,' in Trans. Roy. Soc. South 
Australia, xxxix. 562, 664. 
2 Foreman, Philippine Islands, p. 212. Worcester, Philippine 
Islands, p. 439. 
3 Taintor, Aborigines of Northern Formosa, p. 33. 
4 Strzoda, 'Die Li auf Hainan,' in Zeitschr. f. Ethnol. xliii. 202 s] n HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
85 
mainland, Hirth writes about the people of Sui-ch'i hsien 
that " every year when the time for the feast of lanterns 
(Yiian-hsiao, in February or March) has come, men and 
women would come together from far and near to pull the 
rattan, as they call it, i.e., to meet and have a look at each 
other, and at that time the cities and markets are crowded 
with people."1 I am told by Mr. Y. H. Yao that in a famous 
Chinese poem it is said of a young wife that on a spring 
day she went on the balcony of her house and, when she saw 
the sprouting willow, regretted that she had advised her 
husband to go away to try his fortune. 
In Cambodia M. Mondiere has found that twice a year, 
in April and September, men seenTto experience a " veritable 
rut," and will sometimes even kill women who resist them.2 
Concerning the Hami (Semang) of Hulu Jalor, in the Malay 
Peninsula, Messrs. Annandale and Robinson write :-" When 
questioned about the number of children usually bom to 
a Hami woman, she volunteered the information that the 
children of her tribe were always born at the same season of 
the year, the season, according to some Malays who were 
present, which corresponds with the first month of the Arabic 
calendar, as reckoned in the Peninsula, that is to say about 
March. This would be just after the conclusion of the stormy 
season. The. statement was confirmed by a Malay woman, 
who remarked that the Panghans bred like beasts ; but 
Malay evidence is practically worthless regarding these 
people. We were unable to obtain information concerning 
the number of children usually born, owing to our Hami 
informant's inability to count; but she said that a child 
was bom regularly every year to women of the proper age."3 
In his book on the tribes in the interior of the Malay 
Peninsula, Dr. Martin says that this statement stands quite 
alone, and that the subject therefore requires further in- 
vestigation. 4 Mr. Skeat informs us that the Besisi, belonging 
1 Hirth, ' Peninsula of Lei-chou,' in China Review, ii. 351. 
2 Mondidre, ' Cambodgiens, ' in Diction, des sciences anthropol., 
quoted by Ellis, op. cit. i. 126. 
3 Annandale and Robinson, Fasciculi Malayenses, i. 6 
4 Martin, Die Inlandstamme der Malayischen Halbinsel, p. 897. 86 
THE HISTORY OF HUMAN MARRIAGE chap- 
to the Jakun, " commonly have a regular carnival (at the 
end of the fradi or rice harvest) when (as they say) they are 
' allowed to exchange ' their wives, a practice which recalls 
the wedding law of ancient Peru, by which there was estab- 
lished one universal wedding-day annually throughout the 
land."1 And Dr. Martin himself states that the Besisi 
contract their marriages when the rice harvest has come to 
an end.2 But, for reasons which will be given below, neither 
of these statements seems to be of much importance in the 
present connection. 
In Assam the Bihu festival, which takes place every year 
in April and lasts for seven days, is celebrated with music, 
dancing, and rejoicing, and the people abandon themselves 
freely both to drunkenness and other forms of licentiousness. 
The season of this spring festival " has always been claimed 
by the female sex as a period of considerable licence ; and 
the exercise of their freedom within that period does not 
seem to be attended with any stain, blemish, or loss of 
reputation." An unfortunate youth who has failed to secure 
the consent of the parents of the girl he has selected to be his 
wife has then recourse to a stratagem to effect his object. 
" He lays wait in the road till the damsel passes by to the 
fair or festival with her female relatives, when, with the aid 
of his companions, he carries off the feigning reluctant bride, 
and immediately marries her privately " ; and in a few days 
the parents are obliged to be reconciled and consent to a 
public marriage.3 The Khasis, in the hills between Assam 
and Eastern Bengal, have grand dances in the month of 
March, at the new moon, and at these dances many matches 
are made. It is customary for unmarried girls only to dance ; 
they form a ring, and " the young bachelors run round the 
outside of the ring, moving fans made of feather."4 The 
Bhuiyas, an Orissa hill tribe, have in February a festival 
called Magh Porai, at which they give way to frightful 
1 Skeat and Blagden, Pagan Races of the Malay Peninsula, 
ii. 70. Cf. ibid. ii. 76, 145. a Martin, op. cit. p. 867 sq. 
3 Butler, Travels and Adventures in the Province of Assam, p. 226 sq. 
Endle, Kacharis, pp. 34, 50, 89. 
4 Steel, ' Kasia Tribe,' in Trans. Ethn. Soc. N.S. vii. 309 ii HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
87 
debauchery and intoxication ; it continues for three days, 
" during which all respect for blood relations and husbands 
is set at naught, and even sisters and brothers make in- 
decent jokes regarding each other."1 The Hos in Chota 
Nagpur, according to Colonel Dalton, have every year a 
great feast in January, " when the granaries are full of grain, 
and the people, to use their own expression, full of devilry. 
They have a strange notion that at this period, men and 
women are so over-charged with vicious propensities, that 
it is absolutely necessary for the safety of the person to let 
off steam by allowing for a time full vent to the passions. 
The festival, therefore, becomes a saturnalia, during which 
servants forget their duty to their masters, children their 
reverence for parents, men their respect for women, and 
women all notions of modesty, delicacy, and gentleness." 
Men and women become almost like animals in the indulgence 
of their amorous propensities, and the utmost liberty is given 
to the girls.2 The same writer adds that " it would appear 
that most Hill Tribes have found it necessary to promote 
marriage by stimulating intercourse between the sexes at 
particular seasons of the year."3 The Punjas of Jeypore 
have a festival in the first month of the new year, where men 
and women assemble ; and the lower orders or castes 
observe this festival, which is kept up for a month, by both 
sexes mixing promiscuously and taking partners as their 
choice directs.4 A similar feast, comprising a continuous 
course of debauchery and licentiousness, is held once a year 
by the Kotas of the Nilgiris.5 
Very widely celebrated is the feast which is called Holi, 
Phag, or Phagua in Northern India, Shimga or Hutashana 
in the Deccan and Western India, and Kaman-pandikai 
1 Macmillan, ' Bhuiyas,' in Calcutta Review, ciii. 188. 
2 Dalton, Descriptive Ethnology of Bengal, p. 196 sq. See also 
de Gryse, ' Les premiers habitants du Bengale,' in Les Missions 
Catholiques, 1897, p. 393. 3 Dalton, op. cit. p. 300. 
4 Shortt, Hill Ranges of Southern India, iii. n sq. Idem, ' Con- 
tribution to the Ethnology of Jeypore,' in Trans. Ethn. Soc. N.S. 
vi. 269. 
0 Idem, ' Account of the Hill Tribes of the Neilgherries, ' in Trans. 
Ethn. Soc. N.S, vii. 282. 88 
THE HISTORY OF HUMAN MARRIAGE chap. 
in the south. It is observed in the month of Phalguna 
(February-March) and lasts for at least three days, but in 
some places even for fifteen days or more, of which the last 
three are the most important. It has been styled the festival 
of the god of lust. " Most of the observers of this feast," 
says Natesa Sastri, " imagine that the object they worship 
is Cupid and that the mock-funs they observe are on account 
of Kama, the God of Love. 'x M. Rousselet gives the 
following description of it as celebrated among the Hindus 
of Oudeypur. " The festival of Holi marks the arrival of 
spring, and is held in honour of the goddess Holica, or Vasanti, 
who personifies that season in the Hindu Pantheon. The 
carnival lasts several days, during which time the most 
licentious debauchery and disorder reign throughout every 
class of society. It is the regular saturnalia of India. 
Persons of the greatest respectability, without regard to rank 
or age, are not ashamed to take part in the orgies which 
mark this season of the year. . . Effigies of the most 
revolting indecency are set up at the gates of the town 
and in the principal thoroughfares. Women and children 
crowd round the hideous idols of the feast of Holica, and 
deck them with flowers ; and immorality reigns supreme 
in the streets of the capital."2 Among the Rajputs of 
Me war, according to Tod, the last days of spring are dedi- 
cated to Camdeva, the god of love : " the scorching winds 
of the hot season are already beginning to blow, when 
Flora droops her head, and the 'god of love turns anchorite.' "3 
Among the Aryans who inhabited the plains of the North 
the spring, or vasanta, corresponding to the months of 
March and April, was the season of love and pleasure, 
celebrated in song by the poets, and the time for marriages 
and religious feasts.4 Strabo states that the marriages of the 
Crooke, ' Holi: a Vernal Festival of the Hindus,' in Folk-Lore, 
xxv. 55 sqq. Natesa Sastri, .Hindu Feasts, Fasts, and Ceremonies, 
p. 44 sqq. Padfield, The Hindu at Home, p. 152. 
2 Rousselet, India and its Native Princes, p. 173. 
3 Tod, Annals and Antiquities of Rajast'han, i. 495. 
4 Rec'us, Nouvelle geographie universelie, viii. 70. Cf. Crooke, 
in Folk-Lore, xxv. 56. n HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
89 
ancient Persians were contracted about the vernal equinox.1 
He also tells us that the Amazons for ten months of the 
year lived in seclusion from all male company, but once 
every spring ascended a certain mountain to meet the men 
of a neighbouring people and then during two months 
had sexual intercourse with them for the sake of propagation.2 
The later Romans connected April with Venus.3 In an 
Arabic almanack which has circulation in Northern Africa 
it is said that April and May are most excellent months for 
conjugal intercourse, whereas moderation should be observed 
in August.4 
A mediaeval author, the unknown biographer of St. 
Adalbert, and the monk of the Russian Abbey of Eleasar, 
known by the name of Pamphil, who lived in the sixteenth 
century, speak of the existence of certain yearly festivals 
among some Slavonic peoples at which great licence pre- 
vailed. According to the latter author, such meetings were 
regularly held on the borders of the State of Novgorod on the 
banks of rivers, resembling in that particular the annual 
festivals mentioned in the Chronicle of Nestor at which every 
man carried off a woman with her own consent.6 The 
gatherings took place, as a rule, at the end of June, the day 
before the festival 'of St. John the Baptist, which in pagan 
times was that of a divinity known by the name of Jarilo, 
corresponding to the Priapus of the Greeks.6 M. Volkov 
observes that not long ago these customs still survived in the 
government of Tver, " ou le jour de Yarilo (divinite 
phallique du printemps) les jeunes filles du peuple etaient 
envoyees par leurs parents prendre part a des jeux analogues 
a ceux des anciens Slaves, dans le but de se fiancer."7 In 
1 Strabo, Geographica, xv. 3. 17. 
2 Ibid. xi. 5. 1. 
3 Fowler, Roman Festivals of the Period of the Republic, p. 67. 
Westropp and Wake, Ancient Symbol Worship, p. 26. 
4 I am indebted to my friend Mr. Henry Bishop for this statement. 
5 Nestor, Chronique, p. 10. 
6 Kovalewsky, ' Marriage among the Early Slavs,' in Folk-Lore, 
i. 466 sq. Idem, Modern Customs and Ancient Laws of Russia, p. 10 sq. 
7 Volkov, ' Rites et usages nuptiaux en Ukraine,' in L'Anthro 
pologie, ii. 166 sq. 90 
THE HISTORY OF HUMAN MARRIAGE chap. 
this connection reference may also be made to M. Bagin's 
statement that the Votyak of Kasan still retain the very 
ancient custom of marrying their young at a definite period 
of the year, which is before the hay harvest, about the end 
of June.1 
As Mannhardt pointed out, there have been, and are 
even now, spring and midsummer festivals in various 
European countries, which are celebrated with bonfires, 
music, and dancing, and associated with love- and match- 
making.2 Yet it seems very doubtful whether these festivals 
can be regarded as evidence of an increase of the sexual desire 
in spring or at the beginning of summer, as Kulischer and 
some later writers have maintained,3 considering that 
similar festivals are held at other times of the year as well. 
Thus, for example, we hear that in ancient Russia the yearly 
festivals on Christmas day and on the day of the baptism of 
Christ were likewise characterised by promiscuous inter- 
course of the sexes.4 In another place I have stated facts 
which tend to show that the original object of European, as 
well as North African, midsummer ceremonies has mainly 
been to serve as a protection against evil forces supposed to 
be active at the summer solstice,5 and I venture to believe 
that something similar may be said of other festivals held 
at astronomical crises, such as the winter solstice and the 
vernal equinox. The leading ceremonies on these occasions 
are not generally of a sexual character ; and that love-making 
is associated with holiday gatherings of young people of both 
sexes can easily be explained without the help of the theory 
1 Solotaroff, ' On the Origin of the Family,' in American Anthro- 
pologist, xi. 230. 
2 Mannhardt, Wald- und Feldkulte, vol. i. ch. v. § 8 sqq., especially 
pp. 449, 450, 469, 480 sq. 
3 Kulischer, * Die geschlechtliche Znchtwahl bei den Menschen 
in der Urzeit,' in Zeitschr. f. Ethnol. viii. 152 sqq. Pearson, 
Chances of Death, ii. 104 sq. Ellis, op. cit. i. 130 sqq. I myself 
expressed the same opinion in the earlier editions of this work. 
4 Kovalewsky, in Folk-Lore, i. 467. Idem, Modern Customs and 
Ancient Laws, p. 11. 
5 Westermarck, Ceremonies and Beliefs connected with Agriculture, 
&c. in Morocco, p. 79 sqq. 11 HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
91 
of an ancient sexual season. More importance may be 
attached to the connection of the Russian festival with the 
phallic divinity Jarilo. 
Feasts comprising sexual licentiousness are reported from 
various parts of America1 and Africa. Among the Bororo 
Indians, Mr. Fric was told that at a certain time of the year 
a feast takes place in the bahito, or men's house, at which the 
young men steal girls who have no parents and keep them 
in the bahtto.2 Sir H. H. Johnston mentions the occurrence 
of orgies of a sexual character in British Central Africa 
" at certain seasons,"3 Mr. A. J. Swann among some tribes 
near Lake Nyasa,4 Dr. Fritsch among the Hottentots,5 the 
Rev. H. Rowley among the Kafirs in connection with the 
harvest.6 The Rev. H. T. Cousins wrote to me that among 
the Kafirs inhabiting what is known as Cis-Natalian Kafir- 
land there are more children born in August and September, 
which are the spring months in South Africa, than in any 
other season ; and he ascribes this surplus of births to feasts 
with debauchery and unrestricted intercourse between the 
unmarried people of both sexes. Dr. A. Sims, again, in- 
formed me from Stanley Pool that among the Bateke more 
children arc born in September and October, that is, the 
season of the early rains, than at other times ; and the 
Rev. Ch. E. Ingham, writing from Banza Manteka, stated 
that he believed the same to be the case among the Bakongo. 
These statements, like some others quoted before, were 
answers to questions which I had addressed to persons living 
among various savage peoples. Another answer came from 
Mr. T. Bridges, who informed me that among the Yahgans 
in the southern part of Tierra del Fuego, so far as he knew, 
1 Latchain, ' Ethnology of the Araucanos,' in Jour. Roy. Anthr. 
Inst, xxxix. 354. Bancroft, Native Races of ths Pacific States of 
North America, i. 551 sq. (Keres of New Mexico). Hartland, Primitive 
Paternity, ii. 240 sq. 
2 Fric and Radin, ' Contribution to the Study of the Bororo 
Indians,' in Jour. Anthr. Inst, xxxvi. 390. 
3 Johnston, British Central Africa, p. 408 n.i. 
4 Mr. Swann, in a letter to the author. 
5 Fritsch, Die Eingeborenen Siid-Afrika's, p. 328. 
11 Rowley, Africa Unveiled, p, 165. 92 
THE HISTORY OF HUMAN MARRIAGE chap. 
one month is the same as another with regard to the number 
of births. 
The statements which I have now quoted as evidence in 
favour of a human pairing season are certainly of very un- 
equal value, and some of them might perhaps as well have 
been omitted altogether. I have already referred to the 
doubtful testimony of certain feasts at which sexual licence 
prevails ; and it should be added that even when such licence 
seems to be the chief feature of a feast it is not necessarily 
an indication of an increase of the sexual desire in the season 
when the feast is held. It may be a magical rite intended to 
promote the growth of the crops, especially when found at 
agricultural festivals, which are in many cases associated 
with a more or less promiscuous intercourse of the sexes.1 
Of the Besisi carnival mentioned above Mr. Skeat remarks 
that it is evidently supposed to have some sort of productive 
influence not only upon the crops, but upon all other con- 
tributing sources of food-supply.2 And Mr. Crooke thinks 
there is some reason to believe that the intention to promote 
the fertility of man, animals, and crops supplies the basis of 
the Holi rites.3 
Another custom which calls for the greatest caution is 
that of contracting or celebrating marriages at some special 
time of the year. This custom may be due to purely practical 
motives.4 The Choroti of the Gran Chaco in South America 
usually contract their marriages during the algaroba-season, 
" when there is a good supply of food, and drinking-bouts 
and other festivals are held almost continually."5 Among 
the Kubu of Upper Djambi in Sumatra marriages are con- 
cluded at a time of the year when there is an abundance of 
wild fruits of a certain kind which can be preserved for a 
long time, so that the young couple shall have enough to 
eat.6 Among the Benua of Malacca " marriages are 
1 See Frazer Magic Art, ii. 97 sqq. ; Hartland, Primitive Paternity, 
ii. 240. 2 Skeat and Blagden, op. cit. ii. 121. 
3 Crooke, in Folk-Lore, xxv. 83. 
4 Cf. van Gennep, Les rites de passage, p. 199. 
5 Karsten, Indian Dances in the Gran Chaco (S. America), p. 30. 
6 Hagen, Die Orang Kubu auf Sumatra, p. 131. n HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
93 
ordinarily celebrated about the months of July and August, 
when fruits are plentiful."1 In Morocco they are commonly 
celebrated in the autumn, when the harvest has come to an 
end and the granaries are full of com ; I was told that the 
Ait Warain, south of Fez, never hold a wedding at any other 
time of the year.2 For a similar reason, the peasantry of 
many European peoples celebrate their marriages by 
preference in the autumn or at the beginning of the winter.3 
The weddings of the Wadshagga in East Africa are held at 
the time for the brewing of beer, which again depends on 
the millet-harvest.4 Among the West African Yoruba 
marriages took place in the olden days at the beginning of 
the rains, " but as there was very little food in the house at 
that season the time was changed to the time of the harvest 
of the new yams, which was a time of great rejoicing (Orisha 
Oko), and the second planting season."5 Some other West 
Africans preferred the dry season or the latter part of the 
rainy for their weddings, because there was then an abund- 
ance of fish and the weather was better suited for outdoor 
sports and plays.6 The leisure enjoyed by the people may 
also influence their choice 'of the time when they hold their 
weddings. This, for example, is the case with the Hopi 
Indians of Arizona, who generally marry in autumn or 
winter.7 In Dardistan marriages as a rule take place in 
January and February, because " there is then no agricultural 
work to occupy the men, and the houses are well stored 
with meat. So firmly is the custom established in Nager, 
that a heavy fine is inflicted on marriages held at any other 
season."8 The Kukis usually marry in the same months 
1 Favre, ' Account of the Wild Tribes inhabiting the Malayan 
Peninsula,' in Jour. Indian Archipelago, ii. 263. 
2 Westermarck, Marriage Ceremonies in Morocco, p. 86. 
3 v. Schroeder, Die Hochzeitsgebrduche der Esten, &c., p. 48 sq. 
Weinhold, Die deutschen Frauen in dem Mittelalter, i. 363 sq. Meyer, 
Deutsche Volkskunde, p. 174. 
4 Volkens, Der Kilimandscharo, p. 251. 
5 Dennett, Nigerian Studies, p. 165 sq. 
6 Nassau, Fetichism in West Africa, p. 9. 
7 Voth, ' Oraibi Marriage Customs,' in American Anthropologist, 
N.S. ii. 240. 8 Biddulph, Tribes of the Hindoo Koosh, p. 78. 94 
THE HISTORY OF HUMAN MARRIAGE chap. 
" because they have provisions in the greatest plenty, and 
it is their most idle time."1 The Coorg weddings generally 
" come off during the months of April and May, when the 
rice-valleys are dry, and there is little work to be done."2 
Superstition, too, may play a part in the matter.3 Among 
the Brahmans of India, for instance, there are properly 
but four months in the year in which marriage can be 
celebrated, namely, March, April, May, and June, which are 
considered a fortunate time for marrying. But Dubois 
thinks that the principal motive which originally induced 
them to fix on these months was that the country labours 
were then all closed or suspended on account of excessive 
heat, and the preceding harvest furnished the means of 
supplying what the ceremony required.4 In China the 
spring season and the last month in the year are regarded as 
the most fortunate nuptial periods.5 It is not impossible, 
however, that superstitions of this kind have their root in 
sexual periodicity. 
It may also be said that some of the statements are too 
indefinite to be quite convincing, and that a few others bear 
the stamp of exaggeration. Yet allowing for all such de- 
ficiencies in the material, I think there is sufficient evidence 
left to show that an annual increase of the sexual desire or 
of the reproductive power, generally in spring, is of frequent 
occurrence in mankind. This conclusion derives much sup- 
port from definite statistical data relating to the distribution 
of births over the different months of the year. 
In the eighteenth century Wargentin observed that in 
Sweden more children were born in one month than in 
another.6 The same has since been found to be the case in 
1 Macrae, ' Account of the Kookies,' in Aszatick Researches, 
vii. 194. 
2 Richter, Manual of Coorg, p. 134. 
3 See infra, on Marriage Rites. 
4 Dubois, Description of the Character, &c. of the People oj India, 
p. 103 sq. See also Padfield, op. cit. p. 101 sq. 
6 Wells Williams, The Middle Kingdom, i. 791. 
G Wargentin, ' Uti hvilka Manader flera Miinniskor arligen fodas 
och do i Sverige,' in Kongl. Vetenskaps academiens Handlingar, 
xxviii. 249 sqq. ii HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
95 
other European countries as well. According to Wappaus, 
the number of births in Sardinia, Belgium, Holland, and 
Sweden is subject to a regular increase twice a year, the 
maximum of the first increase occurring in February or 
March and that of the second in September and October.1 
Sormani observed that in the south of Italy there is an 
increase only once a year, but more to the north twice, 
namely, in spring and in autumn.2 Mayr 3 and Beukemann4 
found in Germany two annual maxima-in February or 
March and in September ; and Haycraft states that in the 
eight largest towns of Scotland more children are bom in 
legitimate wedlock in April than in any other month.5 
As a rule, according to Sormani, the first annual increase of 
births has its maximum in Sweden in March ; in France 
and Holland, between February and March ; in Belgium, 
Spain, Austria, and Italy, in February; in Greece, in 
January. Thus it comes earlier in Southern Europe than 
farther north.6 Again, the second annual increase is found 
more considerable the more to the north we go. In South 
Germany it is smaller than the first one, but in North 
Germany generally larger ;7 and in Sweden it is decidedly 
larger.8 
As to non-European countries Wappaus observed that 
in Massachusetts the birth-rate -likewise underwent an 
increase twice a year, the maxima falling in March and 
September; and that in Chili many more children were 
bom in September and October-that is, the beginning of 
spring-than in any other month.9 In Cuba, according to 
1 Wappaus, Allgemeine Bevblkerungsstatistik, i. 237. 
2 Sormani, La feconditd e la mortalitd umana in rapporto alle 
stagioni ed ai clima d'Italia, quoted by Mayr, Die Gesetzmassigkeit 
im Gesellschaftsleben, p. 242. 
3 Mayr, op. cit. p. 240. 
4 Beukemann, Ein Beitrag zur Untevsuchung uber die Vertheilung 
der Geburten nach Monaten, p. 15 sqq. 
5 Haycraft, loo. cit. p. 119 sq. 
6 Sormani, quoted by Mayr, op. cit. p. 241. 
7 Beukemann, op. cit. p. 26. 
■8 Wargentin, loc. cit. p. 252. Wappaus, op. cit. i. 237. 
9 Wappaus, op. cit. i. 250, 237. 96 
THE HISTORY OF HUMAN MARRIAGE chap. 
the records compiled by Dr. Finlay, both whites and coloured 
people exhibit a specially high birth-rate in July, and another, 
though less marked, rise in the birth-rate in November and 
December.1 Mr. S. A. Hill, of Allahabad, has shown that 
among the Hindus of that province the birth-rates likewise 
vary in different months, the minimum falling in June and 
the maximum in September and October.2 According to 
the records of births registered in the district of Coimbatore 
during the years 1888-1892, the birth-rate was lowest in 
February (6*95 per cent.) and highest in October (9'62 per 
cent.) and July (9T5 per cent.).3 In North Arcot it was 
likewise, during the same period, lowest in February (5'98 
per cent.) and highest in August and July (10'40 and 10'13 
per cent, respectively) ;4 whilst in South Canara it was 
lowest in October and September (6'68 and 674 per cent, 
respectively) and highest in July and June (9-80 and 9'69 
per cent, respectively).5 Europeans and Eurasians were 
not included in these records. 
The periodical fluctuations in the birth-rate may no 
doubt be due to various causes. But I think there is 
every reason to believe that the maximum in February and 
March (in Chili, September) is, at least to a large extent, 
due to an increased tendency to procreation in May and 
June (in Chili, December).6 This seems the more likely 
since it is especially illegitimate births that are then com- 
paratively numerous.7 Mr. Heape believes that both the 
autumn and spring maxima of conceptions in Cuba have a 
similar origin.8 
1 Heape, ' Proportion of the Sexes produced by Whites and 
Coloured Peoples in Cuba,' in Philosophical Transactions, ser. B. 
vol. cc. 296. 
2 Hill, ' Life Statistics of an Indian Province,' in Nature, xxxviii. 
250. 
3 Nicholson, Madras District Manuals : Coimbatore, ii. 100. 
4 Cox, Madras District Manuals : North Arcot, ii. 100. 
5 Stuart, Madras District Manuals : South Canara, ii. 92. 
c The same opinion has been expressed by Wappaus (op. cit. 
i. 239, 247), Ploss (Das Weib, i. 414), and many later writers. 
7 See, e.g., Mayr, Bevolkerungsstatistik, p. 171. 
8 Heape, in Philosophical Transactions, ser. B. vol. cc. 296 sqq. ti HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
97 
If we thus find in man, even to this day, an increase 
either of the sexual desire or of the reproductive power in 
spring or at the beginning of summer, I think we may 
look upon it as a survival of a pairing season among our 
early human or pre-human ancestors.1 We are the more 
justified in doing so as a sexual season occurs among the man- 
like apes, and conditions similar to those which, led to it 
in their case may be supposed to have produced the same 
result in the case of primeval man. We may assume that 
he in this respect followed the law which governs the pairing 
of other animals, that his reproductive activity was con- 
trolled by the needs of the next generation. It is true 
that if coition took place in spring or at the beginning of 
summer, the offspring were bom somewhat earlier in the 
year than is the case in the majority of mammalian species. 
But we must remember that the infancy of man is unusually 
long, and that with regard to the time most favourable to 
the subsistence of children, we have to take into consideration 
not only the first days or weeks of their existence, but the 
earlier period of their infancy in general. Besides food 
and warmth there are other factors that may affect the 
welfare of the offspring, and it is not easy to find out all of 
them. We may be unable to say exactly why the badger 
bears its young at the end of February or the beginning of 
March,2 and the reindeer of the Norwegian mountains as 
early as April; 3 but there can be no doubt that these 
seasons are adapted to the requirements of the respective 
species. 
The cause of the winter maximum of conceptions, especially 
considerable among the peoples of Northern Europe, is 
1 That the periodical increase of births is a survival of an ancient 
sexual season was suggested by Kulischer (loc. cit. p. 156 sq.} in 1876, 
and by Rosenstadt (' Zur Frage nach den Ursachen, welche die 
Zahl der Conceptionen beim Menschen in gewissen Monaten des 
Jahres regelmassig steigern,' in Mittheil. aus dem embryologischen 
Institute der K. K. Universitdt in Wien, ser. ii. pt. iv. 95) in 1890 
the year after the present chapter with the same suggestion was 
originally published (in my academical thesis, The Origin of Human 
Marriage, Helsingfors, 1889). 
2 Brehm, op. cit. ii. 149. 3 Ibid. iii. 124. 98 
THE HISTORY OF HUMAN MARRIAGE chap. 
generally sought in social influences, such as the quiet 
ensuing on the harvest time, the better food, the amusements 
of Christmas,1 or the greater frequency of marriages.2 
But although I do not deny that it may have a social origin, 
the explanations given are not free from objections. The 
people certainly recover before December from the labours 
of the field. Already Wargentin remarked that the Christ- 
mas amusements take place at the end of that month 
and far into January, without showing any particular 
influence upon the number of births in October.3 And 
as regards the supposed connection between an increased 
tendency to procreation and abundance of food it is worth 
noticing that in the northern parts of Europe many more 
conceptions take place in May and June, when the conditions 
of life are often rather hard, than in September, October, 
and November, when the supplies of food are comparatively 
plentiful; indeed, in Sweden and the north-western 
provinces of Germany the latter months are characterised 
by a minimum of conceptions.4 Among the Cis-Natalian 
Kafirs more children are conceived in November and 
December than in any other month, although, according 
to Mr. Cousins, food is most abundant among them from 
March to September ; and among the Bateke the maximum 
of conceptions falls in December and January, although 
food is, as I am informed by Dr. Sims, most plentiful in 
the dry'season, that is, from May to the end of August. 
Finally, the opinion that the winter maximum of conceptions 
is due to the greater frequency of marriages is opposed 
to the conclusion arrived at by many investigators, that the 
unequal distribution of marriages over the different months 
exercises little or no influence upon the distribution of births.5 
1 Wappaus, op. cit. i. 241. Mayr, Bevblkerungsstatistik, p. 171. 
2 Mayo-Smith, Science of Statistics, i. 75. 
3 Wargentin, loc. cit. p. 254. 4 Beukemann, op. cit. pp. 18, 28. 
•' Wappaus, op. cit. i. 242. Villerme, quoted by Ploss-Bartels, 
Das Weib, i. 712. Diising, quoted by Rauber, Der Uberschuss an 
Knabengeburten, p. 40. Bertillon, ' Natality (demographic),' in 
Dictionnaire encyclopedique des sciences medicates, ser. ii. vol. xi. 479. 
Rosenstadt, loc. cit. p. 94. Mayr, Bevblkerungsstatistik, p. 171. 
Heape, in Philosophical Transactions, ser. B. vol. cc. 298. n HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
99 
I am far from venturing to express any definite opinion 
as to the cause of the winter maximum of conceptions, 
but I consider it quite possible that this maximum, also, 
is at least in some measure a result of natural selection, 
although of a comparatively late date. Considering that 
the September maximum of births (or December maximum 
of conceptions) in Europe becomes larger the farther north 
we go ; that the agricultural peoples of Northern Europe 
have plenty of food in autumn and during the first part of 
winter, but often suffer some want in spring ; and, finally, 
that the winter cold hardly affects the health of infants, 
the woods giving sufficient material for fuel-it has occurred 
to me that children bom in September may have a particu- 
larly good chance of remaining alive. Dr. Beukemann states 
in fact that " the children born in autumn possess the 
greatest vitality and resisting power against the dangers of 
earliest infancy."1 And more than half a century ago 
Edward Smith found, by taking 3050 cases of English 
children dying within one year of their birth, that the 
percentage of those bom in the different months varied from 
less than 7 per cent, in February and 7 per cent, in September 
to nearly 11 per cent, in June-in other words, that children 
conceived in May and December showed signs of the highest 
vitality and those conceived in September of the lowest.2 
Commenting on this very interesting fact-which was not 
known to me when I first made my suggestion-Dr. Have- 
lock Ellis writes, " As we have seen, May and December are 
precisely the periods when conceptions in Europe generally 
are at a maximum, and September is precisely the period 
when they are at a minimum, so that, if this coincidence is 
not accidental, the strongest children are conceived when 
there is the strongest tendency to procreate, and the feeblest 
children when the tendency is feeblest."3 This would 
perhaps be an adequate explanation either of an increase of 
the sexual desire or of a greater disposition to impregnation 
in December; and it may also explain why the spring and 
early summer maximum of conceptions, instead of dying 
1 Beukemann, op. cit. p. 59. 2 Smith, Health and Disease. p. 172. 
3 Ellis, op. cit. i. 142 sq. 100 
THE HISTORY OF HUMAN MARRIAGE chap. 
out as a useless survival, is still preserved in the midst of 
civilisation. But the cases investigated are too few to 
allow any definite conclusions. 
Perhaps the increase of births among the Hindus of 
Allahabad in September and October-that is, the end of 
the hot season and the beginning of winter-has a similar 
origin. It may be that children bom in those months 
have the best chance of surviving; during the winter the 
granaries get filled, and some of the conditions of life become 
more healthy, although September itself is a very unhealthy 
month.1 Mr. Hill, on the other hand, attributes the increase 
of conceptions to the direct influence of healthy conditions 
with an abundant food supply.2 But, as already said, the 
birth statistics for other countries do not show any such 
influence. 
That man at present is not restricted to any particular 
season for the procreation is, of course, no objection to 
the hypothesis that his primitive ancestors were so. When 
Professor Nicholson says that Darwinism fails to assign any 
adequate cause for this,3 he forgets that natural selection is 
only a negative cause, which checks undesirable variations. 
The more progress man makes in arts and inventions; 
the more he acquires the power of resisting injurious external 
influences ; the more he rids himself of the necessity of 
freezing when it is cold and starving when nature is niggard 
of food; in short, the more independent he becomes of the 
changes of the seasons-the greater is the probability that 
children born at one time of the year will survive almost 
as well as those born at any other. Variations as regards 
the pairing time, which may always occur occasionally, 
will do so the more frequently on account of the changed 
conditions of life, which directly or indirectly cause varia- 
bility of every kind ;4 and these variations will be preserved 
and transmitted to following generations. It is interesting 
to note that the periodical fluctuation in the number of 
1 Hill, loc. cit. p. 250. 2 Ibid. p. 250. 
3 Nicholson, Sexual Selection in Man, p. 9. 
4 Darwin, Variation of Animals and Plants under Domestication, 
ii. 255. ii HUMAN PAIRING SEASON IN PRIMITIVE TIMES 
101 
births, though comparatively inconsiderable in every civilised 
society, is greater in countries predominantly agricultural, 
such as Chili, than in countries predominantly industrial, 
as Saxony; 1 that it is greater in rural districts than in 
towns ; 2 and that it was greater in Sweden in the middle 
of the eighteenth century than it is now.3 The more man 
has abandoned natural life out of doors, the more luxury 
has increased and his habits have got refined, the greater 
is the variability to which his sexual life has become subject, 
and the smaller has been the influence exerted upon it by 
the changes of the seasons. 
In the sexual life of domesticated animals and wild 
animals kept in captivity there is also great variety, and 
the variations are extended not only to varieties of a species, 
but even to individuals of that species under domestication.4 
Among dogs,5 goats,6 asses in southern countries,7 and 
captive cattle and deer,8 the males are indeed prepared to 
propagate at all times, or almost at all times, throughout 
the year. The domestic cat may have three or four sexual 
seasons each year, whereas the wild cat has only one-some 
say two, but this is doubtful:9 Among dogs in the wild 
state the bitch only experiences heat once a year, in the 
spring ; but among domesticated dogs there is not only 
the spring period of heat, early in the year, but also an 
autumn period, about six months after, though the primitive 
period remains the most important one.10 Dr. Hermann 
Muller has observed a canary which laid eggs in autumn and 
winter.11 Natural selection cannot, of course, account 
for such changes : they fall under the law of variation. 
It is the limited pairing season that is a product of this 
1 Wappaus, op. cit. i. 247. 
2 Ibid. i. 246. Quetelet, Treatise on Man, p. 20. Bertillon, 
loc. cit. p. 480. 
3 Wappaus, op. cit. i. 343. 
4 Heape, in Quarterly Jour. Microscop. Science, N.S. vol. xliv. 
pt. i. 21 sqq. 
6 Ibid. p. 64. 6 Brehm, Thierleben, iii. 333. 
7. Ibid. iii. 43. 8 Heape, loc. cit. p. 64. 
9 Ibid. p. 23. 10 Ellis, op. cit. i. 124. 
11 Muller Am Neste, pp 2, 86, 104 102 
THE HISTORY OF HUMAN MARRIAGE ch. ii 
powerful process, which acts with full force only under 
conditions free from domestication or civilisation. 
If the hypothesis set forth in the present chapter holds 
good, it must be admitted that the continued excitement 
of the sexual instinct could not have played a part in the 
origin of human marriage, provided that marriage existed 
among primitive men-as I believe it did. CHAPTER III 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY I 
ALLEGED INSTANCES OF PEOPLES LIVING IN PROMISCUITY 
The inference that marriage and the family consisting 
of parents and children existed among primitive man is 
opposed to the views held by many sociologists who have 
written upon early history. It is often said that the human 
race must have originally lived in a state of promiscuity, 
where individual marriage did not exist, where all the 
men in a horde or tribe had indiscriminately access to all the 
women, and where the children bom of these unions belonged 
to the community at large. This opinion has been expressed 
by Bachofen, McLennan, Morgan, Lord Avebury, Giraud- 
Teulon, Lippert, Kohler, Post, Wilken, Kropotkin, Wilutzky, 
Bloch, and many others.1 It is apparently somewhat less 
1 Bacliofen, Das Mutterrecht, pp. xix. xx. io. Idem, Antiquarische 
Briefe, p. 20 sq. McLennan, Studies in Ancient History, pp. 92, 95. 
Idem, Studies in Ancient History. Second Series, pp. 55, 57. Morgan, 
Systems of Consanguinity and Affinity, pp. 480, 487 sq. Idem, Ancient 
Society, pp. 418, 500 sqq. Avebury, Origin of Civilisation, p. 68 sqq. 
Idem, Marriage, Totemism and Religion, p. 3 sqq. Giraud-Teulon, 
Les origines du mariage et de la famille, p. 70. Lippert, Kultur- 
geschichte der Menschheit, ii. 7. Kohler, * Ein Beitrag zur ethnolo- 
gischen Jurisprudenz,' in Zeitschr. vergl. Rechtswiss. iv. 267. Post, 
Die Geschlechtsgenossenschaft der Urzeit, p. 16 sq. Idem, Die Grundlagen 
des Rechts, p. 183 sq. Idem, Studien zur Entwickelungsgeschichte des 
Familienrechts, p. 54 sq. Wilken, ' Over de primitieve vormen van 
het huwelijk en den oorsprong van het gezin,' in De Indische Gids, 
1880, vol. ii. 611. Idem, Das Matriarchat bei den alten Arabern, p. 7 sq. 
Kropotkin, Mutual Aid, pp. 86, 313 sqq. Wilutzky, Vorgeschichte des 
Rechts, i. 12 sqq. Bloch, Sexual Life of Our Time, p. 188 sqq. Fiske, 
Outlines of Cosmic Philosophy, ii. 345. Kulischer, ' Die geschlecht- 
103 104 
THE HISTORY OF HUMAN MARRIAGE chap. 
popular now than it was at the time when I first wrote this 
book, but it has still as staunch supporters as ever. Dr. 
Bloch, for instance, says that recent ethnological research 
has proved the untenability of my criticism of the doctrine 
of promiscuity, that there can be no doubt whatever that 
in the beginnings of human development a state of sexual 
promiscuity actually prevailed, and that it even seems 
incomprehensible how a dispute could ever have arisen 
in the matter.1 Some other recent writers are almost 
equally positive. I consider it therefore by no means 
superfluous to renew and extend my criticism. And now, 
as before,2 I shall not merely endeavour to show that the 
supposed survivals of ancient promiscuity really are no such 
survivals at all, but also indicate how the customs in question 
may be explained. 
The evidence adduced in support of the hypothesis of 
promiscuity flows from two different sources. First, there 
are in books of ancient and modem writers notices of peoples 
who are said to live or to have lived promiscuously. Second, 
there are certain customs which are assumed to be relics 
from an earlier stage of civilisation when - marriage did not 
liche Zuchtwahl bei den Menschen in der Urzeit,' in Zeitschr. f. 
Ethnol. viii. 140 sq. Gumplowicz, Grundriss der Sociologie, p. 107. 
Engels, Der Ursprung der Familie, des Privateigenthums und des 
Staats, p. 17. Bebel, Woman in the Past, Present, and Future, p. 9- 
Reclus, Primitive Folk, p. 157. Finck, Primitive Love and Love- 
Stories, pp. 79 n. 1, 438. Parkinson, Dreissig Jahre in der Siidsee, 
p.6q6sq. Grandidier, Ethnographic de Madagascar, ii. 175. Herbert 
Spencer (Principles of Sociology, i. 635) infers that even in prehistoric 
times promiscuity was checked by the establishment of individual 
connections, but thinks that in the earliest stages it was but in a 
small degree thus qualified. Miiller-Lyer writes (Die Familie, 
p. 47) :-" Uber die Frage, ob die urzeitlichen Menschen in Promis- 
kuitat gelebt haben, oder ob es in den Urhorden schon Sonder- 
familien und Sonderehen gegeben habe, lassen sich nur Vermutungen 
aufstellen. Wahrscheinlich ist die letzere Annahme nicht." Dr. 
Hartland ('Totemism and Exogamy,'in Folk-Lore, xxii. 367) finds 
the hypothesis of primitive promiscuity ''by no means untenable." 
1 Bloch, op. cit. pp. 188, 194. 
2 I do not understand how Schurtz (Altersklassen und Mdnner- 
biinde, pp. 68, 69, 175 sq.) could accuse me of having been merely 
negative in my criticism. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
105 
exist. In the present chapter I shall deal with the former 
groups of facts. 
McLennan observes that " tradition is found everywhere 
pointing to a time when marriage was unknown, and to 
some legislator to whom it owed its institution : among 
the Egyptians to Menes ; the Chinese to Fohi; the Greeks 
to Cecrops; the Hindus to Svetaketu."1 The Chinese 
annals recount that in the beginning men differed in nothing 
from other animals in their way of life. ' As they wandered 
up and down in the woods and women were in common, it 
happened that children never knew their fathers, but only 
their mothers. But the emperor Fou-hi abolished this in- 
discriminate intercourse of the sexes and instituted marriage.2 
In the Indian epic Mahabharata, Pandu tells his wife Kunti 
that the women were in former times not kept within houses 
and dependent upon husbands and other relatives, but 
enjoyed themselves as best they could. This practice 
was not regarded as sinful, but was applauded by great 
Rishis, and is still in vogue among the Uttarakurus. But 
it was abolished by Svetaketu, the son of the sage Uddalaka, 
when he saw that a strange Brahman in the presence of his 
own father took his own mother away with him.3 According 
to Athenian tradition, the women were of old the common 
property of the men, who coupled with them like beasts, 
so that while every one knew his mother, nobody knew who 
his father was. This communism, however, was abolished 
by Kekrops, the first king of Athens, who established the 
laws and rules of marriage in its place.4 The remote 
Laplanders, also, sing about Njavvis and Attjis, who 
instituted marriage and bound their wives by sacred oaths.5 
1 McLennan, Studies in Ancient History, p. 95. 
2 Goguet, Origin of Laws, Arts, and Sciences, iii. 311, 313. 
3 Mahabharata, i. 122. 4 sqq. Muir, Original Sanskrit Texts, ii. 327. 
4 Clearchus of Soli, quoted by Athenaeus, Dipnosophistee, xiii. 
2, P- 555 D. Charax of Pergamus, Hellenica, 10 (in Fragment a 
Historicorum Gr eecorum, ed, C. Muller, iii. 638). John of Antioch, 
Historia, 13 {ibid. iv. 547). Suidas, Lexicon Greece et Latine, s.v. 
' KfKpo'p,' vol. ii. 198 sq. Justin, Histories Philippicee, ii. 6. 7. 
St. Augustine, De civitate Dei, xviii. 9. See also Frazer, Magic Art, 
ii. 284. 5 v. Duben, Lappland och Lapparne, p. 330, 106 
THE HISTORY OF HUMAN MARRIAGE chap. 
Legends of this sort can no more be regarded as evidence 
of primitive promiscuity than the second chapter of Genesis 
can be quoted in proof of primitive monogamy. They 
may be simply due to the tendency of the popular mind to 
ascribe almost any great institution to a wise legislator or 
ruler, if not to direct divine intervention. But at the 
same time I do not deny that they may be an echo of 
social conditions in the past. The story in the Mahabharata 
may allude to the laxity of morals among the non-Aryan 
people of India and the Himalayas, as the polyandry of the 
five Pandavas is probably an allusion to their polyandrous 
practices.1 Lassen fixed the abode of the Uttarakurus 
beyond that great mountain chain.2 The Athenian legend, 
again, has been represented as a survival of mother-right ;3 
but that matrilineal descent once prevailed in Greece is an 
hypothesis which has been strongly contested.4 Mr. Rose 
suggests that the legend in question may be " the theory of 
primitive promiscuity, no less a theory and no more a fact 
because stated by an anthropologist who lived 2000 years 
ago." He adds that any one who has read Lucretius or 
Ovid knows how much in favour this theory was among 
Greek scientists.5 
Statements about foreign tribes living in a state of 
promiscuity abound in Greek and Roman literature. 
Herodotus 6 and Strabo 7 tell us that among the Massagetae, 
although each man had but one wife, the men used the women 
promiscuously, and that when a Massagetan desired to have 
the company of a woman he hung up his quiver in front 
of her chariot and had intercourse with her without shame. 
1 See infra, on Polyandry. 
2 Lassen, Indische Alterthumskunde, i. 612, 802. See also Oppcrt, 
Original Inhabitants of Bharatavavsa or India, p. 613. 
3 Avebury, Origin of Civilisation, pp. 123, 127 Frazer, Magic 
Art, ii. 284. Miss Harrison, Prolegomena to the Study of Greek 
Religion, p. 262. Hartland, Primitive Paternity, ii. 134 n. 1. 
1 Rose, ' On the alleged Evidence for Mother-right in Early 
Greece,' in Folk-Lore, xxii. 279 sqq. Wilamowitz-Moellendorff 
and Niese, Staat und Gesellschaft der Griechen und Romer, p. 33. 
■' Rose, loc. cil. p. 289. 
Herodotus, i. 216. 7 Strabo, Geographica, xi. 8. 6. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
107 
It should be noticed that in these statements, which have 
constantly been quoted as evidence in favour of early 
promiscuity, individual marriage is expressly said to exist, 
although not only the husband but other men as well had 
access to his wife. According to a later authority, a Chinese 
writer of the thirteenth century by name Ma-touan-lin, 
it was the custom among the Massagetae of Turkestan that 
brothers had one wife between them and that the children 
bom of these unions belonged to the eldest brother ; and 
if a man had no brothers he associated with other men.1 
This account seems to be correct, considering that polyandry 
of the same type has prevailed among other Central Asiatics 
from time immemorial.2 Now it is by no means improbable 
that the statements of Herodotus and Strabo had their 
origin in facts of this kind, and that the monogamous 
marriage mentioned by them was the union of the eldest 
brother or the first male partner, who was the chief or the 
real husband. And it should also be borne in mind that 
polyandry is not infrequently combined with polygyny, 
so that a group of men actually have several wives in 
common.3 
Of the Galactophagi, another Scythian people, Nicolaus 
Damascenus says that they had their property and women 
in common, and that they called the old men fathers, 
the young men sons, and those of equal age brothers.4 
According to the same writer, the Liburnes in Illyria likewise 
had community of wives and brought all the children up 
together till they became five years of age, when each 
child was attributed to the man to whom it bore the strongest 
resemblance ; and thenceforth that man educated the child 
as if it were his own.5 Herodotus states of the Agathyrsi, 
who in his time dwelt in the country now called Transylvania 
and afterwards were driven more to the north,6 that they 
1 R6musat, Nouvcaux melanges asiatiques, i. 245. 
2 See infra, on Polyandry. 
3 See infra, on Group-marriage and other Group-relations. 
4 Nicolaus Damascenus, Morum mirabilium collectio, e Stobaei 
Florilegio, v. 73. » Ibid. xliv. 41. 
6 Rawlinson's translation of Herodotus, iii. 91 n. 6. 108 
THE HISTORY OF HUMAN MARRIAGE chap. 
had wives in common so that they might be all brothers and, 
as members of one family, might neither envy nor hate 
one another.1 Promiscuity is said to have prevailed 
among the Mosyni, on the coast of Pontus.2 
Communism in women and children was ascribed by 
classical writers to various Ethiopian peoples :-the Ichthyo- 
phagi (inhabiting islands in the Red Sea),3 Hylophagi 
and Spermatophagi,4 Garamantians,5 and Troglodytes.® 
Of the last-mentioned people, however, it was said that chiefs 
had their own wives, and that anybody who had intercourse 
with the wife of a chief had to pay a sheep. Concerning 
the Auseans, inhabiting the borders of Lake Tritonis, 
Herodotus writes :-" These people do not marry or live in 
families, but dwell together like the gregarious beasts. 
When their children are full-grown they are brought before 
the assembly of the men, which is held every third month, 
and assigned to those whom they most resemble."7 Of the 
Nasamonians, another Libyan people who dwelt around the 
shores of the Greater Syrtis, he says :-" Every man, by the 
custom of the country, has several wives, and they have 
intercourse with them in common; and much the same 
as the Massagetae, they have intercourse when they have 
set up a staff before them."8 It is of interest to notice 
that Herodotus expressly affirms the existence of indi- 
vidual marriage among these Libyans ;9 for what he says 
about their oath-taking and some other customs is so true 
of the present inhabitants of Northern Africa that it may be 
supposed that his statement concerning their sexual 
1 Herodotus, iv. 104. 
2 Pomponius Mela, Chorographia, i. 19. Cf. Xenophon, Anabasis, 
v. 4. 33 ; Diodorus Siculus, Bibliotheca historica, xiv. 30. 7. 
3 Ex Agatharchide de Mari Erythraeo, v. 51, in Muller, Geographi 
Graeci minores, i. 130. Diodorus Siculus, op. cit. iii. 15. 2. 
4 Ex Agatharchide de Mari Erythraeo, v. 31, in Muller, op. cit. i. 143. 
Diodorus Siculus, op. cit. iii. 24.4. 
5 Solinus, Collectanea rerum memorabilium, xxx. 2. Pliny, 
Historia naturalis, v. 8. 45. 
0 Strabo, op. cit. xvi. 4. 17. Ex Agatharchide de Mari Erythraeo, 
v. 61, in Muller, op. cit. i. 153. Diodorus Siculus, op. cit: iii. 32. I. 
' Herodotus, iv. 180. 8 Ibid. iv. 172. 9 See also ibid. iv. 296. Hi CRITICISM OF HYPOTHESIS OF PROMISCUITY 
109 
communism also had some foundation in fact. What the 
facts actually were it is of course impossible to say. Perhaps 
they had promiscuous intercourse on certain occasions. 
Marmol Caravajal speaks of a very ancient city situated 
on the road leading from Sefru, in the neighbourhood of 
Fez, to the province of Numidia, in which, according to 
African writers, there was a great temple where men and 
women had promiscuous intercourse during one night at a 
certain time of the year, after sacrifices had been made and 
the lights had been extinguished. The women who had 
been present were not allowed to sleep with their husbands 
until it was known whether they had become pregnant, and 
the children resulting from those irregular connections were 
destined to become servants of the temple. But when the 
Arabs invaded Mauritania, they destroyed the city with 
its temple and all its houses.1 Yet similar practices are 
reported to exist in Morocco even to this day. In the 
interior of that country I heard stories about certain tribes 
or communities where people of both sexes are said to assem- 
ble in a mosque once every year in the night. The lights are 
extinguished, everybody lies down, and the men and women 
who are lying near each other have sexual intercourse regard- 
less of all ties of marriage and consanguinity; but before 
the commencement of the orgies the leading man moves a 
bamboo staff over their heads to find out whether anybody 
remains standing and consequently, being a stranger, is to 
be turned out. This was told me not by the people them- 
selves, but by other natives, and must therefore be looked 
upon with considerable suspicion ; but there may be some 
truth in it after all.2 
1 Marmol Caravajal, La description general de Affrica, book iv. 
ch. 109, vol. ii. fol. 163a. 
2 See also Salmon, ' Les B^adoua,' in Archives marocaines, ii. 
362 sq..; Doutte, Merrdkech, p. 366 sq. ; Mouli^ras, Une Tribu 
Zenete anti-musulmane au Maroc (les Zkara,'), p. 100 sq. Similar 
stories are told about the Lycian Tachtadshys by their Turkish 
neighbours ; but whether partially true or not, the stories contain 
obvious inaccuracies (Petersen and Luschan, Reisen in Lykien, 
p. 199)- See also Blau, ' Nachrichten uber kurdische Stamme,' in 
Zeitschr. Deutsch. Morgenldnd. Gesellsch. xvi. 623 sq. 110 
THE HISTORY OF HUMAN MARRIAGE chap. 
Considering how uncertain the information is which 
people give about the sexual relations of their own neigh- 
bours, we must be careful not to accept as trustworthy 
evidence the statements made by classical writers with 
reference to more or less distant tribes of which they evi- 
dently possessed very little knowledge. In the very chapter 
where Pliny states that among the Garamantians men and 
women lived in promiscuous intercourse he tells us of another 
African tribe, the Blemmyans, that they were said to have 
no head and to have the mouth and eyes in the breast.1 I 
have never seen this statement quoted in any book on human 
anatomy, and can see no reason to assume that our author 
was so much better acquainted with the sexual habits of 
the Garamantians than he was with the personal appearance 
of the Blemmyans. Moreover, the statements referred to 
above are so short and ambiguous that different construc- 
tions may be put upon them. The community of women 
mentioned in them does not necessarily imply general 
promiscuity within the horde or tribe but may mean group- 
relations similar to those which are known to prevail among 
certain modern savages ; and if the existence of marriage 
is denied, we must remember that the word " marriage " 
may have many meanings. 
Promiscuity has been ascribed by mediaeval writers to 
certain European peoples. Cosmas of Prague, a Latin 
annalist of the eleventh century, writes of the Bohemians 
or Czechs :-" Ipsa connubia erant illis communia. Nam 
more pecudum singulas ad noctes novos ineunt hymeneos, 
et surgente aurora trium gratiarum copulam et secreta 
amoris rumpunt vincula."2 Professor Kovalewsky says 
that this statement as to promiscuity is directly confirmed 
by that of another mediaeval author, the unknown bio- 
grapher of St. Adalbert-who ascribes the animosity of the 
Bohemian people towards the saint to the fact of his strong 
1 Pliny, op. cit. v. 8 :-" Garamantes matrimoniorum exortes 
passim, cum feminis degunt. . . . Blemmyis traduntur capita 
abesse, ore et oculis pectori adfixis." 
2 Cosmas of Prague, Chronica Bohemorum, i. 3 (Migne, Patrologiae 
cursus, clxvi. 60 sq.}. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
111 
opposition to the shameful promiscuousness which in his 
time prevailed in Bohemia-as also by the monk Pamphil, 
who lived in the sixteenth. century ; both speak of the 
existence of certain yearly festivals at which great licence 
prevailed.1 But to indulge in sexual excesses once a year 
is not the same as to live in promiscuity permanently ; and 
as for Cosmas' statement it should be noticed that he himself 
distinguishes between a mythical age and an historical age, 
and leaves it to his readers to decide which of his statements 
refer to the former and which to the latter.2 Krek points 
out that his Chronicle is almost valueless as an historical 
source and that anybody who looks upon it otherwise shows 
incapacity of discerning between history and myth.8 But 
even if we could be induced to believe that an Aryan people 
a few centuries ago had no marriage, we should certainly 
have to admit that it had lost something which its ancestors 
possessed in the past. In the Chronicle ascribed to Nestor, 
who is supposed to have been a Russian monk of the eleventh 
century, we read of the Drevlians that they lived like wild 
beasts, that marriage was unknown to them, and that they 
carried off young girls on the banks of rivers. Of three 
other Slavonic tribes, the Radimich, the Viatich, and the 
Severs, it is said :-" Marriage did not exist among them, 
but games were held in the outskirts of the villages. They 
met at these games for dancing and every kind of diabolic 
amusement, and there every man carried off the woman with 
whom he had previously come to an agreement. They had 
even two or three wives."4 This description makes it 
clear that when the Russian monk denied the existence of 
marriage among some pagan tribes he did not use the term 
marriage in the sense given to it by the modern sociologist. 
Professor Kovalewsky says that an Arabic traveller of 
1 Kovalewsky, ' Marriage among the Early Slavs,' in Folk-Lore, 
i. 466. Idem, Modern Customs and Ancient Laws oj Russia, p. 10. 
See also Volkov, ' Rites et usages nuptiaux en Ukraine,' in L'Anthro- 
pologie, ii. 164. 
2 Cosmas of Prague, op. cit. i. 13 (Migne, op. cit. clxvi. 79). 
3 Krek, Einleitung in die slavische Literatur^eschichte, p. 199. 
4 Nestor, Chronique, p. 10. 112 
THE HISTORY OF HUMAN MARRIAGE chap. 
the tenth century, Abu-el-Qassim, speaks of the hetairism 
of the married women as a peculiarity of the national life 
of the Circassians ; and he adds that this statement is con- 
firmed by another Arabic writer, Massudi.1 But the former 
simply wrote that the Circassian women " pass for being 
voluptuous,"2 and the latter that they " are said to be sur- 
prisingly beautiful and very voluptuous." 3 The old traveller 
Tavernier, who is also referred to by Professor Kovalewsky, 
states that the more gallants a Circassian woman has the 
more highly is she respected ; but at the same time he speaks 
of their marriages, describes how they are concluded, and 
adds that if no child is born within a few years the husband 
is permitted to take another wife.4 Nor do other travellers 
speak about a state of promiscuity among the Circassians.5 
Lord Avebury quotes the sexual relations of various 
modem savages as instances of promiscuity or, as he calls it, 
" communal marriage."6 " The Bushmen of South Africa," 
he says, " are stated to be entirely without marriage." He 
does not indicate the source from which he has taken this 
statement; all the authorities I have consulted unanimously 
assert the reverse, and, as we have noticed before, the family 
is in fact the chief social institution of this people. Sir 
Edward Belcher tells us that in the Andaman Islands the 
custom is for the man and woman to remain together until 
the child is weaned, when they separate and each seeks a 
new partner.7 Lord Avebury, however, also draws attention 
to Mr. Man's account of the Andaman Islanders, according 
to which " bigamy, polygamy, polyandry, and divorce are 
unknown " among them,8 but seems to maintain that the 
1 Kovalewsky, ' La famille matriarcale au Caucase,' in L'Anthro- 
pologic, iv. 266. 
2 Quoted by Darinsky, ' Die Familie bei den kaukasischen Vol- 
kern,' in Zeitschr. vergl. Rechtswiss. xiv. 175. 
3 Massudi, * Description du Caucase,' in Klaproth, Magasin 
Asiatique, i. 289. 
4 Tavernier, Six Voyages, i. 339. 5 Darinsky, loc. cit. p. 177 sqq. 
G Avebury, Origin of Civilisation, p. 68 sqq. 
1 Belcher, ' Andaman Islands,' in Trans. Ethn. Soc. N.S. v. 45. 
8 Man, ' Aboriginal Inhabitants of the Andaman Islands,' in 
Jour. Anthr. Inst. xii. 135. nt CRITICISM OF HYPOTHESIS OF PROMISCUITY 
113 
two accounts may refer to different tribes and consequently 
both be true. Mr. Portman expressly says that Belcher's 
statement is " quite incorrect " j1 but even if it were other- 
wise it could not be looked upon as evidence of promiscuity, 
suggesting as it does monogamous marriages of short dura- 
tion. Speaking of the Haida of Queen Charlotte Islands, 
Mr. Poole says that among them " the institution of marriage 
is altogether unknown," and that the women " cohabit 
almost promiscuously with their own tribe, though rarely 
with other tribes."2 From the description of Captain Jacob- 
sen's voyage to the north-western coast of North America, 
however, it appears that marriage exists among those 
Indians although the husbands often prostitute their wives ;3 
and Mr. Swanton, who spent ten months among them, gives 
us various details about their marriages and tells us that 
these were often arranged as soon as a child was born. 
Indeed, if a man were unfaithful after marriage, his mother- 
in-law exacted a large amount of property from him, and if 
the wife were unfaithful the husband generally took personal 
revenge.4 In the Californian Peninsula, it is said, the sexes 
met without any formalities, and the vocabulary of the 
people did not even contain the verb " to marry." 5 But 
surely, the want of an equivalent for the verb " to marry " 
does not imply the want of the fact; and Baegert, the 
authority for the statement in question, indicates himself 
that marriage did occur among those Indians when he says 
that " each man took as many wives as he liked, and if 
there were several sisters in a family he married them all 
together."6 
Lord Avebury observes that in the Pacific Islands, accord- 
ing to Mr. Hyde, there was an " utter absence of what we 
mean by the family, the household, and the husband ; the 
only thing possible was to keep distinct the line through the 
1 Portman, History of Our Relations with the Andamanese, p. 519. 
2 Poole, Queen Charlotte Islands, p. 312. 
3 Woldt, Capitain Jacobsen's Reise an der Nordwestkuste Amerikas, 
pp. 20, 21, 28 sq. 
4 Swanton, Haida, p. 50 sqq. 
5 Baegert, ' Aboriginal Inhabitants of the Californian Peninsula,' 
in Smithsonian Report, 1863, p. 368. " Ibid. p. 368. 114 
THE HISTORY OF HUMAN MARRIAGE chap. 
mother, and enumerate the successive generations with the 
several putative fathers."1 As an evidence of "communal 
marriage," this statement is contradicted by all that we 
know about the marriages of the peoples inhabiting the 
various islands of the Pacific Ocean. It is true that some of 
them have been said to have no marriage, but these state- 
ments have been proved to be erroneous. According to one 
narrator, the women of Easter Island had sexual intercourse 
in a promiscuous manner.2 This statement was probably 
based on the fact that, although most girls were married very 
early in life, there were some who remained unmarried and 
acted as a kind of prostitutes, having even as many as five 
lovers, owing to the great preponderance of men in the 
island.3 Geiseler says that each family lived by itself.4 
With reference to the Tahitians, Forster wrote, " We have 
been told a wanton tale of promiscuous embraces, where 
every woman is common to every man : but when we en- 
quired for a confirmation of this story from the natives, we 
were soon convinced that it must, like many others, be con- 
sidered as a groundless invention of a traveller's gay fancy."5 
Nowhere has debauchery been practised more extensively 
than among the Areois of Tahiti, who have also been accused 
of having their women in common.6 Yet Ellis assures us 
that, " although addicted to every kind of licentiousness 
themselves, each Areoi had his own wife ; . . . and so jealous 
were they in this respect that improper conduct towards 
the wife of one of their own number was sometimes punished 
with death."7 Lord Avebury observes that the original 
Hawaian word for " to marry " meant " to try," and that 
the missionaries have been attempting to replace this by 
our word " marry " under a native form ;8 but Fornander 
has objected to this derivation, and Dr. Emerson has 
1 Avebury, Origin of Civilisation, p. 69. 
2 Quoted by Tregear, ' Easter Island,' in Jour. Polynesian Soc 
i. 99. 
3 Geiseler, Die Oster Insel, p. 29. 4 Ibid. p. 41. 
5 Forster, Voyage round the World, ii. 132. 
G Turnbull, Voyage round the World, p. 304. 
7 Ellis, Polynesian Researches, i. 239. 
" Avebury, op. oil. p 6g. hi CRITICISM OF HYPOTHESIS OF PROMISCUITY 
115 
pointed out that the woid meaning " to try " is ho-a'o and 
not ho-ao.1 Contrary to the view that Hawaian society 
was characterised by relations between the sexes which 
approached a condition of complete promiscuity, Dr. Rivers 
maintains that the evidence clearly shows " that in Hawaian 
society in its ancient condition there existed the institution 
of individual' marriage, though undoubtedly accompanied 
by much sexual laxity and with the possession of marital 
rights by others than the husband."2 Lisiansky wrote 
more than a hundred years ago :-" Former voyagers have 
asserted, that the Marquesan men and women have no 
individual attachments, but cohabit promiscuously, as 
inclination may dictate. This, however, is a mistake : the 
marriage state is held nearly as sacred among them as it is 
among any uncivilised people. It is true indeed, fathers 
sometimes offered us their daughters, and husbands their 
wives; but this proceeded from their ardent desire of 
possessing iron, or other European articles. . . . This out 
of the question, jealousy is so prevalent with the men, that 
upon the smallest suspicion of infidelity, they punish their 
wives with severity."3 In a quite recent book of travels 
Mr. Rannie states that he was told by the natives of New 
Ireland themselves that there is no marriage among them ;4 
but this statement is disproved not only by what we arc 
told by other travellers, but by some facts mentioned in his 
own description of those islanders. 
Lord Avebury quotes certain Australian facts as evidence 
for his theory. With these I shall deal in the chapter on 
group-marriage ; for what they prove, or are supposed to 
prove, is not promiscuity or " communal marriage," but 
sexual relations, or marriage, between a definite group of 
men and a definite group of women within the community. 
I presume that no authority on the Australian aborigines 
would deny the accuracy of Mr. Curr's statement that in 
none of their tribes men and women have been found living 
in a state of promiscuous filtercourse, the reverse being a 
1 Rivers, History of Melanesian Society, i. 383. 
2 Ibid. i. 385 sq. 3 Lisiansky, Voyage round the World, p. 82. 
1 Rannie, My Adventures among South Sea Cannibals, p. 265. 116 
THE HISTORY OF HUMAN MARRIAGE chap. 
matter of notoriety.1 Indeed, nowhere is marriage subject 
to stricter rules than among those savages. Lord Avebury 
also mentions the Nayars, Tottiyars, and Todas of Southern 
India; but the statements he refers to only speak of poly- 
andry or group-marriage, and will be considered under these 
forms of marriage. He also says, quoting Watson and Kaye, 
thattheTeehursof Oude live together almost indiscriminately 
in large communities, and that even when two persons are 
regarded as married the tie is but nominal.2 With reference 
to this alleged instance of promiscuity, Mr. Crooke observes:- 
" There is no evidence whatever that anything like com- 
munal marriage prevails among them. The fact seems to be 
that by the necessities of their occupation the husbands leave 
their wives for long periods at a time and go on voyages as 
far as Calcutta. That a high standard of female morality is 
maintained during their absence it would be rash to assert: 
but this is very different from communal marriage."8 
Bernhoft mentions two other peoples of India, the Irulas 
and Kurumbas, as instances of savages who have no mar- 
riage.4 Of the former, Harkness writes that they " have no 
marriage contract, the sexes cohabiting almost indiscrimi- 
nately ; the option of remaining in union, or of separating, 
resting principally with the female. Some among them," he 
adds, " the favourites of fortune, who can afford deliberately 
to expend the sum of four or five rupees on festivities, will 
celebrate their union, by giving a feast to all their friends 
and neighbours."5 That they are not without marriage also 
appears from the following statement made by Ward :- 
" In their marriages perhaps they are singular. This con- 
tract does not take place between the parties cohabiting till 
the second or third child is bom, when the man agrees to pay 
a stipulated sum by instalments as a dowry to the friends of 
1 Curr, Australian Race, i. 126. 
2 Watson and Kaye, People of India, ii. no. 85. 
3 Crooke, Tribes and Castes of the North-Western Provinces and 
Oudh, i. p. clxxxiii. sq. 
4 Bernhoft, ' Altindische Familienorganisation,' in Zeitschr. f. 
vergl. Rechtswiss. ix. 14, quoting Ritter, Erdkunde, p. 1015 sq. 
° Harkness, Description of a Singular Aboriginal Race inhabiting 
the Neilgherry Hills, p. 92. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
117 
Hie woman, who give with her as a portion a buffalo ; the 
contract now becomes binding."1 Of the Kurumbas 
Harkness simply says, " They have no marriage ceremony ; 
but occasionally, when two have been living together for 
some time, they will enter into an agreement, in the presence 
of friends, to remain united for life."2 To be without a 
marriage ceremony is certainly not the same as to be without 
marriage. 
Bastian adds to the list the Keriahs and the Chittagong 
tribes, and some American Indians-the Guaycurus, the 
Arawaks, and the Kutchin.3 Concerning the Keriahs, 
Colonel Dalton only asserts that they have no word for 
marriage in their own language, but he does not deny that 
marriage itself occurs among them; on the contrary, it 
appears that they buy their wives.4 The Chittagong Hill 
tribes, as we shall find later on, practise mostly monogamy. 
Anyone who takes the trouble to read Richardson's,5 
Kirby's,6 or Bancroft's7 account of the Kutchin will notice 
that polygyny, but not promiscuity, is prevalent among 
them, and that the husbands are very jealous of their wives. 
The same is stated by v. Martius about the Arawaks, whose 
blood-feuds are generally due to jealousy and a desire to 
avenge violations of conjugal rights ;8 and the occurrence of 
marriage among them has also been ascertained by Schom- 
burgk9 and Brett.10 Of the Guaycurus, Lozano says that the 
men have not more than one wife, but that nevertheless 
there are no proper marriages among them, since the husband 
may part with his wife and the wife with her husband without 
1 Ward, in Grigg, Manual of the Nilagiri District, Appendix, 
p. Ixxviii. 2 Harkness, op. cit. p. 129. 
3 Bastian, ' Ueber die Eheverhaltnisse,' in Zeitschr. f. Ethnol. 
vi. 406. 
4 Dalton, ' The " Kols " of Chota Nagpore,' in Trans. Ethn. Soc. 
N.S. vi. 25. 
5 Richardson, Arctic Searching Expedition, i. 383. 
0 Kirby, ' Journey to the Youcan,' in Smithson. Rep. 1864, p. 419. 
7 Bancroft, Native Races of the Pacific States, i. 131. 
8 v. Martins, Beitrage zur Ethnographic Amerika's, i. 693. 
9 Schomburgk, Reisen in Britisch-Guiana, xi. 459 sq. 
10 Brett, Indian Tribes of Guiana, p. 98, 118 
THE HISTORY OF HUMAN MARRIAGE chap. 
difficulty and without disgrace.1 That they have marriages 
appears from other accounts as well, and it is even said that 
separations between husbands and wives are rare among 
them.2 Bastian further states that the Jolah on the island 
of St. Mary, according to Hewett, possess their women in 
common,3 and that, according to Magalhaes, the like is true 
of the Kahyapo of Matto Grosso.4 The former of these 
statements I have been unable to check. As regards the 
Kahyapo, Magalhaes says that their " communism " is of 
the following kind. When a woman reaches the age at which 
she is allowed to enter into relations with the other sex she 
may become pregnant by any man she likes. During the 
period of pregnancy and suckling she is supported by the 
father of the child, who also can have intercourse with the 
other women dwelling in the same hut. Subsequently the 
woman is at liberty to conceive again by the same man, 
or she can do so by another one ; but if she does the latter, 
the duty of maintaining the earlier offspring is transferred 
to the new father.5 This can certainly not be called pro- 
miscuity. Krause says that monogamy seems to prevail 
among the Kahyapo.6 
Garcilasso de la Vega asserts that among the natives of 
Passau, in Peru, before the time of the Incas, men had no 
separate wives. He assures us that he saw those Indians 
with his own eyes when on his way to Spain, as the ship 
stopped on their coast for three days.1 Speaking of the 
Yahgans of Tierra del Fuego, Admiral Fitzroy says, " We 
had some reason to think there were parties who lived in 
a promiscuous manner-a few women being with many 
1 Lozano, Description chorographica del tcrrcno, &c. de las Provin- 
cias del Gran Chaco, Gualamba, p. 70. 
2 Charlevoix, History of Paraguay, i. 91 sq. Sanchez Labrador, 
El Paraguay Catolico, ii. 24 sqq. do Prado, ' Historia dos Indios 
Cavalleiros, da Nagao Guaycuru,' in 0 Patriota, 1814, no. 4, p. 20 sq. 
3 Bastian, Rechtsverhaltnisse, p. Ixi. n. 36. 
4 Idem, Die Culturldnder des alien America, ii. 654 n. 4. 
5 Couto de Magalhaes, Trabalho preparatorio para aproveitamento 
do selvagem e do solo por elle occupado no Brazil. 0 selvagem, p. in. 
c Krause, In den Wildnissen Brasiliens, p. 401. 
' Garpilasso dp la- Vega, Royal Commentaries of the Yncas, ii. 443. hi CRITICISM OF HYPOTHESIS OF PROMISCUITY 
119 
men."1 With reference to this statement Mr. Bridges, who had 
lived amongst them for thirty years, wrote to me, " Admiral 
Fitzroy's supposition concerning parties among the natives 
who lived promiscuously is false, and adultery and lewdness 
are condemned as evil, though through the strength of 
animal passions very generally indulged, but never with the 
consent of husbands or wives, or of parents." 
According to Wilken, there is no marriage among the 
Lubus of Sumatra, the Poggy Islanders off the west coast of 
that island, the Olo Ot together with a few other tribes of 
Borneo, the mountaineers of Peling east of Celebes, and the 
Orang Sakai, Orang Semang, and Orang Benua of Malacca.2 
None of these statements can be trusted. Van Ophuijsen 
assures us that among the Lubus a man has to give presents 
to the father of the girl he marries.3 Of the Poggy Islanders, 
Crisp wrote at the end of the eighteenth century, " Simple 
fornication between unmarried persons, is neither a crime 
nor a disgrace : and a young woman is rather- liked the 
better, and more desired in marriage, for having borne a 
child."4 This implies that marriage prevailed among them ; 
and the same is testified by later authorities. According to 
Rosenberg 5 and Hollander,6 a man has only one wife, 
whom he buys from her parents, divorces are unknown, 
and hi the case of adultery both culprits are punished with 
death. Among the Olo Ot, according to Schwaner, the will 
of the girl plays the most important part at the conclusion 
of a marriage ; he adds that the marriage tie is very loose 
among them, but says expressly that the statement which 
denies the existence of marriage among them is based on 
1 King and Fitzroy, Voyages of the Adventure and Beagle, ii. 182. 
2 Wilken, in De Indische Gids, 1880, vol. ii. 610 sq. Idem, Over de 
verwantschap en het huwelijks- en erfrecht bij de volken van het maleische 
ras, pp. 20, 82 note. 
3 van Ophuijsen,.' De Loeboes,' in Tijdschrift voor indische taal-, 
land- en volkenkunde, xxix. 97 sq. 
4 Crisp. ' Account of the Inhabitants of the Poggy Islands,' in 
Asiatick Researches, vi. 87 sq. 
3 Rosenberg, Der malayische Archipel, p. 199. 
0 Hollander, Handleiding bij de beoefening der land- en volkenkunde 
van Ncderlandsch Oost-Indie, i. (1895) p. 615. 120 
THE HISTORY OF HUMAN MARRIAGE chap. 
hearsay only.1 As for the mountaineers of Peling, it is 
significant that, in a later edition of the very work from 
which Wilken derived his information, the sentence in which 
marriage is said to be unknown among them 2 has been 
excluded.3 Wilken also refers to a statement of a person 
who has several times visited Peling and each time stayed 
there for a considerable period, to the effect that the moun- 
taineers there " on the whole have no idea of the marriage 
tie : men and women meet each other only temporarily, 
and mingle without any shame before the eyes of all."4 
But M. de Clercq states, on the other hand, that it is entirely 
untrue that the islanders of Peling have no marriage ; they 
usually have one wife, whom they acquire by purchase.5 
Miklucho-Maclay wrote of the Orang Sakai:-" Communal 
marriage exists, it appears, among the Orang Sakai; at 
least I must conclude so from a great number of accounts. 
A girl having been married to a man for some days or weeks 
goes, with his consent, and voluntarily, to live for a shorter 
or longer period with another man. She thus goes in turn 
to all the men of the party until she comes back to her first 
husband; she does not remain with him however but 
continues to engage in such temporary marriages, which are 
regulated by chance and by her wishes. She is however 
considered the wife of the man who first took her."6 Mr. 
Skeat observes that this is " the only notice of such a 
custom, and resting as it does on second-hand evidence or 
worse, cannot be accepted without corroboration." 7 
1 Schwaner, Borneo, i. 230, 231 note :-" De Koeteinezen verhalen, 
dat hunne Ot geene huwelijken sluiten, geen woningen hebben, 
en als de dieren des wouds door hen gejaagd worden." 
2 Hollander, op. cit. ii. (1864), P- 236 •-" Zelfs het huwelijk ie 
bij hen niet bekend." 
8 The same work, 5th edition, revised by R. van Eck, ii. (1898) 
p. 235. 4 Wilken, in De Indische Gids, 1880, vol. ii. 610. 
6 de Clercq, Bijdragen tot de kennis der residentie Ternate, p. 131. 
G Miklucho-Maclay, ' Ethnological Excursions in the Malay 
Peninsula,' in Jour. Roy. Asiatic Soc. Straits Branch, no. ii. 215. 
7 Skeat and Blagden, Pagan Races of the Malay Peninsula, ii. 56 
n. 2. As regards the untrustworthiness of Malay evidence see 
Favre, Account of the Wild Tribes inhabiting the Malayan Peninsula, 
p. 47, and Annandale and Robinson, pasciculi Malayenses, i. 6. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
121 
Miklucho-Maclay admits himself that his meetings with the 
Orang Sakai were too short to enable him to say much on 
the subject of their mode of living and customs, and that he 
only received his information about their " communal 
marriage " from Malays in Pahang and members of the 
Catholic Mission at Malacca.1 According to Maxwell, the 
stringency which, attaches to the marriage law of the 
Sakai is astounding, and the punishment for adultery is 
death, usually carried out by a relative.2 M. Pleyte, who 
in a violent attack upon my criticism of Wilken's statements 
in the first edition of the present work accuses me of having 
evaded those relating to the Orang Sakai and the moun- 
taineers of Peling,3 could thus no longer maintain that these 
statements are better substantiated than the rest. With 
regard to the Semang, Mr. Skeat writes that they are, as 
far as he could learn, habitually monogamists, and that he 
failed to obtain any sort of. evidence in support of the 
statement which has been more than once advanced, namely, 
that their women were in common like their other property.4 
" This idea," he says, " of the laxity of the marriage-tie 
among the Negritos may possibly arise from the great 
ante-nuptial freedom which appears to be allowed, but there 
is every reason to believe that when once married the 
Semang of both sexes are in the highest degree faithful to 
each other and that cases of unfaithfulness are exceedingly 
rare. That conjugal infidelity is strongly discountenanced 
is shown by the penalty assigned to it."5 Another first-class 
authority, Dr. Martin, writes :-" So far as my own experience 
goes, the pure Senoi and Semang tribes are thoroughly 
monogamous-again a fresh evidence against the formerly 
so highly cherished hypothesis of promiscuity in primitive 
1 Miklucho-Maclay, loc. cit. pp. 211, 215 n. 16. 
2 [Maxwell, in] Jour. Roy. Asiatic Soc. Straits Branch, no. i. 112. 
3 Pleyte, in a review of the first edition of the present work in 
De Indische Gids, 1891, p. 2048. 
4 Newbold wrote (Account of the British Settlements in the Straits of 
Malacca, ii. 379), " The Semang women like those of the ancient 
Massagetae, and the more modern Tartar Kie-Kia-sse tribes, are 
said to be in common like their other property." 
5 Skeat and Blagden, op. cit. ii. 55 sq. See also ibid. ii. 59. 122 
THE HISTORY OF HUMAN MARRIAGE chap. 
forms of human life. Most of the earlier observers also agree 
with this opinion." Polygyny, he adds, has been found 
only among border tribes who have come into closer contact 
with the Malays.1 Among the pure tribes adultery is 
generally punished with death,2 and among the Semang, 
according to Vaughan Stevens, it very rarely happens that 
a man has intercourse with another man's wife.3 Of the 
Semang of Ijoh, Swettenham says not only that they as a 
rule have one wife, but that there is no divorce among them.4 
Wilken's statement that the Orang Benua have no marriage 
-repeated by M. Pleyte and Dr. R. Schmidt5-is likewise 
contradicted by the information given by many authorities, 
who speak about their marriages.6 We are told that poly- 
gamy is prohibited among them7 or exceptional,8 that 
adultery is very rare,9 and that any married person surprised 
in it might be put to death, though if a woman so surprised 
could prove that she was seduced she would not be put to 
death but would be sent away by her husband.10 According 
to Newbold, a man can divorce his wife and take another,11 
according to Favre he lost the dowry given to her in case 
he divorced her,12 according to Vaughan Stevens divorce docs 
not occur among them.13 
M. Pleytc,14 partly followed by Dr. Schmidt,15 and van der 
1 Martin, Die Inlandstamme der Malayischen Halbinsel, p. 864 sq. 
2 Ibid. p. 874. 
3 Stevens, Materialien zur Kenntniss der wilden Stamme auj dcr 
Halbinsel Maldka, ii. 132. 
1 Swettenham, ' Comparative Vocabulary of the Dialects of 
some of the Wild Tribes inhabiting the Malayan Peninsula,' in Jour. 
Roy. Asiatic Soc. Straits Branch, no. v. 156. 
Schmidt, Liebe und Ehe im alten und modemen Indien, p. 331. 
11 Skeat and Blagdon, op. cit. ii. 79 sqq. Newbold, op. cit. 
ii. 4,07 sq. 
' Newbold, op. cit. ii. 408. Favre, in Journal Indian Archi- 
pelago, ii. 269. 
8 Logan, ' Orang Binua of Johore,' in Journal Indian Archipelago, 
i. 270. Griinwedel, ' Die Reisen des Hrn. Vaughan Stevens in 
Malacca,' in Verhandl. Berliner Gesellsch. Anthrop. 1891, p. 833. 
11 Griinwedel, loc. cit. p. 833. 
11 Favre, loc. cit. p. 269. 11 Newbold, op. cit. ii. 408. 
12 Favre, loc. cit. p. 269. 13 Griinwedel, loc. cit. p. 833. 
14 Pleytc, loc. cit. p. 2049 sq. 15 Schmidt, op. cit. p. 330 sq. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
123 
Lith,1 mentions among peoples living in promiscuity the 
Orang Benua of Johor, the Orang Gunung of Biliton, and the 
Dyaks of Sidin, because husbands sometimes exchange their 
wives-which, so far as I can see, presupposes that they 
must have wives to exchange-as also the Dyaks of Sing- 
kawang in Western Borneo, who on the authority of Adriani 
are said to have communal marriage. I have entered into 
all these details because M. Pleyte has accused me of using 
indirect means in my criticism,2 by which he probably means 
that I did not deal with all the statements made by Wilken. 
My subsequent researches have only strengthened my con- 
viction of their exceedingly uncritical character. For the 
benefit of future champions of the hypothesis of promiscuity 
I shall add the following statement, which seems to have 
escaped the vigilance of my Dutch antagonists. Friar 
Odoric of Pordenone, who visited Sumatra in the earlier 
part of the fourteenth century, says that somewhere in that 
island " all the women be in common ; and no one there 
can say, this is my wife, or this is my husband ! But when 
a woman beareth a boy or a girl she giveth the child to whom 
she listeth of those with whom she hath consorted, and 
calleth him the father. The whole of the land likewise is 
in common."3 I cannot prove that the Friar was wrong, 
but he does not convert me. 
Wilutzky asserts that individual marriage is unknown both 
among the forest tribes in the interior of Malacca and " to 
a large extent " (in weitem Umfang) in Africa, for instance in 
Darfur and among the Kafirs.4 That such a statement 
can occur in a book with scientific pretensions written in the 
twentieth century is almost incredible. Otherwise Africa 
has, since the days of ancient Greece and Rome, contributed 
very little to the lists of peoples put together by the advocates 
of primitive promiscuity. Giraud-Teulon5 refers to Dapper's 
1 van der Lith, Spaan, Fokkens, and Snellcman, Encyclopaedic 
van Nederlandsch-Indie, ii. 63. 
2 Pleyte, loc. cit. p. 2051. 
3 ' Travels of Friar Odoric of Pordenone,' in Yule, Cathay and 
the Way thither, ii. 147 sq. 
1 Wilutzky, op. cit. i. 20. 5 Giraud-Teulon, op. oil. p. 72. 124 
THE HISTORY OF HUMAN MARRIAGE chap. 
old description of the kingdom of Bomu, in which the 
people are said to have neither law, nor religion nor any 
proper names, and to possess their women and children in 
common, and the king is said to be so rich that all his 
utensils are made of gold.1 This does not sound very con- 
vincing. Dr. Post has found no people in Africa living in 
a state of promiscuity.2 
I have now mentioned all the cases which are known to 
me of peoples said to live in a state of promiscuity. I 
think that it would be difficult to find a more untrust- 
worthy collection of statements. Some of them are simply 
misrepresentations of theorists in which sexual laxity, 
frequency of separation, polyandry, group-marriage or some- 
thing like it, or absence of a marriage ceremony or of a 
word for " to marry " or of a marriage union similar to our 
own is confounded with promiscuity. Others are based 
upon indefinite evidence which may be interpreted in one 
way or another, or on information proved to be inaccurate. 
And not a single statement can be said to be authoritative or 
even to make the existence of promiscuity at all probable 
in any case. That no known savage people nowadays is, 
or recently has been, living in such a state is quite obvious, 
and this greatly discredits the supposition that promiscuity 
prevailed among any of the peoples mentioned by classical 
or mediaeval writers in their summary and ambiguous 
accounts. 
But even if we had reason to believe that a few peoples 
really have had nothing but promiscuity, it would be a 
great mistake to infer that these utterly exceptional cases 
represented a stage of human development which mankind 
as a whole has gone through. Further, nothing would 
entitle us to consider this promiscuity as a survival of the 
primitive condition of man, or even as a mark of a very 
rude state of society. It is by no means among the very 
lowest races that sexual relations most nearly approach 
to promiscuity : we shall find that many or most of them 
are completely or almost completely monogamous, and 
1 Dapper, Description de l'Afrique, p. 223. 
8 Post, Afrikanische Jurisprudent, i. 304. in CRITICISM OF HYPOTHESIS OF PROMISCUITY 
125 
that among some of them even divorce is said to be unknown. 
Nowhere seems the fidelity of married women to be less 
insisted upon than among various polyandrous peoples who 
are pastoral or agricultural in their habits and certainly 
cannot be looked upon as representatives of the lowest 
type of humanity known to us. Mr. Rowney states that 
among the Bhutias the marriage tie is so loose that chastity 
may be said to be quite unknown, that the husbands are 
indifferent to the honour of their wives, that " the inter- 
course of the sexes is, in fact, promiscuous." But the 
Bhutias are followers of Buddha, and " can hardly be counted 
among the wild tribes of India, for they are, for-the most 
part, in good circumstances, and have a certain amount of 
civilisation among them."1 Dr. Jochelson writes in his 
excellent monograph on the Koryak :-" Sociologists who 
think that all mankind, without exception, have passed 
through the so-called period of promiscuity as a necessary 
stage in the evolution of marital relations, might find this 
stage in the free morals prevalent in the hamlets of the 
Russians or Russianised natives of north-western Siberia. 
It is difficult to find a girl that has reached or even ap- 
proached the age of sexual maturity that is innocent. . . . 
On the Kolyma, where often several families live together 
in one house, it is difficult to say who is whose wife."2 A 
very different state of morals prevails among the Koryak 
who have not been adulterated by the bearers of Russian 
civilisation. 
To sum up : even if there really are or have been peoples 
living in a state of promiscuity, which has never been proved 
and is exceedingly hard to believe, these peoples do not 
afford any evidence whatever for promiscuity having been 
the rule in primitive times. 
1 Rowney, Wild Tribes of India, pp. 142, 143, 140. 
2 Jochelson, Koryak, p. 733 sq. CHAPTER IV 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : 
PRE-NUPTIAL UNCHASTITY 
It is argued that promiscuity is by no means restricted 
to those peoples who are said to have nothing else. Side 
by side with marriage it is found among savages in all parts 
of the world, and very frequently not as a mere fact but as 
a practice permitted by custom. This, we are told, shows 
that sexual intercourse must originally have been unchecked.1 
It is a well-known fact that among many uncivilised 
peoples both sexes enjoy perfect freedom previous to 
marriage. Instances of this have been given by myself 2 
as well as by other writers,3 and I could fill pages with 
fresh materials at my disposal. If we look at the facts a 
little more closely, however, we soon find that many of 
them could not, in any circumstances, be regarded as 
relics of primitive promiscuity-either because they are 
known to be of later growth, or because they do not repre- 
sent promiscuity at'all. 
1 Post, Die Grundlagen des Rechts, p. 187. Wilken, ' Over de 
primitieve vormen van het huwelijk en den oorsprong van het 
gezin,' in De Indische Gids, 1880, vol. ii. 1195. Wilutzky, Vorge- 
schichte des Rechts, i. 26 sqq. Bloch, Sexual Life of Our Time, p. 189 
sqq. Corijj, Mating, Marriage, and the Status of Woman, p. in sq. 
2 Origin and Development of the Moral Ideas, ii. 422 sqq. 
3 See, e.g., Post, Grundriss der ethnologischen Jurisprudenz, i. 
21 sqq. ; Hartland, Primitive Paternity, ch. vi. ; Hobhouse, Wheeler, 
and Ginsberg, Material Culture and Social Institutions of the 
Simpler Peoples, p. 176 sqq. 
126 ch. iv CRITICISM OF HYPOTHESIS OF PROMISCUITY 
127 
In various cases we are told that the wantonness of 
savages is chiefly due to contact with civilised races/ The 
pioneers of a " higher civilisation " are very frequently 
unmarried men who go out to make their living in uncivilised 
lands, and though unwilling to contract regular marriages 
with native women, they may have no objection to corrupt- 
ing their morals. It is strange to hear from a modern student 
of anthropology, and especially from an Australian writer, 
that in sexual licence the savage has never anything to 
learn, and that " all that the lower fringe of civilised men 
can do to harm the uncivilised is to stoop to the level of the 
latter instead of teaching them a better way."1 Mr. Edward 
Stephens has a very different story to tell with reference to 
the tribes which once inhabited the Adelaide Plains in South 
Australia. "As a rule, to which there are no exceptions," 
he says, " if a tribe of blacks is found away from the white 
settlement, the more vicious of the white men are most 
anxious to make the acquaintance of the natives, and that, 
too, solely for purposes of immorality. ... I saw the 
natives and was much with them before those dreadful 
immoralities were well known, . . . and I say it fearlessly, 
that nearly all their evils they owed to the white man's 
immorality and to the white man's drink."2 Mr. Curr 
observes that prior to the coming of the whites the Ban- 
gerang in Victoria as a rule " enforced constancy on the 
part of their wives, and chastity on their unmarried 
daughters."3 
Speaking of some cannibals in New Guinea, Dr. Chalmers 
remarks :-" Why savages should be always spoken of as 
immoral I fail to see. They are not so when compared 
with the more highly civilised countries of the world. 
I am sorry to have to say that it is contact with the 
civilised white that demoralises them, and they then become 
loose and immoral."4 From Fiji "there is a mass of 
1 Sutherland, Origin and Growth of the Moral Instinct, i. 186. 
2 Stephens, ' Aborigines of Australia,' in lour. Roy. Soc. N. S. 
Wales, xxiii. 480. 
3 Curr, Recollections of Squatting in Victoria, p. 249. 
* Chalmers, Pioneer Life and Work in New Guinea, p. 112 128 
THE HISTORY OF HUMAN MARRIAGE chap. 
evidence to show that in heathen times the majority of girls 
were virgins until they married or entered into concubinage, 
because the law of custom allowed them no opportunities 
for secret amours ; whereas, after fifty years of individual 
freedom, it is extremely rare for a girl to preserve her virtue 
to the age of eighteen." It was with the introduction of 
Christianity, says Sir Basil Thomson, that there came a 
change;1 earlier writers speak of the bad influence of Euro- 
pean sailors.2 When visiting the Sandwich Islands with 
Cook, Vancouver saw little or no appearance of wantonness 
among the women. But when he visited them some years 
afterwards it was very conspicuous ; and he ascribes this 
change in their habits to the intercourse with foreigners.3 
Owing to the same influence, it is said, the women of Ponape 
(Caroline Islands) and Tanna (New Hebrides) lost their 
modesty.4 Even in Tahiti, so notorious for the licentious- 
ness of its inhabitants, immorality was formerly less than 
it became subsequently. When a girl who had been be- 
trothed as a child grew up, a small platform of considerable 
elevation was erected for her abode within the dwelling of 
her parents in order that she should not lose her virtue. 
" Here she slept and spent the whole of the time she passed 
within doors. Her parents, or some member of the family, 
attended her by night and by day, supplied her with every 
necessary, and accompanied her whenever she left the 
house. Some of their traditions," Ellis adds, " warrant 
the inference that this mode of life, in early years, 
was observed by other females besides those who were 
betrothed."5 
In Madagascar most of the tribes attach no importance 
to the chastity of a girl.6 But, says Mr. Sibree, " there 
1 Thomson, Fijians. A Study of the Decay of Custom, p. 236 sqq. 
See also ibid. p. 176. 
2 Meinicke, Die Inseln des Stillen Oceans, ii. 45. 
3 Vancouver, Voyage of Discovery, i. 171 sq. 
4 Waitz-Gerland, Anthropologic der Naturvblker, vol. v. pt. ii. 108. 
Brenchley, fottings among the South Sea Islands, p. 208. 
5 Ellis, Polynesian Researches, i. 270. 
6 Sibree, The Great African Island, p. 252. Grandidier, Ethno- 
graphic de Madagascar, ii. 134 sqq. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
129 
are some other tribes, more isolated, as certain of the 
eastern peoples, where a higher standard of morality pre- 
vails, girls being kept scrupulously from any intercourse 
with the other sex until they are married."1 M. Grandidier 
observes that in certain parts of that island the girls were 
in former days much chaster than after they came into 
contact with Europeans.2 The Bantu Kavirondo were 
much more moral than most of the other tribes of the 
Uganda Protectorate " before they became corrupted by 
Swahili porters from the coast, Indians, and white men."3 
Admiral Fitzroy observed that the unchastity of the 
Patagonian women did not correspond with the pure 
character attributed to them at an earlier time by Falkner, 
and he thinks that " their ideas of propriety may have been 
altered by the visits of licentious strangers."4 A more 
recent traveller, Captain Musters, observed indeed little 
immorality amongst the Indians while in their native 
wilds..6 Among the Huitoto Indians of the Putumayo 
region, says Hardenburg, " the women are naturally chaste, 
and it was not until the advent of the rubber-collectors 
that they began to lose this primitive virtue, so generally 
met with among people not yet in contact with white 
men."6 Concerning the Pima Indians of Arizona, Mr. 
Russell says that before they came in contact with " civilisa- 
tion " chastity was the rule among the young women.7 
Among the Yokut of California the freedom of the un- 
married people of both sexes is very great now, but they are 
said to have been comparatively virtuous before the arrival 
of the Americans.8 In British Columbia and Vancouver 
Island, according to Lord, breaches of chastity on the 
part either of married or unmarried females were in primitive 
times often punished by death among the interior tribes, 
1 Sibree, op. cit. p. 252. 2 Grandidier, op. cit. ii. 137. 
3 Johnston, Uganda Protectorate, p. 746. 
4 Ring and Fitzroy, Narrative of the Voyages of the Adventure 
and Beagle, ii. 173. 
6 Musters, At Home with the Patagonians, p. 197. 
6 Hardenburg, Putumayo, p. 154. 
7 Russell, ' Pima Indians,' in Ann. Rep. Bur. Ethnol. xxvi. 182. 
8 Powers, Tribes of California, p. 381. 130 
THE HISTORY OF HUMAN MARRIAGE chap. 
whereas among the fish-eaters of the north-west coast 
" it has no meaning, or, if it has, it appears to be utterly 
disregarded."1 Among the Queen Charlotte Islanders the 
present depravation has, according to Captain Jacobsen, 
been caused by the gold diggers who went there in the middle 
of the last century.2 In Greenland, says Dr. Nansen, 
" the Eskimo women of the larger colonies are far freer in 
their ways than those of the small outlying settlements 
where there are no Europeans."3 
From his experience in Siberia Dr. Jochelson has come 
to the conclusion that " every contact between representa- 
tives of civilised nations and primitive tribes lowers the 
ideal of the latter by destroying those customs which used to 
restrict the freedom of sexual relations among these tribes." 
This is shown by the striking contrast between the relations 
of the sexes among the Koryak, where there is a lack of 
Russian influence, and those prevailing among the neigh- 
bouring tribes-the Kamchadal, Chukchee, Yukaghir, and 
Tungus. Among the Koryak, girls before marriage must 
not have sexual intercourse with men ; this rule is pretty 
strictly observed by them. Young men will not " serve " 
for a dissolute girl; and " should a girl become pregnant 
before marriage it is considered shameful and her parents 
scold her. ... In olden times cohabitation out of wed- 
lock with a girl sometimes led to wars between the families 
to which the young people belonged."4 Vambery writes, 
" The difference in morality which exists between the 
Turks affected by a foreign civilisation and kindred tribes 
inhabiting the steppes, becomes very conspicuous to any 
one living among the Turkomans and Kara-Kalpaks ; for 
whether in Africa or Asia, certain vices are introduced only 
by the so-called bearers of culture."6 Mr. Endle praises 
1 Lord, The Naturalist in Vancouver Island, ii. 233. 
2 Woldt, Capitain Jacobsen's Reise an der Nordwestkuste Amerikas, 
p. 28. 
3 Nansen, First Crossing of Greenland, ii. 329. 
4 Jochelson, Koryak, p. 733 sqq. Idem, Yukaghir, p. 62. See 
also Bogoras, Chukchee, p. 572. 
6 Vambery, Die primitive Cultur des turko-tatarischen Volkes, 
p. 72. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
131 
the Kacharis of Assam for their chastity before marriage 
and their faithfulness to their marriage vows in after-life ; 
but he adds that this holds good of the Kachari in his simple, 
patriarchal, village life, and there only. When contaminated 
by civilisation, much of his innocence disappears : "of this 
sad deterioration of character any man who has been long 
in the country, and learnt to know the people well, must 
have experienced many melancholy and painful illustra- 
tions."1 On the other hand, Mr. Gait, opposing my state- 
ment that contact with a higher culture has proved per- 
nicious to the morality of savage peoples, maintains that 
" on the whole, there can be no doubt that the relations 
of the sexes in India are steadily becoming more regular."2 
The evidence he produces chiefly shows that polyandry is 
dying out, or has died out; but as polyandry is commonly 
combined with general sexual laxity its disappearance may 
have raised the standard of pre-nuptial chastity. It also 
seems that Hindu civilisation has in some degree exercised 
a chastening influence on the Mongolian and Dravidian 
population of India. When speaking of contact with a 
" higher civilisation " I am chiefly thinking of contact with 
Europeans. 
The facts I have quoted are certainly sufficient to prove 
that the pre-nuptial freedom which is found among so many 
simple peoples is not always primitive. Nor must it be 
identified with promiscuity. It does not mean that an un- 
married woman is constantly changing her lovers or an 
unmarried man the objects of his love, or that they can do 
so without reproach. Sexual connections between a boy 
and a girl are very frequently a preliminary to their marriage. 
They may be a regular method of courtship, or they may be 
a trial before establishing more permanent relations. 
Among the Yukaghir, in spite of the sexual licence which 
prevails among their young people, the word aya'bol', 
which is used for a youth who courts several girls, and for 
a girl who is accessible, is considered an insult, and for such 
a girl, according to the songs, the lads do not serve. " The 
1 Endle, Kacharis, p. 3. 
2 Gait, Census of India, 1911, vol. i. (India) Report, p. 248 sq. 132 
THE HISTORY OF HUMAN MARRIAGE chap. 
time of ' free love ' among young people," says Dr. Jochelson, 
" may be regarded as a period of trial."1 Of various tribes 
in India, such as the Tipperahs,2 Oraons,3 and Kolyas,4 
it is said that unmarried girls may cohabit freely with 
young men, but are never found living promiscuously with 
them. The Toungtha of the Chittagong Hills draw " a 
strong distinction between a woman prostituting herself 
habitually as a means of livelihood, and the intercourse 
by mutual consent of two members of opposite sexes, lead- 
ing, as it generally does, to marriage."5 Of some other tribes 
in India we are told that they " allow a probationary period 
of cohabitation," no stigma, however, attaching to the girl 
if this does not culminate in marriage.6 
Among the Sea Dyaks promiscuous immorality is said 
to be unknown. It is true that very often a girl is with 
child before her marriage, but the father generally acknow- 
ledges the child and marries the woman ; intercourse often 
takes place between those who have been betrothed, but 
not formally married, simply to ascertain if the marriage 
will be fruitful.7 Among the Bontoc Igorot of Northern 
Luzon " marriage never takes place prior to sexual intimacy, 
and rarely prior to pregnancy " ; and although it is custom- 
ary for a young man to be sexually intimate with one, two, 
three, or even more girls at the same time, " a girl is almost 
invariably faithful to her temporary lover."8 Among the 
Southern Massim in British New Guinea, according to 
Dr. Seligman, marriage does not normally take place except 
after more or less prolonged sexual connections ; and he 
1 Jochelson, Yukaghir, pp. 62, 63, 65, 66, 87. 
2 Dalton, Descriptive Ethnology of Bengal, p. 203. 
3 Ibid. p. 248. 
4 Watt, ' Aboriginal Tribes of Manipur,' in Jour. Anthr. Inst. 
xvi. 358. 
5 Lewin, Wild Races of South-Eastern India, p. 193*. 
6 Gait, op. cit. p. 243. 
' Gomes, Seventeen Years among the Sea Dyaks of Borneo, pp. 68, 
69, 127. Brooke Low, quoted by Ling Roth, Natives of Sarawak t 
i. 115. See also St. John, Life in the Forests of the Far East, i. 53 
(Dyaks on the Batang Lupar). 
8 Jenks, Bontoc Igorot, p. 66 IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
133 
believes it can be stated " that love affairs undertaken 
lightly, under conditions which render marriage impossible 
or very unlikely, are in a sense excrescences upon a system 
in which intercourse was in theory at any rate the mode of 
courtship."1 In Kaiser Wilhelm Land, also, marriage is 
as a rule preceded by sexual intimacy.2 But among the 
natives of Mailu in British New Guinea a different custom 
prevails.. Dr. Malinowski tells us that when their young 
men have serious matrimonial plans there is no sexual 
intercourse between the boy and the girl, whereas if he does 
not want to marry her intercourse takes place. Yet the 
great freedom enjoyed by the young people does not mean 
" that there is anything like a promiscuous intercourse or 
even anything • approaching licentiousness in sex matters. 
... A girl who changes her lover often is considered 
decidedly open to blame; a girl once betrothed is bound 
to keep chaste, the same rule of conduct applying to a boy."3 
So also in the Tonga Islands, where unmarried women might 
bestow their favours upon whomsoever they pleased without 
any opprobrium, it was thought shameful for a woman 
frequently to change her lover.4 
Speaking of the absence of restraint in sexual matters 
among the natives of Madagascar, M. Grandidier likewise 
remarks, " Malgre cette liberte excessive des moeurs, il est 
honteux pour une jeune fille de prendre tous les jours un 
nouvel amant; mais . . . lorsqu'il y a profit, il n'y a plus 
ni discredit ni honte." Their marriages are preceded by 
shorter or longer periods of free intercourse, in order that the 
couple shall learn to know each other ; this is the case even 
among the Antimorona, who exact absolute chastity from 
their women but nevertheless consider it necessary that 
there shall be a week's probation previous to the celebration 
of the marriage.5 The Thonga, a Bantu tribe on the eastern 
1 Seligman, Melanesians of British New Guinea, p. 499. 
2 Neuhauss, Deutsch Neu-Guinea, i. 160. 
3 Malinowski, ' Natives of Mailu,' in Trans. Roy. Soc. South 
Australia, xxxix. 559, 561. 
4 Mariner, Natives of the Tonga Islands, ii. 174. 
5 Grandidier, op. cit. ii. 72, 136, 161-163, 186. 134 
THE HISTORY OF HUMAN MARRIAGE chap- 
coast of South Africa, have the following custom, which 
they call gangisa. When boys have gone through the 
puberty rites they ask the girls each to choose one of them, 
and when the girl has made her choice her boy plays with 
her as husband and wife, first in building little huts, and so 
forth, but later on in a less platonic way. In fact, nothing 
is prohibited in the relations between the young people of 
both sexes except that the girls must not become'pregnant. 
If this happens the parents of the girl tell the lover to marry 
her, and if he refuses, the child will belong to the family 
of the girl. At the same time the Thonga have the idea 
" that promiscuity of any kind is a bad and a dangerous 
thing " ; and even in the case of gangisa boys are censured 
when two of them court the same girl.1 Nor can the term 
" promiscuity " be applied to the relations between the 
Masai warriors, who are not allowed to marry, and the 
immature girls, or ditos, of their tribe with whom they 
live. The warrior chooses the dito he fancies, and makes 
her mother numerous small presents; As a rule he keeps 
only one or two ditos at a time, and on returning from war, 
more often than not, continues with the same menage. 
If he is dissatisfied with the dito he has chosen he returns 
her to her mother, and selects another, but this is of rare 
occurrence. When the girl is nearing womanhood, she 
generally leaves the warrior and goes back to her mother. 
But if by chance she remains with him and conceives by 
him, he may make up his mind eve^^ually to marry the 
girl; according to Baumann, it is ew 1 customary for him 
to do so, although he may free himself from it by giving a 
present to her father.2 Among the Warega in the Belgian 
Congo continence before marriage is unknown; but if a 
woman too often changes her lover she loses her good name 
and is called kitazi, that is, prostitute.3 Among the West 
African Pangwe the free intercourse between a youth and 
1 Junod, Life of a South African Tribe, i. 96, 97, 195. 
2 Hinde, The Last of the Masai, pp. 55, 72 sq. Johnston, Uganda 
Protectorate, p. 824. Merker, Die Masai, p. 83. Baumann, Durch 
Massailand, p. 161. 
3 Delhaise, Les Warega, p. 167. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
135 
a girl is in many cases only a trial of shorter or longer 
duration, which if satisfactory leads to marriage.1 
Among many peoples there is a regular marriage upon 
trial before the union becomes definite, the bridegroom 
either taking the girl to his own house or going himself to 
stay with her parents for a certain length of time.2 Some- 
thing of the kind existed even in Scotland prior to the 
Reformation, as a genuine custom known as " hand-fasting." 
" At the public fairs men selected female companions with 
whom to cohabit for a year. At the expiry of this period 
both parties were accounted free; they might either 
unite in marriage or live singly."3 A similar custom 
existed in Ireland, in a very rude form ;4 and the Welsh, 
according to Giraldus Cambrensis, did not marry until 
they had tried, by previous cohabitation, the disposition 
and particularly the fecundity of the person with whom 
they were engaged.5 In an earlier chapter we have seen 
that among many uncivilised peoples the free intercourse 
between unmarried persons generally leads to marriage if 
1 Tessmann, Die Pangwe, ii. 260. 
2 Thurston, Castes and Tribes of Southern India, v. 93 (Muduvars). 
Elton, ' Notes on Natives of the Solomon Islands,' in Jour. Anthr. 
Inst. xvii. 94 sq. Rivers, History of Melanesian Society, i. 383 
(Hawaians). Bastian, Afrikanische Reisen, p. 71 (Congo natives). 
Sarytschew, ' Account of a Voyage of Discovery to the North-East 
of Siberia,' in Collection of Modern and Contemporary Voyages, vi. 76 
(Aleut). Simpson, quoted by Murdoch, 'Ethnological Results 
of the Point Barrow Expedition,' in Ann. Rep. Bur. Ethn. ix. 412 
(Eskimo). Parkman, Jesuits in North America in the Seventeenth 
Century, p. xxxiv. (Hurons). Strachey, Historie of Travaile into 
Virginia Britannia, p. no. James, Indians of the Painted Desert 
Region, p. 228 (Havasupai of Arizona). Grubb, A.n Unknown 
People, p. 214 (Lengua Indians of the Paraguayan Chaco). 
Wilken, Das Matriarchat bei den alten Arabern, p. 21 sq. Ploss- 
Bartels, Das Weib, i. 660 sqq. Potter, Sohrab and Rustem, p. 129 sqq. 
Wilutzky, op. cit. i. 21 sq. 
3 Rogers, Scotland Social and Domestic, p. 109. See also Dalyell, 
Darker Superstitions of Scotland, p. 283. 
4 Gomme, ' Exogamy and Polyandry,' in Archaeological Review, 
i- 391- 
5 Giraldus de Barri, Description of Wales, book ii. ch. 6 {Itinerary 
of Archbishop Baldwin through Wales, vol. ii. 346). 136 
THE HISTORY OF HUMAN MARRIAGE chap. 
the girl becomes pregnant or gives birth to a child, or that 
a seducer or lover in such a case is compelled to marry her 
or otherwise has to pay a fine.1 All this presupposes that 
the father of the child is known-which means something 
very different from promiscuity. 
Of course I do not mean to say that pre-nuptial relations 
among simple peoples always have this character. We 
have too many and too positive statements to the contrary 
to allow us to doubt that promiscuity outside marriage does 
exist. Among the Point Barrow Eskimo, according to 
Mr. Murdoch, " promiscuous sexual intercourse between 
married or unmarried people, or even among children, 
appears to be looked upon simply as a matter for amuse- 
ment."2 Concerning the natives of St. Christoval and 
adjacent islands, of the Solomons, we are told that " for 
two or three years after a girl has become eligible for marriage 
she distributes her favours amongst all the young men of 
the village " ; 3 and in Rubiana, belonging to the same 
group, it is considered an honour for an adult girl to have 
intercourse with as many men as possible within a short 
time.4 The Hawaians regarded it as "a meanness for a 
man or woman to refuse a solicitation for sensual gratifica- 
tion."6 Among the common Line Islanders of the Gilbert 
Group and some neighbouring islands marriage was the 
exception and free intercourse the rule, and a woman was 
at liberty to accept as many men as would take her, provided 
they paid for the privilege.6 The young girls of Madison's 
Island, of the Marquesas, " are the wives of all who can 
purchase their favours, and a handsome daughter is con- 
sidered by her parents as a blessing which secures to them, 
for a time, wealth and abundance."7 And besides the 
1 Supra, p. 72 sqq. 
2 Murdoch,' Ethnological Results of the Point Barrow Expedition, ' 
in Ann. Rep. Bur. Ethn. ix. 419. 
3 Guppy, Solomon Islands, p. 43. 
4 Ribbe, Zwei Jahre unter den Kannibalen der Salomo-Inseln, 
p. 270. 
5 Jarves, History of the Hawaiian Islands, p. 42. 
6 Tutuila, ' Line Islanders,' in Jour. Polynesian Soo. i. 270. 
7 Porter, Journal of a Cruise made to the Pacific Ocean, ii. 60. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
137 
prostitution of girls by their parents1 and of married women 
by their husbands2 we also find among various uncivilised 
peoples prostitution carried on by a class of professionals.8 
There are harlots in some of the Melanesian islands ; 
" at Santa Cruz," says Dr. Codrington, " where the separa- 
tion of the sexes is so carefully maintained, there are cer- 
tainly public courtesans."4 Among the Line Islanders a 
niki-rau-raro, who is earning her living by prostitution, 
" is greatly respected and envied if successful in doing 
it."5 A similar demi-monde class {rar an) is found in Ponape, 
of the Caroline Islands.6 In Easter Island, where there 
were many more males than females, some of the young 
women remained unmarried and offered themselves up to 
men; they seem to have been looked upon as a kind of 
public benefactors.7 In the Hawaian Islands, on the other 
hand, " common and frequent prostitution was considered 
in some degree disreputable, and it was enjoined by the 
better class of parents on their sons to avoid it."8 In Green- 
land there were professional harlots already in early times,9 
and similar women have been found among various Indian 
tribes in North and South America.10 Among the Omaha, 
who called them minckeda, they were hardly ever girls but 
1 See, e.g., Lery, Histoire d'un voyage faict en la terre du Brtsil, 
p. 303 (Tupis) ; Hennepin, New Discovery of a Vast Country in 
America, p. 481 (Iroquois, &c.); Stannus, 'Notes on some Tribes of 
British Central Africa,' in Jour. Roy. Anthr. Inst. xl. 299. 
2 See, e.g., Parry, Journal of a Second Voyage for the Discovery of 
a North-West Passage, p. 529 (Eskimo of Melville Peninsula) ; 
Kumlien, Contributions to the Natural History of Arctic America, p. 16 
(Eskimo of Cumberland Sound); infra, i. 332 sq. 
3 See Ploss-Bartels, Das Weib, i. 584, 590. 
4 Codrington, Melanesians, p. 235 sq. See also Powell, Wanderings 
in a Wild Country, p. 261 (natives of New Britain). 
5 Tutuila, in Jour. Polynesian Soc. i. 270. 
6 Christian, Caroline Islands, p. 74. 
7 Geiseler, Die Oster-Insel, p. 29. 
s Jarves, op. cit. p. 43. 
!l Cranz, History of Greenland, i. 176. 
10 Carver, Travels through the Interior Parts of North America, 
p. 375. Davis, El Gringo, p. 221 (Indians of New Mexico). Latcham, 
* Ethnology of the Araucanos,' in Jour. Roy. Anthr. Inst, xxxix. 354. 138 
THE HISTORY OF HUMAN MARRIAGE chap. 
chiefly divorced wives and they were looked down upon.1 
Among the Karaya, on the River Araguaya, they were 
women belonging to other tribes.2 Among the Witoto and 
Boro, in the north-west Amazon region, " very frequently 
widows become the tribal prostitutes, a custom that is 
not recognised, but is tolerated, and is never practised 
openly or immodestly."3 Prostitution prevails in many 
negro countries ;4 and so favourably, we are told, is this 
institution sometimes regarded that rich negro ladies on 
their death-beds buy female slaves and present them to the 
public, " in the same manner as in England they would 
have left a legacy to some public charity."6 In Unyoro 
the king usually supported in connection with his own 
establishment a large number of professional prostitutes- 
perhaps 2,000-whose existence as an organised corps was 
recorded by all travellers in that country from the days of 
Sir Samuel Baker until the complete upsetting of the native 
Government of Unyoro in 1895.6 It does not seem, however, 
that prostitution is common among unadulterated savages. 
In some cases it is expressly said to be due to contact with 
foreigners.7 
Yet however commonly pre-nuptial chastity be disregarded 
in the savage world, we must not suppose that such 
1 James, Account of an Expedition from Pittsburgh, i. 233, 239. 
Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethn. iii. 365. 
2 Krause, In den Wildnissen Brasiliens, p. 327. 
3 Whiffen, North-West Amazons, p. 167. 
4 Monrad, Skildring af Guinea-Kysten, p. 49 (Negroes of Accra). 
Bowdich, Mission to Ashantee, p. 303. Norris, quoted by Ling 
Roth, Great Benin, p. 37 n. 2 (Dahomans). Dennett, Notes on the 
Folk-lore of the Fjort {French Congo), p. 21. Emin Pasha in Central 
Africa, p. 88. 
6 Reade, Savage Africa, p. 547 sq. 
6 Johnston, Uganda Protectorate, p. 590. Emin Pasha in Central 
Africa, p. 87. Wilson and Felkin, Uganda, ii. 49. 
7 Bogoras, Chukchee, p. 610 (the maritime peoples on both coasts 
of Bering Sea). Bainbridge, ' Saorias of the Rajmahal Hills,' in 
Memoirs of the Asiatic Society of Bengal, ii. 58. Condon, ' Contribu- 
tion to the Ethnography of the Basoga-Batamba, Uganda Protec- 
torate,' in Anthropos, vi. 372. Nordenskibld, Indianliv i El Gran 
Chaco (Syd-Amerika), pp. 88, 197. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
139 
disregard is anything like a universal characteristic of the 
lower races. Among very many of them sexual intercourse 
before marriage is said to be rare, if not unknown, at least 
on the part of any girl who is not a prostitute, or to be 
looked upon as a disgrace or punished as a crime ; and in 
such cases not only the girl but the man who seduced her is 
subject to punishment or censure. In other cases, again, 
it is said that the birth of an illegitimate child is followed 
by punishment. As facts of this kind have been too often 
overlooked, I shall produce the evidence. 
Among many South American Indians there is no doubt 
great freedom before marriage.1 But Dobrizhoffer praised 
the Abiponian women for their virtuous life.2 Abbe Ignace 
tells us that fornication is proscribed among the Canelas in 
Maranhao.3 Among the Karaya, another Brazilian tribe, 
virginity is said to be highly esteemed and carefully guarded 
and sexual intercourse out of wedlock to be severely pun- 
ished, sometimes even with death.4 The latter statement 
is not confirmed by Krause, but he admits that the girls 
do their best to preserve their purity and avoid going into 
the forest alone for fear of the bachelors.5 Mr. Whiffen 
also says that among the Witoto and Boro " virginity, as 
with us, is strictly protected so far as is possible."6 During 
his nine years' stay in British Guiana, Appun only heard of 
one illegitimate birth among the natives ; the mother fell 
into disrepute and was avoided, especially by the men.7 
When one of the Chichimec of Central Mexico marries, a 
1 Guevara, ' Folklore Araucano,' in Anales de la Universidad de 
Chile, cxxvii. 626; Latcham, ' Ethnology of the Araucanos,' in 
Jour. Roy. Anthr. Inst, xxxix. 354. Nordenskiold, op. cit. pp. 73, 81 sq. 
(Choroti). Chomd, ' Dritter Brief,' in Stoecklein, Der Neue Welt- 
Bott, vol. iv. pt. xxix. 72 (Chiriguanos of Bolivia). Schmidt, * Ueber 
das Recht der tropischen Naturvolker Siidamerikas,' in Zeitschr. 
vergl. Rechtswiss. xiii. 309. 
2 Dobrizhoffer, Account of the Abipones, ii. 153. 
3 Ignace, ' Les Capiekrans,' in Anthropos, n. 478. 
4 Ehrenreich, Beitvdge zur V olkerhunde Brasiliens, p. 27. 
6 Krause, op. cit. p. 326 sq. c Whiffen, op. cit. p. 156. 
" Appun, ' Die Indianer von Britisch-Guayana,' in Nas Ausland, 
xliv. 833. 140 
THE HISTORY OF HUMAN MARRIAGE chap. 
bride who proves not to be a virgin may be returned to her 
parents.1 Of the Mexican Tepehuane, Dr. Lumholtz in- 
forms us that " outside of her home a woman is absolutely 
forbidden to speak to any man who does not belong to her 
own immediate family. . . . Even at the dancing-place it 
is against the law for her to step aside to exchange a few 
words with any young man. If discovered in such a com- 
promising position, both offenders are immediately arrested, 
and their least punishment is two days' imprisonment. 
If their examination by the judges proves that their con- 
versation was on the forbidden topic of love, they get a 
whipping and may be compelled to marry."2 Among the 
Hupa of California a grown-up girl was not allowed to be 
alone with a man either in the house or outside; she was 
told the results of wrongdoing, and severely punished by 
beating if she were remiss ; and a seducer was obliged to 
marry his victim.3 Among the Omaha extra-matrimonial 
intercourse is, as a rule, practised only with public women, 
or minckeda ; and so strict are they in these matters, " that 
a young girl or even a married woman walking or riding 
alone, would be ruined in character, being liable to be taken 
for a minckeda, and addressed as such."4 
Of many other North American Indians we read that the 
girls were chaste or carefully guarded,5 or that a girl who 
1 Bancroft, Native Tribes of the Pacific States, i. 632. 
2 Lumholtz, Unknown Mexico, i. 467. 
3 Goddard, Life and Culture of the Hupa, p. 55. 
4 Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethn. iii. 365. 
See also Fletcher and La Flesche, ' Omaha Tribe,' ibid, xxvii. 323 ; 
James, Account of an Expedition from Pittsburgh to the Rocky Moun- 
tains, i. 23g. 
6 Cremony, Life among the Apaches, p. 244 sq. Charlevoix, 
Voyage to North-America, ii. 38 sq. ; Parkman, op. cit. p. xxxiv. n, 1 
(Algonkin). Catlin, Illustrations of the Manners, &c. of the North 
American Indians, i. 121 (respectable families of the Mandan). 
Schoolcraft, Archives of Aboriginal Knowledge, v. 654 (Nez Percds 
in Oregon). Sapir, ' Notes on the Takelma Indians of Southwestern 
Oregon,' in American Anthropologist, N.S. ix. 274. Hearne, Journey 
to the Northern Ocean, p. 311 (Northern Indians). Morice, ' Great 
Dene Race,' in Anthropos, ii. 32 (various Dene tribes). IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
141 
was known to have lost her virtue lost with it one of her 
chances of a favourable marriage j1 although this by no 
means applies to all, perhaps not even to the majority, of 
their tribes.2 Among the Tlingit, "if unmarried women 
prove frail the partner of their guilt, if discovered, 
is bound to make reparation to the parents, • soothing 
their wounded honour with handsome presents."3 Among 
the early Aleut, according to Veniaminof, " girls or 
unmarried females who gave birth to illegitimate 
children were to be killed for shame, and hidden."4 
Egede tells us that among the Greenlanders unmarried 
women observed the rules of modesty much better than 
married women. " During fifteen full years that I lived in 
Greenland," he says, " I did not hear of more than two or 
three young women, who were gotten with child unmarried ; 
because it is reckoned the greatest of infamies."5 According 
to Dalager, whose account likewise dates from the eighteenth 
century, the Greenlanders, though no models of virtue, are 
less addicted to wantonness than other peoples. A young 
man marries as soon as the sexual instinct prompts him to 
do so and he is able to maintain a wife, and the girls also 
behave very chastely, since otherwise no bachelor will care 
to marry them; but it is different with young widows and 
divorced wives.6 Modem accounts are less favourable. 
Holm says that on the east coast of Greenland it is no dis- 
grace for an unmarried girl to get a child, but that it is a 
1 Sproat, Scenes and Studies of Savage Life, p. 95 (Aht). Keating, 
Expedition to the Source of St. Peter's River, ii. 169 sq. (Chippewa). 
Heriot, Travels through the Canadas, p. 339. Waitz, Anthropologic 
der N aturvblker, iii. 505. 
2 See Cremony, op. cit. p. 244 (Navaho) ; Bossu, Travels through 
Louisiana, i. 231 (Alibamu) ; Lahontan, New Voyages to North- 
America, ii. 452 sqq. (Indians of Canada). Morice, loc. cit. p. 32 
(Dene and allophylic North Pacific tribes). 
3 Douglas, quoted by Petroff, ' Report on the Population, &c. of 
Alaska,' in Tenth Census of the United States, p. 177. 
1 Veniaminof, quoted ibid. p. 155. 
Egede, Description of Greenland, p. 141. 
11 Dalager, Grdnlandske Relationer, p. 67. 142 
THE HISTORY OF HUMAN MARRIAGE chap. 
disgrace for a married woman to get none.1 Other Eskimo 
are equally indifferent to the chastity of their girls.2 
Among the Siberian and other tribes belonging to the 
former Russian Empire chastity is generally held in little 
or no regard; but, as we have already noticed, this seems 
partly at least to be due to Russian influence. According 
to earlier accounts, the bridegroom might, among several 
of them, claim a fine in case the bride was found to have 
lost her virtue ;8 and among the Chulim, if the Mosaic 
testimony of chastity was wanting, the husband went away 
and did not return before the seducer had made peace with 
him.4 Among the Tungus a seducer was bound to marry 
his victim and pay the price claimed for her, or, if he refused, 
submit to corporal punishment.6 In order to preserve the 
virtue of their daughters, the Yakut employ a chastity girdle, 
which is not removed even at night; but when the bride 
price has been partly or fully paid the parents take no further 
interest in the matter. The Yakut are said to see nothing 
immoral in free love, provided only that nobody suffers 
material loss by it.6 Among the Koryak the girl is even as 
inaccessible to the bridegroom while he serves for her as to 
a stranger; "intercourse of a bride with her bridegroom 
before the termination of his service is deemed a sin."7 
In Circassia an incontinent daughter was generally sold as 
soon as possible, being a disgrace to her parents ; and if not 
a virgin, the bride ran the risk of being put away after the 
1 Holm, ' Konebaads-Expeditionen til Grdnlands Ostkyst, 
1883-85/ in Geografish Tidskrift, viii. 91. 
2 Murdoch, ' Ethnological Results of the Point Barrow Expedi- 
tion,' in Ann. Rep. Bur. Ethn. ix. 419 sq. Turner, ' Ethnology of the 
Ungava District,' ibid. xi. 189 (Koksoagmiut). Parry, Second 
Voyage for the Discovery of a North-West Passage, p. 529 (Eskimo of 
Iglulik and Winter Island). Kumlien, Contributions to the Natural 
History of Arctic America, p. 16 (Eskimo of Cumberland Sound). 
3 Georgi, Beschreibung alter Nationen des russischen Reichs 
pp. 79, 104, 237, 238, 283. 
4 Ibid: p. 232. 
6 Ibid. p. 311. 
0 Sieroshevski, quoted by Miss Czaplicka, Aboriginal Siberia, 
p. 108 ; and trans, in Jour. Anthr. Inst. xxxi. 96. 
7 Jochelson, Koryak, p. 735. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
143 
first night.1 Professor Kovalewsky states that among the 
Ossetes the lack of virginity in a bride is considered a 
disgrace, not only to the husband, but to all his relatives.2 
Among the Chuvash the signum innocentiae is exhibited 
coram fof>ulo.z Vambery says that a fallen girl is unknown 
among the Central Asiatic Turks.4 
In Dardistan young people have continued opportunities 
of meeting each other in the fields or at festive gatherings, 
and love declarations often take place on these occasions; 
" but if any evil intention is perceived the seducer of a girl 
is punished by this savage, but virtuous, race with death."5 
Among many of the uncivilised tribes in India and Indo- 
China the standard of female morality is very low indeed;6 but 
this is not true of all of them. Among the Bodo and DJiimals 
chastity is prized in man or woman, married and unmarried.7 
The mountaineers of the Rajmahal Hills allow a young man 
to show his love for a girl of adult age by sleeping on the 
same bedstead with her; but " should any indiscretion 
arise previous to marriage for the young couple sleeping 
together, they are considered disgraced and are visited with 
fine."^ Among the Santals a youth and a girl are allowed 
to look at but not to speak to each other. If they do, the 
youth is taken to the village council and asked if he wants 
to marry the girl; should he say no, he is beaten and fined, 
but should he say yes, he is only fined.9 The Let-htas, a 
hill tribe of Burma, are more particular still: until married, 
the young people of both sexes are domiciled in two long 
houses at opposite ends of the village, and " when they may 
have occasion to pass each other, they avert their gaze, so 
1 Klemm, Allgemeine Cultur-Geschichte der Menschheit, iv. 26. 
2 Kovalewsky, Coutume contemporaine et loi ancienne, p. 176. 
3 Vambdry, Das Turkenvolk, p. 461. 
4 Ibid. p. 240. 
5 Leitner, Results of a Tour in ' Dardistan, &c.,' iii. 35. 
6 See, e.g., Crooke, Tribes and Castes of the North-Western Provinces 
and Oudh, i. p. clxxxiv. ; Gait, op. cit. p. 243. 
7 Hodgson, Miscellaneous Essays, i. 123. 
8 Sherwill, ' Notes upon a Tour through the Rajmahal Hills,' 
in Jour. Asiatic Soc. Bengal, xx. 557. 
9 Hertel, Indisk Hjemmemission blandt Santalerne, p. 83. 144 
THE HISTORY OF HUMAN MARRIAGE chap. 
that they may not see each other's faces."1 Among the 
Lisu tribes of the Burma-China frontier there is little sexual 
intercourse before marriage, and it is considered a great 
disgrace for a girl to give birth to a child out of wedlock.2 
The Assam tribes differ greatly with regard to pre-nuptial 
chastity. The Nagas give both sexes full freedom before 
marriage ;3 indeed, among the Angami Nagas girls consider 
short hair, the symbol of virginity, a disgrace, and men are 
desirous to have proof that their future wives will not be 
barren.4 Among the Kacharis, Rabhas, and Hajongs, on 
the contrary, sexual intercourse before marriage is rare 
where contact with " civilisation " has not exercised its 
deteriorating influence ; and when it does take place and 
pregnancy follows, the seducer is not only compelled to marry 
the girl but must besides pay an enhanced bride price or a 
fine to the village elders.6 Among the Kukis, according to 
older accounts, a man who seduced a girl had to marry her 
as soon as her parents heard of it ;6 but nowadays their 
customs seem to be less stringent.7 The moral standard of 
unmarried Garo women is, generally speaking, a high one, 
although the matrimonial bonds are loose.8 Among the 
Kammalans, or artisans, in the Cochin State sexual licence 
is in no case tolerated; should it occur, the girl and her 
parents are placed under a ban.9 The Ulladans, belonging 
to the lowest caste among the purely Malayali Hindu and 
animistic castes of the same State, likewise prohibit sexual 
intercourse before marriage; " should an unmarried girl 
1 O'Riley, quoted by Fytche, Burma, i. 343. 
2 Geis, quoted by Rose and Brown, ' Lisu (Yawyin) Tribes,' in 
Memoirs Asiatic Soc. Bengal, iii. 263. 
3 Hodson, Naga Tribes of Manipur, pp. 78, 87. Soppitt, Short 
Account of the Kuki-Lushai Tribes, p. 5. 
4 Prain, 1 Angami Nagas,' in Revue coloniale internationale, 
v. 491 sq. b Endle, Kachans, pp. 3, 30, 31, 85 sq. 
6 Macrae, ' Account of the Rookies,' in Asiatick Researches, 
vii. 193. Butler, Travels and Adventures in the Province of Assam, 
P- 84. 
7 Soppitt, op. cit. p. 5. Shakespear, Lushei Kuki Clans, p. 53. 
8 Playfair, Garos, p. 70. 
9 Anantha Krishna Iyer, Cochin Tribes and Castes, i. 345. iv CRITICISM OF HYPOTHESIS OF PROMISCUITY 
145 
become pregnant and the fact be known, her secret lover is 
summoned by the tribesmen, who compel him to take her 
to wife, as otherwise they are placed under a ban."1 So 
also among the Velans of the Cochin State, the man who has 
made a girl pregnant must marry her, and very often they 
a e both fined.2 The strict morality which characterises 
the Veddas of Ceylon " extends to unmarried girls, who are 
protected by their natural guardians with the keenest sense 
of honour," although it does not extend to widows, however 
young and pretty.3 Of the wild tribes of the Malay Peninsula 
Vaughan Stevens says that irregular connections only occur 
among the Belendas who have been most subject to Malay 
influence.4 
In the Malay Archipelago intercourse between unmarried 
people is among many tribes considered neither a crime nor a 
disgrace ;6 but there are others who look upon it in a different 
light and require that the bride shall be a virgin.6 In Nias 
the pregnancy of an unmarried girl is punished with death, 
inflicted both upon her and upon the seducer.7 Among the 
Hill Dyaks the young men are carefully separated from the 
girls, licentious connections between the sexes being strictly 
prohibited ;8 and the Sibuyaus, a tribe belonging to the Sea 
Dyaks, though they do not consider the sexual intercourse 
of their young people a positive crime, yet attach an idea of 
1 Ibid. i. 60 sq. 
2 Thurston, Castes and Tribes of Southern India, vii. 348. 
3 Nevill, ' Vaeddas of Ceylon,' in Taprobanian, i. 178. 
4 Stevens, ' Mittheilungen aus dem Frauenleben der Orang 
Belendas, der Orang Djakun und der Orang Laut, in Zeitschr. f. 
Ethnol. xxviii. 175. See also Martin, Die Inlandstdmme der Malay- 
ischen Halbinsel, p. 874 ; Knocker, * Aborigines of Sungei Ujong,' 
in Jour. Roy. Anthr. Inst, xxxvii. 293 ; Logan, * Orang Sakimba of 
the Extremity of the Malay Peninsula,' in Jour. Indian Archipelago, 
i. 297. 
6 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. 
iv. 434 sqq. Schmidt, Liebe und Ehe im alten und modemen Indien, 
p. 221 sq. 
G Wilken, loc. cit. p. 446 sqq. Schmidt, op. cit. p. 223 sq. 
i Wilken, loc. cit. p. 444. 8 Low, Sarawak, pp. 300, 247. 
VOL. I L 146 
THE HISTORY OF HUMAN MARRIAGE chap. 
great indecency to irregular connections and are of opinion 
that an unmarried woman with child must be offensive to 
the superior powers.1 Dr. Hagen states that the Kubus of 
Sumatra expect chastity from their young people and that 
the free love of which Boers speaks is exceptional, secret, 
and illegal among them.2 According to Chamisso, some of 
the independent tribes of the Philippines held chastity in 
great honour, not only in the case of women, but also in the 
case of young girls, and protected it by severe laws.3 This 
statement is confirmed by Dr. Hans Meyer4 and Professor 
Blumentritt6 with regard to the Igorot of Luzon. Of 
the Negritos, Mozo wrote in the eighteenth century, " Before 
marriage a false step is hardly heard of among them."6 
With reference to those of the interior of Luzon, Garcia 
states that a girl who became enceinte was punished with 
the greatest severity, even though she at once married the 
seducer.7 
In New Guinea the relations between the unmarried vary 
greatly in different tribes. Among many of them incon- 
tinence seems to be very common,8 whereas among others 
the greatest chastity is said to be maintained.9 Among the 
Mekeo people of British New Guinea, in former days, a girl 
1 St. John, op. cit. i. 52 sq. 
2 Hagen, Die Orang Kubu auf Sumatra, p. 130 sq. For other 
natives of Sumatra see Marsden, History of Sumatra, p. 261. 
3 v. Kotzebue, Voyage of Discovery, iii. 66. 
4 Meyer, ' Die Igorrotes von Luzon,' in Verhandl. Berliner 
Gesellsch. Anthr. 1883, p. 384 sq. 
6 Blumentritt, Versuch einer Ethnographic der Philippinen, p. 27. 
6 Mozo, Noticia hist6rico natural de los gloriosos triumphos por 
los religiosos del orden de N.P.S. Agustin en las Islas Philipinas, 
p. 108. 
7 Garcia, quoted by Bille, Beretning om Corvetten Galathea's 
Reise omkring J or den, ii. 181. 
8 Seligman, op. cit. pp. 134 (Koita), 499 (Southern Massim). 
Malinowski, loc. cit. pp. 559, 561, 563 (natives of Mailu). Williamson, 
Mafulu Mountain People of British New Guinea, p. 172. Neuhauss, 
op. cit. i. 160 (natives of Kaiser Wilhelm Land). Krieger, Neu- 
Guinea, p. 395 (natives of the western part of Dutch New Guinea). 
9 Krieger, op. cit. p. 395. Finsch, Neu-Guinea, pp. 77, 82, 92, 101. 
Earl, Papuans, p. 81. Bink, in Bull. Soc. d'Anthr. Paris, ser. iii. 
vol. xi. 397. iv CRITICISM OF HYPOTHESIS OF PROMISCUITY 
147 
who became with child before marriage ran the risk of being 
killed.1 Among the natives living in the neighbourhood of 
Geelvink Bay, on the northern coast, a seducer is compelled 
to marry the girl and can escape from doing so only by leav- 
ing the country.2 On the Maclay Coast, Miklucho-Maclay 
found extra-matrimonial intercourse to be exceedingly 
rare, probably on account of the early marriages.3 Among 
the Western Islanders of Torres Straits " irregular inter- 
course with women.was invariably spoken of as ' stealing,' 
as the girls were regarded as the property of their fathers 
and the wives as the property of their husbands."4 In some 
parts of the Bismarck Archipelago seduction, if proved by 
witnesses, is severely punished ; the girl may even be killed, 
in which case the man has to pay her. value in shell money.5 
Among good families at Saa in Malanta, of the Solomon 
Islands, " the virginity of a bride is a matter of much 
concern to her friends, not only because the boy's friends will 
not pay what they have promised if her character is ques- 
tionable, but because they value propriety."6 Among some 
of the New Caledonians unchastity is considered dis- 
honourable, and to call a person a bastard is an insult ;7 
whereas among others " les j eunes filles peuvent disposer 
de leur corps."8 The women of Uea, Loyalty Islands, are 
described as " strictly chaste before marriage, and faithful 
wives afterwards " ;9 and in Lifu, belonging to the same 
1 Williamson, Some unrecorded Customs of the Mekeo People 
of British New Guinea,' in Jour. Roy. Anthr. Inst, xliii. 278. 
2 Krieger, op. cit. p. 395. 
3 Miklucho-Maclay, ' Anthropologische Bemerkungen uber die 
Papuas der Maclay-Kiiste in Neu-Guinea,' in Natuur kundig Tijd- 
schrift voor Nederlandsch Indie, xxxiii. 245. Idem, ' Ethnologische 
Bemerkungen uber die Papuas der Maclay-Kiiste in Neu-Guinea,' 
ibid. xxxv. 89. 
4 Haddon, in Reports of the Cambridge Anthropological Expedition 
to Torres Straits, v. 275. 
6 Pfeil, Studien und Beobachtungen aus der Siidsee, p. 31. 
6 Codrington, op. cit. p. 239. 
7 Lambert, Mceurs et superstitions des Neo-Caledomens, p. 72 sq. 
8 Brainne, La Nouvelle-Catedonie, p. 251. 
9 Erskine, Islands of the Western Pacific, p. 341. See also Cheyne, 
Description of Islands in the Western Pacific Ocean, p. 25. 148 
THE HISTORY OF HUMAN MARRIAGE chap. 
group, " a young man meeting or walking alone with and 
speaking to an unmarried or espoused girl might be clubbed 
by her father or other guardian."1 We have previously 
noticed the chastity which in former days characterised the 
girls in Fiji.2 It may be added that the old women of the 
bridegroom's family ascertained whether the bride was a 
virgin ; and if the result of their inquiries was unsatisfactory, 
the feast connected with the clipping of her hair " was made 
the occasion for putting her friends to shame."3 Among the 
nobles of the Line Islanders proof of virginity is required on 
marriage and " it must be conclusive."4 In Tonga the 
nuptial mat was paraded from house to house ;5 and in 
Samoa the innocence of the bride is tested in the sight of 
the whole village by a sort of surgical operation performed 
by the bridegroom, who then raises his hand to show the 
blood on his finger.6 But Turner remarks that although the 
virtue of chastity was ostensibly cultivated in Samoa by 
both sexes, it was more a name than a reality.7 There was 
more laxity of intercourse between boys and girls in Poly- 
nesia than in Melanesia.8 With reference to the latter, Dr. 
Codrington observes that although unchastity was not 
very seriously regarded, " yet it is certain that in these 
islands generally there was by no means that insensibility 
in regard to female virtue with which the natives are so 
commonly charged."9 Among the Maori,10 Marquesas 
1 Ray, ' People and Language of Lifu,' in Jour. Roy. Anthr. Inst. 
xlvii. 280. 
2 Supra, p. 127 sq. See also Erskine, op. cit. p. 255; Meinicke, 
op. cit. ii. 45. 3 Thomson, Fijians, p. 202. 
4 Tutuila, in Jour. Polynesian Soc. i. 271. 
6 Thomson, op. cit. p. 203. 
6 Kramer, Die Samoa-Inseln, i. 36 sqq. Turner, Samoa, p. 95. 
Wilkes, Narrative oj the United States Exploring Expedition, ii. 80. 
7 Turner, Nineteen Years in Polynesia, p. 184. 
8 Cf. Thomson, op. cit. p. 234. 9 Codrington, op. cit. p. 235. 
10 Taylor, Te Ika a Maui, p. 33. Polack, Manners and Customs 
of the New Zealanders, i. 145. Shortland, Traditions and Super- 
stitions of the New Zealanders, p, 120. Thomson, Story of New 
Zealand, i. 178. Brown, Maori and Polynesian, p. 68. Gisborne, 
Colony of New Zealand, p. 27. Tregear, The Maori Race, p. 284. 
Best, ' Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 32, 35. iv CRITICISM OF HYPOTHESIS OF PROMISCUITY 
149 
Islanders,1 Hawaians,2 Marshall Islanders,3 and Pelew 
Islanders,4 there is said to be, or to have been, the greatest 
freedom before marriage. 
The same can certainly not be said of the Australian 
aborigines while in their native state. There is extra- 
matrimonial intercourse among them on certain occasions 
or in certain circumstances, recognised and regulated by 
their customs, but this is something very different from 
irregular unions. For the latter, says Dr. Howitt, the Dieri 
have special terms, and they are condemned and abhorred 
by them.5 Among the Aranda, or Arunta, in Central 
Australia, according to Strehlow, if a grown-up youth has 
sexual intercourse with a grown-up girl or with the wife of 
another man, they are both speared and their bodies thrown 
into the fire ; 6 and the same is said to be the case among the 
Loritja.7 Of the Maroura tribe, on the Lower Darling, we are 
told that before the advent of the whites " their laws were 
strict, especially those regarding young men and young 
women. It was almost death to a young lad or man who had 
sexual intercourse till married."8 Among various tribes in 
Western Victoria " illegitimacy is rare, and is looked upon 
with such abhorrence that the mother is always severely 
beaten by her relatives, and sometimes put to death and 
burned. The father of the child is also punished with the 
greatest severity, and occasionally killed."9 Mr. Moore Davis 
1 Porter, Journal of a Cruise made to the Pacific Ocean, ii. 59 sq. 
Tautain, ' Ltude sur le manage chez les Polyn6siens (Mao'i) des 
lies Marquises,' in L' Anthropologic, vi. 645. See also Christian, 
Eastern Pacific Lands, p. 95. 
2 Jarves, History of the Hawaiian Islands, pp. 18, 42. 
8 Kohler, ' Das Recht der Marschallinsulaner,' in Zeitschr. vergl. 
Rechtswiss. xiv. 417. 
4 Kubary, ' Die Palau-Inseln,' in Jour. d. Museum Godeffroy, 
iv. 53- 
5 Howitt, Native Tribes of South-East Australia, p. 187. 
6 Strehlow, Die Aranda- und Loritja-Stamme in Zentral-Australien, 
vol. iv. pt. i. 93. 
7 Ibid. vol. iv. pt. i. 103. 
8 Holden, in Taplin, Folklore of the South Australian Aborigines, 
p. 19. 
0 Dawson, Australian Aborigines, p. 28. 150 
THE HISTORY OF HUMAN MARRIAGE chap. 
writes :-" Promiscuous intercourse between the sexes is not 
practised by the Aborigines, and their laws on the subject, 
particularly those of New South Wales, are very strict. 
When at camp, all the young unmarried men are stationed 
by themselves at the extreme ends, while the married men, 
each with his family, occupy the centre. No conversation is 
allowed between the single men and the girls or the married 
women. . . . Infractions of these and other laws were 
visited either by punishment by any aggrieved member of 
the tribe, or by the delinquent having to purge himself of 
his crime by standing up protected simply by his shield, or 
a waddy, while five or six warriors threw, from a com- 
paratively short distance, several spears at him."1 Among 
the Euahlayi tribe, says Mrs. Langloh Parker, " unchaste 
women were punished terribly. After we went west even the 
death penalty for wantonness was enforced."2 Among the 
North-West-Central Queensland aborigines, on the other 
hand, women are allowed considerable sexual freedom, 
unless they happen to be betrothed. Yet " after all, morality 
in its broadest sense is certainly a recognised virtue." A 
prostitute, though her frailty is usually due to the death or 
desertion of her husband, is despised, and has a special 
name applied to her ; and the man who habitually consorts 
with such a woman is similarly regarded with contempt, 
and spoken of under a distinct term.3 
Among a large number of African peoples the un- 
married are allowed full liberty before marriage, and 
virginity is neither expected nor found in a bride.4 But those 
1 Moore Davis, quoted by Brough Smyth, Aborigines of Victoria, 
ii. 318. See also Fraser, Aborigines of New South Wales, p. 27. 
2 Mrs. Langloh Parker, Euahlayi Tribe, p. 59 sq. 
3 Roth, North Queensland Ethnography : Bulletin No. 8. Notes 
Government, Morals, and Crime, p. 7. 
4 See, e.g., v. Francois, Nama und Damar a Deutsch-Sud-West- 
Afrika, p. 213 (Hottentots) ; Magyar, Reisen in Sud-Afrika, p. 285 
(Kimbunda) ; Decle, Three Years in Savage Africa, pp. 160 
(Matabele), 346 sq. (Wanyamwezi) ; Johnston, British Central 
Africa, p. 409 note ; Idem, Uganda Protectorate, p. 610 (Bairo) ; 
Roscoe, Northern Bantu, pp. 260, 261 (Bateso), 279, 281 (Nilotic 
Kavirondo) ; Felkin, ' Notes on the For Tribe of Central Africa,' IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
151 
of them among whom the contrary is the case are also very 
numerous. Of some Bushman tribes we are told that pre- 
nuptial intercourse seems hardly to occur,1 or that the 
girls generally are virgins when they enter into marriage 
because most of them marry soon after puberty.2 The 
Kafirs hold chastity in little regard ; yet in some of their 
tribes a girl who has lost her virtue fetches a lower price 
than a virgin,3 or the father of an unmarried woman who 
becomes pregnant can demand a fine of one head of cattle 
from the father of the child,4 or the mere seduction of a virgin 
incurs the fine of three or four head of cattle.5 Mr. Cousins 
wrote to me that between their various feasts the Cis- 
Natalian Kafirs, both men and women, have to live in strict 
continence, the penalty for breaking this rule being banish- 
ment from the tribe. Casalis mentions a curious custom 
prevalent among the Basuto which shows that unchastity 
in a young man is not looked upon with perfect indifference, 
and is even supposed in certain circumstances to expose 
him to supernatural danger.8 Among the Awemba in 
Northern Rhodesia chastity is an unknown quantity in 
young girls over fifteen years of age, but they nevertheless 
have a very definite code of sex-morality : "all know that 
in Proceed. Roy. Soc. Edinburgh, xiii. 233 ; Angus, ' Initiation 
Ceremony of Girls, as performed in Azimba Land, Central Africa,' 
in Verhandl. Berliner Gesellsch. Anthr. 1898, p. 481 ; Munzinger, 
Ostafrikanische Studien, p. 524 (Barea and Kunama) ; Clozel and 
Villamur, Les coutumes indigenes de la C6te d'Ivoire, p. 101 (Baoul6) ; 
Tessmann, Die Pangwe, ii. 253, 254, 258 sq. ; Thomas, Anthropological 
Report on Ibo-speaking Peoples of Nigeria, i. 69; Partridge, Cross 
River Natives, p. 254 ; Torday and Joyce, Les Bushongo, pp. 116, 
266, 272 ; I idem, ' Notes on the Ethnography of the Ba-Huana,' in 
Jour. Anthr. Inst, xxxvi. 288 ; Weeks, Among the Primitive Bakongo, 
p. 163 ; Idem, Among Congo Cannibals, p. 127 (Boloki) ; Idem, 
* Anthropological Notes on the Bangala of the Upper Congo River,' 
in Jour. Roy. Anthr. Inst, xxxix. 442, and xl. 417; Delhaise, Les 
Warega, p. 173. 
1 Kaufmann, ' Die Auin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 156. 
2 Trenk, ' Die Buschleute der Namib,' ibid, xxiii. 169. 
3 Kropf, Das Volk der Xosa-Kaffern, p. 138. 
' Warner, in Maclean, Compendium oj Kafir Laws, p. 64. 
6 Brownlee, ibid. p. 112. 6 Casalis, Basutos, p. 267 sq. 152 
THE HISTORY OF HUMAN MARRIAGE chap. 
immorality is wrong, and that it runs counter to the laws of 
superstitious observances."1 Among the Herero it is con- 
sidered a shame to the parents of a girl if she loses her virtue ; 
hence she is in many cases betrothed already as a child, 
which compels her as well as her fiance to live chastely.2 
And according to one of our authorities a seducer is severely 
punished, nay is even in danger of losing his life.8 
Among the natives near Port Herald in British Central 
Africa, according to Father Torrent, the women are very 
strict in their habits and it seldom happens that a bride is 
not a virgin.4 Among the Konde people, living to the east 
of the northern part of Lake Nyasa, sexual intercourse 
between young people is frequent, because the prevailing 
polygamy makes it impossible for many a young man to 
marry; but if the case becomes known the seducer is com- 
pelled to buy the girl from her father, or, if he cannot pay the 
bride price, is deprived of his spear and has to leave the 
country.5 In Ruanda an adult girl who is known to be 
unchaste is beaten and can never get a husband, and a bride 
who is found not to be a virgin is invariably sent away ; and 
in former days a girl was killed if she became a mother or 
got with child.6 Among the Wapore of Usambara un- 
married daughters are strictly guarded, because a false step 
on their part reduces the price paid for them.7 The Baziba, 
inhabiting a country to the west and south-west of Lake 
Victoria Nyanza, look upon illegitimate intercourse between 
the sexes before marriage as the most serious offence known 
to their laws, although no action is taken until the birth of 
a child; " then the man and woman are bound hand and 
1 Gouldsbury and Sheane, Great Plateau of Northern Rhodesia, 
p. 141. 
2 Bensen, quoted by Kohler, ' Das Recht der Herero,' in Zeitschr. 
f. vergl. Rechtswiss. xiv. 304. Dannert, Zum Rechte der Herero, p. 26. 
3 Meyer, quoted by Kohler in Zeitschr. f. vergl. Rechtswiss. xiv. 
3<>4- 
4 Fiillebom, Das Deutsche Njassa- und Ruwuma-Gebiet, p. 552. 
5 Ibid. p. 345. See also v. Behr, ' Die Volker zwischen Rufiyi 
und Rovuma,' in Mittheil. Deutsch. Schutzgeb. vi. 79. 
0 Schumacher, ' Das Eherecht in Ruanda,' in Anthropos, vii. 4, 6. 
7 Baumann, Usambara, p. 237. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
153 
foot and thrown into Lake Victoria."1 Among the Bantu 
Kavirondo death was formerly meted out to young men and 
girls who were found guilty of fornication.2 Among the 
Bakoki, another tribe in the Uganda Protectorate, the 
seducer had to pay three cows to the father of the girl and 
one to the chief, and the girl was driven from home and 
remained for ever after an outcast.3 Among the Ban- 
yankole, " should any woman commit fornication and have 
a child before marriage, she is disgraced for life. The clan 
condemns and disowns her as soon as the fact is known."4 
It is not surprising therefore to hear that " there is generally 
chastity amongst the young .women before marriage."6 
If a Busoga girl gets a child and the father does not pay the 
bride price and marries her, she is turned adrift in disgrace 
by her brother and a medicine-man is called in t^ kill a goat 
and cook a meal for the inmates, " thus purifying the house 
from any taint and propitiating the god of the family."6 
Of the natives of the Sese Archipelago in the Victoria Nyanza 
we are told that " if a young woman was seduced, the man 
was obliged to marry her, and in addition to pay a fine of 
two goats ' ;7 and of the Lendn, that for the seduction of a 
girl a fine of four cows is payable to her father.8 In the Madi 
or Moru tribe the girls are carefully looked after, " but as 
marriage usually takes place very early, there is not much 
cause for them to go wrong. This also applies to the men."9 
Among the Akikuyu of British East Africa an unmarried 
girl about to become a mother meets with the gravest dis- 
approval from her parents, and her own companions also 
disapprove of her conduct. The man, in such circumstances, 
can either buy the girl, and take the child, or pay ten goats 
and one sheep, in which case the girl and child remain at 
home. For any second child bom in similar circumstances 
1 Cunningham, Uganda, p. 290. 
2 Johnston, Uganda Protectorate, p. 747. 
3 Cunningham, op. cit. p. 102. 4 Roscoe, Northern Bantu, p. 121. 
5 Johnston, op. cit. p, 630. 6 Roscoe, op. cit. p. 233. 
7 Cunningham, op. cit. p. 94 sq. 8 Ibid. p. 334. 
9 Felkin, ' Notes on the Madi or Moru Tribe of Central Africa,' 
in Proceed. Roy. Soc. Edinburgh, xii. 323. 154 
THE HISTORY OF HUMAN MARRIAGE chap. 
only a small compensation is required, say five goats. 
" It was very definitely stated that the marriage value of a 
girl with such a history would be diminished."1 Among the 
Wagiriama of British East Africa " the sexes aie not allowed 
free sexual intercourse with each other before marriage as in 
some tribes."2 Among the Nandi, "in the event of a 
warrior causing a girl to conceive, he has to slaughter an 
ox when the child is bom. He may take the head away 
himself, but the rest of the animal belongs to the girl's 
father. Except with the Toiyoi clan, the girl is punished 
by being put in Coventry, none of her girl friends being 
allowed to speak to or look at her until after the child is 
bom and buried. She is also regarded with contempt for the 
rest of her life and may never look inside a granary for fear 
of spoiling the com."3 Among the Dinka illegitimate 
children are said to be the result of there being many people 
who cannot marry on account of their poverty ; for each child 
the father has to pay the penalty of four head of cattle, but the 
children appertain to the mother's family.4 Galla parents in- 
culcate very emphatically the virtue of chastity upon their 
daughters, and formerly maidens guilty of incontinence were 
thrown into the Sabaki River and drowned.5 Among the 
Beni Amer the unmarried women are very modest, although 
the married ones believe that they are allowed everything ; 
and if a girl becomes a mother she and the father and the 
child are all killed.6 The same is the case among the Marea.7 
Among the Takue a seducer may have to pay the same sum 
as if he had killed the girl, although the fine is generally 
reduced to fifty cows.8 Among the Beni Mzab a man who 
1 Routledge, With a Prehistoric People, p. 126. 
2 Barrett, ' Notes on the Customs and Beliefs of the Wa-Giriama, 
etc., British East Africa,' in Jour. Roy. Anthr. Inst. xli. 21. 
3 Hollis, Nandi, p. 76. 
4 Petherick, Egypt, the Soudan and Central Africa, p. 393. 
5 Wakefield, ' Marriage Customs of the Southern Gallas,' in 
Folk-Lore, xviii. 325. See also Paulitschke, Ethnographic Nordost- 
Afrikas, ii. 11. 
e Munzinger, Ostafrikanische Studien, pp. 322, 326. 
7 Ibid. p. 243. 
8 Ibid. p. 208. iv CRITICISM OF HYPOTHESIS OF PROMISCUITY 
155 
seduces a girl has to pay two hundred francs and is banished 
for four years.1 In North-Eastem Africa girls are frequently 
subject to infibulation in order to remain chaste until they 
marry. This practice occurs among the Beja, Galla, Somal, 
Massaua, Sudanese, Southern Nubians, and Danakil, as 
also in a part of Kordofan and in Sennaar.2 
Of the ancient inhabitants of the Canary Islands we arc 
told that a woman who lost her virtue was ostracised and 
not spoken to for the rest of her life.3 Concerning the 
Algerian Berbers, Hanoteau and Letourneux write that 
their customs do not tolerate any sexual relation out of wed- 
lock, and that an illegitimate child is killed together with 
its mother.4 In Morocco a bride who is found not to be a 
virgin is frequently sent away by the bridegroom, and in 
some tribes she is killed by her own father or brother.5 
It is a common though not universal custom there that the 
garment with the marks of virginity is publicly exhibited-a 
custom also found in Algeria6 and Egypt,7 among the Swahili 
in East Africa,8 and among the Yoruba of the Slave Coast.9 
Among the last-mentioned people, however, virginity in a 
bride is only of paramount importance when the girl has been 
betrothed in childhood, an unbetrothed girl being mistress 
of her own actions.10 Among most of the Ewhe tribes the 
absence of the primitiae is ipso facto a reason for repudiating 
a bride ; the penalty for seduction is marriage and the 
payment of the head-money which would have been 
1 Chavanne, Die Sahara, p. 315. 
2 Stoll, Das Geschlechtsleben in der Volkerpsychologie, p. 548 sqq. 
Gray, ' Circumcision (Introductory),' in Hastings, Encyclopedia of 
Religion and Ethics, iii. 667, 669. 
3 Cook, ' Aborigines of the Canary Islands,' in American Anthro- 
pologist, N.S. ii. 480. 
4 Hanoteau and Letourneux, La Kabylie et les coutumes Kabyles, 
ii. 148, 187. 
5 See Westermarck, Marriage Ceremonies in Morocco, ch. vii. 
6 Villot, Mceurs, coutumes et institutions des indigenes de l'Algerie, 
p. 107. 
7 Burckhardt, Arabic Proverbs, p. 117 sq. 
8 Velten, Sitten und Gebrauche der Suaheli, p. 135. 
9 Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 154. 
10 Ibid. pp. 154, 184. 156 
THE HISTORY OF HUMAN MARRIAGE chap. 
demanded, or a heavy fine without marriage, with the alter- 
native of enslavement.1 On the Gold Coast a similar rule 
prevails with reference to a bride who is found not to be a 
virgin ; and " if a man betrays a virgin, he is compelled to 
marry her or to pay the price of her dowry, if the parents 
will not consent to the marriage."2 According to Fanti 
customary law, " if a man seduce an unmarried woman, he 
is liable to pay to her family damages for the wrong so done 
her and the disgrace brought on her family."3 In Sierra 
Leone a seducer is called on to pay " virgin money."4 
Among the Ibo-speaking people of the Asaba district in 
Nigeria " a good deal of value is laid upon the virginity of 
the bride."5 Among the Nigerian Kagoro " there is very 
little actual connection before marriage even between be- 
trothed couples, and the girdle of string (ivyan) worn by 
girls is a sign of virginity."6 Among the people of Loango, 
according to Proyart, " a youth durst not speak to a girl 
except in her mother's presence," and " the crime of a maid 
who has not resisted seduction, would be sufficient to draw 
down a total ruin on the whole country, were it not expiated 
by a public avowal made to the king."7 Nowadays at least, 
great pre-nuptial freedom is common among the Congo 
tribes. Of the natives of the Lower Congo Mr. Weeks states 
that, in their lodges, " the sexes were allowed to mix as 
freely as their worst passions prompted " ; and when an 
unmarried woman has a child no fine is paid by its father. 
1 Ellis, Efoe-speaking Peoples of the Slave Coast, pp. 156, 157, 201, 
2c206. Klose, Togo unter deutscher Flagge, p. 255. Dr. Grade 
says (in A us alien Welttheilen, xx. 5) that among the Negroes of 
Togoland a much higher price is paid for a bride who is a virgin than 
for any other. 
2 Cruickshank, Eighteen Years on the Gold Coast, ii. 195, 196, 212. 
3 Sarbah, Fanti Customary Laws, p. 48. 
4 Thomas, Anthropological Report on Sierra Leone. Part I. 
Law and Custom of the Timne and other Tribes, pp. 97, 101. 
5 Idem, Anthropological Report on Ibo-speaking Peoples of Nigeria, 
iv. 65. 
6 Tremeame, ' Notes on the Kagoro and other Nigerian Head- 
Hunters,' in four. Roy. Anthr. Inst. xlii. 169. See also ibid. pp. 172, 
190. 
7 Proyart, ' History of Loango,' in Pinkerton, op. cit. xvi. 568. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
157 
Yet at the same time we are told that such a child, which 
belongs to the woman's family, is spoken of as a " chifd of 
adultery," and that the other children taunt him by saying, 
"You have no father, you came from a tree."1 Among 
various Congo tribes a seducer is fined;2 and the Bayaka 
allow a bridegroom to repudiate the bride if he finds that 
she is not a virgin.3 
The facts stated do not, of course, give us an answer to 
the question whether among the uncivilised peoples generally 
pre-nuptial chastity is more often condemned or condoned. 
Messrs. Hobhouse, Wheeler, and Ginsberg have in a recent 
book dealt with this problem and come to the conclusion 
that among the cases examined by them-about 120 in 
number, probable ones reckoned as a half-those in which 
pre-nuptial relations are condemned are nearly as numerous 
as those in which they are condoned, and that consequently 
" there is no general tendency either way."4 Although I 
prefer giving no figures, partly on account of the indefinite- 
ness of many of the statements, I may say that my own 
collection of facts convinces me that the standard of 
savage chastity has at any rate not been overrated by those 
authors. Considering, moreover, the deteriorating influence 
which contact with civilisation has in so many cases exercised 
on the lower races, and the still more important fact that 
" pre-nuptial chastity " includes all kinds of sexual relations 
previous to formal marriage, however exclusive and constant 
they may be-I do not hesitate to affirm that anything like 
promiscuity among the unmarried is an exception in the 
customs of unadulterated savages. We have seen that even 
among peoples who are notorious for their laxity it is a slur 
upon a girl's reputation frequently to change her lover. 
Professor Hobhouse and his collaborators have also 
1 Weeks, Among the Primitive Bakongo, pp. 163, 108. 
2 Tbrday and Joyce, Les Bushongo, p. no. Hdem, 'Notes on 
the Ethnography of the Ba-Yaka,' in Jour. Anthr. Inst, xxxvi. 48. 
forday, Camp and Tramp in African Wilds, p. 203 (Southern 
Bambala). 
3 Torday and Joyce, in Jour. Anthr. Inst, xxxvi. 45. 
4 Hobhouse, Wheeler, and Ginsberg, Material Culture and Social 
Institutions of the Simpler Peoples, p. 167. 158 
THE HISTORY OF HUMAN MARRIAGE chap. 
examined how far the condemnation or condonation of pre- 
nuptial chastity is correlated with economic advance. They 
have found the agricultural and pastoral peoples to be in 
this respect " decidedly above the hunters."1 So far as the 
" lower hunters " are concerned, they mention three cases 
of condemnation and five of condonation.2 All the former 
and three of the latter refer to Australian tribes ; in other 
words, 50 per cent, of these tribes are said to condemn and 
the other 50 per cent, to condone pre-nuptial chastity 
(all cases of " ceremonial " unchastity being excluded). 
This, however, does not agree with my own facts, according 
to which aboriginal Australian custom in general is very 
strict. The two remaining cases of condonation refer to the 
Punans of Borneo and the Andaman Islanders. Messrs. 
Hose and McDougall say of the former :-" Sexual restraint 
is probably maintained at about the same level as among 
the other peoples, the women being more strictly chaste 
after than before marriage. ... A young man will become 
the lover of a girl generally of some other group than his 
own, and when she becomes pregnant the marriage is cele- 
brated."3 Among the Andaman Islanders, according to 
Portman, " there is a freedom of intercourse between the 
sexes before marriage," but their passions are not gratified 
to any great extent until after marriage.4 We have previously 
noticed the high standard of pre-nuptial chastity among 
some Bushman tribes of South Africa, the Veddas of Ceylon, 
the wild tribes of the Malay Peninsula, the Kubus of Sumatra, 
and the Negritos of the Philippines. Some of these tribes 
practise a primitive kind of agriculture, but they all belong 
to the lowest races now existing. Concerning the Yahgans of 
Tierra del Fuego Mr. Bridges wrote that lewdness is con- 
demned among them as evil and never indulged in with the 
consent of parents, but yet very frequent. ■ With reference 
1 Ibid. p. 167. 
2 Ibid. pp. 177, 179, 181. In the summary on p. 167 the number 
of the latter is given as six, but this is not borne out by the details 
in the Appendix. 
3 Hose and McDougall, Pagan Tribes of Borneo, ii. 183. 
4 Portman, History of Our Relations with the Andamanese, i. 29, 39. iv CRITICISM OF HYPOTHESIS OF PROMISCUITY 
159 
to the lowest races, the general statement may nevertheless 
be made that pre-nuptial unchastity among them is neither 
common nor condoned. 
The same cannot be said of the higher hunters. Professor 
Hobhouse and his collaborators have found among them only 
three cases of condemnation and thirteen of condonation, 
nearly all from North America.1 Some of the latter, however, 
refer to peoples whose morality is said to have been lowered 
by contact with white people or of whom earlier travellers 
have given a more favourable account, and the cases of 
condemnation might have been increased by the Tlingit, 
Aht, Nez Perces, and some other Indians. And as the 
lower hunters, generally speaking, have a stricter standard 
of pre-nuptial chastity than the higher ones, so also the 
lowest agricultural stage comes out materially better than 
the two higher stages (9 cases of condemnation and 2 of 
condonation against respectively 16 and 27, and i8| and 15); 
and the higher agricultural tribes also stand considerably 
below the pastoral ones (6 cases of condemnation and 3I of 
condonation).2 
Whatever may be said about the exactness of these figures 
-the authors themselves would no doubt be the first to 
admit that their value is only relative-I think it is perfectly 
obvious at all events that among the " simpler peoples " 
the standard of pre-nuptial chastity in a tribe is not pro- 
portionate to its degree of culture. There is no evidence 
whatever for the broad statement made by Dr. Hartland, 
that " at first, and for a long time, mere passing amours are 
not regarded, or at least . . . not interfered with," and that 
only by-and-by " virginity comes to have a special market- 
value."3 It seems to me, on the contrary, that in the lowest 
tribes chastity is more respected than in the higher ones. 
This is also what might be expected if marriage is the natural 
and normal relation between the sexes in mankind. Pre- 
1 Hobhouse, Wheeler, and Ginsberg, op. cit. pp. 167, 181, 183, 
185, 187. 
2 Ibid. p. 167. See also Hobhouse Morals in Evolution, p. 174 n.i. 
3 Hartland, Primitive Paternity, ii. 93. 160 
THE HISTORY OF HUMAN MARRIAGE chap. 
nuptial chastity or unchastity largely depends on the age 
when marriages are contracted. This holds true of savages, 
as is directly indicated in some of the statements quoted 
above, and it is true of civilised peoples as well. It has been 
proved that in the cities of Europe prostitution increases 
according as the number of marriages decreases j1 and 
Engel and others have shown that the fewer marriages are 
concluded in a year the greater is the ratio of illegitimate 
births.2 At the lower stages of culture celibacy is much 
rarer and marriage is entered into at an earlier age than 
among ourselves. But even in savage life there are circum- 
stances which may compel adult persons to live unmarried 
for a longer or shorter time. A man may be too poor to 
maintain a wife, or where he has to buy her he may be unable 
to pay the price, or the polygyny of some may lead to the 
celibacy of others. These obstacles, however, would occur 
chiefly where some advancement in culture has been made, 
and in a much smaller degree under more primitive con- 
ditions, where consequently there would be less reason for 
pre-nuptial unchastity. 
The difficulty of procuring a wife, however, is not the only 
cause of sexual intercourse before marriage. In many cases, 
as we have seen, the pre-nuptial relation is a kind of trial, 
which, if successful, leads to marriage. Its object may be 
to ascertain that the woman will gratify her lover's desire for 
offspring, and in such a case the marriage is not concluded 
before the birth of a child or until there are signs of pregnancy. 
In his notes on the For tribe of Central Africa, for example, 
Dr. Felkin states that in the Gebel Marah district, where 
barrenness is common, " the men always make sure of a 
woman's fertility before marriage."3 But offspring may 
also be secured in another way, namely, by marrying a 
woman who has conceived by another man. Among the 
Angami Nagas, although youthful licentiousness never ends 
in marriage, men are desirous to have proof in advance that 
1 v. Oettingen, Moralstatistik, pp. 199, 216. 
3 Ibid. p. 327. 
3 Felkin, ' Notes on the For Tribe of Central Africa,' in Proceed. 
Roy. Soc. Edinburgh, xiii. 207 sq. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
161 
their wives will not be barren.1 The Akamba in British 
East Africa " have no respect for maidens, and regard a 
pregnant girl as the most eligible spouse, exactly as if she 
were a cow in calf." 2 Of various peoples we are told that 
a young woman is liked the better and more desired in 
marriage for having borne a child.3 Among the Mong- 
wandi of the Upper Mongala region a grown woman who has 
already been a mother costs quite six times the price of a 
little girl.4 Nay, among the Bagas-Foreh, inhabiting 
the small islands at the mouth of the Rio Nunez in French 
Guinea, a young lady cannot hope to find a husband unless 
she has given birth to two children who are already old 
enough to walk.6 Indeed, a woman who is a mother 
may be a more acceptable wife, not only because she has 
proved that she can bear children, but on account of the 
direct value of the offspring she brings with her.6 
There may be yet other reasons why a man prefers 
marrying a girl who is not a virgin. Of the Indians of Quito 
we are told by Juan and Ulloa that a virgin is never the 
object of their choice ; "for they look on it as a sure sign, 
that she who has not been known to others, can have nothing 
pleasing about her."7 A similar view is taken by some of the 
Ewhe in the interior of Togoland, contrary to the regard in 
which virginity is held by those living on the coast.8 Con- 
cerning the Kamchadal, Steller says that if a girl was given 
in marriage as a virgin-which sometimes happened in 
1 Prain, loc. cit. p. 491 sq. 2 Eliot, East Africa Protectorate, p. 125. 
3 Delafosse, ' Le peuple Sidna ou Senoufo,' in Revue des etudes 
ethnographiques et sociologiques, i. 483. Roscoe, Northern Bantu, 
p. 171 (Bagesu). Buch, ' Die Wotjaken,' in Acta Soc. Scientiarum 
Fennicae, xii. 509. Melnikow, ' Die Burjaten des Irkutskischen 
Gouvemements,' in Verhandl. Berliner Gesellsch. Anthr. 1899, p. 442. 
Crisp, * Account of the Inhabitants of the Poggy Islands,' in Asiatick 
Researches, vi. 87 sq. 
4 Johnston, George Grenfell and the Congo, ii. 677. 
6 Leprince, reviewed in L'Anthropologie, xi. 769 sq. 
8 See Tessmann, op. cit. ii. 258 sq. (Pangwe). 
7 Juan and Ulloa, ' Voyage to South America,' in Pinkerton, 
op. cit. xiv. 521. 
8 Klose, op. cit. p. 255. 162 
THE HISTORY OF HUMAN MARRIAGE chap. 
former times-the bridegroom was dissatisfied and scolded 
her mother because she had neglected her education in the 
art of venery.1 Speaking of Tibetans, Marco Polo observes 
that no man among them " would on any consideration take 
to wife a girl who was a maid ; for they say a wife is nothing 
worth unless she has been used to consort with men. And," 
he adds, " their custom is this, that when travellers come 
that way, the old women of the place get ready, and take 
their unmarried daughters or other girls related to them, 
and go to the strangers who are passing, and make over the 
young women to whomsoever will accept them; and the 
travellers take them accordingly and do their pleasure ; after 
which the girls are restored to the old women."2 This 
statement suggests that we have here to do with a de- 
floration rite of the kind which will be discussed in the next 
chapter: if the girl was considered " worth nothing " as 
long as she remained a virgin, the reason for it was pre- 
sumably the idea that defloration was coupled with danger, 
or that intercourse with a stranger was beneficial, or both 
ideas combined. Notions of this sort have certainly to be 
taken into consideration when we examine the causes of 
pre-nuptial unchastity. They seem, partly at least, to 
account for the custom which requires girls to earn their 
dowries by prostitution before they marry.3 
That promiscuous pre-nuptial unchastity is a survival of 
earlier general promiscuity is a legitimate assumption only if 
we may assume that the causes to which it can be traced 
have operated on an infinitely larger scale in the past than 
they do at present. And for such an assumption there is no 
justification whatever. We have noticed that the obstacles 
to early marriages are mainly due to advancement in culture. 
Trial unions cannot be called promiscuous. Nor do super- 
stitious beliefs concerning defloration generally lead to 
anything like promiscuity. The cases in which the depre- 
ciation of virginity is stated to be due to desire for offspring 
are quite exceptional and cannot possibly be regarded as the 
1 Steller, Beschreibung von dem Lande Kamtschatka, p. 346. 
2 Marco Polo, Kingdoms and Marvels of the East, ii. 44. 
3 See infra, i. 200. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
163 
expression of a tendency which was ever general. On the 
other hand, there is ample evidence of men giving preference 
to virgin brides, and this preference is probably very ancient. 
It seems to spring partly from a feeling akin to jealousy 
towards women who have had previous connections with 
other men,1 and partly from an instinctive appreciation of 
female coyness. Each sex is attracted by the distinctive 
characteristics of the opposite sex, and coyness is a female 
quality. Hence conspicuous eagerness in a woman appears 
to a man unwomanly, repulsive, contemptible ; his ideal is 
the virgin, the libertine he despises. This preference for 
virginity and female coyness has undoubtedly tended to 
keep pre-nuptial intercourse, especially of a promiscuous 
kind, in check, and has at the same time influenced the moral 
judgment of it.2 
One more cause of pre-nuptial unchastity remains to be 
considered, namely, the taste for variety. This has been 
much emphasised by Dr. Bloch, who bases on it his chief 
argument in favour of original promiscuity. It is, he says, 
" perfectly clear that the human need for sexual variety, 
which is an established anthropological phenomenon, must 
in primitive times have been much stronger and more 
unbridled, in proportion as the whole of life had not hitherto 
risen above the needs of purely physical requirements. 
Since even in our time, in a state of the most advanced 
civilisation, after the development of a sexual morality 
penetrating and influencing our entire social life, this natural 
need for variety continues to manifest itself in almost 
undiminished strength, we can hardly regard it as necessary 
to prove that in primitive conditions sexual promiscuity 
was a more original, and, indeed, a more natural, state than 
marriage. From the purely anthropological standpoint . . . 
permanent marriage appears a thoroughly artificial institu- 
tion, which even to-day fails to do justice to the human need 
for sexual variety."3 That the sexual instinct is stimulated 
1 See infra, i. 316 sq. 
2 See Westermarck, Origin and Development of the Moral Ideas, 
ii. 434 sqq. 
3 Bloch, op. cit. p. 192. 164 
THE HISTORY OF HUMAN MARRIAGE chap. 
by a change of its object is an undeniable fact. That this 
taste for variety is a cause of much extra-matrimonial inter- 
course of a more or less promiscuous character is also well 
known. But the assumption that it dominated primitive 
man to such an extent as to exclude all unions of a greater 
durability is warranted by nothing that is known either 
about anthropoid apes or savage men. Indeed, promis- 
cuity was not the only means by which he might have 
satisfied his " need for sexual variety " ; if he got tired of his 
wife he might take another one. This would seem to have 
been the more sensible course to choose ; for a wife is not 
only a source of sexual pleasure but a helpmate, a food- 
provider, a cook, and a mother of children. This is a point 
of view which Dr. Bloch and some other authors are too 
apt to overlook when they speak of early marriage. 
If, as I maintain, marriage is based on an instinct acquired 
as a means of preserving the life of the progeny, sexual 
intercourse must no doubt originally, as a general rule, have 
been followed by a prolonged union between the parties. 
But this does not imply that the connection between sexual 
intercourse and marriage would always have remained 
equally close. As a social fact and a social institution 
marriage has been greatly influenced by the progress of 
civilisation, and extra-matrimonial relations can be easily 
explained without recourse to the hypothesis that in the 
beginning all relations between the sexes were promiscuous. 
When we consider that in our own midst prostitution has 
shown a tendency to increase in a higher ratio than popula- 
tion,1 and that in spite of the general infertility of prosti- 
tutes there are towns in Europe where the illegitimate births 
outnumber the legitimate ones,2 it is nothing less than absurd 
to speak of the unchastity of unmarried savages as a relic of 
an alleged primitive stage of promiscuity. 
As a survival of ancient communism in women has 
also been regarded the fact that courtesans have sometimes 
been held in greater estimation than married women.3 
1 v. Oettingen, op. cit. p. 199. 2 Ibid. p. 317. 
3 Avebury, Origin of Civilisation, p. 107. Giraud-Teulon, Les 
origines du mariage et de la famille, p. 43 sqq. Wilutzky, op. cit. i 27. IV CRITICISM OF HYPOTHESIS OF PROMISCUITY 
165 
Lord Avebury observes that they were highly respected at 
Athens, and gives instances in which high rank attached 
to the " Chief of the Courtesans " in the Indian city of 
Vesali. Moreover, " in Java we are told that the courte- 
sans are by no means despised, and in some parts of Western 
Africa the negroes are stated to look on them with respect."1 
Such feelings, Lord Avebury argues, would naturally arise 
when the courtesans were originally fellow-countrywomen 
or relations, and the special wives were captives and slaves ; 
and they would long survive the circumstances from which 
they sprang.2 The courtesans are thus regarded as repre- 
sentatives of the communal wives of primitive times. This 
conclusion is particularly startling when we consider the 
state of culture of the chief peoples concerned. McLennan 
justly remarks, with reference to those " communal wives," 
that " if any inference is to be made from their standing in 
Athens, in the brilliant age of Pericles, as to the state of 
matters in the primitive groups, proof of primitive commu- 
nism in women might as well be sought in London or Paris 
in our own day. Far back in the interval between savagery 
and the age of Pericles are the heroes of Homer, with their 
noble wedded wives."3 If courtesans were respected 
and sought after even by the principal men in Athens, the 
simple reason was that they were the only educated women 
there. In India prostitutes are attached to temples, and, 
according to Dubois, are the only females who may learn 
to read, to sing, and to dance; 4 and it is said to be the 
custom for the Hindus in large towns to frequent the society 
of courtesans for the charm of their witty and pointed 
conversation.5 
1 Avebury, op. cit. p. 438 sq. 2 Ibid. p. 107. 
3 McLennan, Studies in Ancient History, p. 343. 
4 Dubois, Description of the Character, Manners, and Customs 
of the People of India, p. 295. 
0 Russell, Tribes and Castes of the Central Provinces of India, iii. 
379- CHAPTER V 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : 
the Jus Primae Noctis 
The hypothesis of promiscuity or " communal marriage " 
is supposed to derive much support from certain customs 
which are interpreted as acts of expiation for individual 
marriage. In many cases, we are told, the exclusive posses- 
sion of a wife could only be legally acquired by a temporary 
recognition of the pre-existing communal rights. As a 
recognition of this kind is regarded the jus primae noctis 
accorded to a priest, king, chief, or nobleman, who is then 
looked upon as a representative of the community after 
the ancient right was taken away from its male members 
in general.1 
Among some of the Brazilian Indians the jus primae 
noctis is said to be granted to the medicine-man2 or to the 
1 Bachofen, Das Mutterrecht, pp. 12, 13, 17, 18, &c. Avebury, 
Origin of Civilisation, p. 437 sq. Giraud-Teulon, Les origines du 
manage et de la famille, pp. 32, &c. Kulischer, ' Die communale 
" Zeitehe " und ihre Ueberreste,' in Archiv f. Anthrop. xi. 223. 
Post, Die ■ Geschlechtsgenossenschaft der Urzeit, p. 37. Wilken, 
' Over de primitieve vormen van het huwelijk en den oorsprong van 
het gezin,' in De Indische Gids, 1880, vol. ii. 1196. Liebrecht, 
Zur Volkskunde. p. 423 sq. Wilutzky, Vorgeschichte des Rechts, 
i. 34 sqq. Bloch, Sexual Life of Our Time, p. 190. See Schmidt, 
' Das Streit uber das jus primae noctis,' in Zeitschr. f. Ethnol. 
xvi. 44 sq. 
2 v. Martius, Beitrage zur Ethnographie Amerika's, i. 113 (Juris, 
Passes). 
166 ch. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
167 
chief.1 Oviedo y Valdes states that among the Arawaks 
and some peoples in the province of Paria a virgin bride had 
to spend the first night with a fiache, or priest.2 In Cumana, 
in the present Venezuela, legitimate wives, but not con- 
cubines, were deflowered by the priests, and it was considered 
a great crime not to conform to this custom.3 So also the 
Caribs of Cuba strictly prohibited a bridegroom from lying 
with the bride during the first night of their marriage. If 
it was a cacique that married, he invited other chiefs to lie 
with his bride, and if the bridegroom was a man of somewhat 
lower rank he invited his equals ; but people of the lowest 
class " borrowed on this occasion the charitable cares of 
their caciques and of their priests."4 In Guatemala it was 
customary for the high-priest to spend the first night with 
the bride.5 Among the Nicaraguans, according to Anda- 
goya, " a man whom they held as a pope, and who lived in a 
temple, had to sleep with the bride on the previous night " ;6 
whilst according to Gomara, many virgins were given to the 
caciques to be deflowered.7 Castaneda de Najera wrote 
in the middle of the sixteenth century that among the 
Tahus in the province of Culiacan, in the New Kingdom of 
Galicia (210 leagues west of Mexico), it was the custom for 
the husbands to buy the women whom they married of their 
fathers and relatives at a high price, and then to take them 
to a chief who was considered to be a priest, to deflower 
1 v. Spix and v. Martins, Reise in Brasilien, iii. 1182 (Jumanas), 
1189 (Culinos). 
2 Oviedo y Valdas, Historia general y natural de las Indias, book 
xxiv. ch. 3, pt. ii. vol. i. 222. 
3 Gomara, ' Primera parte de la historia general de las Indias,' 
in Biblioteca de autores espanoles, xxii. 206. Coreal, Voyages aux 
Indes Occidentales, i. 139 sq. Herrera, General History of the West 
Indies, iii. 310. Simon, Primera parte de las Noticias historiales 
de las Conquistas de tierra firme en las Indias Occidentales, iv. 26. 3, 
p. 320. 
4 Coreal, op. cit. i. 10 sq. Carli, Le lettere americane, i. 71. 
5 Herrera, op. cit. iii. 341. 
0 Andagoya, Narrative of the Proceedings of Pedrarias Davila 
in the Province of Tierra Firm?, p. 33 sq. (the early part of the 
sixteenth century). 
7 Gomara, loc. cit. p. 283. 168 
THE HISTORY OF HUMAN MARRIAGE chap. 
them and see if they were virgins.1 Among the Tarahumare 
in modem Mexico, according to Lumholtz, " the shamans 
avail themselves of jus primae noctis." 2 Among the Kinipetu 
Eskimo near Hudson's Bay the same right is said to belong 
to the ankut, or priest.3 
Among the Ballante of Senegal the king has not only the 
power of life and death over his subjects, but also le droit du 
seigneur everywhere in the tribe, and no girl can marry 
before she has been deflowered by him.4 The same right 
is said to have belonged to the chief of the Bagele in Adam- 
awa, a native kingdom in Northern Nigeria and Kamerun,5 
and, in the days of Herodotus, to the king of the Adyrmachi- 
dae, a Libyan tribe reckoned to Egypt.6 Of some of the 
ancient inhabitants of the Canary Islands we are told that 
their chiefs had the maidenheads of all virgins that married ;7 
or, that " the night before the bride was presented to her 
husband, she was delivered to the Guanarteme (or king), 
who, if he did not chuse to lie with her himself, gave her to 
the Faycag (a priest or lawyer next in dignity to the king), 
or to some other noble person of his intimate acquaintance, 
to enjoy her."8 In 1632, however, Abreu de Galindo wrote 
that the natives denied that such a custom ever existed 
among their ancestors.8 But there are rumours of the exist- 
ence of something similar among certain modem Berbers. 
Two Berbers from the tribe Ath Ubahthi in the eastern part 
of Morocco, near Ujda, told me that among their neigh- 
bours, the Ath Zihri (Zkara), it is the custom for the chief of 
1 Castafieda de Najera, ' Relation de la Jornada de Cibola,' in 
Ann. Rep. Bur. Ethn. xiv. pt. i. 448, 513 sq. 
2 Lumholtz, Unknown Mexico, i. 270. 
3 Klutschak, Ais Eskimo unter den Eskimos, p. 234. 
4 Marche, Trois voyages dans l'Afrique Occidental, p. 70. 
5 Barth, Reisen in Nord- und Central-Afrika, ii. 571 n.* 
6 Herodotus, iv. 168. 
7 Alvise Cadamosto, ' Delle navigation!,' in Ramusio, Navi- 
gationi et viaggi, i. 106. Barros, L'Asia, i. 12, vol. i. 24. Faria y 
Sousa, Asia Portoguesa, i. i. i. 12, vol. i. 14. Walckenaer, Histoire 
ginirale des voyages, i. 76. 
8 Abreu de Galindo, History of the Discovery and Conquest of the 
Canary Islands, p. 69. 
9 Ibid. p. 69. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
169 
the village to be secretly sent for by the bridegroom's family 
to have intercourse with the bride the first night, unless she 
be a widow or a divorced wife. They said that this is well 
known to all the neighbouring tribes, though denied by the 
Ath Zihri themselves. This statement agrees with the in- 
formation obtained by M. Moulieras, although, according to 
him, the person who is said to perform the act belongs to the 
caste of the rusma, or spiritual leaders of the tribe.1 It is 
impossible for me to decide whether there is any truth or 
not in these accounts, but they may perhaps derive some 
support from certain curious customs among other Berbers 
in Morocco,2 as also from some statements relating to the 
Beni Ulid, an Arab-speaking mountain tribe in the neigh- 
bourhood of Fez. An old man from the Hiaina, another tribe 
near Fez, told me that formerly it was the custom among 
them that on the evening of the day when the bride arrived 
at her new home the bridegroom, accompanied by members 
of his family and women, went to a neighbouring village to 
fetch from there a man to spend the first night with the 
bride and have intercourse with her. When they fetched 
him the women sang, " Rejoice O lady and continue to 
rejoice, and the perforator has come to you from the Beni 
Ulid." According to another account given me by two 
Berbers from other tribes in the same part of Morocco, 
it was the custom among the Beni Ulid that, if the bride- 
groom was not strong enough to deprive the bride of her 
virginity, he asked his best-man to inform his father about it 
and tell him to send for some man who was reputed for his 
virility. A messenger on horseback was despatched to 
fetch such a man, who was then received by the women 
singing a ditty similar to the one already mentioned. He 
was paid for his service. But the accuracy of these state- 
ments, too, is doubtful, coming as they do from neighbours, 
who are not always trustworthy informants.3 
Among the Tachtadshys in Lycia there are tribes in 
1 Mouli6ras, Une tribu Z enite anti-musulmane au Maroc (les 
Zkara), p. 85 sqq. 
2 See Westermarck, Marriage Ceremonies in Morocco, p. 273. 
3 Ibid. p. 271 sq. 170 
THE HISTORY OF HUMAN MARRIAGE chap. 
which the religious head, called dede, possesses the jus 
primae noctis, even though he does not regularly exercise it, 
and other tribes in which he is entitled to choose any woman 
he likes at the yearly religious assemblies.1 The Dusik- 
Kurds in the Dersim mountains south of Erzingan have 
every year a great feast, when the men and the married 
women assemble in a large room and have promiscuous 
intercourse. After the congregation have kissed the chodsha, 
or priest, on the hand, he cries out, " I am the great bull, 
not a fattened ox 1 " The latest married among the women, 
by preference one who has had her wedding on the same day, 
steps forward to him and says, " I am the young cow ! " 
When she says this, the lights are extinguished and the 
orgies begin.2 Among the Zikris, an heretical Muhammadan 
sect in Baluchistan, the Mulla has the jus primae noctis, 
although this right can be redeemed by a small money 
payment to him.3 
In the old kingdom of Cambodia, according to a Chinese 
account from the end of the thirteenth century, parents 
chose Buddhist or Taouist priests to deprive their daughters 
of their virginity before their marriage. This ceremony, 
which was called chin-than, was performed on a certain day 
once a year, which was fixed by the magistrate of the place. 
Each priest was allowed to deflower one girl only every year, 
and he was handsomely rewarded for his service. Owing 
to the difficulty of procuring the necessary gifts, poor girls 
might have to wait till their eleventh year before they could 
marry, whereas rich girls generally married at the age of 
seven to nine. But there were people who supplied poor 
parents with the money required for the defloration of their 
daughters, and this was looked upon as a meritorious deed.4 
M. Aymonier maintains that this account is too strange to 
1 Petersen and Euschan, Reisen in Lykien, Milyas und Kibyratis, 
p. 199 n. 1. 
2 Blau, * Nachrichten uber kurdische Stamme,' in Zeitschr. 
Deutsch. Morgenland. Gesellsch. xvi. 623 sq. 
3 Hughes-Buller, Census of India, 1901, vol. v- (Baluchistan) 
Report, p. 45. 
4 Remusat, Nouveaux melanges asiatiques, i. Ii6^sq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
171 
deserve credit / but the existence of very similar customs 
among other Asiatic peoples spoils his argument. 
On the coast of Malabar, Brahmans have for a long time 
acted as deflowerers of brides. Barthema wrote in the 
beginning of the sixteenth century that when the king of 
Calicut takes a wife he selects the most worthy and the 
most honoured of the Brahmans and makes him sleep the 
first night with his wife in order that he may deflower her. 
But, he says, " do not imagine that the Brahman goes 
willingly to perform this operation ; the king is even obliged 
to pay him four hundred or five hundred ducats." Barthema 
adds that the king only, and no other person in Calicut, 
adopts this practice ;2 but other travellers give us different 
information. Hamilton, who lived in the East Indies in 
the latter part of the seventeenth and the earlier part of the 
eighteenth century, writes :-" When the Samorin marries, 
he must not cohabit with his bride till the Nambourie or 
chief priest has enjoyed her, and, if he pleases, may have 
three nights of her company, because the first fruits of her 
nuptials must be an holy oblation to the god she worships : 
and some of the nobles are so complaisant as to allow the 
clergy the same tribute; but the common people cannot 
have that compliment paid to them, but are forced to supply 
the priests' places themselves."3 Admiral Verhoeven, who 
visited Calicut in 1608, also states that the brides of 
lords and noblemen, but not of commoners, were often 
deflowered by Brahmans; 4 whilst Sir Thomas Herbert, 
who sailed for the East Indies in 1626, says that it was 
formerly the custom among the Nayars that a Brahman 
had the first night's company with the bride, but that 
this practice had been wholly abrogated.6 On the other 
1 Aymonier, Le Cambodge, iii. 625. 
2 de Barthema, Itinerario nello Egypte, &c. fol. li.a. Travels 
of Ludovico di Varthema, p. 140. 
3 Hamilton, ' New Account of the East Indies,' in Pinkerton, 
Collection of Voyages, viii. 374. 
4 Verhoeven, Kurtze Beschreibung einer Reyse in die Ost Indien, 
p. 56. 
6 Herbert, Travels into Divers parts of Africa, and Asia the Great, 
P- 337- 172 
THE HISTORY OF HUMAN MARRIAGE chap. 
hand, we are told by Mandelsloe, who stayed there a few 
years later (1638-1639), that the Brahmans of Malabar 
were held in such veneration that they had the first fruits of 
all the brides, and that the richer people invited them to 
perform this task with very considerable presents.1 Accord- 
ing to Roggewein, the Brahmans had introduced into 
Cochin the custom that when any man married he was 
absolutely forbidden to sleep with his wife the first night, 
and that this function was to be performed in his stead by 
one of the Brahmans or, if there were none, by some other 
man.2 And various other writers state in a general way 
that among the inhabitants of Cochin or Malabar it was 
the custom for Brahmans to deflower the brides.3 But in 
several accounts of extra-matrimonial defloration of brides in 
Malabar no mention at all is made of Brahmans.4 Lopez 
de Castanheda states of the sisters of the Samorin and 
other kings of Malabar that when any of them reaches 
the age of ten their kindred send for a young man of 
the Nayar, or military, caste out of the kingdom, and 
give him presents to induce him to deprive the young 
virgin of her maidenhood ;5 and in another early 
Portuguese account it is said that when a Nayar woman has 
two or three young daughters she, for the same purpose, 
chooses a Nayar for each of them.6 According to Navarette's 
1 ' Remarks and Observations made by J. A. de Mandelsloe, 
in his Passage through several Countries of the Indies,' in Harris, 
Navigantium atque I tiner antium Bibliotheca, i. 767. 
2 Roggewein, ' Account of Commodore Roggewein's Expedition 
for the Discovery of Southern Lands,' in Harris, op. cit. i. 297. 
3 Balbi, Viaggio dell' Indie Orientali. foil. 75b, 137a. Schouten, 
Ost-Indische Reyse, p. 168. Sonnerat, Voyage aux Indes Orientates, 
i. 68. Guyon, New History of the East-Indies, i. 431. de Gubernatis, 
Memoria intorno ai viaggiatori italiani nelle Indie Orientali dal 
secolo XIII a tutto il XVI, p. 137. See also Gaya, Ceremonies 
nuptiales de toutes les nations, p. 56 sq. 
4 See infra, i. 184 sqq. 
5 Lopez de Castanheda, Historia do descobrimento e conqvista da 
India pelos Portvgveses, i. 45. 
0 ' Sommario di tutti li regni, citta, & popoli orientali,' in Ramusio, 
op. cit. i. (1563) 331 D. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
173 
description, from the latter part of the seventeenth century, 
persons who were acquainted with the coast of Malabar 
said " that when some Persons marry'd, the Husband 
carry'd his Wife before he had to do with her himself, to the 
King, who kept her eight Days in his Palace, making use of 
her at his pleasure ; and that time being expir'd, the Man 
came for his Wife, taking it as a great Honour and Favour 
that his King would make use of her. In other places they 
carry them to the Temples of the Idolatrous Priests, and 
left them there the same number of Days to the same 
purpose."1 Admiral Jakob van Neck states that when one 
of the great lords of Goa married, it was the custom for him 
to take his bride to the sovereign and ask him to sleep with 
her tor the first three nights.2 
The Bhutanese, in the eastern part of the Himalayas, 
according to Mr. Claude White, " follow certain curious 
customs, such as the right of the head man when girls 
marry " ; but, he adds, " this is being put a stop to by the 
present Tongsa."3 The ruin of the ancient city of Harappa, 
in the Punjab, is ascribed by the people to the vengeance of 
God on its governor, " who claimed certain privileges on the 
marriage of every couple in his city, and in the course of his 
sensualities, was guilty of incest."4 In his book on the 
tribes of the Hindu-Kush, Major Biddulph wrote in 1880 
with reference to the people of Hunza, who have only the 
laxest form of Muhammadanism, " The droit du seigneur was 
exercised by the father of the present ruler, and though 
the custom has been allowed to fall into disuse, it is evident 
from the accounts given of weekly orgies held by Ghazan 
Khan that the right is only held in abeyance and not for- 
mally renounced."5 From ancient Arabia we have the 
story of king 'Amliq of Tasm, who compelled the subject 
1 Navarette, ' Account of the Empire of China,' in Churchill, 
Collection of Voyages and Travels, i. 320. 
2 van Neck, quoted by Schmidt, Jus primae noctis, p. 312. 
3 White, Sikhim and Bhutan, p. 13. 
4 Burnes, Travels into Bokhara, iii. 137. 
5 Biddulph, Tribes of the Hindoo Koosh, p. 77. 174 
THE HISTORY OF HUMAN MARRIAGE chap. 
tribe of Djadis to take to him every bride before she had 
connection with her husband.1 And according to another 
story, the last king of Saba, by name Sharahbil, allowed no 
girl in his country to marry without being deflowered by 
him.2 
It has been a widespread popular belief that a droit du 
seigneur existed in feudal times in Europe, and it is said 
that in some places a similar right was accorded to the clergy.3 
The existence of such rights has also been reported by 
historians. Thus old writers on the history of Scotland 
tell us that King Evenus HI., contemporary with Augustus, 
made a law by which he and his successors in the throne 
were authorised to He with every bride, if a woman of 
quality, before her husband could approach her ; and in 
consequence of this law the great men of the nation had 
a power of the same kind over the brides of their vassals 
and servants. The law was strictly observed throughout 
the kingdom, and was only discontinued or repealed more 
than ten whole centuries afterwards, when the importunities 
of St. Margaret prevailed with her husband, Malcolm 
Cammor (or Canmore), to abofish this unjustifiable custom. 
From that time forward, the vassal or servant was allowed 
to redeem the first night of his bride by paying a tax in 
money, which was called mercheta mulierum.-The story was 
first told by Boece, or Boethius,4 and subsequently repeated 
by other historians.5 It was also referred to by an ItaHan 
writer fiving at the beginning of the seventeenth century, 
Bonifacio Vannozzi, who asserted that a similar custom 
1 Abulfeda, Historia anteislamica, p. 182 sq. Rasmussen, Historia 
prcecipuorum Arabum regnorum, p. 81 sq. Caussin de Perceval, 
Essai sur I'histoire des Arabes, i. 28 sq. See also Wellhausen, ' Die 
Ehe bei den Arabern,' in Nachrichten d. Konigl. Gesellsch. d. Wissensch. 
zu Gottingen, 1893, p. 464. 
2 Weil, Biblische Legenden der Muselmdnner, p. 253. 
3 Schmidt, Jus primae noctis, p. 43 sqq. 
4 Boethius, Scotorum historiae a prima gentis'origine, book iii. 
fol. 34 b sq. 
5 Leslie, De origine moribus, et rebus gestis Scotorum, p. 96. 
Buchanan, Rerum Scoticarum historia, foil. 31 b, 64 a. Mackenzie, 
Lives and Characters of the most Eminent Writers of the Scots Nation, 
i. 132. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
175 
formerly existed in Piedmont, and that a cardinal of the 
house of Rovere had told him how he himself had burned 
a charter conveying to his house the privilege of deflowering 
the brides whom their vassals married. This story afterwards 
obtained currency through Bayle, who retold it in his great 
Dictionary.1 In Russia landlords were said to claim this 
right as late as in the nineteenth century.2 
On the other hand, the existence of a droit du seigneur in 
any form in Europe has been eagerly disputed.3 In the 
middle of the eighteenth century this was done by Grupen, 
who reviewed the materials then available, beginning with 
Lactantius' statement 4 that the Emperor Maximinus did 
xOt allow any marriage to take place without his consent 
so that he should be able to be the praegustator at the 
wedding.5 A century later Veuillot tried to explain away the 
chief evidence on which the popular view of the subject 
was based ;6 and subsequently Dr. Karl Schmidt in a very 
learned work, after a penetrating discussion of the whole 
question, arrived at the conclusion that the belief in the 
droit du seigneur was nothing but " ein gelehrter Aberglaube." 
The law, he says, which is believed to have extended over a 
large part of Europe, has left no evidence of its existence in 
law-books, charters, decretals, trials, or glossaries.7 In the 
latter part of the eighteenth century the Scotch story 
connected with the mythical King Evenus III. was denounced 
as unhistorical by Macpherson8 and Lord Hailes.9 And of 
1 Bayle, Dictionnaire historique et critique, xiii. 335, ' Sixte IV.' 
2 Schmidt, op. cit. p. 221. 
3 For opinions in favour of and against the existence of such a 
right see Schmidt, op. cit. p. 1 sqq. ; Idem, ' Das Streit fiber das jus 
primae noctis,' in Zeitschr. f. Ethnol. xvi. 19 sqq. 
4 Lactantius, De mortibus persecutorum, 38 (Migne, Patrologiae 
cursus, vii. 255). 
6 Grupen, De uxore theotisca, p. 1 sqq. 
G Veuillot, Le droit du seigneur au moyen age, passim. 
7 Schmidt, Jus primae noctis, pp. 379, &c. 
8 Macpherson, Critical Dissertations on the Origin, &c. of the 
Antient Caledonians, p. 175 sqq. 
9 Lord Hailes (Sir David Dalrymple), Annals of Scotland, vol. iii. 
Appendix I., 'Of the Law of E venus, and the Mercheta mulierum,' 
p. 1 sqq. 176 
THE HISTORY OF HUMAN MARRIAGE chap. 
the alleged droit du seigneur in Russia Alexander Herzen 
wrote that no such right ever existed there, and that if the 
law were put in execution it would punish with the same 
severity the violation of a female serf as that of a free 
woman. Yet he admits that it was an easy matter for a 
landlord to outrage the daughters and wives of his serfs 
with impunity.1 
The belief in a mediaeval jus primae noctis has been traced 
to two main sources. One is the so-called marchet or merchet, 
a fine paid by a vassal to his lord on the marriage of his, 
the tenant's, daughter, which has been misinterpreted as a 
pecuniary commutation for a former right claimed by a 
feudal lord of sleeping with his vassal's daughter on their 
wedding night. Lord Hailes gave the following explanation 
of this fine, as existing in old Scots law :-" Persons of low 
rank, residing on an estate, were generally either ascripti 
glebae, or were subjected to some species of servitude similar 
to that of the ascripti glebae. On that estate they were 
bound to reside, and to perform certain services to the lord. 
As women necessarily followed the residence of their 
husbands, the consequence was, that when a woman of that 
rank married a stranger, the lord was deprived of that part 
of his live stock. He would not submit to this loss, without 
requiring indemnification; at first, the sum paid by the 
father of the young woman would nearly amount to an 
1 Herzen, Le peuple russe et le socialisms. Letire d M. J. Michelet, 
p. 38. I have to thank Mr. S. Rapoport for referring me to this 
pamphlet.-How Russian landlords might behave before the 
emancipation of the serfs is shown in an article named 3anncKH 
ce.ibcKaro cBHiueHHiiKa (' Notes of a Country Clergyman ') in Pyccuaa Crapaiia 
Russian Antiquity,' xxvii. 63, 77), where it is said of one of them : 
" Often N. I- tsh would stroll late in the evening about his village 
to admire the prosperous condition of his peasants ; he would stop 
at some cottage, look in at the window, and tap on the pane with his 
finger. This tapping was well known to everybody, and in a moment 
the best-looking woman of the family went out to him." Another 
landlord, whenever he visited his estate, demanded from the manager, 
immediately after his arrival, a list of all the grown-up girls. " Then, 
the author continues, " the master took to his service each of the girls 
for three or four days, and as soon as the list was finished, he went 
off to another village. This occurred regularly every year." v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
177 
estimated indemnification; and as the villains were 
grievously under the power of their lord, it would be often 
exorbitant and oppressive. In process of time, the lord 
would discover, that as the young women of his estate were 
exported, the young men of his estate would import others ; 
so that, upon the whole, no great prejudice could arise 
from extra-territorial marriages. Hence the indemnification 
would be converted into a small pecuniary composition, 
acknowledging the old usage, and the right of the master. 
As the intrinsic and marketable value of money decreased, 
this stated composition would be gradually omitted out of 
terriers and rent-rolls, or would be thrown into the aggregate 
sum of rent." 1 As to the merchet in English law a similar 
suggestion has been made by Maitland, namely, that the 
idea at the root of it was " much rather that of preserving 
the live stock oil the manor than that of a jus primae noctis." 
He points out that the merchet was often higher for marriage 
out of the manor than for marriage within the manor ; that 
fines for marrying sons out of the manor were not unknown ; 
and that " the merchet is often mentioned in close connection 
with a prohibition against giving sons a clerical education- 
an education which would enable them to take orders and 
so escape from bondage." 2 
As another proof of the existence of a jus primae noctis 
has been regarded the fee which a husband had to pay to his 
bishop or other ecclesiastical authority for the privilege 
of sleeping with his wife on the first night or nights of their 
marriage. Thus we are told that Philip VI. and Charles VI. 
in the fourteenth century made vain efforts to induce the 
Bishops of Amiens to give up the old custom of demanding 
from every newly married couple in their town and diocese 
a considerable fine for granting them permission to have 
conjugal intercourse during the first three nights of their 
marriage.3 But it is perfectly obvious that the fee was meant, 
1 Lord Hailes, op. cit. iii. 12 sq. 
2 Maitland, in a letter quoted by Frazer, Folk-Lore in the Old 
Testament, i. 487 sq. 
3 Sugenheim, Geschichte der Aufhebung der Leibeigenschaft und 
Horigkeit in Europa, p. 104. 178 
THE HISTORY OF HUMAN MARRIAGE chap. 
not as compensation for the relinquishment of a right 
accorded to the ecclesiastical authority, but as payment for 
a dispensation. The Catholic Church had prescribed that 
newly married couples should observe chastity on their 
wedding night or, with special reference to the example set 
by Tobias and Sarah, even during the first three nights;1 
and in time the clergy found it expedient to mitigate the 
rigour of the canon and to grant husbands the right of 
lying with their own wives on the first night of marriage, 
provided that they paid a moderate fee for the privilege. 
" This," says Sir James G. Frazer, "was the true jus primae 
noctis, a right accorded, not to a licentious feudal superior, 
but to a woman's lawful husband." 2 
Yet even though the fine or fee paid to the feudal lord or 
the ecclesiastical authority has been grossly misinterpreted, 
there must have been some reason for the misinterpretation. 
It is not easy to see how the idea of a jus primae noctis could 
have entered the heads of the people if there had not been 
some tradition of such a right; travellers' tales from distant 
parts of the world could hardly account for an idea of this 
sort. Dr. Pfannenschmid has pointed out that the belief 
1 See infra, on Marriage Rites. 
2 Frazer, op. cit. i. 501. In the middle of the eighteenth century 
the Dutch historian van Loon suggested that the merchet had a 
similar origin. He writes {Beschryving der aloude Regeeringwyze van 
Holland, iii. 165 sqq.) :-" In the fourth council of Carthage, held in 
the year 398, it was ordained that all new-married persons, out of 
respect for the sacerdotal benediction, eadem node in^virginitate 
permaneant. This species of continence was not only enforced by 
the general constitutions of the Kings of the Franks, but also 
prolonged for three nights, after the example of Tobias. . . . 
Nevertheless, when, about the beginning of the twelfth century, 
the office of judges in the tribunal of a hundred (centenae) had become 
hereditary, instead of elective, as in the days of the Kings of the 
Franks, and when in the following century, the jurisdiction of the 
Counts {Graven} became feudal, the ancient constitutions of these 
Kings, touching the abstinence for three nights, &c., were neglected 
by the new Lords of the country; and, if they were not totally 
abolished, at least the redemption of this inconvenient custom was 
permitted ; just as in Brabant at this day persons newly betrothed 
are permitted to purchase an exemption from their bans thrice 
proclaimed." v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
179 
in the seignorial privilege has been found especially in 
districts and places which have been inhabited for long 
by a Celtic or partly pro-celtic population, and thinks that 
it hints at an ancient right, even though a jus primae noctis 
in the traditional sense of the term has not existed in the 
Middle Ages.1 In support of this suggestion I wish to draw 
attention not only to the legendary statements of old 
Scottish historians but, in particular, to ancient Irish 
documents. In the description of the battle of Gabhra, 
fought a.d. 283, we read that Cairbre, the son of Art, had 
a fair, mild-eyed, and modest daughter, whom the son of 
the king or lord of the Decies came to seek as his wife. But 
when Fionn and the Fenians of Ireland heard of this, they 
dispatched messengers to Cairbre to remind him to pay the 
tribute consisting of twenty ounces of gold or the right of 
cohabiting with the princess the night previous to her 
marriage.2 In the old Irish manuscript ' The Book of 
Leinster,' which was compiled in part about the middle of 
the twelfth century, it is said that great honour was bestowed 
on Conchobar, King of Ulster, who lived at the time when 
Christ was born ; for every man who had a marriageable 
daughter let her sleep the first night with him.3 In another 
manuscript, ' Leabhar na h-Uidhri,' which contains a collec- 
tion of pieces in the Irish language compiled about a.d. 
1100, it is likewise stated that the same king deflowered 
all the virgins of Ulster ; and it is indicated that the deflora- 
tion of maidens even was a duty incumbent upon him.4 
This suggests that we have here to do, not with a mere 
privilege exacted by a ruler, but with a custom rooted in 
some popular idea similar to those found in other parts of 
1 Pfannenschmid, ' Jus primae noctis,' in Das Ausland, Ivi. 150. 
2 O'Kearney, Battle of Gabhra, pp. 135, 137. 
8 Book of Leinster, p. 106, col. 2 ; cf. Atkinson's Introduction, 
p. 27. A translation is given by Schmidt, in Zeitschr. f. Ethnol. 
xvi. 50. See also d'Arbois de Jubainville, ' Le droit du roi dans 
l'6pop6e irlandaise,' in Revue archeologique, xlii. 332 ; Idem, L'epopee 
celtique en Irlande, i. 7, 29 n. 1. 
* Ledbhar na h-Uidhri, p. 127, col. 1. See Schmidt, in Zeitschr. f. 
Ethnol. xvi. 50 ; d'Arbois de Jubainville, in Revue archeologique, 
xlii. 333 sq. 180 
THE HISTORY OF HUMAN MARRIAGE chap. 
the world in connection with the defloration of brides or 
maidens. 
A very recent instance of the jus primae noctis in Europe is 
reported from Albania. Mr. Brailsford wrote in 1903 :- 
"For a certain sum paid annually, an Albanian chief will 
undertake to protect a tributary village, or if the village is 
outside the Albanian sphere of influence, it is generally 
obliged to have its own resident brigands, who may or may 
not be Albanians. If the village belongs to a Turkish 
landlord, these men are generally chosen from among his 
retainers. They are known under the name of bekchi, or 
rural guards. . . . The rural guard exacts a substantial 
ransom in cash for his services. He levies certain traditional 
dues, e.g., blackmail upon every maid who marries. The 
sum varies with the ability of her father and her husband 
to pay, and in default of payment, the bekchi will exercise 
the jus primae noctis." 1 
We shall now consider how the facts stated may be ex- 
plained. Before we are entitled to assume that the jus 
primae noctis accorded to a priest or a headman is a survival 
of ancient communal rights we must investigate whether 
it may be accounted for by feelings or ideas existing among 
peoples who recognise such a right or are addicted to prac- 
tices of a kindred nature. This is the method which should 
be adopted in the study of any custom, and is particularly 
called for in a case like the present one where a custom is 
used as evidence for the existence of a previous state which 
is itself entirely hypothetical. 
The first fact that attracts our attention is a frequent 
reluctance on the part of the bridegroom to deflower the 
bride, or to do so in the manner indicated by nature. Dr. 
Rivet writes of the Jibaros in Ecuador, " Suivant certains 
informateurs indiens, le fiance ouvrirait le vagin de sa 
femme a 1'aide d'os prepares pour cet usage, de fa^on a la 
rendre apte au coit." 2 Among the natives at Alice Springs 
in Central Australia the husband sometimes performs the 
1 Brailsford, ' The Macedonian Revolt,' in Fortnightly Review, 
N.S. Ixxiv. 431 sq. 
2 Rivet, ' Les Indiens Jibaros,' in L'Anthropologic, xviii. 607. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
181 
same operation with a stick after he has taken his new wife 
to his own camp.1 In Samoa, as we have seen, public 
defloration of the bride is a regular feature of the wedding 
ritual, the bridegroom performing it with his fore-finger.2 
A very similar practice exists in Egypt. In his description 
of that country, dating from 1840, Clot-Bey wrote :-" Le 
mari deflore l'epousee avec le doigt indicateur de la main 
droite, enveloppe d'un mouchoir de mousseline blanche. . . . 
Le mouchoir, teint du sang de la jeune victime, est presente 
aux parents, qui la felicitent de sa chastete et temoignent 
hautement leur joie. Cette preuve sanglante de la purete 
de 1'epouse est presentee ensuite aux invites de la noce."3 
Mr. El-Bakry, a native of Egypt, tells me that this is still 
true of the fellahin of his country. 
In other cases girls are artificially deflowered before 
marriage by somebody else than their future husband. 
Among the Sakalava of Madagascar " les j eunes filles 
se deflorent elles-memes quand elles n'ont pas ete 
deflorees des leur bas age par leur mere, et un pere ne 
marie jamais sa fille avant que cette operation ait ete 
menee a bonne fin par 1'une ou par l'autre. Les 
princesses seules restent intactes."4 In some provinces of 
ancient Peru, when a girl was sought in marriage, she was 
brought out in public and, in presence of the relations who 
had made the contract, her mother deflowered her with her 
own hand, to show to all present the proof of the care that 
had been taken of her.5 Among the Kamchadal a man 
blamed his mother-in-law if he found his wife to be a virgin ; 
hence her mother deflowered her in her early youth.6 In 
1 Gillen, ' Notes on some Manners and Customs of the Aborigines 
of the McDonnell Ranges belonging to the Arunta Tribe,' in Report 
on the Work of the Horn Scientific Expedition to Central Australia, 
iv. 165. 2 Kramer, Die Samoa-Inseln, i. 36 sqq. 
3 Clot-Bey, Aperpu general sur I'igypte, ii. 44. 
4 Noel, ' He de Madagascar,' in Bulletin de la Soci^te de Geographic, 
ser. ii. vol. xx. 294. 
5 Garcilasso de la Vega, First Part of the Royal Commentaries of 
the Yncas, i. 59. Cieza de Leon,' La Cr6nica del Peril [parte primera],' 
in Biblioteca de autores espanoles, xxvi. 402. 
6 Steller, Beschreibung von dem Lande Kamtschatka, p. 346. 182 
THE HISTORY OF HUMAN MARRIAGE chap. 
the Philippine Islands old women often performed the same 
operation on young girls.1 Among the Jatt of Baluchistan 
an old woman privately deflowers the bride with a razor 
a few hours previous to consummation, and it is believed 
that consummation is the only permanent cure for the 
wound.2 Among the Wamegi in Central Africa "the 
girls are deflowered by certain old women.''3 Among the 
Wayao in British Central Africa they are' taken away to the 
bush by elderly women to be initiated, and the initiation 
ceremonies comprise, it is said, " a forcible vaginae dilatatio 
by mechanical means, an operation which the girls are 
enjoined to bear bravely. At the same time they are told 
that it must be followed by cohabitation with a man. 
This is regarded by the Yao as a necessity to render the girl 
marriageable before the age of puberty. The girls and their 
mothers believe that if after these initiation ceremonies 
nisi cum mare coitus fiet they will die or at any rate will not 
bear children when eventually married. Pater puellae 
virum robustum (saepe attamen senem) legit atque ei 
pecuniam dat ut puellae virginitatem adimat. Hoc ante 
pubertatem fieri necesse, ne coitum conceptio sequatur."4 
In many Australian tribes the girls are artificially deflowered 
by other men than their future husbands,5 and, as we shall 
see subsequently, the defloration is followed by sexual inter- 
course. 
Among various peoples young women or girls are de- 
flowered by extra-matrimonial intercourse in circum- 
stances which clearly show that the act by no means im- 
plies the exercise of a right on the part of him who performs 
it, but is an operation which the husband is anxious to 
1 Mallat, Les Philippines, i. 61. 
2 Bray, Census of India, 1911, vol. iv. (Baluchistan) Report, 
p. 106. 
3 Roscoe, ' Notes on the Manners and Customs of the Baganda,' 
in Jour. Anthr. Inst. xxxi. 121. 
4 Johnston, British Central Africa, p. 410. 
5 Hill, quoted by Miklucho-Maclay, ' Uber die Mika-Operation 
in Central-Australien,' in Verhandl. Berliner Gesellsch. Anthr. 1880, 
p. 89 (natives of New South Wales). Gason, ' Of the Tribes, Dieyerie, 
&c.,' in Jour. Anthr. Inst. xxiv. 169. Infra, i. 200 sqq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
183 
avoid. In Azimba Land, in Central Africa, when an un- 
married girl is " danced," or initiated, her father " has to 
hire a man to sleep with his daughter the night after she is 
' danced ' and to deflower her ' Ka chatoa massita ' (take 
away fat). To this man he has to pay one fowl, one bowl 
of flour, and a small bowl of beer, and after the girl has slept 
with this man, she is supposed to have no future intercourse 
with him."1 Among the tribes near Fort Johnston in 
British Central Africa " a virgin on her marriage is ' broken ' 
by a friend of the bridegroom before the latter cohabits 
with her. The friend is said ' to eat new things '-Kudia 
tt'jobvu."2 In New Caledonia, " lorsqu'un mari ne peut ou 
ne veut deflorer sa femme, il se trouve, en payant, certains 
individus qui s'en acquittent a sa place. Ce sont des 
Jerceurs attitres."3 In the Philippine Islands also, there 
were at the time of their discovery men who had for their 
employment to take away the virginity of damsels and who 
were paid for doing so, since virginity in a girl was looked 
upon as a hindrance to her marriage ;4 but these professionals 
seem to have disappeared in the course of the seventeenth 
century.5 Among the Todas of the Nilgiris, in Southern 
India, a man of strong physique, who may belong to any 
clan except that of the girl, " comes and stays in the village 
for one night and has intercourse with the girl. This must 
take place before puberty, and it seemed that there were 
few things regarded as more disgraceful than that this cere- 
mony should be delayed till after this period. It might be 
a subject of reproach and abuse for the remainder of the 
woman's life, and it was even said that men might refuse 
to marry her if this ceremony had not been performed at the 
proper time."6 
1 Angus, ' The " Chensamwali " or Initiation Ceremony of Girls, 
as performed in Azimba Land,' in Verhandl. Berliner Gesellsch. Anthr. 
1898, p. 481. 
2 Stannus, ' Notes on some Tribes of British Central Africa,' in 
Jour. Roy. Anthr. Inst. xl. 309. 
3 Moncelon, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. ix. 368. 
4 Morga, Philippine Islands, p. 304 sq. Mallat, op. cit. i. 61. 
5 Gemelli-Careri, Giro del mondo, v. 87. Hertz, Gesammelte 
Abhandlungen, p. 198. 6 Rivers, Todas, p. 503. 184 
THE HISTORY OF HUMAN MARRIAGE chap. 
We have seen that the defloration of girls in Malabar is 
spoken of by various writers as a right belonging to the 
Brahmans; but other authorities, whose accounts date 
from the beginning of the sixteenth century, represent it 
in a very different light. Alvares Cabral states that the 
Nayar women beg the men to deprive them of their virginity, 
since they can find no husbands as long as they remain 
virgins j1 and according to Barbosa, the mother likewise 
asks some young men to deflower her daughter, because the 
Nayars hold it an unclean thing and almost a disgrace to 
deflower women.2 Hieronimo di Santo Stefano wrote of 
them at the close of the fifteenth century that " the men 
never marry any woman who is a virgin ; but if one, being 
a virgin, is betrothed, she is delivered over before the 
nuptials to some other person for fifteen or twenty days in 
order that she may be deflowered."3 It seems probable 
that the tali kettu, or mock marriage ceremony, existing 
among the Nayars and some other castes of the same tract 
is a relic of such pre-nuptial defloration. Every girl among 
them, before she arrives at maturity, is subject to this rite, 
the essential incident of which consists in the nominal 
husband tying a tali, or tiny plate of gold, round her neck. 
Having played his part in the ritual and received the cus- 
tomary fee, the man goes his way; he has no conjugal 
rights over the girl, nay in some places at least the fact 
that he has tied the tali round her neck constitutes an insur- 
mountable barrier to his becoming her husband in later life.4 
1 ' Navigation del captino Pedro Alvares,' in Ramusio, op. cit. 
h J37- 
2 Barbosa, Description of the Coasts of East Africa and Malabar 
in the Beginning of the Sixteenth Century, p. 126. 
3 Account of the Journey of Hieronimo di Santo Stefano, p. 5. 
4 Gopal Panikkar, Malabar and its Folk, p. 143. Mary Billington, 
Woman in India, p. 80. Among the Kammalans, or artisan classes, 
in the Cochin State, however, " if the tali tier of a girl wishes to keep 
her as wife, he has the preference to anybody else, and to none else 
may she be given without his consent " (Anantha Krishna Iyer, 
Cochin Tribes and Castes, i. 346). Mr. Gait says (Census of India, 
1911, vol. i. [India] Report, p. 242) that " in some parts the bride- 
groom is considered to have some sort of claim to the girl and may 
afterwards enter into sambandham (that is, actual marriage) with v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
185 
Various theories have been suggested to explain this 
ceremony.1 Sir Henry Winterbotham has expressed the 
opinion that it is a relic from the time when the Malabar 
Brahmans were entitled to the first fruits and it was con- 
sidered the high privilege of every Nayar maid to be intro- 
duced by them to womanhood.2 To this the objection has 
been made3 that the rite is also performed among castes of 
a lower status with whose women no Brahman would 
cohabit, and that in the earliest accounts of it there is no 
mention of Brahmans being employed as " bridegrooms," 
although at present Nayar girls often have the tali tied round 
their neck by some elderly Brahman.4 This, however, 
does not prevent the tali kettu ceremony from being a survival 
of the defloration of girls which, according to so many 
accounts, was once customary among the Nayars, but, 
as we have seen, was not necessarily performed by Brahmans. 
In his description of the defloration of the king's sister by a 
young man of the Nayar caste Castanheda expressly 
connects it with a tying ceremony : after the act is 
done the young man hangs a jewel round the woman's neck, 
and this she wears all the rest of her life, as a sign that 
she is now at liberty to dispose of herself to anyone she 
pleases as long as she lives, because without this ceremony 
nobody can take a husband.5 According to another early 
Portuguese account, referred to above, the Nayar who 
deflowers a girl of his own caste spends with her four days, 
and as a token of the defloration hangs round her neck 
a so-called quote made of gold.6 Barbosa, whose description 
her." For the tali kettu ceremony in general see Risley, People of 
India, p. 209 ; Gait, op. cit. p. 242 ; Thnrston, Ethnographic Notes 
in Southern India, p. 121 sqq. ; Anantha Krishna Iyer, op. cit. 
ii. 22 sqq. 
1 See Risley, op. cit. p. 209 ; Gait, op. cit. p. 242 ; Anantha 
Krishna Iyer, op. cit. ii. 27 sqq. 
2 Moore, Malabar Law and Custom, p. 70. 
3 Gait, op. cit. p. 242. 
4 Moore, op. cit. p. 70. Gait, op. cit. p. 242. 
5 Lopez de Castanheda, op. cit. i. 45 sq. 
0 * Sommario di tutti li regni, &c. orientali,' in Ramusio, op. cit. 
k (1563) 33t_D. 186 
THE HISTORY OF HUMAN MARRIAGE chap. 
dates from the same age, is less positive. He says that 
when a Nayar girl becomes ten or twelve years old or more 
her mother begs some relation or friend to marry her 
daughter. The bridegroom puts a small gold jewel round 
the girl's neck, " which she always has to wear as a sign that 
she may now do what she pleases," and goes then away 
without touching her, " on account of being her relation." 
But if he is not so, " he may remain with her if he wish it, 
but he is not bound to do so if he do not desire it. And 
from that time forward the mother goes begging some 
young men to deflower that daughter."1 Even according 
to this account there is a connection between the tying of 
the tali and the defloration of the girl, although they may, 
and, if the tali-tier is a relative, must, be performed by 
different men. Buchanan, who visited Malabar at a much 
later date-in 1800-wrote :-" The Nairs marry before 
they arc ten years of age, in order that the girl may not 
be deflowered by the regular operations of nature ; but 
the husband never afterwards cohabits with his wife. Such 
a circumstance, indeed, would be considered as very in- 
decent."2 About half a century later Graul informs us 
that the man who ties the tali round the girl's neck then 
spends four nights with her, and on the fifth morning 
receives a present from her uncle or brother, after which 
he goes away; but if he is a semi-Brahman-which is 
looked upon as a special honour-his connection with the 
girl ceases already at the moment he has tied the tali.3 
It seems that even now the tali-tym^ ceremony is in some 
cases, for instance, with the Paduvals, followed by a mock 
consummation; and the Izhavans, or Tiyyans, of Cochin 
allow or allowed not long ago the tali kettti bridegroom to 
spend several days in the bride's house.4 It has also been 
observed that the ritual resembles in certain respects 
that which is used for the consecration of a deva-dasi, or 
1 Barbosa, op. cit. p. 124 sqq. 
2 Buchanan, ' Journey from Madras,' in Pinkerton, Collection of 
Voyages and Travels, viii. 737. 
8 Graul, Reise nach Ostindien, iii. 337. 
4 Anantha Krishna Iyer, op. cit. i. 287 sq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
187 
temple prostitute.1 If the hypothesis which I am here 
advocating is correct, the object of tying the tali round 
the girl's neck may be explained as a means of protection 
or purification rendered necessary by the act of defloration, 
which, as Barbosa says, was held to be " an unclean thing."2 
To this day it is considered a religious impurity for a girl 
to attain puberty before the performance of this ceremony.3 
In some cases the defloration is performed by a foreigner. 
Verhoeven states that when a lord or nobleman in Calicut 
marries, he hires either a Brahman or a white man to spend 
the first night with the bride and pays him about four or 
five hundred florins for doing so.4 Speaking of the Cochin 
custom according to which a bride is deflowered by one of 
the Brahmans or if there be none at hand by some other 
man, Roggewein adds that " this was formerly a very 
considerable advantage to such foreigners as were settled 
here, the Malabars making choice of them, rather than their 
own countrymen; and on such occasions they made very 
large presents, which sometimes amounted to 500 or 600 
florins : but, of late days, this source is quite dried up; 
for the Bramins are become so very religious, that they 
take care never to be out of the way when this part of 
their duty is to be performed."5 The inhabitants of Malacca, 
according to Olearius, were fond of foreigners and even 
begged them to sleep the first night with their brides to 
deprive the latter of their virginity.6 Barthema observes 
that the king of Tenasserim let his wife be deflowered by 
a white man, either a Christian or a Moor, and not, like the 
king of Calicut, by a Brahman ; and he illustrates this by 
the personal experience of his party.7 Barthema's English 
1 Gait, op. cit. p. 242. Risley, op. cit. p. 209. Thurston, Ethno- 
graphic Notes, p. 121. See also Francis, Census of India, 1901, vol. 
xv. (Madras) Report, p. 151 ; Thyagaraja Aiyar, Census of India, 
1911, vol. xxi. (Mysore) Report, p. 99. 
2 Barbosa, op. cit. p. 126. 3 Anantha Krishna Iyer, op. cit. ii. 22. 
4 Verhoeven, op. cit. p. 56. 5 Roggewein, loc. cit. p. 297. 
« Olearius, in Mandelsloe, Morgenlandische Reise-Beschreibung, 
p. 144. See also Francisci, Neu-polirler Geschicht- Kunst- und Sitten- 
Spiegel auslandischer Volcker, p. 937. 
7 Barthema, Itinerario ncllo Egypto, &c. fol. Ixviii. 188 
THE HISTORY OF HUMAN MARRIAGE chap. 
translator adds that the prevalence of a similar custom in the 
Burmese provinces is confirmed by writers of a later date, 
and that evidence is not wanting of its existence up to a 
very recent period.1 Van Linschoten, who sailed for India 
in 1583, writes that in the kingdom of Pegu, " when any 
gentleman or nobleman will marrie with a maide, hee goeth to 
seeke [one of his friendes, or] a straunger [and entreateth him 
to lie with] his bride the first night of their mariage, and to 
take her maydenhead from her, which he esteemeth as a great 
pleasure and honour [done unto him], that another man 
wold take upon him to ease him of so much payne. This 
custome is not onely used among the Gentlemen and chiefe 
nobilitie of the lande, but by the King himselfe."2 Of the 
inhabitants of Aracan in the eighteenth century Richard 
says :-" Virginity is not an esteemed virtue with them. 
Husbands prefer running the risk of fathering the children 
of others, rather than marry a novice. It is generally 
Dutch sailors who are liberally paid for this infamous 
prostitution."3 We have previously spoken of the Tibetan 
custom mentioned by Marco Polo of old women taking their 
unmarried daughters or other girls related to them to 
strangers who were passing, and making them over to whom- 
soever would accept them.4 
Among certain peoples it is said to be the custom for a 
father to deflower his daughter.5 This has been represented 
as a right belonging to him. Herport wrote in the latter 
part of the seventeenth century that when a Sinhalese gave 
his daughter in marriage, he first slept with her himself on 
the plea of having a right to the first fruit of the tree he had 
planted.6 Miklucho-Maclay was told that among the Orang 
Sakai of the Malay Peninsula the fathers of grown-up 
daughters claimed for themselves the jus frimae noctis; 
1 Jones, in Travels of Ludovico di Varthema, p. Ixxix. 
2 Voyage of J. H. van Linschoten to the East Indies, i. 99. 
3 Richard, ' History of Tonquin,' in Pinkerton, op. cit. ix. 760 sq. 
4 Marco Polo, Kingdoms and Marvels of the East, ii. 44. Supra, 
p. 102. 
■' See Schmidt, Liebe und Ehe im alien und modernen Indien, 
p. 227 (Alfoors of Tonsawang in Minahassa, some Battas). 
6 Herport, Ost-Indianische Reiss-Beschreibung, p. 178 sq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
189 
and he heard of the existence of the same custom in the 
Eastern Moluccas.1 It is difficult, however, to believe 
that the custom in question really represents a right claimed 
by the father. His intercourse with the daughter more 
probably serves the object of making her acceptable to her 
husband.2 
Why is a bridegroom reluctant to have intercourse with a 
virgin bride ? One answer which has been given to this 
question is that he shuns the trouble.3 This explanation 
gains some support from statements made by a few first- 
hand authorities,4 but can at most have only a very limited 
application. The chief reason for his reluctance is no 
doubt superstitious fear.5 Among the Yahgans of Tierra 
del Fuego, if the husband wants to continue to eat the flesh 
of the guanaco or the seal, he must on the morning after the 
consummation of his marriage purify himself by bathing in 
the sea, which in many cases, especially in the winter, makes 
him ill.6 The Nayars, as we have seen, hold defloration 
to be polluting.7 In the most popular mediaeval book of 
travel, ' The Voiage and Travaile of Sir John Maundeville,' 
which was composed soon after the middle of the fourteenth 
century, we read of an island in the Far East where it is the 
custom for the bridegroom not to spend the first night with 
the bride but to make another man do so and reward him 
1 Miklucho-Maclay, ' Ethnological Excursions in the Malay 
Peninsula,' in Jour. Straits Branch Roy. Asiatic Soc. no. 2, p. 216. 
2 Cf. Hertz, op. cit. p. 218 ; Ploss-Bartels, Das Weib, i. 691. 
3 Francisci, op. cit. p. 936. Virey, De la femme sous ses rapports 
physiologique, moral et litteraire, p. 165 sq. 
4 van Linschoten, op. cit. i. 99. Hill, quoted by Miklucho-Maclay, 
in Verhandl. Berliner Gesellsch. Anthr. 1880, p. 89. Roth, Ethno- 
logical Studies among the North-West-Central Queensland Aborigines, 
P-/75- 
5 Cf. Rosenbaum, Geschichte der Lustseuche im Alterlhume, i. 54 ; 
Hertz, 'op. cit. p. 211 sqq. ; Ploss-Bartels, op. cit. i. 691 ; Crawley, 
Mystic Rose, p. 348 sq. 
6 Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 132. 
7 Supra, p. 187. Among the Nattu Malayans, a jungle tribe in 
the Cochin State, " a girl who has reached the age of puberty as a 
virgin is considered impure, and no person will take her for wife " 
(Anantha Krishna Iyer, op. cit. i. 32). 190 
THE HISTORY OF HUMAN MARRIAGE chap. 
with money and thanks. In every town there are such men, 
whose sole occupation is to deflower brides, and in conse- 
quence arc called, in the language of the country, " Cade- 
beriz, that is to seyne, the Foies of Wanhope. For thei of 
the Contree holden it so gret a thing and so perilous, for to 
have the Maydenhode of a Woman, that hun semethe that 
thei that haven first the Maydenhode, puttethe him in aven- 
ture of his Lif." The inhabitants explained this custom as 
an inheritance from ancient times, when " men hadden ben 
dede for defkmrynge of May denes, that hadden Serpentes in 
hire Bodyes, that stongen men upon hire Zerdes, that thei 
dyeden anon."1 I have not been able to find the authority 
for this superstition, but it cannot be supposed to have 
originated in the mediaeval author's own imagination. 
The fear of defloration is no doubt closely connected with 
the fear of hymeneal blood. In the Vedic literature the blood 
of the bridal night is represented as a poison and a seat of 
danger ;2 and a similar belief probably accounts for the old 
German custom of bringing new clothes to the married 
couple on the first morning of their wedded life.3 The woman, 
as well as the man, may be supposed to be in peril. Rochas 
writes of the New Caledonians :-" Le premier rapproche- 
ment intime est plus effrayant pour les caledoniennes que 
pour les femmes de nos societes policies dont 1'esprit eclaire 
est a 1'abri de toute crainte superstitieuse. Il n'a jamais lieu 
sans une purification legale, avec une eau lustrale consacree 
par un sorcier et dite eau virginale." The hymeneal blood 
may also be regarded as dangerous to the offspring. In 
Andjra, in Morocco, there are bridegrooms who take care 
that no offspring can result from the defloration of the bride, 
since many people believe that the child would be diseased 
if the semen came into contact with the blood.6 Among the 
Swahili there is on the first day of the wedding only immissio 
1 Voiage and Travails of Sir John Maundeville, p. 285 sq. 
2 Weber, ' Vedische Hochzeitsspriiche,' in Indische Studien, 
v. 189, 190, 211 sqq. 
3 Weinhold, Die acuischen Frauen in dem Mittelalter, i. 401 sq. 
Hertz, op. cit. p. 213. 4 de Rochas, La Nouvelle Calddonie, p. 235. 
5 See Westermarck, Marriage Ceremonies in Morocco, p. 265 sq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
191 
fenis, not seminis.1 And in Ukrainia, according to 
Tchoubinsky, the bridegroom likewise refrains from con- 
summating the marriage when he deflowers the bride.2 
If defloration is considered a dangerous act it m^y be 
asked how other men can be induced to perform it on behalf 
of the bridegroom. This is easy to explain, even apart from 
the fact that they are often paid for it. As will be shown in a 
subsequent chapter, a bridegroom is commonly held to be in 
a state of danger, extremely susceptible to evil influences; 
hence an act which is supposed to be dangerous to him may 
be supposed to be much less dangerous or quite harmless to 
other men. Sometimes the defloration is performed by 
foreigners, who probably do not share the native dread of it; 
and in other cases it is performed by a holy man, whose 
sanctity allows him to do with safety what is perilous to 
ordinary persons. According to notions expressed in the 
ancient literature of India, the priest alone can purify the 
garment of the bride, just as he is the only one who is not 
polluted by contact with sacrificial blood.3 
The operation in question may be performed by a holy 
man not only because it is supposed to entail no risk for 
him but because it is supposed to be good for the bride or a 
blessing for the married couple. Sexual intercourse with 
such a person is frequently held to be highly beneficial. 
Egede informs us that the native women of Greenland 
thought themselves fortunate if an Angekokk, or " prophet," 
honoured them with his caresses ; and some husbands even 
paid him for having intercourse with their wives, since they 
believed that the child of such a holy man could not but be 
happier and better than others.4 Chenier speaks of a saint 
in Tetuan, in Morocco, who seized a young woman and had 
commerce with her in the midst of the street; " her com- 
panions, who surrounded her, uttered exclamations of joy, 
felicitated her on her good fortune, and the husband himself 
1 Zache,' Sitten und Gebrauche der Suaheli,' in Zeitschr.f. Ethnal. 
xxxi. 75, 84. 
2 Volkov, ' Rites et usages nuptiaux en Ukraine,' in L'Anthro- 
pologic, ii. 578 sq. 3 Weber, loc. cit. p. 190. 
4 Egede, Description of Greenland, p. 140. 192 
THE HISTORY OF HUMAN MARRIAGE chap. 
received complimentary visits on this occasion."1 Among 
certain tribes of the Tachtadshys in Lycia,where the dede is at 
their yearly religious assemblies entitled to have intercourse 
with any woman he chooses, her husband " feels 
considerably honoured by this distinction."2 When the 
prelates of the Armenian Yezidees travel about on their 
visitation tours, they " no sooner arrive in a village, where 
they intend passing a day or two, than they at once get 
married; and the young lady selected for this honour 
becomes, in consequence, so holy that she is looked upon as 
a kind of saint or demon. If she happens to have a son, he 
becomes one of the priestly caste." The prelate may never 
see or even think about his spouse after the wedding day, 
although his colleague or successor is compelled to make a 
very heavy disbursement before attaining his predecessor's 
divorce.3 In the Mahabharata we read of princes and heroes 
who go and present their wives or daughters to some pious 
hermit that he may deign to accord them a son of his good 
works. It may indeed be said that " all India is imbued 
with the belief that sacerdotal blood is gifted with regenerat- 
ing virtues." 
Barthema states that when the King of Calicut travels, 
one of the Brahmans, " although he might be only 
twenty years of age, remains in the house with the 
queen, and the king would consider it to be the greatest 
favour that these Brahmins should be familiar with the 
queen."4 According to Mandelsloe, there is in Malabar 
scarcely a man of note who, when he is to be absent from 
home for any length of time, does not recommend his 
family, and especially his wife, to the care of a Brahman, to 
supply his place.5 In agreement with these views and 
practices the intercourse of a Brahman with a bride during 
1 Chenier, Present State of the Empire of Morocco, i. 187. See also 
Westermarck, Moorish Conception of Holiness (Baraka), p. 85. 
2 Petersen and Luschan, op. cit. p. 199 n. 1. 
3 Creagh, Armenians, Koords, and Turks, i. 154 sqq. 
4 Travels of Ludovico di Varthema, p. 144. 
5 Mandelsloe, in Harris, op. cit. i. 767. See also Guyon, op. cit. 
i. 431. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
193 
the first night is also regarded as beneficial. Sir Thomas 
Herbert states that the Nayars supposed " the ground of 
better value by that holy seed, as they call it."1 Guyon 
informs us that the young spouse in Malabar frequently 
rewards the Brahman to whom he has carried his bride, 
because he is persuaded that a marriage begun by a Brahman 
" cannot fail of being happy."2 According to Navarette, 
the brides who are taken to the temples of idolatrous priests 
to be deflowered are held to be sanctified thereby, " and the 
Husbands carry them home well pleas'd."3 So also the 
Zikris consider the bride who has intercourse with the Mulla 
to be sanctified and cleansed by the association with him.4 
In the case of a bride the first and most obvious benefit 
derived from intercourse with a holy man is purification or 
the removal of danger. It should be added that benefits 
may also be supposed to result from defloration by a stranger, 
who is often looked upon almost as a supernatural being. 
There can be no doubt that the so-called jus primae 
noctis granted to a priest is, largely at least, based upon 
similar ideas. This is indicated by the facts that it is found 
among peoples where such ideas prevail; that, where the 
practice of defloration occurs, it is spoken of sometimes as 
the exercise of a right and sometimes as a favour or an 
onus to be paid for ; and that even when it is called a right, 
he who deprives the bride of her virginity may expect a fee 
for it. Speaking of the jus primae noctis of the paje among 
certain Brazilian tribes, von Martius observes that it is 
probably founded on the belief in woman's uncleanness, 
which prevails among so many rude peoples.5 What is 
or has been merely a habit may be interpreted as, or actually 
become, a right.6 The same maybe said of the jus primae 
noctis of a chief or a king, whose services may be sought for 
on grounds similar to those which have led to defloration by 
priests. With reference to le droit du seigneur of the king of 
the Ballante, M. Marche remarks :-" Ce r'est meme pas 
1 Herbert, op. cit. p. 337. 2 Guyon, op. cit. i. 431. 
3 Navarette, loc. cit. p. 320. 4 Hughes-Buller, op. cit. p. 45. 
5 v. Martius, op. cit. i. 113 sq. 
6 Cf. Hertz, op. cit. p. 214 sq. ; Ploss-Bartels, op. cit. i. 691. 194 
THE HISTORY OF HUMAN MARRIAGE chap. 
pour lui, a proprement parler, un droit, mais une obligation, 
car sans cette formalite une jeune fille ne pourrait se marier. 
Cela oblige le pere qui a des filles peu avenantes a aller faire 
au souverain un beau cadeau, en le suppliant d'avoir pitie 
de ses filles qui attendent de lui le droit de prendre un 
epoux."1 And in ' Leabhar na h-Uidhri ' the druid Cathbad 
is represented as saying that King Conchobar was obliged 
to sleep the first night with Cuchulaind's bride.2 
The defloration of a bride, however, could never have 
come to be looked upon as a right unless the act had been 
attractive. It is not to be believed that the chief or the 
priest slept with another man's bride from unselfish motives 
alone; and there may be cases in which the right to do so 
was nothing but a consequence of might. The jus primae 
noctis of a chief may have the same origin as the right of 
certain chiefs to cohabit with their female subjects at any 
time. Among the Guarani, according to Charlevoix, the 
caciques had a right to the use of the daughters of their 
subjects when they required it.3 In the Marshall Islands 
the chief has only to express a wish, and the subjects place 
anything they possess at his disposal, even their wives and 
children ;4 and on Jaluit, the southernmost island of the 
group, the men of the higher classes have a right to appro- 
priate the wives of the lower people.5 So also in the 
Hawaian Islands the chief had a claim over the virgins of 
his districts,6 and " the wives of the country people were 
sometimes appropriated by the men about court."7 Among 
the Marquesans " les grands chefs du meme rang (ou peut- 
etre d'un rang superieur) que les jeunes maries pouvaient 
posseder 1'epouse."8 In Tonga the women of the lower 
1 Marche, op. cit. p. 70. 
2 Leabhar na h-Uidhri, p. 127, col. 1. Schmidt, in Zeitschr. f. 
Ethnol. xvi. 50 sq. 3 Charlevoix, History of Paraguay, i. 202. 
4 Erdland, ' Die Stellung d^r Frauen in den Hauptlingsfamilien 
der Marshallinseln,' in Anthropos, iv. 106. 
5 Hernsheim, Sudsee-Erinnerungen, p. 81. 
fi Rivers, History of Melanesian Society, i. 386. 
7 Malo, Hawaiian Antiquities, p. 91. 
8 Tautain, ' Stude sur le mariage chez les Polynesiens des iles 
Marquises,' in L'Anthropologic, vi. 645. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
195 
people were at the disposal of the chiefs, who even used to 
shoot the husbands if they made resistance.1 Among the 
Maori, when a chief desired to take to himself a wife, he 
fixed his attention upon one and took her, if need be by 
force, without consulting her feelings and wishes or those of 
anyone else.2 In Madagascar kings and princes had the 
power to dispose of all women in their dominions.3 Con- 
cerning the Barotse Dr. Holub writes, " Quite indisputed is 
the king's power to put to death, or to make a slave of any 
one of his subjects in any way he chooses; he may take a 
man's wife simply by providing him with another wife as 
a substitute."4 Among the Negroes of Fida, according to 
Bosman, the captains of the king, who have to supply 
him with fresh wives, immediately present to him any 
beautiful virgin they may see ; and none of his subjects 
dare presume to offer objections.5 In Dahomey all 
women belong to the king, who can cause any girl 
to be brought to him before marriage and, if he pleases, 
retain her in the palace.6 Speaking of the ancient kingdom 
of Chamba (embracing a large portion of what is now called 
Cochin China), Marco Polo says, "No woman is allowed to 
marry until the King shall have seen her ; if the woman 
pleases him then he takes her to wife ; if she does not, he 
gives her dowry to get her a husband withal."7 There is 
reason to believe that in comparatively modern times some 
of the Rajas of Riwa, a native state bordering on the North- 
Western Provinces of India, in their annual progresses, 
insisted on a supply of girls from the lower tribes.8 Among 
the Kukis " all the women of the village, married or single, 
are at the pleasure of the rajah," who is regarded by his 
1 Waitz-Gerland, Anthropologic der Naturublker, vi. 184. 
2 Yate, Account of New Zealand, p. 96. 
3 Grandidier, Ethnographic de Madagascar, ii. 154. 
4 Holub, Seven Years in South Africa, ii. 160 sq. 
° Bosman, ' Description of the Coast of Guinea,' in Pinkerton, 
op. cit. xvi. 480. 
6 Bastian, Der Mensch in der Geschichte, iii. 302. Burton, Mission 
to Gelele, ii. 67. 7 Marco Polo, op. cit. ii. 268. 
8 Crooke, Tribes and Castes of the North-Western Provinces and 
Oudh, i. p. clxxxv. 
O 2 196 
THE HISTORY OF HUMAN MARRIAGE chap. 
people with almost superstitious veneration.1 We hear of 
similar rights granted to priests. When one of the itinerant 
priests of Siva, known under the name of djaugoumas, who 
are mostly celibates, does an adept the honour to enter his 
house, every male who inhabits it is obliged to go away and 
lodge elsewhere, leaving wives and daughters with the holy 
personage, who prolongs his sojourn as seems good to him.2 
In an old monograph it is said of the Tottiyars, a Telugu 
agricultural caste, that " they have Gurus of their own caste, 
instead of Brahmans, one of whose privileges appears to be 
to lie with the wives of their disciples whenever they feel 
inclined."3 Among the Maori " a man who held the power 
that a priest did might claim almost any girl he desired, 
with a very good chance of getting her." Our informant, 
Mr. Best, adds that he has seen the same sort of thing in 
Mexico A Such privileges may be granted to priests not 
only on account of the miraculous effects attributed to their 
embraces, but for the purpose of gratifying their own desires, 
so as to avert their anger or gain their good-will or as a 
genuine mark of esteem. 
Whether the jus frimae noctis belonging to a chief or 
priest ultimately springs from ordinary persons' fear of 
hymeneal blood or from hope of benefits resulting from inter- 
course with a holy or superior person or from the sexual 
appetite of the man who has the right, it is always the con- 
sequence of his own personal qualities or authority and can- 
not, therefore, be regarded as the relic of an ancient com- 
munal right. But there are other cases in which sexual 
intercourse with a bride or an unmarried girl, involving 
defloration, is accorded to several men, who are not neces- 
sarily either chiefs or priests, and these cases have likewise 
been regarded as expiation for individual marriage dating 
from a time when women were held in common.5 
1 Dalton, Descriptive Ethnology of Bengal, p. 45. 
2 Reclus, Primitive Folk, p. 172. 
3 Nelson, View of the Hindu Law, p. 141. 
4 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Inst, xxxvi. 64. 
'• Avebury, Origin of Civilisation, p. 437 sq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
197 
Herodotus states that " when a Nasamonian first marries, 
it is the custom for the bride on the first night to lie with all 
the guests in turn, and each, when he has intercourse with 
her, gives her some present which he has brought from home."1 
Of the Augilae in the Cyrenaica, another Libyan people, 
Pomponius Mela writes that it is a solemn custom amongst 
them that their women the first night they are married shall 
abandon themselves to the common abuse of all men who 
come with reward, and that it is counted the greatest honour 
to have had to do with many on that occasion, although 
ever after the women keep themselves exceedingly chaste.2 
Solinus refers to the same custom among them, saying that 
the wives are compelled to have adulterous intercourse 
the first night.3 In the Balearic Islands, according to 
Diodorus Siculus, the oldest of the relatives and friends who 
were present at the nuptials cohabited with the bride first 
and then the rest, one after another according to their age, 
until at last the bridegroom had the honour to lie with her.4 
In some provinces of Peru, says Garcilasso de la Vega, " the 
nearest relations of the bride and her most intimate friends 
had connection with her, and on this condition the marriage 
was agreed to, and she was thus received by the husband." 
He adds that Pedro de Cieza says the same ; 5 but the state- 
ment of the latter seems to refer not to the Indians of Pera, 
but to those of New Granada.6 In another chapter of his 
' Commentaries ' Garcilasso writes of the natives of Manta 
and its districts that " their marriages took place under 
the condition that the relations and friends of the bridegroom 
should enjoy the bride before her husband."7 Among the 
Caribs of Cuba, as we have already noticed, the bride of a 
cacique or of a man of somewhat lower rank had during the 
1 Herodotus, iv. 172. 
2 Pomponius Mela, Chorographia, i. 46. 
3 Solinus, op. cit. xxxi. 4, Mommsen's edition, p. 137. 
4 Diodorus Siculus, op. cit. v. 1. 
6 Garcilasso de la Vega, op. cit. i. 59. 
G Cieza de Leon, ' La Cr6nica del Peril [parte primera],' ch. 49 
in Biblioteca de autores espanoles, xxvi. 402. Cf. Markham in 
Garcilasso de la Vega, op. cit. i. 59 n*. 
7 Garcilasso de la Vega, op. cit. ii. 442. 198 
THE HISTORY OF HUMAN MARRIAGE chap. 
first night intercourse, not with the bridegroom, but with his 
equals who were invited to the wedding.1 Among the 
aborigines of Porto Rico " the right of the first night, when 
the bride had connection with other men of the rank of her 
husband, was practised not only by caciques and their depend- 
ent chiefs, but also by the common people."2 Von Langsdorf 
states that in Nukahiva, one of the Marquesas Islands, " if 
the daughter of a person of distinctionmarries, a number of 
swine are' killed, and all the friends and acquaintances are 
invited to the feast. Every guest at the nuptials has then 
a right, with the consent of the bride, to share the pleasures 
of the nuptial night with the bridegroom. The feasting 
commonly lasts two or three days, till all the swine are eaten, 
and from that time the wife must abstain from all inter- 
course with any other man except her husband."3 Dr. 
Tautain mentions the existence of a similar custom among the 
Marquesas Islanders :-" Sur un signal du marie tons les 
hommes presents se reunissaient, formaient une file en 
chantant et en dansant, et, chacun a son tour, defilait 
devant 1'epouse, qui, couchee dans un coin du paepae des 
koika* la tete appuyee sur les genoux de son mari, les 
traitait tons en epoux. Le defile, qui avait commence 
par les plus vieux, les moins nobles, se terminait par les 
grands chefs et, en dernier lieu, par le mari qui ensuite 
emmenait sa femme dans leur case."5 
From these statements by themselves it is impossible 
to find the meaning of the customs they describe, but there 
is every reason to believe that those customs are closely 
connected with others which have already come under our 
notice. It is significant that Cieza de Leon and Garcilasso 
de la Vega speak of the pre-nuptial intercourse between the 
bride and her relatives and friends and the artificial deflora- 
tion of her by her mother as alternative customs occurring 
1 Supra, p. 167. 
2 Fewkes, ' Aborigines of Porto Rico and Neighboring Islands,' 
in Ann. Rep. Bur. Ethn. xxv. 48. 
3 v. Langsdorf, Voyages and Travels, i. 153. 
4 Koika, " fete." Paepae, " plateforme 61ev6e en pierres seches." 
5 Tautain, loc. cit. p. 642. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
199 
among different tribes of the same country. The Cuban 
custom just referred to is mentioned together with the 
defloration by caciques and priests in the case of brides of 
the lowest class, and it is expressly said that it was incumbent 
upon the invited guests to have connection with the bride.1 
Among Libyan peoples we find sexual relations between the 
bride and wedding guests alternating with defloration by the 
king. If intercourse with a priest or chief is supposed to 
remove the danger to which the bridegroom otherwise were 
exposed, a similar result may be expected from intercourse 
with the male guests at the wedding. It will perhaps be 
argued that the Libyan custom according to which they had 
to give presents to the bride ill agrees with the idea that they 
rendered the bridegroom a service. But presents may be 
regarded as bearers of good luck-this is distinctly the case 
with the silver coins given at weddings in Morocco ;2 and 
besides, the old custom may have survived the idea from 
which it sprang, and a mercenary'motive may have been 
attached to it. Those who maintain that the inter- 
course of the wedding guests with the bride is the relic 
of communal marriage would also have to explain how it is 
that the exercise of a right has to be paid for. With refer- 
ence to Lord Avebury's interpretation of the customs in 
question as acts of expiation for individual marriage, 
McLennan remarked that they are not cases of privileges 
accorded to the men of the bridegroom's group only, which 
they should be if they referred to an ancient communal 
right.3 In most cases we are not told to what group the men 
belonged ; but in Manta they are said to have been relatives 
and friends of the bridegroom. Those, however, who are 
inclined to regard this statement as evidence of Lord Ave- 
bury's theory should notice that in other places belonging 
to the same region it was " the nearest relations of the bride 
and her most intimate friends " who had intercourse with 
her, or her own mother who deflowered her. 
1 Carli, op. cit. i. 71. 
2 See Westermarck, Marriage Ceremonies in Morocco, particularly 
p. 347 and the pages referred to in n. 5. 
3 McLennan, Studies in Ancient History, p. 341, 200 
THE HISTORY OF HUMAN MARRIAGE chap. 
Among various peoples the girls are said to be accustomed 
to earn a dowry by prostitution before they marry. This 
is, or has been, the case with the Ulad Naid in Algeria1 and 
among the ancient Phoenicians,2 Cyprians,3 Lydians,4 
and Etruscans,5 as also in the New World, among the 
Natchez of Louisiana6 and in Nicaragua7 and Guatemala.8 
We have reason to believe, however, that this institution 
did not merely serve an economic object, but was in origin 
and purpose akin to the wedding ceremonies just considered. 
It has been especially found among races or in regions where 
other forms of defloration by somebody else than the husband 
have been prevalent. In ' The Testament of Judah ' we are 
told that " it was a law of the Amorites, that she who was 
about to marry should sit in fornication seven days by the 
gate."8 Among the Mfiote, on the coast of Loango, it is 
the custom to take marriageable girls, dressed in long robes, 
from village to village with dancing and singing and offer 
the jus primae noctis for sale.10 
In this connection certain Australian customs must be 
mentioned. Among some Queensland tribes, when a young 
woman shows signs of puberty, two or three men take her 
away in the bush, and, throwing her down, one of them 
forcibly enlarges the vaginal orifice with his fingers ; then 
" other men come forward from all directions, and the 
struggling victim has to submit in rotation to promiscuous 
coition with all the ' bucks ' present. . . . She now has 
1 Soleillet, L'Afrique occidentals, p. 118. 
2 St. Augustine, De civitate Dei, iv. io. 
3 Justin, Historiae Philippicae, xviii. 5. As regards this custom 
see further infra, i. 208. 
4 Herodotus, i. 93. Clearchus, quoted by Athenaeus, Deipno- 
sophistae, xii. 11, p. 516. See also Aelian, Varias historias, iv. 1, 
though the collection of a dowry is not here mentioned as the motive 
for the prostitution of the girls. 
5 Plautus, Cistellaria, ii. 3. 20 sq. 
6 Le Page du Pratz, History of Louisiana, p. 343. 
7 Andagoya, op. cit. p. 33. 8 Herrera, op. cit. iii. 340. 
9 ' Testament of Judah,' in Testaments of the Twelve Patriarchs, 
trans, by Charles, p. 81. 
111 Soyaux, Aus West-Afrika, p. 161. See also Bastian, Die 
dcutsche Expedition an der Loango-Kuste, i. 152, 177. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
201 
attained the degree in which she is allowed to marry. ' '1 Similar 
customs are found in many parts of Australia.2 Spencer 
and Gillen say that in all tribes examined by them from the 
Urabunna in the south right through the centre of the 
continent to the western shores of the Gulf of Carpentaria a 
woman is sooner or later after the performance of the rite 
of cutting open the vagina handed over to certain definite 
men who have access to her before she becomes the property 
of one man ; and among these are not only men belonging to 
the same class as her husband, but individuals to whom she 
at other times is strictly forbidden, in the majority of tribes 
even tribal brothers.3 We are told that these customs 
" certainly point back to a time when there existed wider 
marital relations than obtain at the present day-wider, 
in fact, than those which are shown in the form of group 
marriage from which the present system is derived," and that 
even if they do not afford direct evidence of the former 
existence of actual promiscuity " they do afford evidence 
leading in that direction."4 This, however, is a mere 
assumption suggested by Lord Avebury's expiation theory, 
and as to the natives' own views we are left in the dark. Dr. 
Roth observes that the commonest reason assigned by the 
aborigines for the practice of lacerting the vaginal orifice, 
or female introcision, " is to make him ' big fellow,' not only 
for the convenience of the escaping progeny, as the men will 
allege, but also for the progenitor, as the women will say."5 
We are not told whether the subsequent coitus is considered 
1 Roth, op. cit. p. 174 sq. 
3 Schomburgk, ' Uber einige Sitten und Gebrauche der tief im 
Innern Siidaustraliens, am Peake-Flusse und dessen Umgebung, 
hausenden Stamme,' in Verhandl. Berliner Gesellsch. Anthr. 1879, 
P« 235 Purcell, ' Rites and Customs of Australian Aborigines,' 
ibid. 1893, P- 288. Gillen, in Report on the Work of the Horn 
Scientific Expedition, iv. 165 (Arunta). 
3 Spencer and Gillen, Northern Tribes of Central Australia, 
P- 133 Udem, Native Tribes of Central Australia, p. 92 sqq. 
4 lidem, Native Tribes, p. m. 
6 Roth, op. cit. p. 175. See also Hill, quoted by Miklucho-Maclay, 
in Verhandl. Berliner Gesellsch. Anthr. 1880, p. 89 (natives of New 
South Wales) ; infra, i. 560 sq. 202 
THE HISTORY OF HUMAN MARRIAGE . chap. 
to serve similar purposes. In certain cases mentioned above 
it is for some curious reason held necessary that the deflora- 
tion or vaginae dilatatio by mechanical means should be 
followed by cohabitation with a man. So also the sexual 
indulgence after introcision in Australia might be due to 
peculiar native beliefs, although these have escaped our 
informants' notice. On some other occasions magic signi- 
ficance is avowedly attributed to extra-matrimonial inter- 
course. According to Spencer and Gillen, it is very usual 
" to allow considerable licence during the performance of 
certain ceremonies when a large number of natives, some of 
them coming often from distant parts, are gathered together 
-in fact on such occasions all of the ordinary marital rules 
seem to be more or less set aside for the time being. . . . The 
idea is that the sexual intercourse assists in some way in the 
proper performance of the ceremony, causing everything to 
work smoothly."1 With reference to the Dieri, Gason 
speaks of some " indescribable customs," so obscene and 
disgusting that he must pass them over by only thus briefly 
referring to them :-" that of causing a plentiful supply of 
wild dogs, that of creating a plenty of snakes, that of giving 
strength to young men."2 Indeed, Dr. Roth expressly 
states that the Queensland natives ascribe curative qualities 
to the mixture of blood and semen in the girl who has been 
subject to introcision and subsequent promiscuous coition : 
it is drunk by any sick individual who is in camp on the 
occasion.3 But besides superstition, voluptuousness may 
also have its share in the sexual indulgence following on the 
operation in question, as well as in certain other cases. Dr. 
Eylmann observes that although the old men- have several 
wives each and most of these are young women, their sen- 
suality goes so far that they from time to time arrange 
erotic feasts at which all the ordinary rules of morality are 
suspended and adultery is not only allowed but compulsory.4 
1 Spencer and Gillen, Northern Tribes, p. 136 sq. lidem, Native 
Tribes, p. 96 sq. 
8 Gason, ' Manners and Customs of the Dieyerie Tribe,' in Woods, 
Native Tribes of South Australia, p. 280. 3 Roth, op. cit. p. 175. 
4 Eylmann, Die Eingeborenen der Kolonie Siidaustralien, p. 152 sq. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
203 
Among the Dieri, " on the young women coming to maturity, 
there is a ceremony called Wilpadrina, in which the elder 
men claim, and exercise, a right to the young women."1 
Mr. Mathew says that among the Kabi in Queensland there 
were occasions when some right, analogous to the jus 
primae noctis, seems to have been exacted by the seniors 
in the camp ;2 and Mr. D. Campbell states that the elders of 
the tribe claim the same right in the South Gregory District.3 
In Easter Island, also, the old men are said to have the right 
of deflowering all the girls.4 
There are also, in Australia, cases of a different type in 
which a party of men have intercourse with a woman before 
the husband. " In New South Wales and about Riverina," 
says Mr. J. M. Davis, " when a young man is entitled to have 
a lubra, he organises a party of his friends, and they make 
a journey into the territories of some other tribe, and there 
lie in wait, generally in the evening, by a water-hole where 
the lubras come for water. Such of the lubras as may be 
required are then pounced upon, and, if they attempt to 
make any resistance, are struck down insensible and dragged 
off. There is also this peculiarity, that in any instance 
where the abduction has taken place for the benefit of some 
one individual, each of the members of the party claims, as 
a right, a privilege which the intended husband has no 
power to refuse."5 Dr. Howitt states that among the 
Kuinmurbura the men of the husband's class and totem 
who assisted him in carrying off the woman have access to 
her as a right; and this Howitt looks upon as a kind of 
expiation for individual marriage, " indicating a time when 
there was group-marriage."6 Similar customs have been 
1 Howitt, Native Tribes of South-East Australia, p. 664. Idem, 
' Diery and other kindred Tribes of Central Australia,' in Jour. 
Anthr. Inst. xx. 87. 
2 Mathew, Two Representative Tribes of Queensland, p. 161. 
3 Idem, Eaglehawk and Crow, p. 108. 
4 Martinez, quoted by Knoche, ' Einige Beobachtungen uber 
Geschlechtsleben und Niederkunft auf der Osterinsel,' in Zeitschr. 
f. Ethnol. xliv. 659. 
5 Davis, in Brough Smyth, Aborigines of Victoria, ii. 316. 
0 Howitt, Native Tribes, p. 219. 204 
THE HISTORY OF HUMAN MARRIAGE chap. 
recorded from various other Australian tribes.1 For my 
own part I have, in the earlier editions of the present work, 
explained the access which in such cases is granted to the 
husband's friends as a reward for a good turn done, or, 
perhaps, as McLennan suggests,2 a common war-right, 
exercised by the captors of the woman. Spencer and Gillen 
admit that there is much to be said in favour of this explana- 
tion, but try to throw doubt on it by the remark that, " so 
far as Australia is concerned, it is founded upon such vague 
statements as that quoted by Brough Smyth upon the 
authority of Mr. J. M. Davis."3 By this they evidently 
mean that the members of the party organised by the young 
man, who claim access to the girl, are merely said to be his 
" friends " ; they suspect that those friends are men 
belonging to the intended husband's own class, and seem 
to maintain that this would invalidate my explanation, since 
the claim made by them would be based on an ancient group- 
right. In a following chapter I shall try to show that no 
such group-rights have ever been proved to exist; whereas 
it is only natural that the reward in question could not be 
given to any men who are prohibited by tribal custom from 
having intercourse with the girl. The idea of reward is 
very conspicuous in the following statement referring to the 
Narrinyeri:-" In the cases of elopement the young man 
might call in the aid of his comrades, who then had the right 
of access to the girl, and his male relatives would only defend 
him from the girl's kindred on the condition of access to 
her."4 
A custom very similar to the Australian one is found in 
1 Ibid. pp. 193, 205, 206, 261. Mathew, Eaglehawk and Crow, 
p. 108 (natives at the Clarence River, New South Wales). Idem, 
' Australian Aborigines,' in Jour, and Proceed. Roy. Soc. New South 
Wales, xxiii. 404 (Kabi). 
2 McLennan, Studies in Ancient History, p. 337 note. Cf. Mathew, 
in Jour. Roy. Soc. N. S. Wales, xxiii. 404. Davis says (loc. cit. p. 316) 
that in New South Wales and about Riverina " in cases where one 
tribe has attacked another and carried off a lot of the lubras, those 
unfortunates are common property till they are gradually annexed 
by the best warriors of the tribe." 
3 Spencer and Gillen, Native Tribes, p. 103. 
4 Howitt, Native Tribes, p. 261. v CRITICISM OF HYPOTHESIS OF PROMISCUITY 
205 
East Africa. Here, however, the capture of the girl is 
ceremonial only. Among the Wataveta, after the gi^l has 
been bought by the bridegroom she runs away and hides. 
She is pursued by four of his friends whose services have been 
solicited by him, and is soon captured with much affectation 
of resistance. Then she is conveyed to the hut of the bride- 
groom's mother, where she is kept a close prisoner for five 
days, and is now accessible to the four friends. Only after 
they have indulged in their privilege the legitimate husband 
claims his bride, and they are then visible to all their 
acquaintances.1 A similar custom occurs among the 
Wateita.2 Among the Masai, too, according to Merker, 
one or two of the bridegroom's old companions in arms have 
often a right of access to the bride. In case he opposes their 
claims, he is called by insulting epithets and told that if 
they rob him of some of his cattle he has nothing to com- 
plain of. He may escape the danger, however, by conclud- 
ing his marriage without any ceremony, simply buying the 
bride and taking her quietly home to his hut.3 Mr. Crawley 
1 French-Sheldon, ' Customs among the Natives of East Africa,' 
in Jour. Anthr. Inst. xxi. 366. See also Johnston, Kilima-njaro 
Expedition, p. 431. 
2 Thomson, Through Masai Land, p. 51. 
3 Merker, Die Masai, pp. 49, 232. In Abyssinia the bridegroom 
has a number of arkees, varying in number from six to twelve, who 
were chosen among themselves when boys. When playmates 
together, they agreed that when either of them marries they shall 
reciprocally act as arhees, or bridesmen, to each other. After the 
bridegroom has accepted the bride they are made to promise that 
they will " faithfully and truly fulfil towards her the part of brethren : 
wait on her ; should she hunger, furnish her with food ; or should 
she thirst, with drink. But they have more than this to do. Three 
or four of them generally sleep in the same room with the newly- 
married couple, to furnish them with anything they may require 
during the night." And, our informant adds, " they have a variety 
of other interesting and curious little offices to perform, which it 
appears are considered as not quite fit to print in English " (Parkyns, 
Life in Abyssinia, ii. 52, 56). Among certain Berbers of Morocco 
there are wedding customs of a somewhat similar nature, which I 
have recorded in my book on Marriage Ceremonies in Morocco (see 
particularly p. 273). Thus in one tribe the best-man is present 
when the bridegroom has intercourse with the bride and jokingly 
claims his share of the pleasure. 206 
THE HISTORY OF HUMAN MARRIAGE ch. v 
sees in the Australian and East African customs " the last 
detail in the preparation of the bride for her husband," 
and regards the idea of a reward given on the part of the 
husband to the friends who have assisted him as a secondary- 
development. 1 He may be right. But here again we should 
remember that side by side with the notion of danger there 
is the feeling of sexual pleasure ; and this feeling may surely 
be as primitive a motive as any superstitious belief. 
1 Crawley, The Mystic Rose, p. 349. CHAPTER VI 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : 
RELIGIOUS PROSTITUTION-THE LENDING AND EXCHANGE 
OF WIVES-FEASTS 
Very similar to some of the practices discussed in the 
preceding chapter is a kind of religious prostitution, which 
has likewise been regarded as expiation for individual 
marriage.1 
Herodotus tells us that at Babylon every woman was 
obliged once in her life to go and sit down in the precinct 
of Mylitta-that is, of Ishtar or Astarte-and there consort 
with a stranger. A woman who had once taken her seat 
was not allowed to return home till one of the strangers 
threw a silver coin, which might be of any size, into her 
lap and took her with him beyond the holy ground. The 
silver coin could not be refused : that was forbidden by the 
law, since once thrown it was sacred. The woman went 
with the first man who threw her money, rejecting no one. 
When she had gone with him and so satisfied the goddess, 
she returned home, and from that time forth no gift, however 
great, would prevail with her. The money in question 
was dedicated to the goddess.2 This custom is also men- 
tioned'by Strabo,3 but his account is simply borrowed from 
1 Avebury, Origin of Civilisation, p. 437. Wilutzky, Vorgeschichte 
des Rechts, p. 37 sqq. 
2 Herodotus, i. 199. 3 Strabo, xvi. 1 20. 
207 208 
THE HISTORY OF HUMAN MARRIAGE chap. 
Herodotus. Some independent evidence 1 is found in the 
apocryphal ' Epistle of Jeremy,' which seems to have been 
written about the year 300 b.c.,2 evidently by a person 
who was well informed.3 It is said there that the Babylonian 
women, " with cords about them sit in the ways, burning 
bran for incense : but if any of them, drawn by some that 
passeth by, lie with him, she reproacheth her fellow, that 
she was not thought as worthy as herself, nor her cord 
broken."4 In the cuneiform literature we are told of 
sacred prostitution carried on at Babylonian temples by 
particular women, but no confirmation of Herodotus' 
account has hitherto been found in it. This, however, does 
not justify Zimmern's accusation that he was guilty of 
gross exaggeration.5 Herodotus' statement derives much 
support from the fact that more or less similar customs 
have been reported from other places within the same 
culture area. 
Herodotus says himself that the same rite was performed 
in Cyprus. In a later account, that by Justin, we read 
that " it was a custom among the Cyprians to send their 
daughters, on stated days before their marriage, to the 
sea-shore, to prostitute themselves, and thus procure 
money for their marriage portions, and to pay, at the same 
time, offerings to Venus for the preservation of their chastity 
in time to come."6 This custom closely resembles certain 
others which have been mentioned in another connection,7 
but the reference to Venus suggests that the practice was 
not devoid of religious significance in spite of the mercenary 
motive attributed to it. This also holds good of the descrip- 
tion given by Lactantius, who writes that Venus " first 
1 Cf. Naumann, Untersuchungen uber den apokryphen Jeremias- 
brief, p. 19 sq. 
* Apocrypha and Pseudepigrapha of the Old Testament in English, 
edited by Charles, i. 596. 
3 See Naumann, op. cit. p. 2. 
4 ' Epistle of Jeremy,' v. 43, in Apocrypha and Pseudepigrapha 
of the Old Testament, i. 606. 
6 Zimmern, in Schrader, Die Keilinschriften und das Alte Testa- 
ment, p. 423. 
6 Justin, Historiae Philippicae, xviii. 5. 7 Supra, p. 200. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
209 
instituted the art of courtezanship, as is contained in the 
sacred history, and taught women in Cyprus to seek gain by 
prostitution, which she commanded for this purpose, 
that she alone might not appear unchaste and a courter of 
men beyond other females."1 With reference to the 
inhabitants of Heliopolis or Baalbec, in Syria, the ecclesi- 
astical historian Socrates states that " their virgins were 
presented for prostitution to the strangers who visited 
them " ;8 whilst Sozomenus speaks of their ancient custom 
of yielding up virgins to prostitution with any chance 
comer before being united in marriage to their betrothed. 
He adds that this custom was prohibited by a law enacted 
by Constantine, after he had destroyed the temple of 
Aphrodite at Heliopolis-which implies that the practice 
had a religious aspect.8 So also an earlier authority, 
Eusebius, in a chapter of his * Life of Constantine,' entitled 
* How he destroyed the Temple of Aphrodite at Heliopolis,' 
says that in that city " those who dignify licentious pleasure 
with a distinguishing title of honour, had permitted their 
wives and daughters to commit shameless fornication."4 
At Byblus, where the people shaved their heads in the annual 
mourning for Adonis, women who refused thus to sacrifice 
their hair had to give themselves up to strangers on a certain 
day of the festival, and the money which they earned thereby 
was devoted to the goddess.5 This custom bears a still closer 
resemblance to the practices just mentioned, if Sir James G. 
Frazer is right in his suggestion that it is a mitigation of an 
older rule which compelled every woman without exception 
to sacrifice her virtue in the service of religion.6 In the 
worship of Anaitis the Armenians, even of the highest 
families, prostituted their daughters before they gave them in 
1 Lactantius, Divinae Institutiones, i. 17 (Migne, Patrologiae 
cursus, vi. 207). 
2 Socrates, Historia ecclesiastica, i. 18 (Migne, op. cit. Ser. Graeca, 
Ixvii. 123). 
3 Sozomenus, Historia ecclesiastica, v. 10 (in Migne, op. cit. Ser. 
Graeca, Ixvii. 1243). 
4 Eusebius, Vita Constantini, iii. 58 (Migne, op. cit. Ser. Graeca, 
xx. 1124). 
5 Lucian, De Syria dea 6. G Frazer, Adonis Attis Osiris, i. 38. 210 
the history of human marriage chap. 
marriage, and this was regarded as no bar to an honourable 
match; but we are also told that the girls of rich families 
often gave their lovers more than they received from them.1 
It has been supposed that a similar custom is alluded to in a 
Lydian inscription from the second century, found atTralleis, 
in which a lady by name Aurelia Aemilia proclaims with pride 
that she had prostituted herself in the temple service at the 
command of an oracle, in the same manner as her female 
ancestors had done.2 With reference to these statements 
Dr. Farnell observes that it seems as if in Armenia and Lydia 
there had been a fusion of two institutions which elsewhere 
were distinct one from another, namely, harlot service for a 
prolonged period in a temple, and the consecration of each 
maiden's virginity as a preliminary to marriage.3 
Various theories have been set forth to explain these 
customs. The religious prostitution of the Babylonian type 
has been supposed to be nothing but ordinary immorality 
practised under the cloak of religion.4 It has been repre- 
sented as an act by which a worshipper sacrificed her most 
precious possession to the deity.5 According to M. Cumont, 
it is "a modified form, become utilitarian, of an ancient 
exogamy " : the virgin had to be given to a stranger before 
she was allowed to marry a man of her own race.6 These 
theories seem altogether too unsatisfactory to require any 
special consideration. Mannhardt explained the custom as 
a development of vegetation-ritual,7 and Frazer accepts 
his explanation.8 " We may conclude," the latter says, 
" that a great Mother Goddess, the personification of all 
the reproductive energies of nature, was worshipped under 
different names but with a substantial similarity of myth 
1 Strabo, xi. 14. 16. 
2 Ramsay, Cities and Bishoprics of Phrygia, i. 95. Farnell, Greece 
and Babylon, p. 275. 
3 Farnell, op. cit. p. 275. 
4 Jeremias, Izdubar-Nimrod, p. 60. 
6 Curtiss, Primitive Semitic Religion To-day, p. 155. 
6 Cumont, Les religions orientates dans le paganisms romain, 
p. 287 sq. 
7 Mannhardt, Wald- und Feldkulte, ii. 283 sqq. 
8 Frazer, op. cit. i. 36 n. 6. vi CRITICISM OF HYPOTHESIS OF PROMISCUITY 
211 
and ritual by many peoples of Western Asia ; that associated 
with her was a lover, or rather series of lovers, divine yet 
mortal, with whom she mated year by year, their commerce 
being deemed essential to the propagation of animals and 
plants, each in their several kind; and further, that the 
fabulous union of the divine pair was simulated and, as it 
were, multiplied on earth by the real, though temporary, 
union of the human sexes at the sanctuary of the goddess 
for the sake of thereby ensuring the fruitfulness of the ground 
and the increase of man and beast."1 In the statements 
themselves, however, there is nothing whatever which 
indicates that the practices mentioned were supposed, 
on the principle of homoeopathic magic, to ensure " the 
fertility of the ground and the increase of man and beast." 
Ishtar was no doubt a mother goddess, but her connection 
with vegetation was by no means prominent in the Baby- 
lonian religion.2 And many of our authorities so em- 
phatically represent the prostitution of the women as a 
preliminary to their marriage that this side of the rite cannot 
possibly be ignored. Frazer thinks perhaps that he meets 
this objection by another hypothesis which he combines 
with the former one. He says that if the conception of a 
Mother Goddess " dates, as seems probable, from a time 
when the institution of marriage was either unknown or at 
most barely tolerated as an immoral infringement of old 
communal rights, we can understand both why the goddess 
herself was regularly supposed to be at once unmarried and 
unchaste, and why her worshippers were obliged to imitate 
her more or less completely in these respects. . . . Formerly, 
perhaps, every woman was obliged to submit at least once in 
her life to the exercise of those marital rights which at a 
still earlier period had theoretically belonged in permanence 
to all the males of the tribe."3 Here he falls back upon 
Lord Avebury's theory that religious prostitution is a sur- 
vival from an earlier stage of " communal marriage " and an 
expiation for the infringement of old communal rights-the 
theory with which we are now in the first place concerned. 
1 Frazer, op. cit. i. 39. 2 Schrader, op. cit. p. 430. 
8 Frazer, op. cit. i. 39 sq. 212 
THE HISTORY OF HUMAN MARRIAGE chap. 
Whatever else may be said against this hypothesis, there 
is one formidable objection which, in my opinion, deprives 
it of every chance of success. In many of the statements 
it is expressly said that the women offered themselves up to 
strangers; Lucian is even careful to point out that none 
but strangers were allowed to enjoy them. If their prosti- 
tution really had been an expiation for an infringement of old 
communal rights, we might at all events expect the persons 
with whom they had intercourse to have been representatives 
of the community and not individuals who even under the 
old regime had no marital rights whatsoever. This objec- 
tion has previously been raised by Dr. Farnell1 and by Dr. 
Hartland.2 
Dr. Hartland's own opinion is that the Babylonian rite 
was a puberty rite, and that a maiden was not admitted to 
the status and privileges of adult life until she had thus 
been ceremonially deflowered. Among those privileges, and 
the chief of them, was the gratification of the sexual instinct; 
hence the rite in question was a pre-requisite to marriage.3 
This explanation presupposes that it was confined to un- 
married girls-an assumption which will be examined sub- 
sequently ; but even if it was so, Dr. Hartland's statement 
does not go to the root of the question. As Dr. Farnell 
justly observes,4 it does not explain why the loss of virginity 
should be considered desirable in a puberty ceremony or as 
a preliminary to marriage. Dr. Farnell himself sees in the 
rite the removal of a dangerous tabu by a religious act 
securing the divinity's sanction for the removal; just as 
the ripe cornfield must not be reaped before religious rites, 
such as the consecration of first-fruits, have loosened the 
tabu upon it. " So," he says, " the Babylonian safeguards 
1 Farnell, ' Sociological Hypotheses concerning the Position of 
Women in Ancient Religion,' in Archiv f. Religionswissenschaft, 
vii. 81. Idem, Greece and Babylon, p. 278. 
2 Hartland, ' Concerning the Rite at the Temple of Mylitta,' in 
Anthropological Essays presented to E. B. Tylor, p. 191 sq. Idem, 
Ritual and Belief, p. 269 sq. 
8 Idem, in Anthropological Essays presented to E. B. Tylor, p. 189 
sqq. Idem, Ritual and Belief, p. 278. 
4 Farnell, Greece and Babylon, p. 281. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
213 
the coming marriage by offering the first-fruits of his daughter 
to the goddess who presides over the powers and processes 
of life and birth. Under her protection, after appeal to her, 
the process loses its special danger; or if there is danger 
still, it falls upon the head of the stranger. For I can find no 
other way of accounting for his presence as a necessary 
agent, in the ritual of at least four widely separate com- 
munities of Semitic race."1 Although I cannot agree with 
all the details of this statement, I think it quite probable 
that the bridegroom's fear of deflowering, the bride was one 
motive for the rite. But in no case does it provide us with 
the full explanation of the customs concerned. 
Let us consider their various aspects. In many of the 
statements the women who prostitute themselves are said to 
be virgins or girls who are about to be married. This is 
said of the women of Cyprus by Justin, of those of Helio- 
polis by Socrates and by Sozomenus, and of the Armenian 
women. It has been assumed that Herodotus likewise 
meant virgins in his description of the Babylonian rite, 
although he spoke of women.2 This rite, it is argued, 
was an isolated act and therefore to be distinguished from 
temple prostitution of indefinite duration; the same rite 
was performed in Cyprus, where, as Justin clearly attests, 
the prostitution consisted in the defloration of virgins; 
and the accounts of the practice in other places render it 
fairly certain that only unmarried women were subjected to 
it. These arguments, however, do not seem to me to be 
conclusive. That every Babylonian woman was obliged 
once in her life to prostitute herself in the precinct of 
Mylitta does not necessarily imply that she had to do so 
while a maiden; Justin's description of prostitution in 
Cyprus differs so greatly from that of Herodotus that it 
may be doubted whether they really refer to the same 
1 Ibid. p. 279 sq. An essentially similar view was expressed by 
Farnell in his article in Archiv f. Religionswiss. (vii. 88), and was 
shared by Nilsson (Griechische Feste von religidser Bedeutung, p. 366 
2 Hartland, Ritual and Belief, p. 27Farnell, Greece and Babylon, 
pp. 271/280 n. Nilsson, op. cit. p. 366. 214 
THE HISTORY OF HUMAN MARRIAGE chap. 
practice; and women, not virgins, are mentioned in 
other statements also, referring to Cyprus, Byblus, and 
Babylon itself, while Eusebius expressly speaks of the 
fornication of matrons as well as of unmarried women at 
Heliopolis. This last statement has been treated far too 
lightly by writers who see in the Babylonian and kindred 
practices nothing but a defloration rite j1 Frazer justly 
remarks that Eusebius was bom and spent, his life in Syria, 
was a contemporary of the practices he describes, and 
thus had the best opportunity of informing himself about 
them.2 It may also be asked why Herodotus in the present 
case spoke of women if he meant maidens, although he 
shortly before had been describing the Babylonian custom 
of selling " maidens " by public auction as wives.3 By all 
this I do not wish to deny that the Babylonian rite may have 
been essentially a preliminary to marriage, I only maintain 
that we cannot take for granted that it was so.4 Nor 
must we assume that the prostitution of virgins was re- 
stricted to the act of defloration. It is nowhere said to 
have been thus restricted. Of the Armenian damsels 
we are told that they acted as prostitutes for a long time 
before they were given in marriage.6 The Cyprian maidens, 
according to Justin, procured money for their marriage 
portions by prostituting themselves. The Amorite virgins 
had to " sit in fornication " seven days.6 
The men to whom the women offered themselves are 
said to have been strangers, in Babylon and Cyprus by 
Herodotus, at Heliopolis by Socrates, at Byblus by Lucian. 
Justin's statement that the Cyprians sent their daughters 
to the sea-shore to prostitute themselves may imply the 
same ; Dr. Farnell finds this supposition confirmed by the 
legend given by Apollodorus,7 that the daughter of Kinyras, 
1 Hartland, Ritual and Belief, p. 271. Farnell, Greece and Babylon, 
p. 280 n. Nilsson, op. cit. p. 366 n. 2. 
2 Frazer, op. cit. i. 37 n. 2. 
• Herodotus, i. 196. Cf. Corin, Mating, Marriage, and the Status 
of Woman, p. 63. 
4 Cf. Frazer, op. cit. i. 58. 5 Strabo, xi. 14. 16. 
6 Supra, p. 200. 7 Apollodorus, Bibliotheca, iii. 14. 3. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
215 
owing to the wrath of Aphrodite, had sexual intercourse 
with strangers.1 According to Sozomenus, the virgins of 
Heliopolis were yielded up to prostitution with any chance 
comer, which seems to indicate that the stranger spoken 
of by Socrates was not necessarily a foreigner. In the 
' Epistle of Jeremy ' the Babylonian women are said to 
lie with passers-by; whilst the Amorite virgin who was 
about to marry should sit in fornication by the gate. In 
several statements payment is spoken of. The Cyprian 
girls, according to Justin, prostituted themselves to procure 
money for their marriage portions and to pay, at the same 
time, offerings to Venus. Herodotus states that in Babylon 
(and Cyprus) the money was devoted to the goddess, and 
the same was the case at Byblus. Of the Armenian girls 
we are told that they, if they were rich, often gave their 
lovers more than they received from them. At all the places 
concerned the prostitution of women is said by one or 
another of our authorities to be connected with the cult 
of the goddess of love and procreation, and in some of them, 
at any rate, it took place in the precinct of her temple. 
When the rite consists of, or implies, the defloration of 
a virgin we naturally suspect that it, like other defloration 
practices, may have something to do with the belief that 
it is dangerous for the bridegroom to perform that act 
himself. It may be asked-indeed, in an earlier work 
I have myself asked2-Why should the stranger have been 
more willing than the bridegroom to expose himself to this 
danger ? Dr. Hartland, who has raised the same objection, 
observes that the strangers who visited Babylon or 
Heliopolis could hardly have been on a plane of civilisa- 
tion so far removed from that of the natives that they 
were either ignorant of, or indifferent to, the native ideas ;8 
nay, as we have just noticed, it seems that the stranger 
might have been almost any traveller or chance comer. 
Foreigners have been asked to deflower brides in Calicut 
and Cochin and elsewhere, but they were white men • 
1 Farnell, Greece and Babylon, p. 274. 
2 Westermarck, Origin and Development of the Moral Ideas, ii 446. 
3 Hartland, Ritual and Belief, p. 282 sq. 216 
THE HISTORY OF HUMAN MARRIAGE chap. 
and they were often paid for doing it. So also natives have, 
in various countries, been induced by presents to under- 
take the task. But in the cases we are now discussing the 
man who performs the act, instead of being rewarded, has 
himself to pay. Why should he have to pay for rendering 
a dangerous service ? This argument has been adduced 
by Hertz1 and by Frazer.2 
After reconsidering the whole question I cannot regard 
these objections as conclusive. As I observed in the last 
chapter, where the act is considered to be dangerous for the 
bridegroom to perform, it is not necessarily considered to 
be so for other men, since the bridegroom is commonly 
supposed to be in a delicate condition. Moreover, the 
danger attached to it might be removed by its being 
accomplished as a religious rite, just as it loses its perilous 
character by being performed by a holy individual. As 
for the money paid by the man, it may be said that if the 
act was thought to be harmless there would no doubt have 
been men who were quite willing to pay for it. Sometimes 
he even got back more than he paid. But when the money 
was dedicated to the goddess, it is not certain that it was 
Really meant as a reward. It may have been a prophylactic, 
as money is on certain dangerous occasions in Morocco, or 
it may have served as a bearer of good luck, like silver 
coins in various Moorish ceremonies.3 It is noteworthy 
that it was a silver coin " of any size " that the stranger 
threw into the Babylonian woman's lap. And when he 
did so he said to her, " I invoke the goddess Mylitta to be 
gracious to you."4 
This detail in Herodotus' account is in my opinion of 
the greatest importance for the right interpretation of the 
1 Hertz, Gesammelte Abhandlungen, p. 217. 
2 Frazer, op. cit. i. 59. 3 See supra, p. 199. 
4 This is the hitherto accepted meaning of the Greek phrase, 
(■niKaXtw roi tt)v 0ebv Mv\irra (see, e.g., Liddell and Scott, Greek-English 
Lexicon, p. 535 ; Rawlinson [History of Herodotus, vol. i. 324] 
translates it, " The goddess Mylitta prosper thee "). Dr. Farnell 
maintains (Greece and Babylon, p. 278) that it could as naturally 
mean, " I claim thee in the name of the goddess." This I cannot 
accept. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
217 
rite, because it reveals to us something like the people's 
own views on the subject. The stranger confers upon the 
woman a blessing, invoking the Mother Goddess at her own 
sanctuary. By doing so he evidently meant to ensure- 
not " the fruitfulness of the ground and the increase of man 
and beast "-but the fecundity of the woman herself and 
probably also an easy delivery. For to promote the latter 
was a function of the Mother Goddess; in fact, the name 
Mylitta is said to come from Mu'allidtu, which means 
" midwife."1 Now we can understand the part played by 
the stranger in the rite. If no danger is in the circum- 
stances supposed to be connected with the act, it might as 
well be performed by somebody else; but a stranger is, 
according to early ideas, almost a supernatural being, his 
blessings are particularly efficacious, and great benefits 
may be expected from his love.2 Among the Hebrew 
traditions there is the tale of Lot and the two angels who 
visited him in Sodom, and in the book of Tobit the stranger 
is the angel Raphael.3 The author of the F-pistle to the 
Hebrews writes, " Be not forgetful to entertain strangers 
for thereby some have entertained angels unawares."4 
The embraces of a stranger may therefore be highly appreci- 
ated not only in the case of virgins-from whom he besides 
presumably removes a tabu-but also by matrons. It 
is worth noticing that the stranger is mentioned by 
Herodotus and by Lucian, who speak of women in general; 
and although Eusebius makes no reference to him in 
1 Schrader, op. cit. pp. 423 n. 7, 428. 
2 Westermarck, Origin and Development of the Moral Ideas, i. 
583 sqq. 
3 These facts are pointed out by Dr. Hartland, who in his essay 
on the Mylitta rite in Ritual and Belief (p. 285 sqq.) avails himself 
of my observation that a stranger is regarded as a semi-super- 
natural being. The explanation of the rite which I gave in my Origin 
and Development of the Moral Ideas was subsequently accepted by 
M. van Gennep (Les rites de passage, p. 242 sq.). It is in substantial 
agreement with the present explanation, except that I there re- 
stricted the purpose of the rite to the attainment of the blessing 
of fertility, entirely rejecting Dr. Farnell's suggestion that it was the 
removal of a tabu. 4 Hebrews, xiii. 2. 218 
THE HISTORY OF HUMAN MARRIAGE chap. 
connection with the fornication of matrons and virgins at 
Heliopolis, he figures in other accounts relating to that 
city. 
This explanation of the Babylonian and kindred rites 
derives further probability from various statements quoted 
above which show that intercourse with a holy man is not 
only looked upon as a safeguard, in the case of brides, but 
also as a source of more positive benefits. And yet another 
set of practices should be noticed in this connection. From 
various parts of India we hear of virgins or brides being 
deflowered in temples by means of the phallus of an idol- 
according to Schouten, in order that their marriage shall 
be blessed j1 and similar practices were sometimes resorted 
to by married women as a cure for sterility.2 In Rome, 
according to Christian writers, the bride was placed on the 
phallus of a Priapus or, as Lactantius expresses himself, 
in sinu pudendo of Tutunus, who was identical with Priapus.3 
But Arnobius tells us that matrons also were put on " the 
huge members and horrent fascinus " of Tutunus because 
it was thought auspicious-that is, in order that they should 
become mothers.4 Here then we have instances of sexual 
rites being avowedly performed at sanctuaries partly for 
the purpose of deflowering virgins, and partly to ensure 
fertility in marriage. These are close parallels to the 
Semitic rites, as I conceive them. 
1 Balbi, Viaggio dell' Indie Orientali, fol. 68 ; Voyage of J. H. van 
Linschoten to the East Indies, i. 224 ; Mocquet, Travels and Voyages 
into Africa, Asia, and America, fol. 240 b. (Goa). Schouten, Ost- 
indische Reyse, p. 161 (Canara). Barbosa, Description of the Coasts 
of East Africa and Malabar, p» 96 (Southern Deccan). Fryer {New 
Account of East-India and Persia, p. 179) states that at Semissar, 
in Deccan, a large number of women prostituted themselves to an 
idol; they " reckon it a great Honour, and the Husband thinks 
himself happy in his Cornucopia." 
2 Burnell, in Voyages of J. H. van Linschoten to the East Indies, 
i. 224 n. 3. Cf. Schouten, op. cit. p. 161. 
3 St. Augustine, De civitate Dei, vi. 9. 3 ; vii. 24. 2 (Migne, 
Patrologiae cursus, xli. 188, 215). Lactantius, Divinae institutiones, 
i. 20 {ibid. vi. 227). Rossbach, Untersuchungen uber die romische 
Ehe, p. 369 sqq. See also Hertz, op. cit. p. 272. 
4 Arnobius, Adversus gentes, iv. 7 (Migne, op. cit. v. 1015). VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
219 
There is also another form of religious prostitution besides 
that which we have now considered-not the temporary 
prostitution of a girl before, marriage or of a matron, but 
the more or less permanent prostitution of women attached 
to a temple. Sir James G. Frazer suggests that this kind of 
prostitution likewise may be a survival of early communism. 
" In course of time," he says, " as the institution of individual 
marriage grew in favour, and the old communism fell more 
and more into discredit, the revival of the ancient practice 
even for a single occasion in a woman's Efe became ever 
more repugnant to the moral sense of the people, and 
accordingly they resorted to various expedients for evading 
in practice the obligation which they still acknowledged in 
theory. One of these evasions was to let the woman offer 
her hair instead of her person ; another apparently was to 
substitute an obscene symbol for the obscene act. But 
while the majority of women thus contrived to observe the 
forms of reEgion without sacrificing their virtue, it was 
still thought necessary to the general welfare that a certain 
number of them should discharge the old obEgation in the 
old way. These became prostitutes either for Efe or for a 
term of years at one of the temples : dedicated to the service 
of religion, they were invested with a sacred character, and 
their vocation, far from being deemed infamous, was probably 
long regarded by the laity as an exercise of more than 
common virtue."1 And not only does Frazer consider 
these two kinds of reEgious prostitution as relics of the 
same state of ancient communism, but he also attributes 
to them the same function in the Efe of the people :-" In 
their Ecentious intercourse at the temples the women, 
whether maidens or matrons or professional harlots, imitated 
the Ecentious conduct of a great goddess of fertility for the 
purpose of ensuring the fruitfulness of fields and trees, of 
man and beast."2 For my own part I believe that the two 
kinds of religious prostitution essentially differ from each 
other in origin and purpose. 
Among the Ewhe- and Tshi-speaking peoples on the 
west coast of Africa there are priestesses or sacred women 
1 Frazer, op. cit. i. 40 sq. 2 Ibid. i. 71. 220 
THE HISTORY OF HUMAN MARRIAGE chap. 
who are forbidden to marry, apparently because they are 
considered to be the wives of, or to belong to, the god they 
serve. But this by no means implies that they are debarred 
from sexual intercourse. Major Ellis tells us that among 
the Ewhe-speaking peoples of the Slave Coast the chief 
business of the female kosi, or wife of the god to whom she 
is dedicated, is prostitution. " In every town there is at 
least one institution in which the best-looking girls, between 
ten and twelve years of age, are received. Here they remain 
for three years, learning the chants and dances peculiar to 
the worship of the gods, and prostituting themselves to 
the priests and the inmates of the male seminaries; and 
at the termination of their novitiate they become public 
prostitutes. This condition, however, is not regarded as 
one for reproach; they are considered to be married to 
the god, and their excesses are supposed to be caused and 
directed by him. Properly speaking, their libertinage 
should be confined to the male worshippers at the temple 
of the god, but practically it is indiscriminate. Children 
who are bom from such unions belong to the god."1 So 
also the priestesses among the Tshi-speaking peoples of 
the Gold Coast " are ordinarily most licentious, and custom 
allows them to gratify their passions with any man who 
may chance to take their fancy. A priestess who is 
favourably impressed by a man sends for him to her house, 
and this command he is sure to obey, through fear of the 
consequences of exciting her anger. She then tells him 
that the god she serves has directed her to love him, and, the 
man thereupon fives with her until she grows tired of him, 
or a new object takes her fancy. Some priestesses have as 
many as half a dozen men in their train at one time, and 
may on great occasions be seen walking in state, followed 
by them. Their life is one continual record of debaucheiy 
and sensuality, and when excited by the dance they fre- 
quently abandon themselves to the wildest excesses."2 
In India dancing-girls are, or have been, attached to 
a PTPat manv tpmnlpA Acrnrdinpf in Ward who wratp hr' 
1 Ellis, Ewe-speaking Peoples, p. 140 sqq. 
2 Idem, Tshi-speaking Peoples, p. 121 sq. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
221 
account of the Hindus a hundred years ago, there were, 
for example, at Jugunnat'hu-kshutru in Orissa a number 
of women of infamous character employed to dance and 
sing before the god; the Brahmans who officiated 'there 
continually had connection with them, but they also 
prostituted themselves to visitors.1 With reference to 
Southern India, Dubois wrote that every temple, according 
to its size, entertains a band of so-called d^va-dasi, that is, 
" servants or slaves of the gods." to the number of eight, 
twelve, or more. They perform their religious duties, 
consisting of dancing and singing, twice a day, morning and 
evening. They are also obliged to assist at all the public 
ceremonies, which they enliven with their dance and merry 
song. But as soon as their public business is over, " they 
open their cells of infamy, and frequently convert the 
temple itself into a stew." They are bred to this profligate 
life from their infancy. They are taken from any caste, 
and are frequently of respectable birth. "It is nothing 
uncommon to hear of pregnant women, in the belief that it 
will tend to their happy delivery, making a vow, with the 
consent of their husbands, to devote the child in the womb, 
if it should turn out a girl, to the service of the Pagoda. 
And, in doing so, they imagine they are performing a 
meritorious duty."2 In the Census Report of Mysore for 
1911 Mr. Thyagaraja Aiyar writes :-" The practice of 
dedicating girls to temples or as public women (styled as 
Basavi) obtains in a few of the * lower ' castes, but it is 
gradually getting into disfavour. Among Kurubas when 
there are no sons in a family, the eldest girl is occasionally 
so dedicated. . . ? Among Voddas, if an adult female 
cannot get any one to marry her, she may be dedicated to 
a free life in the name of Yellamma who is their patron 
deity."3 According to Buchanan, even married women 
who wearied of their husbands or widows who grew tired 
1 Ward, View of the History, &c. of the Hindoos, ii. 134. 
8 Dubois, Description of the Character, &c. of the People of India, 
P- 294 sq. 
3 Thyagaraja Aiyar, Census of India, 1911, vol. xxi. (Mysore 
Report, p. 99. 222 
THE HISTORY OF HUMAN MARRIAGE chap. 
of a life of celibacy went to a temple and prostituted them- 
selves.1 Frequently the ceremony of dedication closely 
resembles that of a formal marriage,2 and the woman is 
actually regarded as the wife of the god to whom she is 
devoted. Thus the dancing girls who serve in the pagodas of 
Kartikeya, the Hindu God of war, are betrothed and 
married to him, after which they may prostitute themselves; 
and similarly the Murlis, or dancing girls in Maratha 
temples, are married to Khandoba, the Maratha god of war.3 
There were harlots connected with many Semitic cults. 
In the Gilgamesh-epos, Ishtar is represented as gathering 
round her dissolute girls and harlots, and as a goddess of 
prostitution the epithet " consecrated " is applied to her.4 
Hammurabi makes a distinction between consecrated 
women, or " votaries,"5 who evidently were the daughters 
of good families dedicated by their fathers to religion, 
and so-called qadishtu, or genuine temple harlots ;6 to the 
former class belonged the woman designated the " priestess " 
or " votary " of Marduk.7 In the Canaanite cults there 
were qedeshoth consecrated to the deity with whose temple 
they were associated and at the same time acting as prosti- 
tutes ;8 and at the local shrines of North Israel the worship 
of Yahveh itself seems to have been deeply affected by 
these practices, which were forbidden in the Deuteronomic 
code.9 We hear of women " of the congregation of the 
1 Buchanan, ' Journey to Madras,' in Pinkerton, Collection of 
Voyages and Travels, viii. 749. 
2 Supra, p. 186 sq. 
3 Russell, Tribes and Castes of the Central Provinces of India, iii. 376. 
4 Naumann, op. cit. p. 21. 
5 Die Gesetze Hammurabis, trans, by Winckler, § no, p. 31. 
Code of Laws promulgated by Hammurabi, trans, by Johns, § no, 
p. 20. 
8 Winckler's translation, § 178 sqq., pp. 53, 55. 
7 Ibid. § 182, p. 55. Johns' translation, § 182, p. 40. Cf. Dhorme, 
La religion assyro-babylonienne, p. 300 sq. 
8 Driver, Commentary on Deuteronomy, p. 264. Cheyne, ' Harlot,' 
in Cheyne and Black, Encyclopedia Biblica, ii. 1965. 
,J Hosea, iv. 14. Deuteronomy, xxiii. 17 sq. Cf. Cheyne, loc. cit. 
p. 1965. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
223 
people of Astarte," at Carthage,1 and of numbers of dedi- 
cated slave women in the cult of Aphrodite at Eryx,2 which 
was at least semi-Semitic; and it is likely that some of 
these, at any rate, were dedicated to the impure religious 
practice.3 As for non-Semitic cults, it is clearly attested 
of the worship of Ma at Comana in Pontus4 and of Aphrodite 
in Corinth ;5 but in these cases we have the right to assume 
Semitic influences at work.6 The practice survived in 
Lydia in the later period of the Graeco-Roman culture.7 
Sir William Ramsay speaks of it as service performed to a 
god ; but Dr. Farnell observes that the inscription referring 
to it neither mentions nor implies a god, and that the service 
in Asia Minor, so far as we know, was always to a goddess.8 
The difference between the two types of religious prosti- 
tution which have come under our notice is obvious. In 
the one case a woman, most frequently a maiden, offers 
herself up temporarily to a stranger at the sanctuary of the 
Mother Goddess. In the other case a woman is more or 
less permanently attached to a male god as his wife or 
concubine or to a goddess, presumably as her servant, and 
offers herself up to his or her worshippers. Being regarded 
as sacred and inspired, a woman of the latter class may 
easily persuade a believer that the god has directed her to 
have intercourse with him, and he on his part is naturally 
not slow to accept the invitation. This explanation is 
directly suggested by statements relating to religious 
prostitution in West Africa, and it may have a wider appli- 
cation ; we hear from India of dancing-girls who are con- 
sidered to be possessed by the god and are consulted by the 
people as soothsayers.9 But I venture to suggest that the 
chief explanation lies in the fact that sexual intercourse 
with a holy person is supposed to be beneficial, and may 
1 Corpus inscriptionum semiticarum, pt. i. vol. i. no. 263, p. 340. 
2 Strabo, vi. 2. 6. 3 Farnell, Greece and Babylon, p. 272. 
4 Strabo, xii. 3. 36. 
6 Ibid. viii. 6. 20. Farnell, Cults of the Greek States, ii. 746. 
6 Farnell, Greece and Babylon, p. 272. 
7 Ramsay, op. cit. i. 95. 
8 Farnell, Greece and Babylon, p. 273. 
u Balfour, Cyclopaedia of India, ii. 1012. 224 
THE HISTORY OF HUMAN MARRIAGE chap. 
on that account be a regular feature of the cult. It is note- 
worthy that in some of the Semitic cults there was a prosti- 
tution of men. A clause in Hammurabi's code seems to 
refer to it j1 and it. is known that male prostitutes were 
serving Ishtar at Erech.2 Qedeshim were attached to 
Canaanite temples.3 The word properly denotes men 
dedicated to a deity,4 but has, no doubt for good reason, 
been translated " sodomites " in the English version of the 
Old Testament, It appears that such men were consecrated 
to the mother of the gods, the famous Dea Syria, whose 
priests or devotees they were considered to be ;B and they 
are frequently alluded to by Hebrew writers, especially 
in the period of the monarchy, when rites of foreign origin 
made their way into both Israel and Judah.6 The sodomitic 
acts committed with these sacred men may be explained 
as an outcome of the same belief which I have suggested 
to be the chief cause of the prostitution of the temple women. 
In Morocco supernatural benefits are to this day expected 
not only from heterosexual but also from homosexual 
intercourse with a holy person.7 But I entirely fail to see 
how the function of the male prostitutes could be either a 
rite intended to ensure " the fruitfulness of fields and trees, 
of man and beast," or a survival of " communal marriage." 
That the temple prostitutes are expiating an infringement 
of old communal rights is a suggestion which in some cases 
is absolutely inconsistent with facts, and in all cases a mere 
guess for which no evidence has ever been produced. 
1 Winckler's translation, § 187, p. 57. 
2 Schrader, op. cit. p. 422 sq. 
3 Deuteronomy, xxiii. 17. Driver, op. cit. p. 264. 
4 Driver, op. cit. p. 264 sq. Selbie, ' Sodomite,' in Hastings, 
Dictionary of the Bible, iv. 559. Cf. Frazer, Adonis Attis Osiris, 
i. 73 n. 1. 
6 St. Jerome, In Osee, i. 4. 14 (Migne, op. cit. xxv. 851). Cook's 
note to 1 Kings, xiv. 24, in his edition of Holy Bible, ii. 571. See 
also Lucian, Lucius, 38; Eusebius, Vita Constantini, iii. 55 
(Migne, op. cit. Ser. Graeca, xx. 1120). 
6 1 Kings, xiv. 24.: xv. 12 ; xxii. 46. 2 Kings, xxiii. 7. Job, 
xxxvi. 14. Driver, op. cit. p. 265. 
7 See Westermarck, Moorish Conception of Holiness (Baraka), 
p. 85. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
225 
Another custom which has been adduced as evidence 
of former communism in women is that which requires a man 
to offer his wife or one of his wives to a guest.1 To Lord 
Avebury it seems to involve the recognition of " a right 
inherent in every member of the community, and to visitors 
as temporary members."2 Were this so, we should certainly 
have to conclude that " communal marriage " has been 
widespread in the human race, the custom of lending 
wives occurring among many peoples in different parts 
of the world.3 But we might as well look upon the offer 
1 Avebury, Origin of Civilisation, p. 107. Post, Die Geschlechts- 
genossenschaft der Urzeit, p. 34 sq. Le Bon, L'homme et les societes, 
ii. 292. Lippert, Kulturgeschichte der Menschheit, ii. 17. Kohler, 
' Ueber das Recht der Australneger,' in Zeitschr. vergl. Rechtswiss. 
vii. 327. Wilutzky, op. cit. i. 44 sqq. 
8 Avebury, op. cit. p, 107. 
3 The practice of lending a wife to a guest has been found, for 
example, among the Guarani of Paraguay (Hernandez, Organization 
social de las doctrinas Guaranies, i. 84), Indians of Brazil (v. Martins, 
Beitrage zur Ethnographic Amerika's, i. 118), Apache (Bancroft, 
Native Races of the Pacific States, i. 514), Comanche (Schoolcraft, 
Archives of Aboriginal Knowledge, v. 684), Californian Indians 
(Powers, Tribes of California, p. 153), Omaha (James, Account of an 
Expedition from Pittsburgh to the Rocky Mountains, i. 233), coast 
tribes of British Columbia (Sproat, Scenes and Studies of Savage 
Life, p. 95), Cree (Mackenzie, Voyages to the Frozen and Pacific Oceans, 
p. xcvi.), Eskimo (Richardson, Arctic Searching Expedition, i. 356 ; 
Bancroft, op. cit. i. 65), Aleut (Veniaminof, quoted by Bogoras, 
Chukchee, p. 607, n. 1), Tungus (Patkanov, quoted by Miss Czaplicka, 
Aboriginal Siberia, p. 107), Votyak (Buch, Die Wotjaken, p. 48), 
Tibetans (Marco Polo, Kingdoms and Marvels of the East, ii. 54 ; 
Desgodins, Le Thibet, p. 244; Grenard, Tibet, p. 260), Pathan 
clans of Baluchistan (formerly; Gait, Census of India, 1911, 
vol. i. [India] Report, p. 248), Sinhalese (Pridham, Historical 
Account of Ceylon, i. 250), Orang Sakai of Malacca and Dyaks of 
Sidin in Western Borneo (Wilken, ' Plechtigheden en gebruiken 
bij verlovingen en huwelijken bij de volken van den Indischen 
Archipel,' in Bijdragen tot de taal-, land- en volkenkunde van Neder- 
landsch-Indie, ser. v. vol. iv. 451), Australian aborigines (Curr, The 
Australian Race, i. no ; Angas, Savage Life and Scenes in Australia, 
i. 93 ; Wilkes, Narrative of the United States Exploring Expedition, 
ii- 195 J Spencer and Gillen, Native Tribes of Central Australia, 
p. 63 ; I idem, Northern Tribes of Central A ustralia, p. 140 ; Howitt, 
Native Tribes of South-East Australia, pp. 208, 224 ; Malinowski, 226 
THE HISTORY OF HUMAN MARRIAGE chap. 
of a meal and a bed to a guest as a relic from a time when 
no man had any food or shelter which he could call his 
own. For I think there can be no doubt that the custom 
which requires a host to lend his wife to his guest is only 
an incident of the general rule of hospitality, which in 
some form or other seems to prevail universally at the 
lower stages of civilisation.1 It is not always the wife that 
is offered: it may be a daughter, a sister, or a servant.2 
This sort of hospitality is frequently mentioned in Irish 
The Family among the Australian Aborigines, p. 103), Melanesians 
of the New Hebrides (Codrington, Melanesians, p. 24), Caroline 
Islanders (Kotzebue, Voyage of Discovery into the South Sea, iii. 212), 
Marquesas Islanders (Lamont, Wild Life among the Pacific Islanders, 
p. 42), Hawaians (Jarves, History of the Hawaiian Islands, p. 42 ; 
Meinicke, Die Inseln des Stillen Oceans, ii. 305), Malagasy (Grandi- 
dier. Ethnographic de Madagascar, ii. 202), Masai (Merker, Die Masai, 
p. 50 n. 1 ; Johnston, Uganda Protectorate, p. 825 ; Kaiser, reviewed 
in L'Anthropologic, xvii. 710; Hobley, Ethnology of A-Kamba, 
p. 64), Akamba (ibid. p. 64), Nandi (Hollis, Nandi, p. 77), 
Wasania (Barrett, ' Notes on the Customs and Beliefs of the Wa- 
Giriama, etc., British East Africa,' in Jour. Roy. Anthr. Inst. xli. 31), 
Baganda (Felkin, ' Notes on the Waganda Tribe,' in Proceed. Roy. 
Soc. Edinburgh, xiii. 711 sq.), Banyankole (Roscoe, ' Bahima,' in 
Jour. Roy. Anthr. Inst, xxxvii. 105 ; Idem, Northern Bantu, p. 121 
sq.), Basuto (Rolland, quoted by Theal, History of the Boers, p. 19), 
Kafirs (v. Weber, Vier Jahre in Afrika, ii. 218), Baya (Poupon, 
' fitude ethnographique des Baya,' in L' Anthropologic, xxvi. 126)( 
Yoruba (Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 182), 
Canarians (Cook, ' Aborigines of the Canary Islands,' in American 
Anthropologist, N.S. ii. 479 sq.), ancient Arabs (Wellhausen, ' Die 
Ehe bei den Arabern,' in Nachrichten von der Kbhigl. Gesellsch. f. 
Wissensch. Gottingen, 1893, no. n, p. 462). In many of these cases, 
however, the lending of the wife is only spoken of as a more or less 
frequent practice, not as a duty incumbent upon the host. 
1 See Westermarck, Origin and Development of the Moral Ideas, 
i. 572 sqq. 
2 Waitz, Anthropologie der Naturvolker, iii. in (North American 
Indians). Coxe, Russian Discoveries between Asia and America, 
p. 245. Marco Polo, op. cit. ii. 54 (people of Caindu, Eastern Tibet). 
Sauer, Expedition to the Northern Parts of Russia, p. 49 ; Patkanov, 
quoted by Miss Czaplicka, op. cit. p. 107 (Tungus). Regnard, 
' Journey to Lapland,' in Pinkerton, op. cit. i. 166 sq. Rochon, 
' Voyage to Madagascar, ' ibid. xvi. 747. Moore, Marriage Customs, &c. 
of the Various Nations of the Universe, p. 267. Post, Die Geschlechts- 
genossenschaft der Urzeit, p. 34 sq. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
227 
heroic tales,1 and M. M£ray has called attention to a number 
of instances occurring in French mediaeval literature which 
point to the former existence of the custom in France.2 
Among the Maori " it was a point of hospitality that when 
a strange chief of high rank paid a visit his entertainer 
should send a temporary wife or wives to his guests ; gener- 
ally his own daughter as a special honour";3 whereas 
married women were never offered to visitors.4 When we 
read that among the coast tribes of British Columbia " the 
temporary present of a wife is one of the greatest honours 
that can be shown there to a guest " ;6 or that such an 
offer was considered by the Eskimo " as an act of generous 
hospitality " ; 6 or that in the Hawaian Islands visitors 
who were their hosts' equals or superiors in rank were 
accommodated with women as a necessary exercise of hospi- 
tality7-I cannot see why we should look for a deeper 
meaning in these practices than that which the words 
imply. Among certain tribes the wife is said to be offered 
to a guest who belongs to the husband's clan8 or who is 
his friend.;9 but frequently it is represented as a particular 
mark of distinction,10 or as a favour bestowed even upon 
European visitors.11 And to refuse it may be considered 
an insult12 or make the guest " despised by the men and 
scorned by the women."13 
1 Potter, Sohrab and Rustem, p. 148. 
2 M6ray, La vie au temps des Trouv^res, p. 77 sqq. 
3 Tregear, The Maori Race, p. 298. 
4 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 63. 5 Sproat, op. cit. p. 95. 
0 Richardson, op. cit. p. 356. 7 Jarves, op. cit. p. 42. 
8 Hobley, op. cit. p. 64 (various East African tribes). 
9 Ibid. p. 64 (Akamba). Roscoe, in Jour. Roy. Anthr. Inst, xxxvii. 
105 (Banyankole). Poupon, loc. cit. p. 126 (Baya). 
10 Hernandez, op. cit. i. 84 (Guarani). Lamont, cp. cit. p. 42 
(Marquesas Islanders). Grandidier, op. cit. ii. 202 (Malagasy). 
d'Arbois de Jubainville, L'epopee celtique en Irlande, i. 7 sq. (refer- 
ring to Conchobar, king of Ulster). 
11 James, op. cit. i. 233 (Omaha). Spencer and Gillen, Native 
Tribes of Central Australia, p. 101 sq. Felkin, loc. cit. p. 712 (Baganda). 
12 Potter, op. cit. p. 147. Cook, in American Anthropologist, N.S. 
ii. 480 (Canarians). 13 Felkin, loc. cit. p. 712 (Baganda). 
Q 2 228 
THE HISTORY OF HUMAN MARRIAGE chap. 
To offer one's wife to a friend may not merely be an 
expression of good-will. Generally speaking, when the 
visitor belongs to a community with which there is reciprocity 
of intercourse, it is good policy to give him a hearty recep- 
tion ; for he who is the host to-day may be the guest to- 
morrow. " If the Red Indians are hospitable," says 
Domenech, " they also look for their hospitality being 
returned with the same marks of respect and consideration."1 
So, too, it may be prudent for the host to show unusual 
regard for a powerful or influential man, even though he 
cannot expect to be paid back exactly in his own coin. 
But, as I have shown in another work, the custom of 
hospitality is also associated with superstitious beliefs.2 
The unknown stranger, like everything unknown and 
everything strange, arouses a feeling of mysterious awe in 
simple minds. The Ainu say, " Do not treat strangers 
slightingly, for you never know whom you are entertaining."3 
According to Homeric, notions, " the gods, in the likeness 
of strangers from far countries, put on all manner of shapes, 
and wander through the cities, beholding the violence and 
the righteousness of men." 4 It is significant that in the 
writings of ancient Greece, Rome, and India guests are 
mentioned next after gods as due objects of regard.5 When 
properly treated, the stranger may bring with him great 
blessings; for if efficacy is ascribed to the blessings even 
of an ordinary man, those of a stranger are naturally 
supposed to be still more - powerful. And there is yet 
another reason for pleasing him. He is regarded not only 
as a potential benefactor, but as a potential source of evil. 
He is commonly -believed to be versed in magic ; and the 
evil wishes and curses of a stranger are greatly feared, owing 
partly to his quasi-supernatural character, partly to his 
1 Domenech, Seven Years' Residence in the Great Deserts of North 
America, ii. 319. See also Westermarck, Origin and Development 
of the Moral Ideas, i. 581 n. 2. 
2 Origin and Development of the Moral Ideas, i. 582 sqq. 
3 Batchelor, Ainu and their Folk-Lore, p. 259. 
4 Odyssea, xvii. 485 sqq. 
5 Gellius, Nodes Atticae, v. 13. 5. Anugitd, 3, 31 (Sacred Books 
of the East, viii. 243, 361). See also Hitopadesa, MitralabM, 65. vi CRITICISM OF HYPOTHESIS OF PROMISCUITY 
229 
close contact with the host and his belongings, which makes 
it easy for him to transfer evil to them. In the work re- 
ferred to I said that the custom which requires a host to lend 
his wife to a guest " becomes more intelligible when we 
consider the supposed danger of the stranger's evil eye or his 
curses, as also the benefits which maybe supposed to result 
from his love " ; but I added that I could adduce no direct 
evidence for my supposition.1 The following statements, 
which had escaped my notice, seem to confirm it; the 
first one shows that superstition may induce a man to lend 
his wife to another man, even though he be no stranger. 
Concerning the North-West-Central Queensland aborigines 
Dr. Roth states :-" If an aboriginal requires a woman 
temporarily for venery he either borrows a wife from her 
husband for a night or two in exchange for boomerangs, 
a shield, food, &c., or else violates the female when unpro- 
tected, when away from camp out in the bush. In the 
former case, the husband looks upon the matter as a point 
of honour to oblige his friend, the greatest compliment that 
can be paid him, provided that permission is previously 
asked. On the other hand, were he to refuse, he has the 
fear hanging over him that the petitioner might get a death- 
bone pointed at him-and so, after all, his apparent courtesy 
may be only Hobson's choice."2 The belief in the death- 
bone and its property of causing sickness and death is a 
universal superstition among those natives and one which 
fills them with the utmost fear.3 With reference to the 
people of Caindu, in Eastern Tibet, Marco Polo wrote:- 
" No man considers himself wronged if a foreigner, or any 
other man, dishonour his wife, or daughter, or sister, or any 
woman of his family, but on the contrary he deems such 
intercourse a piece of good fortune. And they say that it 
1 Origin and Development of the Moral Ideas, i. 593. 
2 Roth, Ethnological Studies among the North-West-Central Queens- 
land Aborigines, p. 182. It is said that among the Tottiyars in 
Southern India newly-married women are compelled to cohabit 
with their husband's near relatives, and that ill-luck is believed to 
attend any refusal to do so (Thurston, Castes and Tribes of Southern 
India, vii. 186). 
3 Roth, op. cit. p. 152. 230 
THE HISTORY OF HUMAN MARRIAGE chap. 
brings the favour of their gods and idols, and great increase 
of temporal prosperity. For this reason they bestow their 
wives on foreigners and other people."1 Mr. Potter has 
been told that North American Indians, in lending their 
wives, have often been actuated by a desire to procure 
nobler offspring.2 
Besides the custom of lending wives to visitors, the 
temporary exchange of wives, which is also a frequent, 
though not equally frequent, practice,3 has been repre- 
sented as a survival of ancient promiscuity.4 Sometimes 
the two practices are combined. Among the Banyankole 
or Bahima in Central Africa, " when a man and his wife 
visit a friend, they invariably exchange wives during the 
time of the visit " ;5 and a similar custom prevailed in the 
Hawaian Islands.6 
1 Marco Polo, op. cit. ii. 53 sq. 2 Potter, op. cit. p. 150. 
3 It has been found among the Jibaros of Ecuador (Orton, Andes 
and the Amazon, p. 172 ; Simson [Travels in the Wilds of Ecuador, 
p. 89], however, says that he has not heard of this custom among 
them), Eskimo (Nansen, Eskimo Life, p. 169 ; Holm, ' Konebaads- 
Expeditionen til Grdnlands 0stkyst 1883-85,' in Geografisk Tid- 
skrift, viii. 92 ; Lyon, Private fournal during the Voyage of Discovery 
under Captain Parry, p. 354), Northern D6n6 (Hearne, Journey from 
Prince of Wales's Fort to the Northern Ocean, p. 129 ; Morice, ' The 
Great Dene Race,' in Anthropos, ii. 33), Aleut (v. Langsdorf, Voyages 
and Travels, ii. 47 ; Jochelson, Koryak, p. 756), Kamchadal (Steller, 
Beschreibung von Kamtschatka, p. 347), Himalayans (Stulpnagel, 
' Polyandry in the Himalayas,' in Indian Antiquary, vii. 134), 
Nayars (Travels of Ludovico di Varthema, p. 145 sq.), pagan Arabs 
(Alberuni's India, i. 109), Baya (Poupon, in L' Anthropologic, xxvi. 
126), Ayao (Stigand, ' Notes on the Natives of Nyassaland, 
N.E. Rhodesia, and Portuguese Zambezia,' in Jour. Roy. Anthr. 
Inst, xxxvii. 122), Wagiriama (Barrett, * Notes on the Customs and 
Beliefs of the Wa-Giriama, etc., British East Africa,' ibid. xli. 22), 
Australian aborigines (Howitt, op. cit. pp. 170, 195, 216, 217, 224, 
260, 276 sq. ; Eylmann, op. cit. p. 153), and various South Sea 
Islanders (Christian, Caroline Islands, p. ' 74; Guppy, Solomon 
Islands, p. 43 ; Parkinson, Dreissig Jahre in der Siidsee, p. 267 
[New Ireland and New Hanover]; Waitz-Gerland, Anthropologic 
der Naturvolker, vi. 130, 131, 622). See also below. 
4 Wilutzky, op. cit. i. 20 sq. 
6 Roscoe, in Jour. Roy. Anthr.'Inst, xxxvii. 105. Idem, Northern 
Bantu, p. 122. 6 Remy, Ka Mooolelo Hawaii, p. xli. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
231 
An exchange of wives may take place simply for the sake 
of enjoyment. Merolla da Sorrento says that the Negroes 
of Angola, who used to exchange their wives with each 
other for a certain time, excused themselves, when re- 
proached, on the ground that " they were not able to eat 
always of the same dish."1 Among the Baya of West 
Africa, " un camarade, qui regoit un des ses bons amis, lui 
prete sa femme pour la nuit; s'ils ont chacun du gout pour 
la femme de l'autre, ils en font 1'echange."2 On the east 
coast of Greenland the natives, while living in their winter 
houses, often play " a wife-exchanging or lamp-extinguish- 
ing game," in which the unmarried also take part; and 
Holm tells us that " a good host always has the lamps put 
out at night when there are guests in the house."3 Among 
the Eskimo of Hudson's Bay, " an exchange of wives is 
frequenr, either party being often happy to be released 
for a time, and returning without concern."4 
Among the Eskimo men also exchange their wives for 
purely practical reasons.6 Of those living round Repulse 
Bay we are told that " if a man who is going on a journey 
has a wife encumbered with a child that would make travel- 
ling unpleasant, he exchanges wives with some friend who 
remains in camp and has no such inconvenience. Some- 
times a man will want a younger wife to travel with, and in 
that case effects an exchange, and sometimes such exchanges 
are made for no especial reason, and among friends it is a 
usual thing to exchange wives for a week or two about 
every two months."6 Murdoch speaks of a Point Barrow 
Eskimo of his acquaintance who planned to go to the river 
deer-hunting in the summer of 1882, and borrowed his 
1 Merolla da Sorrento, ' Voyage to Congo,' in Pinkerton, op. cit. 
xvi. 299. 
2 Poupon, in L' Anthropologic, xxvi. 126. 
3 Holm, in Geografisk Tidskrift, viii. 92. Nansen, op. cit. p. 169. 
Cf. Dalager, Gr^nlandske Relationer, p. 67 sq. 
4 Turner, ' Ethnology of the Ungava District,' in Ann. Rep. Bur. 
Ethn. xi. 189. 
6 See Parry, Journal of a Second Voyage, p. 528 ; Hawkes, The 
Labrador Eskimo, p. 116. 
6 Gilder, Schwatka's Search, p. 251. 232 
THE HISTORY OF HUMAN MARRIAGE chap. 
cousin's wife for the expedition, as she was a good shot 
and a good hand at deer-hunting, while his own wife went 
with his cousin on a trading expedition to the eastward. 
On their return the wives went back to their respective 
husbands. But sometimes it happens in similar cases that 
the couples find themselves better placed with their new 
mates than with the former association and that, in conse- 
quence, the exchange is made permanent.1 Among the 
Eskimo at Fury and Hekla Straits some of the young men 
informed Parry and his party that " when two of them 
were absent together on a sealing excursion, they often 
exchanged wives for the time, as a matter of friendly con- 
venience."2 
In some cases anger or disgust is said to be the motive 
for exchanging wives. Among the Himalayan mountaineers, 
according to Dr. Stulpnagel, "it is nothing extraordinary 
to hear that two men disgusted with their wives have 
agreed to interchange them, hoping that a new arrangement 
in their domestic affairs would conduce to greater peace 
and comfort."3 Among the Bangala of the Upper Congo 
River, " sometimes, in anger, two men would exchange their 
wives, especially if one man's wife was continually running 
after the other man."4 Among the Ayao, south of Lake 
Nyasa, if a man's wife admits that she has committed 
adultery, the co-respondent sometimes, instead of giving a 
present, lends his wife for the same number of nights as he 
slept with the other man's wife.5 Among the Darling 
tribes of Ne.w South Wales it is an occasional custom 
" that two tribal brothers having quarrelled, and wishing for 
a reconciliation, the one sends his wife to the other's camp, 
and a temporary change is effected."6 
1 Murdoch, ' Ethnological Results of the Point Barrow Expedition,' 
in Ann. Rep. Bur. Ethn. ix. 413. 
8 Parry, op. cit. p. 529. 
3 Stulpnagel, in Indian Antiquary, vii. 134. 
4 Weeks, * Anthropological Notes on the Bangala of the Upper 
Congo River,' in Jour. Roy. Anthr. Inst, xxxix. 442. 
5 Stigand, in Jour. Roy. Anthr. Inst, xxxvii. 122. 
6 Cameron, * Notes on some Tribes of New South Wales,' in Jour. 
Anthr. Inst. xiv. 353. VI CRITICISM OF HYPOTHESIS OF PROMISCUITY 
233 
Sometimes the exchange of wives is described as a sign 
of friendship. Thus among the Northern Dene, according 
to Father Morice, " the momentary exchange of wives was 
regarded, not as a breach of propriety, but on the contrary 
as an unsurpassed token of friendship " ;x and among the 
Eskimo of Davis Strait and Cumberland Sound, according 
to Dr. Boas, " a strange custom permits a man to lend his 
wife to a friend for a whole season or even longer and to 
exchange wives as a sign of friendship."2 If our informants 
had entered more deeply into the native mind, it is possible 
that they would have found the practice to be something 
more than a mere sign of intimacy ; some magic significance 
may have been ascribed to it. The South Sea Islanders 
maintain that an exchange of names, which is a common 
practice among them, establishes a kind of relationship, 
not only between the parties themselves, but between their 
people generally.3 And, as we shall see. subsequently, 
some savages have the idea that if a man commits adultery 
with another man's wife, the fate of the husband is thereby 
in a mysterious manner associated with that of the adulterer,, 
because they have had intercourse with the-same woman.4 
In certain cases the practice of exchanging wives is 
undoubtedly connected with some superstition. Writing 
of shamanistic performances among the Eskimo of the 
western coast of Hudson's Bay, Dr. Boas states:-"It seems 
that the incantations of the angakut [pl. of angakok] are 
always performed in the evening. After each of these 
ceremonies the people must exchange wives. The women 
must spend the night in the huts of the men to whom they 
are assigned. If any woman should refuse to go to the man 
to whom she is assigned she would be sure to be taken sick. 
The man and the woman assigned to him, however, must 
not be near relatives."5 Dr. Howitt tells us that among the 
Wiimbaio, who occupy the country at the junction of the 
1 Morice, in Anthropos, ii. 33. 
2 Boas, ' Central Eskimo,' in Ann. Rep. Bur. Ethn. vi. 579. 
3 Lamont, op. oil. p. 33. Cf. Melville, Typee, p. 103. 
4 Infra, i. 315. 
6 Boas, quoted by Hartland, Primitive Paternity, ii 144. 234 
THE HISTORY OF HUMAN MARRIAGE chap. 
Darling and Murray Rivers, wives were exchanged, not 
only at times when there were great tribal gatherings, but 
also in order to avert some great trouble which they fancied 
was about to come upon them, for example, a great sick- 
ness j1 and among the Kumai of Gippsland the old men 
ordered an exchange of wives when the Aurora Australis 
was seen, since " this was thought by them to be a sign of 
Mungan's anger."2 In these cases the deeper meaning of 
the practice is obscure. Perhaps the Australian rite was 
a form of homoeopathic magic, the change of wives being 
supposed to effect the change of a dangerous situation. 
The practice of exchanging wives may thus be traced 
to many different causes. Most frequently it is a perfectly 
voluntary arrangement between the husbands or between 
all the parties concerned,3 and it never has the character of 
a claim which one man has to another man's wife. There 
is consequently no reason whatever to regard it as the 
survival of an ancient communal right. 
Among various peoples promiscuous intercourse is in- 
dulged in at certain feasts ; and this, too, has been regarded 
as a survival of ancient promiscuity. We have previously 
noticed several instances of such feasts,4 and others might be 
added. Sometimes they are held at definite seasons, and 
sometimes they are connected with particular events in 
the social life of the people. Speaking of the Koko-nor 
Tibetans, Mr. Rockhill says that, " in the lamaseries in 
Amdo, there is held at different times a feast known to the 
Chinese as t'iao mao hui, ' the hat-choosing festival.' During 
the two or three days the feast lasts a man may carry 
off the cap of any girl or woman he meets in the temple 
grounds who pleases him, and she is obliged to come at 
1 Howitt, op. cit. p. 195. See also Cameron, in Jour. Anthr. Inst. 
xiv. 353. 
2 Howitt, op. cit. pp. 170, 276 sq. 
3 Among the Wagiriama of British East Africa, for instance, 
two men cannot exchange their wives without the agreement of the 
latter (Barrett, in Jour. Roy. Anthr. Inst. xli. 22). 
4 Supra, ch. ii. vi CRITICISM OF HYPOTHESIS OF PROMISCUITY 
235 
night and redeem the pledge."1 In the Philippines, at the 
time of their discovery, marriages were celebrated with 
extremely obscene dances, " apres quoi 1'on se couchait 
pele-mele."2 In Madagascar orgies of great licentiousness 
formerly accompanied the birth of a child in the royal 
family ; on such an occasion the streets and lanes of the 
capital appeared like one vast brothel, and the period during 
which the debauchery lasted was called andro-tsi-maty, 
that is, a time when the law could not condemn nor the 
penalty of death be inflicted.3 
There is reason to believe that at some of the feasts, at 
least, the promiscuous intercourse has the character of a 
magical rite.4 And in no case can I regard debauchery or 
a loosening of the marriage tie on some specific occasion as 
evidence of a time when there was no marriage tie at all. 
1 Rockhill, Land of the Lamas, p. 80 sq. 
2 Mallat, Les Philippines, i. 59. 
3 Ellis, History of Madagascar, i. 150 sq. Sibree, The Great African 
Island, p. 253. Sir James G. Frazer {Totemism and Exogamy, ii. 638) 
thinks that a trace of " an older custom of sexual promiscuity, or 
of something like it," may perhaps be detected in these orgies. 
4 See supra, p. 92. . CHAPTER VII 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY THE 
CLASSIFICATORY SYSTEM OF RELATIONSHIP 
Nearly fifty years ago the American anthropologist, 
Lewis Morgan, published the terms of relationship in use 
among no fewer than 139 different peoples or tribes. He 
divided the systems of nomenclature into two great classes, 
the " descriptive " and the " classificatory," which he 
regarded as radically distinct. " The fust," he says, 
" which is that of the Aryan, Semitic, and Uralian families, 
rejecting the classification of kindred, except so far as it 
is in accordance with the numerical system, describes 
collateral consanguinei, for the most part, by an augmenta- 
tion or combination of the primary terms of relationship. 
These terms, which are those for husband and wife, father and 
mother, brother and sister, and son and daughter, to which 
must be added, in such languages as possess them, grand- 
father and grandmother, and grandson and granddaughter, 
are thus restricted to the primary sense in which they are 
here employed. All other terms are secondary. Each 
relationship is thus made independent and distinct from every 
other. But the second, which is that of the Turanian, 
American Indian, and Malayan families, rejecting de- 
scriptive phrases in every instance, and reducing con- 
sanguinei to great classes, by a series of apparently arbitrary 
generalisations, applies the same terms to all the members 
236 ch. vii CRITICISM OF HYPOTHESIS OF PROMISCUITY 
237 
of the same class. It thus confounds relationships, which, 
under the descriptive system, are distinct, and enlarges the 
signification both of the primary and secondary terms 
beyond their seemingly appropriate sense."1 
Since the publication of Morgan's work many fresh in- 
stances of the classificatory system have been discovered 
in different parts of the world. It is known to exist among 
the North American tribes, with the exception of the Eskimo. 
It is universal in Polynesia, Melanesia, New Guinea, and 
Australia. It is found in India and Northern Asia, and among 
the Bantu peoples of Africa. And vestiges of it have been 
discovered elsewhere, even in some parts of Europe.2 On 
the other hand, objections have been made to the generally 
accepted distinction between descriptive and classificatory 
systems of terms of relationship. Every language, Pro- 
fessor Kroeber remarks, groups together under single 
designations many distinct degrees and kinds of relation- 
ship. Thus our word brother includes both the older and 
the younger brother and the brother of a man and of a 
woman ; whilst the English word cousin denotes thirty-two 
different relationships, and if the term is not strictly limited 
to the significance of first cousin the number of distinct 
ideas which it is capable of expressing is many times thirty- 
two.3 Dr. Rivers finds in particular fault with the term 
" descriptive system " : those designations which apply to 
one person only may be called descriptive if you please, 
but when we pass beyond these our terms are no whit 
more descriptive than those of the classificatory system. 
According to him, there are really three main varieties of 
system of relationship in place of the two which have 
hitherto been recognised, one having its origin in the clan, 
another one in the family in the narrow sense, and a third 
1 Morgan, Systems of Consanguinity and Affinity of the Human 
Family, p. 12. 
2 See Rivers, ' On the Origin of the Classificatory System of 
Relationships,' in Anthropological Essays presented to E. B. Tylor, 
p. 310. Idem, ' Kin, Kinship,' in Hastings, Encyclopedia of Religion 
and Ethics, vii. 703 sq. 
3 Kroeber, ' Classificatory Systems of Relationship,' in Jour. Roy. 
Anthr. Inst, xxxix. p. 77. 238 
THE HISTORY OF HUMAN MARRIAGE chap. 
in the extended or patriarchal family.1 The first is the 
classificatory system, which might consequently also be 
called the " clan system."2 But whatever may be said 
against Morgan's terminology, there can hardly be any 
danger in making use of it in the present discussion. 
The simplest, and according to Morgan the earliest, 
system of the classificatory group is that which he called 
the " Malayan " system. As Mr. Wallace wrote to me, the 
term " Malayan " is badly chosen, since the system in ques- 
tion is not found among true Malays. It occurs among 
the Hawaians and the Maori of New Zealand, and probably 
also in other parts of Polynesia.3 I shall call it the Hawaian 
system. All consanguine!, near and remote, are in this 
system classified into five categories. My brothers and 
sisters and my first, second, third, and more remote male 
and female cousins, are the first category. To all these 
without distinction I apply the same term. My father and 
mother, together with their brothers and sisters, and their 
first, second, and more remote cousins, are the second 
category. To all these without distinction I apply likewise 
the same term. The brothers, sisters, and several cousins 
of my grandparents I denominate as if they were my grand- 
parents ; the cousins of my sons and daughters, as if they 
were my sons and daughters ; the grandchildren of my 
brothers and sisters and their several cousins, as if they were 
my own grandchildren. All the individuals of the same 
category address each other as if they were brothers and 
sisters.4 Altogether there are in the Hawaian Islands only 
fifteen terms of relationship, if relatives by marriage are 
excluded. The other systems of the classificatory group 
are more discriminating. In nearly all of them the father's 
brother is classed with the father and the mother's sister 
with the mother, but very generally a rigorous distinction 
1 Rivers, Kinship and Social Organisation, p. 76 sq. 
2 Ibid, p. 71. 
3 Morgan, op. cit. p. 450 sq. Best, ' Maori Marriage Customs,' 
in Trans, and Proceed, of the New Zealand Institute, xxxvi. 28. 
Rivers, in Hastings, op. cit. vii. 702. 
4 Morgan, Ancient Society, p. 403 sq. Idem, Systems, p. 482 sq. vn CRITICISM OF HYPOTHESIS OF PROMISCUITY 
239 
is made between relatives through the father and relatives 
through the mother-between the father's brother and the 
mother's brother, between the father's sister and the mother's 
sister, and between the children of brothers or of sisters 
and the children of brother ■ and. sister.1 Various other 
details, of less frequent occurrence, will be noticed below. 
From the classificatory terms of relationship far-reaching 
conclusions have been drawn with reference to earlier 
marriage customs. Morgan assumed that the Hawaian 
system of nomenclature was the root from which all the 
others belonging to the classificatory group had gradually 
developed ; and from the Hawaian system itself he inferred 
the former prevalence of " marriage in a group " of all 
brothers and sisters and cousins of the same grade or 
generation-or, more strictly, his case was, that if we 
can explain the system in question on the assumption that 
such a general custom once existed, then we must believe 
that it did formerly exist. " Without this custom," he 
says, "it is impossible to explain the origin of the system 
from the nature of descents. There is, therefore, a necessity 
for the prevalence of this custom amongst the remote 
ancestors of all the nations which now possess the classifi- 
catory system, if the system itself is to be regarded as having 
a natural origin."2 The family resulting from this custom 
he calls, in his later work, the "consanguine family," and 
in this, consisting of a body of kinsfolk, within which there 
prevailed promiscuity, or " communal marriage," between 
all men and women of the same generation, the family in 
its first stage is recognised. Morgan believes, however, 
that as a necessary condition antecedent to this form of 
the family, promiscuity, in a wider sense of the term, may 
be theoretically deduced, though, as he says, " it lies con- 
cealed in the misty antiquity of mankind beyond the reach 
of positive knowledge." 3 
It is needless to examine the last conclusion unless the 
1 See Rivers, in Hastings, op. cit. vii. 702 ; Idem, Kinship and 
Social Organisation, p. 66. 
2 Morgan Systems, p. 488. 
3 Idem, Ancient Society, p. 502. Cf. Idem, Systems, p. 487 sq. 240 
THE HISTORY OF HUMAN MARRIAGE chap. 
hypothesis of the " consanguine family," from which it is 
drawn, is found acceptable; and so far as I can see, this 
hypothesis is not only unfounded but contrary to all reason- 
able assumptions. Among other things it presupposes 
unrestricted sexual intercourse between brothers and sisters, 
which is found nowhere among existing savages and is 
utterly inconsistent with the strict exogamy which prevails 
among most peoples who have a classificatory system of 
relationship terms. Morgan seems to assume that the 
fewer terms such a system contains the more ancient it 
must be, that a system which makes no distinction between 
relatives through father and mother must be earlier than 
one which makes such a distinction. We might just as 
well conclude that the present English nomenclature, 
which classes the brothers and sisters of the father together 
with the brothers and sisters of the mother, must be more 
ancient than the Anglo-Saxon and Latin systems, which 
distinguished between paternal and maternal uncles and 
aunts. Dr. Rivers points out instances of classificatory 
systems undergoing changes in the direction of the Hawaian 
form of nomenclature, and believes that this form, far from 
being primitive, " rather represents a late stage in the history 
of the more ordinary forms of the classificatory system"- 
which, indeed, we might expect from the relatively high 
development of Polynesian society.1 
At the same time Dr. Rivers thinks it possible that the 
Hawaian system " may have had its source in promiscuity, 
even though this condition be late rather than primitive."2 
Still more conciliatory is his attitude towards Morgan's 
hypothesis as regards the form of the family following upon 
the " consanguine family," the so-called " Punaluan family," 
which was founded upon intermarriage of several sisters and 
female cousins with each other's husbands (or several 
brothers and male cousins with each other's wives) in a 
group, the joint husbands (or wives) not being necessarily 
1 Rivers, in Anthropological Essays presented to E. B. Tylor, p. 311 
sqq. Idem, Kinship and Social Organisation, p. 85. Cf. Schmidt, 
Der Ursprung der Gottesidee, p. 184 n. 2. 
8 Rivers, Kinship and Social Organisation, p. 85. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
241 
akin to each other, although often so.1 In an article pub- 
lished in 1907 Dr. Rivers supports the view that the features 
of the classificatory system of relationship as we find them 
at the present time have arisen out of a state of group- 
marriage-a form of marriage in which definite groups of 
men are the husbands of definite groups of women-although 
he points out that the classificatory system lends no support 
to the view that the state of group-marriage was preceded 
by one of wholly unregulated promiscuity.2 In a later 
essay he is somewhat more cautious. He throws aside the 
term group-marriage as only confusing the issue, and speaks 
rather of a state of organised sexual communism, which he 
(in his latest work) defines as " a social condition in which 
it is recognised as legitimate that sexual relations shall 
take place between a group of men and a group of women."3 
But he is still of opinion that " the classificatory system has 
several features which would follow naturally from such a 
condition of sexual communism," and that " the wide dis- 
tribution of the classificatory system would suggest that 
this communism has been very general," although it need 
not have been universal.4 That the classificatory system has 
originated in group-marriage is a view which is held by many 
other writers.5 According to Sir James G. Frazer, the classi- 
ficatory terms express group-relationships, and the only 
reasonable and probable explanation of such a system of 
group-relationships is that it originated in a system of group- 
marriage.6 Professor Kohler expresses the same opinion in 
1 Morgan, Ancient Society, p. 384. 
2 Rivers, in Anthropological Essays presented to E. B. Tylor, p. 323. 
3 Idem, History of Melanesian Society, ii. 127. 
4 Idem, Kinship and Social Organisation, p. 86. In his article on 
' Marriage,' in Hastings' Encyclopedia of Religion and Ethics (viii. 
432), Dr. Rivers says that " the nature of the classificatory system of 
relationship is most naturally explained by its origin in communistic 
conditions." 
5 See, e.g., Fison and Howitt, Kamilaroi and Kumai, p. 50 ; 
Howitt, * Native Tribes in South-East Australia,' in Folk-Lore, 
xvii. 189 ; Spencer and Gillen, Native Tribes of Central Australia, 
P- 59- 
0 Frazer, Totemism and Exogamy, i. 303, 304, 501 ; ii. 69 sqq. 
See also Idem, Folk-Lore in the Old Testament, ii. 311 sqq. 242 
THE HISTORY OF HUMAN MARRIAGE chap. 
words which show that he looks upon it as a scientific truth 
so firmly established that it allows of no doubt whatever.1 
Morgan's hypothesis of the " consanguine family," and 
to some extent also the theory that the classificatory 
nomenclature has its origin in group-marriage, are based 
on the assumption that this nomenclature was originally 
meant to express the degree and kind of blood-relationship 
as definitely as the fatherhood of individuals could be 
known. If a person applied the same term to his father and 
to certain other men as well, he did so because he could not 
know who of them was his father. If he applied the same 
term to his children and to certain other persons also, he 
did so because he could not know who of them were his 
children. And the like holds true of other terms of relation- 
ship. We shall now consider the assumption which is at the 
bottom of these inferences. It may, first, be asked whether 
the meaning of the terms lends any support to it. 
The answer to this question must necessarily be very in- 
complete, since it is only in rare cases that a term is said to 
have any special meaning apart from its being a designation 
for a certain person or a certain class of persons. But we 
know, at all events, one great source from which terms of re- 
lationship, whether " descriptive " or " classificatory," have 
been derived. Professor Buschmann has given us a list of 
the names for father and mother in many different tongues, 
and the similarity of the terms is striking. Thus pa, papa, 
or baba, means father in several languages of the Old and 
the New World, and ma, mama, means mother.2 The Tupis 
have paia for father and maia for mother; the Uaragua^u, 
respectively, paptko and mamko3 In other languages 
1 Kohler, ' Rechtsphilosophie und Universalrechtsgeschichte,' in 
v. Holtzendorff, Enzyklopadie dev Rechtswissenschaft, i. 27. See 
also Idem, ' Zur Urgeschichte der Ehe,' in Zeitschr. vergl. Rechtswiss. 
xii. 321, &c. 
2 Buschmann, ' Ueber den Naturlaut,' in Philologische und 
historische Abhandlungen der Konigl. Akademie der Wissenschaften 
zu Berlin, 1852, p. 391 sqq. Independently of him Lord Avebury 
has compiled a similar table in The Origin of Civilisation, p. 346 sqq. 
8 v. Martius, Beitrdge zur Ethnographie und Sprachenkunde 
Amerika's, ii. 9, 10, 18. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
243 
the terms for father are ab, aba, a^a, ada, ata, tata ; those 
for mother ama, ema, ana, ena, &c. According to Busch- 
mann, there are four typical forms of words for each of these 
ideas : for father, pa, ta, ap, at; for mother, ma, na, am, an. 
Sometimes, however, the meaning of the types is reversed. 
Thus in the language of the West Australian Kariera,1 
the Mahaga language of Ysabel,2 the Koita language of 
British New Guinea,3 and Georgian,4 mama stands for father ; 
whilst the Tuluvas of Southern India call the father amme 
and the mother appe.5 
In many cases the terms used fall outside the types 
mentioned. In the Lifu tongue, for example, one term for 
father is kaka* in the Duauru language of Baladea chichaj 
in the Marean tongue chacha or checked Among the Chalcha 
Mongols and some related peoples mother is eke.9 In the 
Kanuri language of Central Africa the mother is called ya ;10 
while the Kechua in Brazil call the father yay a.11 Among 
the Bakongo, as I am informed by Mr. Ingham, se means 
father ; in Finnish, isd. Again, in the Brazilian Bakairf 
the mother is called ise,12 and in the language of Aneiteum, 
in the New Hebrides, risi.13 
Similar terms are often used for other relationships. The 
Greek TrairiTo^ signifies grandfather and p-dppa grand- 
mother. In the Umbaia tribe of Central Australia an elder 
brother is called pappa, and in the Tjingilli tribe the same 
word is used for children.14 In Lifuan mama means brother,15 
1 Brown, ' Three Tribes of Western Australia,' in Jour. Roy. 
Anthr. Inst, xliii. 149. 
2 von der Gabelentz, Die melanesischen Sprachen, ii. 139. 
3 Seligman, Melanesians of British New Guinea, p. 66. 
1 Hunter, Comparative Dictionary of the Languages of India and 
High Asia, p. 122. 
5 Ibid. pp. 122, 143. 
G von der Gabelentz, op. cit. p. ii. 52. " Ibid. i. 215. 
s Ibid. i. 172. 9 Klaproth, Asia Polyglotta, p. 281. 
10 Barth, Central-afrikanische Vokabularien, p. 212. 
11 v. Martius, op. cit. ii. 293. 
12 von der Steinen, Durch Central-Brasilien, p. 341. 
13 von der Gabelentz, op. cit. i. 71. 
14 Spencer and Gillen, Northern Tribes of Central Australia, pp. 81, 
84. 15 von der Gabelentz, op. cit. ii. 52. 244 
THE HISTORY OF HUMAN MARRIAGE chap. 
and in the language of the Yanadis of the Madras Presidency 
it means maternal uncle.1 
The origin of such terms is obvious. They are formed 
from the easiest sounds a child can pronounce. " Pa-fa, 
ma-ma, tata, and apa, ama, ata," Professor Preyer observes, 
" emerge originally SDontaneously, the way of the breath 
being barred at the expiration, either by the lips (p, w), 
or by the tongue (d, t)."2 Yet the different races vary con- 
siderably with regard to the ease with which they produce 
certain sounds. Thus the pronunciation of the labials is 
very difficult to many Indians ;a and for this reason, it seems, 
their terms for father, mother, or other near kinsfolk, often 
differ much from the types given by Buschmann. 
The terms of the type which we have now considered may 
be " classificatory " as well as " descriptive." To take a few 
instances. A West African Yoruba applies the term baba 
both to his father and to " uncles on both sides of the house."4 
A Maciisi of British Guiana names his paternal uncle papa 
as well as his father,5 and a Maori calls by the same term his 
father, the brothers of his parents, and the sons of his 
parents' uncles.6 In one Australian tongue, the Umbaia, 
pappa stands both for elder brother and father's elder 
brother's son, and in another, the Tjingilli, it stands for 
children, brother's children, and son's son's children.7 In the 
Kariera tribe of Western Australia mama is used for father, 
father's brother, mother's sister's husband, and various 
other relatives.8 Among the Koita of British New Guinea 
mama is the term for father and paternal uncle, nena for 
mother and maternal aunt, and nana for elder brother, 
1 
1 Ranga Rao, ' Yanadis of the Nellore District,' in the Madras 
Government Museum's Bulletin, iv. 99. 
2 Preyer, Die Seele des Kindes, p. 321. 
3 Avebury, Origin of Civilisation, p. 349. 
4 Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 178. 
6 Schomburgk, Reisen in Britisch-Guiana, ii. 318. Appun, ' Die 
Indianer von Britisch-Guayana,' in Das Ausland, xliv. 447. 
6 Best, in Trans, and Proceed, of the New Zealand Institute, 
xxxvi. 28. 
7 Spencer and Gillen, op. cit. pp. 81, 84. 
8 Brown, in Jour. Roy. Anthr. Inst, xliii. 43. 
xxxvi. 28. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
245 
sister, or cousin.1 An Efatese names his father and all the 
" tribe brothers " of his father ava or tama.2 The Dakota 
apply the term ahta not only to the father, but to the father's 
brother, to the mother's sister's husband, to the father's 
father's brother's son, &c., and the term enah not only to 
the mother, but to the mother's sister, to the mother's 
mother's sister's daughter, &c.3 
The terms which have been derived from the babble of 
infants have, of course, been selected, and the use of them 
has been fixed, by grown-up persons.4 They may therefore, 
no doubt, throw some light upon the principles of nomen- 
clature. Terms of this kind may even, besides being de- 
signations of certain definite persons, acquire a meaning of a 
more general and abstract character. Mr. A. J. Swann wrote 
to me from Kavala Island, in Lake Tanganyika, that 
among the Waguha the words for father, baba and tat a, 
also have the meaning of protector and provider. Among 
the Ewhe-speaking peoples of the Slave Coast, according to 
Major Ellis, the words now used to express " father," to and 
fofo, mean respectively, " he who owns " and " he who main- 
tains," neither of them having any relation to the act of 
begetting.5 The Semitic word for father, ab (abu), is used in 
a wide range of senses, although, to quote Robertson Smith, 
it is in all the dialects used " in senses quite inconsistent with 
the idea that procreator is the radical meaning of the word."6 
Moreover, there is a large number of terms of relationship 
which do not consist of sounds borrowed from the lips of 
infants, and some of these are known to have an inde- 
pendent meaning, apart from being terms of relationship, 
or to have been derived from words having an independent 
meaning. But I do not know a single classificatory term 
which directly refers to the act of begetting or to the fact 
1 Seligman, op. cit. p. 66 sq. 
2 Macdonald, Oceania, pp. 126, 186. 
3 Morgan, Systems, pp. 295, 313, 339, 348, 358, 362, 368, 374. 
4 Cf. Lang, ' Origin of Terms of Human Relationship,' in Proceed, 
of the British Academy, 1907-1908, p. 149. 
5 Ellis, Ewe-speaking Peoples of the Slave Coast, pp. 211, 213. 
0 Robertson Smith, Kinship and Marriage in Early Arabia, 
p. 117 sq. 246 
THE HISTORY OF HUMAN MARRIAGE chap. 
of having been begotten-in other words, the meaning of 
which suggests that uncles are classed together with the 
father, or nephews and nieces with sons and daughters, or 
cousins with brothers and sisters, because there is uncertainty 
as regards fatherhood. Such an assumption is not sup- 
ported by the intrinsic meaning of terms, so far as it is 
known. We shall now see whether it may be deduced from 
the result gained by an examination of the qualities which 
individuals named by the same term have in common. 
In every classification similarities of some kind or other arc 
grouped together. What are the similarities between the 
persons classed together by the classificatory terms of 
relationship ? 
The persons named by the same classificatory term are 
generally of the same sex. Yet there are cases in which the 
sex of the person is indicated by a special word added to the 
relationship term. In the Hawaian system father and other 
kinsmen of the same generation are called makua kana ; 
mother, mother's sisters, father's sisters, &c., makua 
waheena ; kana and waheena being the terms for male and 
female respectively. A son is called kaikee kana, a daughter 
kaikee waheena, whilst kana alone is applied to husband, 
husband's brother, and sister's husband, and waheena to 
wife, wife's sister, brother's wife, &c. Certain terms may 
also be influenced by the sex of the' speaker. In many 
systems two brothers have a term or terms which they use 
in addressing or speaking of one another, and the same term 
or terms may be used by two sisters, or these may have terms 
special to themselves ; whereas a brother and sister use a 
different term or different terms.1 In the Omaha language 
different terms were used when a father or mother was spoken 
to by a son and when they were addressed by a daughter.2 
The persons classed together generally belong to the 
same generation, but at the same time the term applied, 
1 Several instances of this are found in Morgan's tables. See 
also Rivers, History of Melanesian Society, i. 9; Macdonald, 
Africana, i. 143 (Eastern Central Africans). 
2 Alice Fletcher and La Flesche, ' Omaha Tribe,* in Ann. Rep. Bur, 
Ethn. xxvii. 314. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
247 
in an equal degree, depends on the generation of the other 
party. Within the same generation, also, difference of age 
very frequently influences the nomenclature. The Hawaians, 
according to Judge Andrews, have no definite general word 
for brother in common use. But kaikuadna signifies any 
one of my brothers or male cousins, older than myself, I 
being a male, and any one of my sisters or female cousins, 
older than myself, I being a female; whilst kaikaina 
signifies the younger brother of a brother, or the younger 
sister of a sister.1 Such distinguishing terms applied to 
older and younger are very frequent in the case of the 
relationship between brothers and between sisters, whereas 
only one term, irrespective of age, is nearly always used for 
brother by a sister and for sister by a brother. Often, too, 
the brothers of the father, and less frequently the sisters of 
the mother, are denoted differently according as they are 
older or younger than the father or mother ; but a similar 
practice is exceptional in the case of the mother's brothers 
or the father's sisters.2 In the Fulfiilde tongue (Central 
Africa) the age of the uncles is so minutely specified that 
the first, second, third, fourth, and fifth uncle, on both 
the father's and the mother's side, are each called by a 
particular name.3 
The influence of age shows itself in yet other ways, both 
in " classificatory " and in " descriptive " systems of 
nomenclature. The wider use of many terms of relationship 
and the meaning of certain terms, or of words related to 
them, bear witness in this direction. Among the North 
American Indians old people are very commonly named 
grandfathers and grandmothers.4 The Finnish immi does 
not signify grandmother only, but old woman in general.5 
Among many uncivilised peoples old men are addressed as 
1 Morgan, Systems, p. 453 note. 
2 See Rivers, in Hastings, op. cit. vii. 702 ; Idem, History of 
Melanesian Society, i. 9 sq. 
3 Barth, op. cit. p. 216. 
4 Waitz, Anthropologic der Natuvvolker, iii. 116. Harrington, 
' Tewa Relationship Terms,' in American Anthropologist, N.S. xiv. 
492. 
5 Ahlqvist, Die Kulturworter der westfinnischen Sprachen, p. 209. 248 
THE HISTORY OF HUMAN MARRIAGE chap. 
" fathers " and old women as " mothers."1 The Swedish 
far and mor and the Russian batushka and matushka 
are often used in the same way ; but the Russian of the 
lower classes also greets men superior to him in years by the 
name of " uncle,"2 and in Swedish the word farbror (father's 
brother) is a familiar term for a man belonging to an older 
generation than the speaker, whilst broder (brother) is 
reciprocally used by male friends of the same generation. 
Among the Basuto, according to Casalis, " in addressing 
a person older than one's self, one says, ' My father, my 
mother ' ; to an equal, ' My brother ' ; and to inferiors, 
' My children.' "3 Mr. Cousins wrote to me that among the 
natives of Cis-Natalian Kafirland the terms for father, 
mother, brother, and sister are not restricted to them only, 
but are applied equally to other persons of a similar age, 
whether related or otherwise. " Bawo," he says, " means 
elder or older, bawo-kulu means a big-father, one older than 
father." The Brazilian Uainuma call a father paii, but also 
pechyry, which means old.4 The Finnish isd and the Votyak 
ai, father, the Lappish of a and the Esthonian di, grandfather, 
are evidently related to the Finnish iso and dija, which mean 
big.5 In German the parents are die Eltern, the older (die 
Aelleren), and they are also called familiarly die Alten, 
the father der Alte, and the mother die Alie or Altschef 
In Hungarian, where bdtya stands for elder brother, an uncle 
is nagybdtya, that is, a big elder brother.7 In ^agatai an 
elder sister is called egeci, which, according to Vambery, 
actually means old woman (ege, old, big ; eci, woman, sister).8 
1 Collins, Account of the English Colony in New South Wales, i. 544. 
Salvado, Mdmoires historiques sur l'Australie, p. 277. Robertson, 
Erromanga, p. 403. Sibree, The Great African Island, p. 244 sq. 
(Hova). Mr. A. J. Swann, in a letter (Waguha). Hinde, The Last 
of the Masai, p. 51. Reade, Savage Africa, p. 248. 
2 Leroy-Beaulieu, Empire of the Tsars, i. 489. 
3 Casalis, BasutOs, p. 207. 4 v. Martius, op. cit. ii. 247 sq. 
•' Ahlqvist, op. cit. p. 209. 
6 Deecke, Die deutschen Verwandtschaftsnamen, p. 79. 
7 Ahlqvist, op. cit. p. 212. 
8 Vambery, Die primitive Cultur des turko-tatarischen Volkes, 
p. 65. vii CRITICISM OF HYPOTHESIS OF PROMISCUITY 
249 
In the Galibi language of Brazil tigami signifies young 
brother, son, and little child indiscriminately.1 Several 
languages have no other words for son and daughter than 
those for lad and girl.2 Thus in Hawaian a son is called 
male child, or little male, and a daughter female child or 
girl.3 
It should be added that in some cases the terms are deter- 
mined, not by the relative ages of those who use the terms, 
but by the ages of the children of some more or less distant 
ancestor.4 Among the Maori, for example, a man will 
address another as tuakana, " elder brother " or " elder 
cousin," if the latter belongs to an elder branch of the 
family, and as taina, " younger brother " or " younger 
cousin," if he belongs to a younger branch.5 
Although terms belonging to classificatory systems are 
often used in addressing strangers, they are in the first place 
applied to kinsfolk. Where the clan exists, they are used to 
denote members of the clan. " All the men of the clan of the 
speaker and of his own generation are classed in terminology 
with his brothers. If the clan is patrilineal, all men of the 
previous generation of his clan are classed with his father, 
and all of the succeeding generation with his sons. Simi- 
larly, all the men of his mother's clan and of her generation 
are classed with his mother's brother, and all the men of 
the succeeding generation with his mother's, brother's 
children."6 Again, when a classificatory nomenclature is 
used by a people among whom the clan-organisation does 
not exist-as is also frequently the case-the terms are 
1 von den Steinen, op. cit. p. 341. 
2 Ahlqvist, op. cit. p. 210. von der Gabelentz, op. cit. i. 172. 
3 Morgan, Systems, p. 452 note. Cf. the German Junge. 
4 See Rivers, History of Melanesian Society, i. 10; Idem, in Hastings, 
op. cit. vii. 702. 
6 Best, loc. cit. p. 28. 
6 Rivers, in Hastings, op. cit. vii. 703. Idem, History of Melanesian 
Society, i. 7. In the latter work (i. 7 n. 1) Dr. Rivers defines the 
clan as "an exogamous group within a tribe or other community, 
all the members of which are held to be related to one another, and 
bound together by a common tie. In general, this tie is either a 
belief in common descent from some ancestor, real or mythical, 
or the common possession of a totem.'' 250 
THE HISTORY OF HUMAN MARRIAGE chap. 
usually limited to those with whom some genealogical 
connection can be traced or with whom the tradition of such 
a connection persists.1 This, however, does not exactly 
prove that the classificatory terms are, or. originally were, 
expressions of blood-relationship. There are special rights 
and duties connected with kinship, and various facts indicate 
that the terms in classificatory systems are fundamentally 
influenced by such social relationships, although it is quite 
probable that vague ideas of consanguinity from the begin- 
ning were attached at least to some of them. Not only 
are different designations applied to strangers and kinsfolk, 
but the comprehensiveness of terms used for kindred of the 
same sex and generation is influenced by their social relations 
or functions, the terms, generally speaking, being less com- 
prehensive the more differentiated these functions are. 
This was pointed out already in the first edition of the present 
work (p. 93 sqq.), and a similar view has been taken by some 
later writers. I wrote (p. 95) that " a maternal uncle is 
almost always distinguished from a father by a separate 
term, whilst this is not the case with an uncle on the father's 
side, the former generally living in another community from 
his nephew, and, besides, very frequently standing to him in 
a quite peculiar relationship through the rules of succession. 
It may be fairly assumed, too, that a mother's sister much 
oftener than a father's sister is called a mother, because 
sisters, among savages, keep, as a rule, far more closely 
together, when married, than brothers and sisters ; some- 
times even, especially among the North American Indians, 
they are the wives of the same man. If we add to this that 
a father's brother's son and a mother's sister's son are more 
commonly addressed as brothers than a father's sister's son 
and a mother's brother's son, it becomes obvious to how great 
an extent the nomenclature is influenced by external (that 
is, social2) relations. But as a certain kind of external 
relationship is invariably connected with a certain degree, 
or certain degrees, of blood-relationship, the designations 
1 Idem, History of Melanesian Society, i. 8. 
2 In other places (pp. go, 93) I used the expression "social 
relationship." vn CRITICISM OF HYPOTHESIS OF PROMISCUITY 
251 
given with reference to the former have been taken as terms 
for the latter." 
I have made this reference to my previous writing to 
defend myself against an accusation which has been brought 
against me by Dr. Rivers. Referring to McLennan and 
myself, he writes that the older objections to Morgan's 
theory were based on the idea that the classificatory system 
is only a table of terms of address, " a view which by no 
means removes the necessity for a theory of its origin " ; 
whereas the tendency of more recent objectors (Lang, 
Thomas1) has been to show that the terms of the system 
are expressive of status and duties and not of consanguinity 
or affinity.2 What I had written was that the terms of 
relationship were most probably in their origin terms of 
address. As such they are used even now. " The American 
Indians," says Morgan, " always speak to each other, when 
related, by the term of relationship, and never by the 
personal name of the individual addressed."3 In Melanesia 
also, according to Dr. Rivers, such terms are used for 
address, personal names often being prohibited.4 Among 
the Bangerang tribe of Victoria grown-up males " invariably 
addressed each other either by a term of relationship, or 
by the name of the class in the tribe to which each person 
belonged " ;5 and in the Kariera tribe of Western Australia 
a man or woman never addresses anyone except young 
children by a personal name, but uses the appropriate 
1 Lang, Social Origins, p. 102 ; see also Idem, in Proceed, of the 
British Academy, 1907-1908, pp. 141, 142, 152 sq. Thomas, Kinship 
Organisations and Group Marriage in Australia, p. 123 sqq. 
2 Rivers, in Anthropological Essays presented to E. B. Tylor, 
p. 318 sq. In his History of Melanesian Society (i. 6) Dr. Rivers 
writes of the classificatory system, " McLennan supposed it to be 
merely a system of titles of address and Westermarck and others 
have held the same opinion and have attached little or no weight 
to the value of the system as evidence of social conditions." 
3 Morgan, Systems, p. 132. See also Charlevoix, Voyage to North- 
America, ii. 42 (Algonkin) ; Whiffen, North-West Amazons, p. 153. 
4 Rivers, History of Melanesian Society, ii. 36 sq. See also 
Seligman, Melanesians of British Neiv Guinea, pp. 60, 436 (Koita, 
Southern Massim). 
6 Curr, Recollections of Squatting in Victoria, p. 268. 252 
THE HISTORY OF HUMAN MARRIAGE chap. 
relationship term.1 At the same time classificatory terms 
are also used in speaking 0/persons, and it may be that one 
term is used when addressing a relative and another term 
when speaking of him.2 But the question whether the 
classificatory terms were originally terms of address or not 
is of no importance in the present connection, when we are 
trying to find the principles according to which the classi- 
fications have been made. And this question was certainly 
not ignored by me, the result of my investigation (pp. 90-96) 
being that the names are given " chiefly with reference to 
sex and age, as also to the external, or social, relationship 
in which the speaker stands to the person whom he 
addresses." 
I am glad to find that this conclusion, which was arrived 
at about thirty years ago, gains so much support from 
Dr. Rivers' recent indefatigable research into the classi- 
ficatory system, which has led him to the conviction that 
" the terminology of relationship has been rigorously deter- 
mined by social conditions."3 Certain facts which he gives 
with reference to the terms for the mother's brother are 
particularly instructive. In Polynesia both the Hawaians 
and the inhabitants of Niue class this individual with the 
father, and in neither case was Dr. Rivers able to discover 
that there were any special duties, privileges, or restrictions 
ascribed to him. In the Polynesian islands of Tonga and 
Tikopia, on the other hand, where there are special terms 
for the mother's brother, this relative has also special 
functions. And the only place in Melanesia where Dr. 
Rivers failed to find a special term for the mother's brother, 
namely, the Western Solomon Islands, was also the only 
part of Melanesia where he failed to find any trace of special 
social functions ascribed to that relative.4 He also observes 
that the general character of the classificatory system is 
exactly such as would be the consequence of its origin in a 
1 Brown, in Jour. Roy. Anthr. Inst, xliii. 150. 
2 Rivers, in Hastings, op. cit. vii. 701. Idem, Todas, pp. 483, 
493« 
3 Idem, Kinship and Social Organisation, p. 1 ; &c. 
4 ibid. p. 15. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
253 
social structure founded on an exogamous social group, 
and that many details of these systems point in the same 
direction. Thus, " the distinctions between the father's 
brother and the mother's brother, between the father's 
sister and the mother's sister and between the children of 
these relatives are themselves the natural consequences of 
the origin of the system in a clan-organisation, for owing to 
the practice of exogamy, the brothers of the father must 
always belong to a clan different from that of the brothers of 
the mother and so with the other examples mentioned. 
Whenever one finds that these distinctions do not exist, 
it is also found that the clan-organisation is absent or is in 
course of profound modification."1 Among some North 
American Indians distinct terms are found for relatives of 
the uncle- and aunt-group after the death of a parent ; 
and the reason for this is no doubt, as Professor Kroeber 
observes, that the uncle's relation to his orphaned nephew 
tends to be somewhat different from his relation to the 
same boy while his natural protector, his father, was 
alive.2 
The social factor has also presumably made itself felt in 
the distinctions between persons of different sex and of 
different age. As Wundt remarks in his discussion of the 
classificatory system, persons of the same sex are more 
closely associated than are men and women. " In the 
men's houses a companion of the same group is a brother, 
one of the next elder group, a father. Together with these 
men the individual goes to war and to the hunt."3 That 
the terms for brother and sister in many cases depend on the 
sex of the speaker as well as on the sex of the person addressed 
is explained by Dr. Rivers as a consequence of the early 
separation of brothers and sisters, which is a frequent 
custom among savage peoples. While two brothers are thus 
constant companions, and also two sisters, a brother and 
sister will at an early age come to belong to different social 
1 Ibid. p. 71 sqq. Rivers, History of Melanesian Society, i. 8. 
2 Kroeber, loc. cit. p. 79. 
8 Wundt, Elemente der Vblkerpsychologie, p. 42 (English trans- 
lation, p. 41). 254 
THE HISTORY OF HUMAN MARRIAGE chap. 
groups within the clan ; and this separation will be accen- 
tuated when the girl marries and, perhaps at an early age, 
moves to and becomes for all practical purposes a member 
of another clan.1 As to the distinctions of age, it should be 
noticed that among uncivilised peoples old age commands 
respect and gives authority, and that, apart from this, 
there are rights incident to superiority of age as well.2 
Fhe Australian natives have a well-regulated order of 
precedence and authority. " When the individual reaches 
the* full development of puberty, he or she undergoes a 
ceremony which entitles him or her on its successful com- 
pletion to a certain social rank or status in the community. 
As life progresses, other and higher ranks are progressively 
attainable for each sex, until the highest and most honourable 
grade, that enjoyed by an old man, or an old woman, is 
reached."3 Of another race which in its entirety has the 
classificatory system, the North American Indians, it is 
said that " superior age gives, authority ; and every person 
is taught from childhood to obey his superiors and to rule 
over his inferiors. The superiors are those of greater age ; 
the inferiors, those who are younger."4 The same influence 
makes itself felt in the relations between elder and younger 
brothers and sisters ; the eldest brother comes next to the 
father in authority, and, in case of his death, succeeds him 
as the head of the family. Dr. Jochelson observes that the 
distinct denomination of the elder brother and sister among 
the Koryak shows their position in the family,5 and that the 
case is similar with the inclusion of the mother's elder 
brother or the father's elder sister in one term with the 
grandfathers or grandmothers among the Yukaghir. 
" Among the female members of a family, the father's 
elder sister occupies the first position, after the father's 
1 Rivers, History of Melanesian Society, i. 9. 
2 See Westermarck, Origin and Development of the Moral Ideas, 
i 603 sqq. 
3 Eoth, Ethnological Studies among the North-Wes^-Central Queens- 
land Aborigines, p. 169. 
' Powell, ' Sociology,' in American Anthropologist, N.S. i. 700. 
(/. Idem, in Ann. Rep. Bur. Ethn. iii. p. Iviii. 
'■ JcchelMin, K aval: p. 760. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
255 
mother, in respect to the household ; and the mother's 
elder brother, after the mother's father, is the head of the 
family."1 Among many Ural-Altaic peoples 'the same term 
is applied to an elder brother as to an uncle, and the same 
term to an elder sister as to an aunt.2 
In an earlier work I have emphasised the immense in- 
fluence which living together, apart from any blood- 
relationship, has exercised upon social relations; when 
unsupported by local proximity kinship itself loses much of 
its social strength, nay we may even say that the social 
force of kinship is ultimately derived from near relatives' 
habit of living together.3 The local influence may therefore 
also be expected to show itself in the nomenclature. In 
fact, there can be little doubt that the classifications, even 
though directly associated with a clan-organisation, are 
often influenced by certain relatives' habit of living in 
the same or in a different locality. We shall see that the 
clan-organisation itself is probably in a large measure 
dependent on local connections.4 Moreover, there are 
instances of classificatory terms being regularly applied to 
all persons of a certain class in the same locality, even though 
they are not kindred. In the Central Provinces of India, 
according to Mr. Russell, it is commonly found that a man 
will address all the men in his village who belong to the 
generation above his own as uncle, though they may be of 
different castes, and the children of the generation below his 
own as nephew or niece. TVIoreover, when a girl is married, 
all the old men of the village call her husband " son-in-law." 
And this extends even to the impure castes who cannot be 
touched.5 In Erromanga a man may call another " his 
brother " if they as children lived and played together in the 
same village.6 
The importance of the social factor is also evident in the 
1 Idem, Yukaghir, p. 72. 2 Ahlqvist, op. cit. p. 211. 
3 Origin and Development of the Moral Ideas,. ii. 197 sqq. 
4 Infra, on Exogamy. 
5 Russell, Tribes and Castes of the Central Provinces of India, 
i- 155- 
0 Robertson, op. cit. p. 403 sq. 256 
THE HISTORY OF HUMAN MARRIAGE chap. 
terms often used in addressing strangers. The Yoruba 
of the Slave Coast apply the terms baba, father, and iya, 
mother, to strangers of the generation next above the 
speaker, "when it is desired to show respect."1 Among the 
Hovas of Madagascar the words for brother and sister " are 
also used widely for any person whom one meets and desires 
to act towards in a friendly manner." 2 In Erromanga, 
" to call a man your father, nate, your brother, avugsai, 
or your son, netug, is the greatest token of your love or 
respect."3 Concerning the Algonkin Indians, Charlevoix 
states that " when there is between them no relation or 
affinity, they use the term of brother, uncle, nephew, or 
cousin, according to each other's age ; or according to the 
value they have for the person they address." 4 The 
Yahgans of Tierra del Fuego, according to Mr. Bridges, 
form certain kinds of friendships, and " speak of aunts, 
uncles, brothers, sisters, cousins, nieces and nephews, &c., 
which are only so through the friendships established." 5 
At the same time they have different appellations for 
nephews and nieces on the brother's side and nephews and 
nieces on the sister's side, and their words for uncle and aunt 
differ according as this relationship is paternal or maternal.6 
Dr. von den Steinen was called by the Brazilian Bakairi 
sometimes " elder brother," sometimes " grandfather," and 
by the Mehinaku " maternal uncle " ; whilst his travelling 
companions were addressed as " younger brothers " 
or " cousins."7 While living in a Pueblo village Miss 
Freire-Marreco was called " grandmother " by the younger 
generation and " aunt " or " female cousin " by the older 
people.8 Mr. Hartshorne was called by the Veddas hura, 
1 Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 178. 
2 Sibree, op. cit. p. 247. 3 Robertson, op. cit. p. 404. 
4 Charlevoix, op. cit. ii. 42. 
5 Bridges, ' Manners and Customs of the Firelanders,' in A Voice 
for South America, xiii. 212. 
6 Idem, in a letter to the author dated Downeast, Tierra del 
Fuego, August 28th, 1888. 
7 von den Steinen, Unter den Naturvblkern Zentral-Brasiliens, 
P- 331* 
8 Harrington, loc. cit. p. 492. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
257 
or cousin.1 When the natives of Botany Bay, Port Jackson, 
and Broken Bay, in the time when Sydney was established, 
perceived the authority with which Governor Phillip com- 
manded, and the obedience which he exacted, they bestowed 
on him the distinguishing appellation of be-anna, or father.2 
Facts like these testify that classificatory terms embody 
feelings of friendliness or respect, normally felt towards 
kindred, and consequently are expressive of relations of a 
social character. 
Yet, however greatly the nomenclature may have been 
influenced by social conditions, the correlation between 
the presence of a term of relationship and special social 
relations or functions associated with it is by no means 
complete. There are cases in which the same term is 
applied to relatives who stand in considerably different 
social relations to the speaker. Dr. Rivers points out 
that while in Tonga and Tikopia the father's sister is denoted 
by a special term, because she has very definite rights and 
duties, there is no special term for her in the Banks Islands, 
although there also she has very definite and important 
functions; she is here classed with the mother, as is the 
case in most Polynesian islands, " although she possesses 
rights and privileges and is subject to restrictions quite 
different from those of the mother."3 Father Egidi in- 
formed Dr. Rivers that in Mekeo the mother's brother has 
the duty of putting on the first perineal garment of his 
nephew ; yet he has no special term, but is classed with the 
father.4 But besides such minor cases in which relatives 
with special functions are not denoted by special terms, 
we are confronted with the extremely frequent custom of 
classing the father's brother with the father and the mother's 
sister with the mother, and with the occasional custom of 
1 Hartshorne, ' Weddas,' in Indian Antiquary, viii. 320. Accord- 
ing to Mr. Le Mesurier (' Veddas of Ceylon,' in Jour. Roy. Asiatic 
Soc. Ceylon Branch, ix. 347), the Rock or Hill Veddas use the word 
for brother, aluwa, when they speak of or to any person with whom 
they are in friendship. 2 Collins, op. cit. i. 544. 
3 Rivers, History of Melanesian Society, ii. 21, 45. Idem, Kinship 
and Social Organisation, p. 16. 
4 Idem, Kinship and Social Organisation, p. 16. 258 
THE HISTORY OF HUMAN MARRIAGE chap. 
classing all uncles with the father and all aunts with the 
mother. This is done in spite of the very special functions 
of the father and the mother. Dr. Rivers himself admits 
that " in general, it would seem that the relations between 
parents and children associated with the classificatory system 
are much like those which exist among ourselves. The 
father and mother provide for the child, feed, clothe, and 
train him, w'hile the child obeys his parents and assists them 
in their occupations."1 In the Kariera tribe, where, 
according to Mr. Brown, the relationship system is " pre- 
eminently a system of reciprocal rights and duties," though 
a man owns certain duties to all the men he calls " father," 
he must observe them more particularly in regard to his 
real father or his father's brothers than in regard to a 
distant cousin of his father ; and the case is similar with 
every other relationship.2 
This remarkable lack of discrimination in the terms 
used for father and mother may possibly be due to the 
fact that these terms have been particularly influenced by 
the babble of infants, who are apt to use the same sounds 
for any man of their father's age and for any woman of their 
mother's age. Another explanation, suggested by two 
American writers with special reference to the Omaha 
Indians, is that those terms were applied to " what may be 
called potential relationships, that is, relationships that 
would be established through marriage made in accordance 
with tribal custom. If the wife had sisters, these women 
held a potential relationship to her husband, as they might 
become his wives either during his wife's lifetime or at her 
death " ; whilst a man was under obligation to marry his 
brother's widow. " Because of these potential relationships 
the children of the wife called all those whom their father 
might marry ' mother ' and all their father's brothers 
' father.' "3 All that we can say with a fair amount of 
certainty is that the classifications in question are connected 
with that greater strength of kinship ties beyond the family 
1 Idem, in Hastings, op. cit. vii. 705. 
2 Brown, in Jour. Roy. Anthr. Inst, xliii. 157. 
3 Alice Fletcher and .La Flesche, loc. cit. p. 313. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
259 
in the strictest sense of the word which is found among most 
uncivilised peoples when compared with ourselves and 
which lessens the difference in status between the members 
of the family and other kindred. But we are by no means 
entitled to conclude that they are survivals of a social 
organisation where there was no difference between the 
social functions of a father and an uncle and between those 
of a mother and an aunt-in other words, where there was no 
family of the present type. Existing social conditions 
may throw light on the terms of relationship, but to infer, 
vice versa, the earlier existence of certain social conditions 
from certain terms of relationship can in most cases be 
nothing more than a guess. 
It should, first, be noticed that if some term or terms of 
relationship are regularly found together with some special 
social institution, it may be that the latter is no more the 
cause of the former than the lightning is the cause of the 
thunder: they may both be effects of a common cause. 
The frequent co-existence of classificatory terms with exo- 
gamic marriage prohibitions may be an instance of this : 
the terms are not necessarily caused by the prohibitions, but 
the strong feeling of kinship which has, for example, led to 
the classification of cousins with brothers and sisters may 
also have led to the rule which treats sexual intercourse 
between them as incest. If a classificatory nomenclature 
thus co-exists with a certain social rule or institution as 
an effect of the same cause as has produced the latter, 
we can never be certain that where we find one of the effects 
the other one must also have occurred in former times. 
Nor does an established causal connection between the 
presence of a term of relationship and some special function 
associated with the relationship prove the earlier existence 
of the social function whenever the term is found alone, 
since the latter may have a different origin in different cases. 
Suppose that in Mekeo the mother's brother, whose duty it 
is to put on the first perineal garment of his nephew, 
instead of being classed with the father had a special term, 
as he ought to have according to the general law formulated 
by Dr. Rivers, that would not prove that whenever the 260 
THE HISTORY OF HUMAN MARRIAGE chap. 
mother's brother has a special term-even if it happened on 
a neighbouring island-it is or has been his duty to put on 
the perineal garment of his nephew. The special term 
merely suggests in a broad way that the relation in which he 
stands to his nephew differs somehow, or has differed some- 
how, from the relation in which the father stands to his son. 
But this is a very empty suggestion, which practically 
teaches us nothing about the present or the past. 
In certain respects, however, classificatory terms have 
been supposed to give us very definite information about 
earlier social conditions. I shall first deal with a case to 
which Dr. Rivers attaches much importance from the 
point of view of method. In some Melanesian islands he 
has found the practice of the cross-cousin marriage- 
that is, the marriage of a man with the daughter either of his 
mother's brother or of his father's sister-combined with a 
nomenclature which is " just such as would follow from this 
form of marriage."1 In the Mbau dialect of Fiji, for 
example, the word vungo is applied to the mother's brother, 
the husband of the father's sister, and the father-in-law. The 
word nganei is used for the father's sister, the mother's 
brother's wife, and the mother-in-law. The term tavalc is 
used by a man for the son of the mother's brother or of the 
father's sister as well as for the wife's brother and the sister's 
husband. Ndavola is used not only for the child of the 
mother's brother or father's sister when differing in sex from 
the speaker, but this word is also used by a man for his 
wife's sister and his brother's wife, and by a woman for her 
husband's brother and her sister's husband. " Every one of 
these details of the Mbau system," says Dr. Rivers, " is the 
direct and inevitable consequence of the cross-cousin 
marriage, if it become an established and habitual practice." 
He then concludes that if there be found among any other 
people all the characteristic features of this system, there 
could be no reasonable doubt as to the former existence of 
the cross-cousin marriage, since " it would seem almost 
inconceivable that there should ever have existed any other 
conditions, whether social or psychological, which could have 
1 Idem, Kinship and Social Organisation, p. 22 sq. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
261 
produced this special combination of peculiar uses of terms of 
relationship."1 So also the terms of relationship used in 
three of the chief languages spoken by the people of South 
India " are exactly such as would follow from the cross-cousin 
marriage." So far as we know, this kind of marriage is not 
now practised by the vast majority of the peoples who use 
those terms, but it still persists in many parts of South 
India ; and we are told that there can be no doubt that the 
terminology is a survival of an ancient social condition in 
which the cross-cousin marriage was habitual.2 
Dr. Rivers bases his conclusions on the assumption that 
when the cross-cousin marriage is found together with a 
nomenclature which is " just such as would follow from this 
form of marriage," the nomenclature must necessarily be 
the result of the cross-cousin marriage. Now it seems to me 
difficult to admit that the terminology in question exactly 
corresponds with the consequences of a cross-cousin marriage. 
Dr. Rivers says himself that although all the peoples who 
are known to practise this kind of marriage use the classi- 
ficatory system of relationship, the marriage is not usually 
between cross-cousins in the wide classificatory sense, but 
between the children of " own brother and sister."3 If an 
ancient Arab called his beloved one bint 'amm, that is, 
father's brother's daughter, and his father-in-law 'amm, 
father's brother, even though they were not related to each 
other by blood,4 it is obvious that he did so because a man 
was held to have a right to his bint 'amm's hand ; 5 but it is 
not equally obvious that a Fijian calls a large number of 
men (mother's brother, father's sister's husband, father-in- 
law) vungo and a large number of women (father's sister, 
mother's brother's wife, mother-in-law) nganei because he 
marries, or is likely to marry, the daughter of one definite 
member of the group. I venture to believe that the ter- 
minology allows of a more natural explanation than that 
given by Dr. Rivers. In Fiii and other places where the 
1 Ibid. p. 43. 2 Ibid. p. 47 sqq. 
3 Rivers, ' Marriage,' in Hastings, op. cit. viii. 426. 
4 Goldziher, ' Endogamy and Polygamy among the Arabs, ' 
in Academy, xviii. 26. 
5 Westermarck, Marriage Ceremonies in Morocco, p. 55 sq. 262 
THE HISTORY OF HUMAN MARRIAGE chap. 
cross-cousin marriage exists, a man is not allowed to marry 
the daughter either of his father's brother or of his mother's 
sister, whereas marriage between cross-cousins in the classi- 
ficatory sense is permitted, although a man usually marries 
the actual daughter of his father's sister or of his mother's 
brother. Persons to whom a man may become related by 
marriage and others to whom he may not become thus related 
stand in a different relation to him and are therefore apt to 
be distinguished from each other by different terms. In 
other words, the terminology may, instead of being due 
to habitual cross-cousin marriage, be simply con- 
nected with the marriageability of cross-cousins in the 
classificatory sense. This explanation reconciles the exist- 
ence of classificatory terms in a wide sense with the preva- 
lence of marriages between cross-cousins in our sense of 
the term. It further accounts for the fact that a terminology 
similar to that which occurs among the peoples who have the 
cross-cousin marriage is also found among many peoples to 
whom this institution is unknown, but who share with the 
former the same rules as to the legitimacy or illegitimacy of 
marriages between different kinds of cousins; and it 
accounts for this fact without postulating the earlier exist- 
ence of habitual cross-cousin marriage among them. This I 
consider to be a great advantage ; a theory which can in 
a satisfactory manner explain a social phenomenon by 
existing conditions must certainly take precedence of one 
which explains it as a survival of something hypothetical 
in the past. It is interesting to note that a terminology 
of relationship which according to Dr. Rivers " bears 
evidence of the cross-cousin marriage " has been found 
among many North American Indians, although this form 
of marriage is known by him to occur only among one 
or two of their tribes. He consequently admits that the 
existing evidence is inconclusive.1 
1 Rivers, Kinship and Social Organisation, p. 49 sqq. Frazer, who 
has in detail studied the prevalence of cross-cousin marriage among 
different peoples, observes {Folk-Lore in the Old Testament, ii. 145 sq.) 
that the Western Tinne present " the only clear case of preference 
for marriage with a first cousin, the daughter of a mother's brother, 
which has been recorded in the whole of North America." vn CRITICISM OF HYPOTHESIS OF PROMISCUITY 
263 
It may seem that I have devoted too much space to a 
detail of comparatively small importance. In this con- 
nection, however, it is not the facts themselves that are of 
interest, but the method of dealing with them. I think I have 
done justice to the school of Morgan by choosing for my 
criticism the case which one of its most careful and pains- 
taking exponents considers to demonstrate more clearly 
and conclusively than anything else the dependence of the 
classificatory terminology upon forms of marriage. Indeed, 
Dr. Rivers doubts whether, apart from quantitative veri- 
fication, " it would be possible in the whole range of science 
to find a case where we can be more confident that one 
phenomenon has been conditioned by another."1 For my 
own part, I have endeavoured to show that even the co- 
existence of a common nomenclature for two or more rela- 
tives with a form of marriage which more or less agrees with 
it does not prove that the form of marriage is the cause of 
the nomenclature. What, then, if the common nomen- 
clature is found alone ? According to Dr. Rivers, " one may 
say that the probability that the common nomenclature 
for two relatives is the survival of a form of marriage becomes 
the greater, the more similar is the general culture in which 
the supposed survival is found to that of a people who prac- 
tise this form of marriage."2 But this is, in my opinion, 
a very insufficient precaution, in the first place because 
co-existence is not causation, and in the second place because 
a classificatory term may have a different origin in different 
cases. It shows, however, like many other statements by 
Dr. Rivers, that he is aware of the perplexity of a subject 
which most of Morgan's followers, like the master himself, 
have treated with the greatest nonchalance.3 When 
1 Rivers, Kinship and Social Organisation, p. 25. 2 Ibid. p. 46. 
3 At the same time Dr. Rivers seems to maintain that we have 
a right to infer the existence of a certain form of marriage from 
certain features of terminology, even though we are ignorant of 
its existence anywhere. For he says {Kinship and Social Organisa- 
tion, p. 26) that Morgan lost a great opportunity by not predicting 
the cross-cousin marriage although he knew such features of termi- 
nology as follow from it. " He might have predicted a form of 
marriage which would soon afterwards have been independently 264 
THE HISTORY OF HUMAN MARRIAGE chap. 
Professor Kohler, for example, finds that among the Wad- 
shagga a man uses the same term for his father's sister as 
for his own sister, he at once concludes that this must be a 
survival from a time when nephews were in the habit of 
marrying their aunts.1 It would be interesting to learn 
what inferences would, according to the same principle, 
be made from the following facts with reference to early 
European marriage customs. In English the word '' nephew ' ' 
was in the seventeenth century frequently used for a grand- 
son, and the word "niece" for a granddaughter.2 In 
Dutch the term neef is applied to a nephew (brother's or 
sister's son), a male cousin (uncle's or aunt's son), and a 
descendant of any such relative or of a niece or a female 
cousin; and the term nicht is applied to a niece (brother's 
or sister's daughter), to a female cousin (uncle's or aunt's 
daughter), and to the descendant of any such relative or of 
a nephew or male cousin.3 Here the same terms are thus 
used for persons belonging to three or more different genera- 
tions. The Latin nepos meant both grandson and nephew ; 
and Herodotus used the word ave^Los both for a nephew 4 
and a cousin.5 
discovered. Such an example of successful prediction would have 
forced the social significance of the terminology of relationship 
upon the attention of students in such a way that we should have 
been spared much of the controversy which has so long obstructed 
progress in this branch of sociology." For my own part I believe 
that the progress of sociology is more hampered by the use of an 
illegitimate method than by the absence of predictions, even though 
these sometimes happen to be true ; and I sincerely hope that 
the new impetus given by Dr. Rivers' investigations to the study 
of the'terms of relationship will not lead sociologists to waste their 
time on forming all sorts of hypotheses on ancient marriage customs 
because the terms of relationship are " just such as would follow " 
from them. There are unfortunately signs that the temptation to 
do so is strong. 
1 Kohler, * Das Banturecht in Ostafrika,' in Zeitschr. f. vergl. 
Rechtswiss. xv. 13. 
2 Craigie, in New English Dictionary on Historical Principles, 
vol. vi. pt. iv. 91, 135. 
3 van Dale, Groot Woordenboek der Nederlandsch taal, pp. 1176 
1183 
4 Herodotus, vii. 5. $ Ibid. v. 30, vii CRITICISM OF HYPOTHESIS OF PROMISCUITY 
265 
If a common nomenclature for persons representing two 
or more relationships really were an indication of a certain 
form of marriage in the past, the English terminology might 
lead to the interesting discovery that the English have been 
in the habit of marrying their sisters. The same term, 
" uncle," is used for the father's brother and the mother's 
brother ; the same term, " aunt," for the father's sister and 
the mother's sister ; the same term, " grandfather," for the 
father's father and the mother's father ; the same term, 
" grandmother," for the father's mother and the mother's 
mother ; the same term, " cousin," for the father's brother's 
son or daughter and for the mother's brother's son or daughter, 
as also for the father's sister's son or daughter and for the 
mother's sister's son or daughter ; the same term, " nephew," 
for the brother's son and the sister's son ; the same term, 
" niece," for the brother's daughter and the sister's daughter ; 
the same term, " grandson," for the son's son and the 
daughter's son ; the same term, " granddaughter," for the 
son's daughter and the daughter's daughter. If we assume 
that the common nomenclature indicates the former com- 
bination of two relationships in one and the same person, 
we come to the conclusion that the lather's brother and sister 
have also been the mother's brother and sister, the father's 
parents also the mother's parents, the father's brother's 
children also the mother's brother's children, the father's 
sister's children also the mother's sister's children, the 
brother's children also the sistei's children, the son's children 
also.the daughter's children. It is true that the English 
have the terms father- and mother-in-law, brother- and 
sister-in-law, and son- and daughter-in-law. But these 
terms would also have been necessary in the days of the 
brother-and-sister marriage, because a man who had no 
sister to marry, and a woman who had no brother, could not 
on that account be expected to remain celibates. The 
Hawaians, too, have terms for relatives by marriage,1 
in spite of their prehistoric " consanguine family," which 
really ought to have made such terms unnecessary ; indeed, 
in place of our word " brother-in-law " they have two 
1 Morgan, Systems, p. 561 sqq. 266 
THE HISTORY OF HUMAN MARRIAGE chap. 
expressions, one used by a man for his wife's brother and 
his sister's husband, and another used by a woman for her 
husband's brother and her sister's husband.1 But even now 
the father- and mother-in-law are in familiar English often 
called " father " and " mother," the brother- and sister-in- 
law " brother " and " sister " (Dr. Rivers, however, seems to 
connect this expression with the prohibition of marrying the 
deceased wife's sister),2 and the son- and daughter-in-law 
" son " and " daughter," and if a man wants to be on good 
terms with his wife's uncles and aunts, he may call them 
" uncle " and " aunt." Not the least interesting fact con- 
nected with the brother-and-sister marriage deduced from the 
English terms of relationship is that it would have occurred in 
comparatively recent times, since both the Anglo-Saxon and 
the Latin systems, as has been mentioned above, distin- 
guished between paternal and maternal uncles and aunts. 
This, however, is rather awkward, because, it also gives the 
historian a voice in the matter. He who studies the history 
of a savage society need not fear any such interference in 
his deductions from the systems of relationship, in which, 
according to Dr. Rivers, he has, " like fossils, the hidden 
indications of ancient social institutions."3 
Why is it a fallacy to conclude that a term which may be 
applied to persons representing different relationships proves 
the previous existence of a certain form of marriage ? The 
fallacy lies in the assumption that the use of such a term 
combines the different relationships in one and the same 
person. A classificatory term is' taken to be used as if it 
stood for two descriptive terms. If I say that a man is the 
father's brother and the mother's brother of another person, 
we may at once draw the conclusion that the latter is the 
offspring of brother and sister ; but the case is different if I 
say that a man is another person's uncle, simply because 
uncle means father's brother or mother's brother, not 
father's brother and mother's brother. Some of those who 
maintain that the classificatory system proves the former 
1 Morgan, Systems, p. 565. 
2 Rivers, Kinship and Social Organisation, p. 65. 
3 Idem, History of Melanesian Society, i. 3. vn CRITICISM OF HYPOTHESIS OF PROMISCUITY 
267 
existence of a " consanguine family " or of group-marriage 
have been guilty of an even worse confusion. They have 
practically assumed that a classificatory term stands for 
one descriptive term, which, however, is applied to different 
relatives because any one of them may actually represent 
the relationship expressed by the term. The same term, 
it is argued, is applied to the father and to the father's 
brother and cousins, or (in the Hawaian system) to the 
mother's brothers and cousins as well, because it is uncertain 
who of them is the father, in our descriptive sense of the 
term. The same terms are applied to brothers and sisters 
and to paternal or maternal cousins or to both paternal and 
maternal cousins, because it is uncertain who are brothers 
or sisters and who are cousins. And so forth. The whole 
of this argument, however, is overthrown by the fact that 
the same term is also applied to the mother and to the 
mother's sisters and cousins or (in the Hawaian system) to 
the father's sisters and cousins as well, and that a woman 
applies the same terms to her sisters' and cousins' children 
as to her own sons and daughters. It is conceivable that 
uncertainty as regards fatherhood might have led a savage 
to call several men his father, but uncertainty as regards 
motherhood could never have led him to call several women 
his mother, or could never have led a woman to call other 
women's children her sons and daughters.1 
Dr. Rivers, however, thinks that there may be two 
answers to this objection. " It may be," he says, " that there 
was once a definite term for the individual relation between 
mother and child, and that the term became extended at a 
later stage of evolution so as to fall into line with other 
kinship terms." The other answer, which he considers to be 
more likely to be true, is that, in such a state of society as 
that we must assume, when the system of relationships was 
in process of development, the special relationship between 
mother and child would hardly have persisted beyond the 
time of weaning. The separation would then have occurred 
1 Cf. Darwin, Descent of Man, ii. 391 ; McLennan, Studies in 
Ancient History, p. 259; Macdonald, Oceania, p. 188 ; Thomas, 
op. cit. p. 123. 268 
THE HISTORY OF HUMAN MARRIAGE chap. 
before the age at which the child began to learn the terms of 
relationship to any great extent; and "it is even possible 
that in this early stage of culture the duty of suckling may 
have been shared by other women of the group, and that 
at the time of weaning, the child might not have been in the 
position to differentiate between its own mother and the 
other child-bearing women of the group."1 The first answer 
is merely a guess, and the second is almost worse. Where 
in the world has a society been found in which it is the 
custom for infants to be taken away from their mothers 
when they are weaned, or for mothers to desert their infants ? 
And even if it were worth while inventing such a society in 
the interest of the cherished institution of group-marriage, 
it would be impossible to find mothers who were equally 
ignorant of their children as the children were of their 
mothers.2 
Sir James Frazer, again, thinks that group-marriage may 
be saved without the support of Dr. Rivers' suggestions. 
He says that the difficulty of understanding how a person 
should ever come to be treated as the child of many mothers 
" only exists so long as we confuse our word ' mother ' with 
the corresponding but by no means equivalent terms in the 
languages of savages who have the classificatory system. 
We mean by ' mother ' a woman who has given birth to a 
child ; the Australian savages mean by ' mother ' a woman 
who stands in a certain social relation to a group of men 
and women, whether she has given birth to any one of them 
or not. She is * mother ' to that group even when she is an 
infant in arms."3 But if the term for " mother " has nothing 
to do with the idea of consanguinity, why should the term 
for " father " have anything to do with it ? If the former is 
applied to a woman who is known not to have given birth 
to the child, why should the latter term be applied to a 
man because he may possibly have begotten it ? Frazer's 
explanation is, in fact, an admission that the classificatory 
terms are not indications of blood-ties; he says himself 
1 Rivers, in Anthropological Essays presented to E. BSTylor, p. 317. 
2 Cf. Lang, in Proceed. British Academy, 1907-1908, p. 140. 
3 Frazer, Totemism and Exogamy, i. 304. vii CRITICISM OF HYPOTHESIS OF PROMISCUITY 
269 
that " the classificatory system is based on the marital, 
not on the parental, relation."1 
Much stress has been laid upon the fact that the Australian 
savage "applies the same term to the woman or women whom 
he actually marries and to all the women whom he might 
lawfully marry, that is, who " belong to the right group."2 
This has been interpreted as a consequence of earlier group- 
marriage ; and the same is the case with the more general 
fact that in the classificatory systems of many peoples in 
different parts of the world the wife's sisters and the brothers' 
wives (male speaking) are classed in nomenclature with the 
wife, and reciprocally, the sisters' husbands (woman speak- 
ing) and husband's brothers are classed with the husband. 
But Frazer's warning with reference to the classificatory 
term for " mother " applies with equal strength to the 
classificatory term for " wife " we must not confuse our 
word " wife " with " the corresponding but by no means 
equivalent terms in the languages of savages who have the 
classificatory system." When they use the same term for a 
wife, in our sense of the word, and for certain other women 
as well, they do so, not on account of earlier group-marriage, 
but simply because the women who may be a man's wives 
and those who cannot possibly be so stand in a widely 
different relation to him. This circumstance, to which I 
shall revert in the chapter on group-marriage, corroborates 
the explanation I have given above of the co-existence of 
certain classificatory terras and cross-cousin marriage. And 
in the present case Dr. Rivers seems on the whole to agree 
with me.3 
Dr. Rivers' attitude towards the dependence of the 
classificatory system upon group-marriage or sexual com- 
munism is somewhat different from that of most other anthro- 
pologists who believe in such a dependence. As we have 
already noticed, he has expressed the view that the features 
of the classificatory system have arisen out of a state of 
1 Ibid. i. 305. 
2 Spencer and Gillen, Native Tribes of Central Australia, p. 57. 
3 Rivers, History of Melanesian Society, ii. 128 sq. I had expressed 
this view already in the second edition of the present work (p. 56). 270 
THE HISTORY OF HUMAN MARRIAGE chap. 
group-marriage, and maintains still, at any rate, that this 
system has several features which would follow naturally 
from a condition of sexual communism. But in spite of his 
suggestion as regards the origin of the classificatory terms 
for " mother," he says that the view which explains the 
classificatory system as the result of group-marriage implies 
that this system was in its origin expressive of status rather 
than of consanguinity and affinity : " the terms would stand 
for certain relations within the group to which only the 
vaguest ideas of consanguinity need have been attached."1 
And Sir James Frazer thinks that he is probably right in this 
contention.2 Then we come back to our old problem how 
far we may infer the earlier existence of certain social con- 
ditions from certain terms of relationship. We have seen 
how dangerous any such inference is even if the terms of 
relationship are regularly found together with some special 
institution. But in the present case there is no such regular 
coincidence. Whilst the classificatory system is extremely 
common, " group-marriage " or sexual communism is com- 
paratively very rare. Dr. Rivers says that he has evidence 
from Melanesia " which places beyond question the former 
presence of such a condition, with features of culture which 
become readily explicable if they be the survivals of such a 
state of sexual communism as is suggested by the terminology 
of the classificatory system."3 Other anthropologists have 
in Australia found evidence of the dependence of the classi- 
ficatory nomenclature upon " group-marriage " still in 
existence. But, as I shall show in another place, neither 
the Melanesian nor the Australian terminology corresponds 
with those particular forms of " group-marriage," or sexual 
communism, which exist or are supposed to have existed 
among the peoples who have the terminology.4 Nay, even 
if they did correspond, it would be a great error of method to 
conclude that the classificatory system has everywhere 
originated in " group-marriage " or sexual communism. 
1 Idem, in Anthropological Essays presented to E. B. Tylor, 
PP- 319, 322. 
2 Frazer, op. cit. i. 307. 
3 Rivers, Kinship and Social Organisation, p. 86. 
4 See infra, on Group-marriage and other Group-relations. VII CRITICISM OF HYPOTHESIS OF PROMISCUITY 
271 
It would not be legitimate even according to the principles 
laid down by Dr. Rivers, as this is certainly one of th@se 
cases in which " differences of culture or the absence of 
intermediate links make it unjustifiable to infer the ancient 
existence of the forms of marriage from which features of 
terminology might be derived."1 Yet he has himself asserted 
that the classificatory system has probably had its origin 
in some universal, or almost universal, stage of social develop- 
ment characterised by group-marriage ;2 although he has 
subsequently somewhat modified this opinion.3 
For the rest, I entirely fail to understand how the classi- 
ficatory system can be regarded as the result of sexual 
communism if, as Dr. Rivers maintains,4 the classificatory 
terms for husband and wife lend no support to the hypo- 
thesis that they are survivals of such communism, but are 
more naturally explained as the result of the status of certain 
men and women as potential husbands and wives. What 
other terms, then, could be expected to tell us that a 
certain group of men have, or have had, the right of access 
to a certain group cf women ? The conclusion that the 
classificatory system is caused by group-marriage or sexual 
communism is not justified if the terms stand for certain 
social relations to which " only the vaguest ideas of con- 
sanguinity " have been attached. They must have expressed 
the degree and kind of blood-relationship as definitely as the 
fatherhood of individuals could be known. Although 
fundamentally influenced by social relations, as already said, 
they probably from the beginning expressed in a general way 
the idea of consanguinity as well, being in the first place 
used as terms for kindred ; but it is simply impossible to 
believe that they were ever meant to be " descriptive " of 
one definite kind of relationship-fatherhood, motherhood, 
and so forth-either actual or hypothetical. 
It has been argued that if the classificatory terms do not 
denote definite relationships, then those who use them have 
1 Rivers, Kinship and Social Organisation, p. 59. 
2 Idem, in Anthropological Essays presented to E. B. Tylor, p. 323. 
3 See supra, p. 241. 
4 Rivers, History of Melanesian Society, ii. 129. 272 
THE HISTORY OF HUMAN MARRIAGE chap. 
no terms at all for such relationships.1 This argument has 
no weight. The English use classificatory terms for all 
except the nearest degrees of relationship, but if they want 
to define any particular relationship they have always 
means of doing so. And so have savages. Mr. Washington 
Matthews writes of the Hidatsa Indians, belonging to the 
Dakota:-" When they wish to distinguish between an 
actual father and a father's brothers, they use the adjective 
ka'ti, true, real, in speaking of the former. . . . There is a 
special word to designate the real mother, although she is 
commonly called by the terms which apply as well to her 
sisters. There are two names for wife ; one for a wife by 
actual marriage, the other for an actual wife as well as what 
might be called a potential wife, i.e., a wife's sisters. . . . 
It must not be supposed, from the wide significance of some 
of their terms, that they do not discriminate between all 
grades and conditions of kinship. When they have no single 
word to define the relationship, they employ two or more 
words."2 In Mota, one of the Banks Islands, according to 
Dr. Codrington, the terms tamai and veve are used for 
father and mother respectively, but are also applied " to all 
of the same generation with the parents who are ' near ' 
and belong to the family connection." Yet this wide use 
of the terms " does not at all signify any looseness in the 
actual view of proper paternity and maternity; they are 
content with one word for father and uncle, for mother and 
aunt, when the special relation of the kinship of the mother's 
brother does not come in ; but the one who speaks has no 
confusion as to paternity in his mind, and will correct a 
misconception with the explanation, ' my own child, tur 
natuk ; his real father, tur tamana ; tur tasina, his brother 
not his cousin.' "3 The Maori, who apply the term papa 
to the father, the uncles, and the sons of the parent's uncles, 
have also the term papara, denoting the real father, although 
1 Fison, ' Classificatory System of Relationship,' in Jour. Anthr. 
Inst. xxiv. 369. 
2 Matthews, Ethnography and Philology of the Hidatsa Indians, 
P- 57- 
3 Codrington, Melanesians, p. 36 sq. vn CRITICISM OF HYPOTHESIS OF PROMISCUITY 
273 
it is not often heard.1 In the Kariera tribe, according to Mr. 
Brown, each term has what we may call a primary or specific 
meaning, which corresponds very closely with the use of 
relationship in English. Thus, although a given person 
applies the name mama to a large number of individuals, 
if he is asked " Who is your mama ? " he immediately replies 
by giving the name of his actual father, unless his own father 
died during his infancy, in which case he gives the name of 
his foster father, a man's nearest relative of this kind being 
not necessarily the man who gave him birth, but the man 
under whose care he lived as a child.2 Messrs. Spencer and 
Gillen emphatically assert that among the Central Australian 
natives described by them there is no " special term applied 
to the special wife, apart from the general one given in com- 
mon to her and other women of her group whom it is lawful 
for a man to marry and outside of whom he may not marry."3 
But Strehlow says that whilst the Arunta apply the term noa 
both to an actual wife and to a potential wife, they use the 
term nd at j a (contraction of noa and atja) for the former only ;4 
and the Finke River natives, according to Schulze, call the 
actual wife noa iltja.5 
Finally I beg to recommend to the consideration of those 
who maintain that the classificatory system of the Australian 
savages has originated in group-marriage the interesting 
statement made by Spencer and Gillen and other writers, 
that in many of their tribes procreation is not known' to be 
the result of sexual intercourse.6 If the same ignorance 
prevailed in those days when their classificatory terms were 
formed, it is obvious that the terms for the father and the 
men classed with him and, generally, all terms of relationship 
1 Best, loo. cit. p. 28. 
2 Brown, in Jour. Roy. Anthr. Inst, xliii. 150. See also ibid. 
P- I57* 
3 Spencer and Gillen, Native Tribes of Central Austra.ia, pp. 95, 
96, 106. lidem, Northern Tribes of Central Australia, p. 140. 
4 Strehlow, Die Aranda- und Loritja-Stdmme in Zentral-Australien, 
vol. iv. pt. i. 65. 
5 Schulze, ' Aborigines of the Upper and Middle Finke River,' in 
Trans. Roy. Soc. South Australia, xiv. 224. See Eylmann, Die 
Eingeborenen d. Kolonie Sudaustralien, p. 171. 6 Seeinfra.i. 288 sqq. 
VOL. I T 274 
THE HISTORY OF HUMAN MARRIAGE ch. vn 
through a male, cannot possibly have any reference to con- 
sanguinity, either actual or hypothetical. The theory of 
group-marriage is thereby deprived of the last shadow of 
support which it could conceivably derive from the classi- 
ficatory terms used for blood-relatives. 
If the classificatory system, for the reasons which I have 
set forth in this chapter, is no indication of earlier group- 
marriage or sexual communism, it is no more an indication 
of a " consanguine family " or a previous state of absolute 
promiscuity. But even if it really were so-contrary to all 
our arguments-that would not prove that such a state ever 
has been universal in the social development of mankind ; 
for we do not know that the classificatory system has pre- 
vailed among all the races of the world, although it has been 
suggested that it has.1 Nor do we know that it is the earliest 
system of nomenclature among those peoples who have it. 
On the contrary, if man originally lived in families or small 
family groups, the classificatory system could hardly have 
been primitive, but would only have emerged after the 
families had expanded into larger bodies. 
It has recently been vindicated with great fervour that 
" systems of relationship furnish us with a most valuable 
instrument in studying the history of social institutions," 
including the institution of marriage, and that the study of 
those systems " is essential for advance in our knowledge 
of prehistoric sociology." 2 For my own part, however, I 
cannot help thinking that the endeavour hitherto made to 
use the classificatory terms of relationship as a means of 
disclosing the secrets of ancient marriage customs has, in 
spite of all the labour and ingenuity bestowed upon it, been 
a source of error rather than knowledge. 
1 E.g., by Rivers, in Anthropological Essays presented to E. B. 
Tylor, p. 310. 
2 Idem, Kinship and Social Organisation, p. 1, and passim. 
Idem, History of Melanesian Society, i. 3, and passim. CHAPTER VIII 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : 
MOTHER-RIGHT 
In a work of prodigious learning, published in 1861,1 the 
Swiss jurist, Dr. Bachofen, directed attention to the fact that 
a system of M utter? echt, " mother-right," prevailed among 
several ancient peoples. Moreover, partly from actual 
statements of old writers, partly from traditions and myths, 
he drew the conclusion that such a system everywhere 
preceded the rise of " father-right." A few years later, 
though quite independently of him, McLennan set forth 
the same hypothesis, being led to it chiefly by his studies 
in modem ethnology. But whilst Bachofen explained the 
mother-right as a consequence of the supremacy of women, 
McLennan regarded what he called " kinship through females 
only " as due to uncertain paternity resulting from early 
promiscuity or polyandry. " It is inconceivable," he says, 
" that anything but the want of certainty on that point 
could have long prevented the acknowledgment of kinship 
through males ; and in such cases we shall be able to con- 
clude that such certainty has formerly been wanting-that 
more or less promiscuous intercourse between the sexes has 
formerly prevailed. The connection between these two 
things-uncertain paternity and kinship through females 
only, seems so necessary-that of cause and effect--that 
1 Bachofen, Das Mutterrecht. 
275 276 
THE HISTORY OF HUMAN MARRIAGE chap. 
we may confidently infer the one where we find the other."1 
This opinion has been shared by various later writers. 
Dr. Bloch, for instance, who is one of the latest, says that 
mother-right " was the typical expression of the uncertainty 
of paternity which resulted from sexual promiscuity."2 
The term "mother-right " has in recent years been adopted 
by English anthropologists, and there can be no harm in 
using it in the present connection if it is clearly understood 
that it is not identical with " matriarchy," or mother-rule. 
It always implies that descent is reckoned exclusively 
through the mother ; a person belongs by birth to the social 
group-the clan or whatever it may be-of his mother, 
not to that of his fa'"her, and this, of course, carries with it 
specific rights and duties. Inheritance and succession, 
however, do not always follow the line of descent. Where 
descent is matrilineal, a man's nearest heir is certainly very 
often his brother by the same mother or his mother's sister's 
son, not his own son ; but there are also cases of mother-right 
in which property and rank are transmitted from father to son. 
Among various peoples with mother-right the mother's 
brother is said to have greater rights over a child than the 
father, or the latter's authority is said to be very slight or 
even nil; but, as I have pointed out before, it is necessary 
to receive with some caution statements which attribute 
unqualified power to the maternal uncle to the exclusion 
of the father.3 There can be no doubt that among most 
peoples with matrilineal descent the father is distinctly 
the head of the family.4 On the other hand, there are also 
peoples with patrilineal descent among whom the maternal 
uncle plays a prominent part. This is, for instance, the case 
in various parts of India, where a special bond between 
maternal uncle and nephew is to be found, particularly 
in the part taken by the former in ceremonial connected with 
his nephew.5 Among the Koita of British New Guinea, 
1 McLennan, Studies in Ancient History, p. 88. 
2 Bloch, Sexual Life of Our Time, p. 189. 
3 Supra, p. 40 sqq. 4 See supra, p. 43 sq. 
See Rivers, ' Marriage of Cousins in India,' in Jour. Roy. Asiatic 
Soc. 1907, p. 611 sqq. vin CRITICISM OF HYPOTHESIS OF PROMISCUITY 
277 
though the descent is reckoned through the father, a boy 
helps his maternal uncle in his garden work, and a maternal 
uncle is supposed to exert more authority than the father or 
paternal uncle; a man or boy might refuse to accompany 
his father or paternal uncle, but would always go with his 
maternal uncle without demur.1 Among the patrilineal 
Thonga in South-East Africa, although " father's right is 
paramount," the uterine nephew is all through his career 
the object of special care on the part of his uncle. When 
weaned, he goes to stay in the village of his mother's rela- 
tives for many years, and a girl sometimes stays there till 
she is grown up ; and when she marries, the maternal uncle 
claims a right on the total amount of the bride price.2 
Frequently matrilineal descent is combined with matrilocal 
marriage, which implies that the married couple live with 
the wife's people ; but in very many cases the marriage is 
patrilocal although the descent is matrilineal. 
Thus the only permanent feature of " mother-right," as 
I understand it, is that descent is reckoned through the 
mother, not through the father ; and as I have said before, 
there is no reason whatever to assume that all the peoples 
who now have matrilineal descent in earlier times also had 
the other customs which are often, but by no means invari- 
ably, combined with it. Nor must we take for granted that 
a certain prominence given to the maternal uncle by a 
people with patrilineal descent is an indication of the former 
prevalence of maternal descent. The reckoning of descent 
either through the father or through the mother does not 
imply that the nearest blood-ties on the other side are 
ignored. 
Mother-right is very widespread in the savage world3- 
1 Seligman, Melanesians of British New Guinea, pp. 49, 67. 
2 Junod, Life of a South African Tribe, i. 253. 
3 See, besides the works of Bachofen and McLennan, Avebury, 
Origin of Civilisation, p. 121 sqq. ; Giraud-Teulon, Les origines du 
mariage et de la famille, chs. vii.-x.; Idem, La m^re chez certains 
peuples de l'antiquite, passim; Bastian, Die Rechtsverhaltnisse bei 
verschiedenen Vblhern der Erde, p. 183 sq. ; Lippert, Die Geschichte 
der Familie, sec. i. ; Idem, Kulturgeschichte der Menschheit, vol. ii. 
ch. ii.; Dargun, Mutterrecht und Vaterrecht, passim; Post, Die 278 
THE HISTORY OF HUMAN MARRIAGE chap. 
and so is father-right j1 whilst a large number of low tribes 
may be said to be neither matrilineal nor patrilineal or, 
rather, both at the same time.2 In many areas either father- 
right or mother-right predominates or prevails exclusively ; 
but it also happens quite frequently that of two neighbouring 
tribes of the same race and on the same level of civilisation 
one has father-right and the other mother-right. Many 
later writers have followed Bachofen and McLennan in 
assuming that mother-right has everywhere preceded father- 
right ;3 and Dr. Hartland, as said above, has even declared 
Geschlechtsgenossenschaft der Urzeit, p. 93 sq. ; Idem, Der Ur sprung 
des Rechts, p. 37 sq. ; Idem, Bausteine fur eine allgemeine Rechts- 
wissenschaft, i. 77 sq. ; Idem, Grundriss der ethnologischen Juris- 
prudenz, i. 71 sqq. ; Starcke, Primitive Family, sec. i. chs. i.-v. ; 
Wilken, ' Over de primitieve vormen van het huwelijk en den 
oorsprong van het gezin,' in De Indische Gids, 1881, vol. ii. p. 244 
sqq. ; Friedrichs, ' Ueber den Ursprung des Matriarchats,' in 
Zeitschr. f. vergl. Rechtswiss. viii. 382 sq. ; Letourneau, Devolution du 
mariage et de la famille, chs. xvi.-xviii. ; Wake, Development of 
Marriage and Kinship, ch. viii. sq. ; Potter, Sohrab and Rustem, 
p. 113 sqq. ; Gasquoine Hartley (Mrs. Gallichan), Position of Woman 
in Primitive Society, p. 103 sqq ; Hartland, Primitive Paternity, 
ch. iv. sq. ; Frazer, Totemism and Exogamy, passim; Rivers, 
' Mother-right,' in Hastings, Encyclopaedia of Religion and Ethics, 
viii. 852 sqq. 
1 When I wrote the first edition of the present work the prevalence 
of mother-right was so much emphasised by most of the leading 
anthropologists of the time that I thought it desirable to give a list 
of uncivilised peoples among whom descent is reckoned through the 
father or property or rank is transmitted from father to son (p. 98 
sqq.}. As the general distribution of father- and mother-right does 
not affect the question I am here discussing, I have not included 
any such list in the present edition. 
2 Cf. Swanton, review of Frazer's Lectures on the Early History of 
Kingship, in American Anthropologist, N.S. viii. 160 ; Rivers, in 
Hastings, op. cit. viii. 852. 
3 Haberlandt, Ethnology, p. 52. Wilutzky, Vorgeschichte des 
Rechts, i. 23. Bloch, op. cit. p. 194. W. I. Thomas, Sex and Society, 
p. 67. Wundt, Elemente der Vblkerpsychologie, p. 172 (English trans- 
lation, p. 173). Kohler, ' Rechtsphilosophie und Universalrechtsge- 
schichte,' in Holtzendorff, Enzyklopddie der Rechtswissenschaft, i. 27. 
Russell, Tribes and Castes of the Central Provinces of India, i. 139. 
Andrew Lang says (Social Origins, p. 21) he is inclined to the 
opinion that the reckoning of descent through the woman is the 
more archaic method. vin CRITICISM OF HYPOTHESIS OF PROMISCUITY 
279 
that the result of anthropological investigations during the 
past half-century has been to show that this assumption 
is correct.1 My own impression is that the assumption of 
a universal stage of mother-right, against which I fought 
in the previous editions of this work, is at present rather 
losing than gaining ground,2 owing partly to our extended 
knowledge of facts and partly to a more general reluctance 
to unmethodical conclusions in anthropology as in other 
sciences. 
Dr. Hartland writes :-" It may be said without fear of 
contradiction that while no case is known where matrilineal 
reckoning betrays evidence of having been preceded by 
paternal descent, the converse has been observed in every 
part of the world. Cases may exist of tribes reckoning 
descent through the father in which no trace remains of 
reckoning through the mother. The mere existence of such 
cases is wholly insufficient to disprove a prior stage of 
motherright, or even to shift the burden of proof."3 I 
should have thought that the burden of proof lies on him 
who asserts the former existence of something which does 
not exist at present, and not on him who professes ignorance 
as regards the past; otherwise any fiction might claim to be 
1 Hartland, Primitive Paternity, i. 256 sq. 
2 See, e.g., Crawley, Mystic Rose, p. 460 sq.; Schmidt, Der 
Ursprung der Gottesidee, p. 182 ; Idem, review of Buschan's Illus- 
trierte Vblkerkunde, vol. i., in Anthropos, v. 586 ; Swanton, ' Social 
Organization of American Tribes,' in American Anthropologist, 
N.S. vii. 664 sqq. ; Idem, ibid. N.S. viii. 160 ; Idem, review of 
Thomas' Kinship Organisations and Group Marriage in Australia, 
ibid. N.S. ix. 743 ; Goldenweiser, review of Hartland's Primitive 
Paternity, ibid. N.S. xiii. 603. Mr. Horatio Hale (Science, xix. 30) 
thinks that in North America father-right and mother-right are both 
primitive; and M. van Gennep (Mythes et legendes d'Australie, 
p. xxxii.) believes that the same is the case in Australia. Mr. N. W. 
Thomas (Kinship Organisations and Group Marriage in Australia, 
p. 15) admits that " patrilineal descent may have been directly 
evolved without the intermediate stage of reckoning through females, " 
although he seems to consider it more probable that it has every- 
where been preceded by female descent. 
3 Hartland, op. cit. ii. 3. See also Idem, ' Matrilineal Kinship, 
and the Question of its Priority,' in Memoirs of the American Anthro- 
pological Association, iv. 87. 280 
THE HISTORY OF HUMAN MARRIAGE chap. 
true if it cannot be actually disproved. As regards the 
statement that cases where father-right betrays evidence of 
having been preceded by mother-right have been observed 
in every part of the world, I must confess that, although 
there are such cases, I only too often find the so-called 
evidence to consist of more or less arbitrary inferences drawn 
from customs which are interpreted as relics of earlier 
mother-right. And although it may be true that no con- 
clusive evidence has been produced of a transition in the 
opposite direction, American ethnologists maintain that 
cases exist in which such a transition has taken place or in 
which a clan system with matrilineal descent has evolved 
where there was previously no clan organisation at all. 
Thus the Kwakiutl present us, according to Dr. Boas, 
with a tribe who had formerly a purely paternal organisation, 
but under the influence of matrilineal neighbours have 
adopted a purely female law of descent, although only through 
the medium of the husband : " the woman brings as a dower 
her father's position and privileges to her husband, who, how- 
ever, is not allowed to use them himself but acquires them for 
the use of his son."1 Dr. Swanton writes that " instead of 
being primitive, a study of the north Pacific area convinces 
one that the maternal clan system is itself evolved, for there 
is every indication that it there grew up in one small area 
at the mouths of the Nass and Skeena rivers and was spread- 
ing northward, southward, and inland at the time these 
tribes first came to the notice of Europeans. That an evolu- 
tion has taken place in the Southwest is indicated by 
Fewkes' study of Hopi clans, as well as everything that we 
have learned of the relation of Navaho culture to that of the 
Pueblos."2 
If mother-right had everywhere preceded father-right, 
we might expect to find it particularly prevalent among 
the lowest savages. But this is not the case. Father 
Schmidt observes that the Pygmy tribes, so far as we have 
1 Boas, ' Social Organization and the Secret Societies of the 
Kwakiutl Indians,' in Ann. Rep. Smithsonian Institution, 1895, 
p. 334 sq. Cf. Rivers, in Hastings, op. cit. viii. 858 sq. 
2 Swanton, in American Anthropologist, N.S. vii. 670 sq. vin CRITICISM OF HYPOTHESIS OF PROMISCUITY 
281 
any information about their social organisation, seem to 
have father-right.1 Among the Auin, belonging to the 
Kalahari Bushmen, a man's property-which is never 
considerable-is inherited by his eldest son.2 Among the 
Pygmies of Central Africa children inherit their father.3 
Among the wild tribes of the Malay Peninsula I have found 
no trace of mother-right. " Les Sakayes," says de Morgan, 
" vivent a 1'etat patriarcal. Le groupe de la famille est 
soumis a 1'autorite du p£re de famille, qui a tous les droits 
sur ses enfants, sauf celui de les tuer."4 Among the Negritos 
of the Philippines, according to Schadenberg, family life 
is likewise patriarchal, and the father has unlimited power 
over the members of his family.5 Among the Kubus of 
South Sumatra a husband inherits his wife and vice versa, 
or the children inherit their parents.6 Among the Veddas 
Dr. Seligman found in some places matrilineal descent, and 
in others, although less often, patrilineal descent, or in the 
same place sometimes one and sometimes the other. " It 
seemed that adult sons inherited most of their fathers' 
personal property, but certainly the sons-in-law had the 
right to receive something."7 According to Dr. P. Sarasin, 
it is really not possible to speak either of father-right or 
mother-right among the Veddas, but if a choice has to be 
made the balance is more in favour of father-right.8 In 
1 Schmidt, op. cit. p. 182. Idem, Die Stellung dev Pygmaenvolker 
in der Entwicklungsgeschichte des Menschen, pp. 181, 187. 
2 Kaufmann, ' Die Auin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 155. 
3 Hutereau, Notes sur la vie familiale et juridique de quelques 
populations du Congo Beige, p. 6 sq. 
4 de Morgan, ' Moeurs, coutumes & langages des Negritos de 
I'int&ieur de la presqu'ile Malaise,' in Bull. Socidtd normande de 
Geographic, vii. 421. See also Idem, ' Negritos de la presqu'ile 
Malaise,' in L'homme, ii. 558. 
5 Schadenberg, ' Ueber die Negritos der Philippinen,' in Zeitschr. f. 
Ethnol. xii. 137. 
G Volz, ' Zur Kenntniss der Kubus in Sudsumatra,' in Archiv f. 
Anthrop. N.S. vii. 104. 
7 Seligman, Veddas, pp. 74, 76 sqq., 118. 
8 Sarasin, in Verhandlungen des II. Intern. Kongr. f. Allgem. 
Religionswiss. Basel, 1905, vol. ii. 134, quoted by Schmidt, Der 
Ursprung der Gottesidee, p. 182. 282 
THE HISTORY OF HUMAN MARRIAGE chap. 
Australia the area inhabited by tribes with a " dichotomous 
kinship organisation "-broadly speaking, the tribes of the 
whole of the known area of Australia, certain coast regions 
of comparatively small extent excepted-is, according to 
Mr. Thomas, divided almost equally between matrilineal 
and patrilineal tribes ; whilst in the small remainder of the 
continent the rule of descent is mostly matrilineal.1 
With reference to the Yahgans of Tierra del Fuego Mr. 
Bridges wrote to me :-" A child belongs equally to the clan 
of its father and mother as regards duty of revenge, but is 
always reckoned a member of the father's clan only. Children 
are generally named after their grandparents, paternal or 
maternal indifferently. They are quite as much attached 
to their mother's relatives, and these to them, as to their 
paternal relatives; the only difference is that they are in- 
tegral parts of the father's clan, not of the mother's." 
Speaking of the same people, M. Hyades remarks, " L'heri- 
tage se transmet a 1'epoux survivant, ou a defaut, au fils 
aine."2 In North America, according to Dr. Goldenweiser, 
we find no correlation between higher culture and tribes 
with patrilineal descent and between lower culture and 
matrilineal tribes, but the reverse is true. The two groups 
of tribes whose culture ranks among the highest of the 
continent, the Indians of the North-West Coast and the 
Iroquois-he might have added most of the Pueblo tribes-8 
count descent through the mother. The Eskimo, on the 
other hand, the North Athapascans (excepting those affected 
by the culture of the coast), the Interior and part of the 
Coast Salish, the tribes of Washington and Oregon, the 
Shoshoni-all tribes of a relatively low culture-are either 
patrilineal or reckon descent " in the modem civilised fashion 
through both parents."4 In North America mother-right 
is often connected with a comparatively high development 
of agriculture ; Dr. Swanton states that typical matrilineal 
1 N. W. Thomas, op. cit. p. 38 sq. 
2 Hyades, ' Ethnographic des Fuegiens,' in Bull. Soc. d'Anthr. 
Paris, ser. iii. vol. x. 334. 
3 Hodge, Handbook of American Indians, ii. 323. 
4 Goldenweiser, in American Anthropologist, N.S. xiii. 603. vni CRITICISM OF HYPOTHESIS OF PROMISCUITY 
283 
clan systems are found especially in the maize country.1 
And Dr. Rivers observes that in Africa mother-right seems 
to be present especially among peoples who live chiefly by 
agriculture, while father-right is associated with pastoral 
life, although the association is by no means universal.2 
Professor Hobhouse, Mr. Wheeler, and Mr. Ginsberg, again, 
by statistically investigating the prevalence of mother- and 
father-right among hunting, pastoral, and agricultural 
savages, came to the conclusion that the maternal principle 
predominates among the hunting peoples, the paternal in 
the pastoral stage, while among agricultural peoples the two 
are nearly balanced.3 The statement referring to hunters, 
however, is distinctly open to criticism. The large majority 
of " lower hunters " included in the list are Australian tribes, 
and of these Australian cases eighteen are represented as 
matrilineal and only eight as patrilineal. Under the latter 
heading there are no other cases at all, whilst the non- 
Australian matrilineal peoples are the Punans, Fuegians, 
Batua, Lushongo Batua, and Veddas.4 A comparison 
between these statements and those which I have given 
above shows a considerable discrepancy, which must sub- 
stantially affect the conclusions drawn from them. But 
there can be no doubt, I think, that Professor Hobhouse 
and his collaborators are right in their inference that the 
patrilineal system is much more common among pastoral 
than among agricultural savages. This fact certainly lends 
no support to the view that the matrilineal system is the 
more primitive of the two systems. 
The prevalence of mother-right among a large number 
of simple peoples cannot, then, be regarded as evidence of 
a universal stage of promiscuity even on the assumption 
that mother-right is due to uncertain paternity. Nor 
would the accuracy of this assumption justify the conclusion 
that promiscuity was the primitive condition of those peoples 
1 Swanton, in American Anthropologist, N.S. vii. 671. 
2 Rivers, in Hastings, op. cit. viii. 857. 
3 Hobhouse, Wheeler, and Ginsberg, Material Culture and Social 
Institutions of the Simpler Peoples, p. 153. 
4 Ibid. p. 152. 284 
THE HISTORY OF HUMAN MARRIAGE chap. 
who have mother-right. Nay, it would not even prove that 
they ever lived in a state of promiscuity. The separation 
of husband and wife, adultery on the part of the woman, 
and the practice of lending wives to visitors are among 
many savages frequent enough to make good the saying, 
" It is a wise child that knows his own father." Mr. Ingham 
wrote to me from Banza Manteka that the Bakongo, who 
reckon descent through the mother, assert uncertain 
paternity as a reason for this custom ; 1 but nevertheless 
he says that they would be horrified at the idea of promis- 
cuous intercourse. Uncertainty of fatherhood is the cause 
which has been assigned for the mother-right of various 
other peoples either by the natives themselves or by those 
who have written on their customs, in spite of the fact that 
marriage does exist among them.2 With reference to the 
Alibamu, for example, Bossu writes that " all the Indians 
deduce their lineage from the women, alledging that they 
may be sure of their origin on that side, and accordingly of 
their having their mother's blood in their veins; but that 
claiming their descent from the men was uncertain."3 
A similar explanation has been given by old travellers of 
the rules of inheritance among the Nayars,4 and of the 
succession to the throne on the Coast of Malabar;5 and in 
these cases it might have been justified by the extreme 
looseness of the sexual relations of the Nayars and by the 
intercourse which Brahmans had with the queens of Malabar. 
No one, however, has yet exhibited any general coincidence 
of what we consider moral and immoral habits with the 
1 Cf. Phillips, ' Lower Congo,' in Jour. Anthr. Inst. xvii. 229. 
2 Bosman, ' Description of the Coast of Guinea,' in Pinkerton, 
Collection of Voyages and Travels, xvi. 421. Grade, in Aus alien 
Welttheilen, xx. 5. Solinus, Collectanea rerum memorabilium, xxx. 
2 sq. Jarves, History of the Hawaiian Islands, p. 18. Powell, 
Wanderings in a Wild Country, p. 60. 
3 Bossu, Travels through Louisiana, i. 232. 
4 Schouten, Ost-Indische Reyse, p. 168. 
5 Balbi, Viaggio del! Indie Orientali, foil. 75 b, 76 a, 137 a. 
Baldaeus, ' Description of the East-India Coasts of Malabar and 
Coromandel,' in Churchill, Collection of Voyages and Travels, iii. 
561 sq. Vin CRITICISM OF HYPOTHESIS OF PROMISCUITY 
285 
prevalence of father- and mother-right among existing 
savages. Among the Barea1 and the Negroes of Loango,2 
inheritance goes through the mother only, although adultery 
is said to be extremely rare ; whereas father-right prevails 
or preponderates among the Chukchee3 and the Todas of 
Southern India,4 although paternity could nowhere be more 
uncertain than among these peoples. The former have cicis- 
beism and group-relations,5 and among the latter there is 
little restriction of any kind on sexual intercourse either 
before or after marriage.6 Other instances have been given 
by Dr. Hartland, who shares my view, expressed in the 
earlier editions of this work, that we have no right to 
ascribe mother-right to uncertain paternity, " nay, that 
such an assumption is not even probably true." He 
writes :-" Motherright then is found not merely where 
paternity is uncertain, but also where it is practically certain. 
Fatherright on the other hand is found not merely where 
paternity is certain, but also where it is uncertain and even 
where the legal father is known not to have begotten the 
children. Nay, the institutions of fatherright often require 
provision for, and very generally permit, the procreation by 
other men of children for the nominal father.7 It follow- 
therefore that the uncertainty of paternity cannot be 
historically the reason for the reckoning of descent ex- 
clusively through the mother."8 
If mother-right has not originated in uncertainty of 
fatherhood, how is it to be explained ? One answer might 
perhaps be that it is due to ignorance of the fact that the 
birth of a child is the result of sexual intercourse between 
man and woman. Paternity, as Sir Henry Maine remarks, 
is " matter of inference, as opposed to maternity, which is 
1 Munzinger, Ostafrikanische Studien, pp. 484, 490. 
2 Proyart, ' History of Loango,' in Pinkerton, op. cit. xvi. 571. 
3 Bogoras, Chukchee, p. 537 sq. 
4 Rivers, Todas, p. 546 sq. 
6 See infra, on Polyandry and on Group-marriage and other 
Group-relations. 
0 Rivers, Todas, p. 529 sqq. 
1 On this subject see infra, on Polyandry. 
8 Hartland, Primitive Paternity, i. 301 sqq., particularly p. 325. 286 
THE HISTORY OF HUMAN MARRIAGE chap. 
matter of observation.''1 Hence it is almost beyond doubt 
that the father's participation in parentage was not recog- 
nised as soon as the mother's.2 When I first dealt with 
the subject of mother-right, I was anxious to learn whether 
savages might still be found who preserved this primitive 
ignorance, and I made an inquiry about it in letters addressed 
to more than a hundred gentlemen, mostly missionaries, 
living in different parts of the world. I received only a few 
answers, and none of them bore testimony to ignorance of 
fatherhood. In reply to my question whether the natives of 
Tierra del Fuego believe that a child descends exclusively 
or predominantly from either of the parents, Mr. Bridges 
wrote that, according to his idea, the Yahgans " consider 
the maternal tie much more important than the paternal, 
and the duties connected with it of mutual help, defence, 
and vengeance are held very sacred." It is doubtful, 
however, whether this refers to the mere physiological 
connection between the child and its parents, in spite of 
the statement made by the same authority which I have 
quoted above, that a Yahgan child is always reckoned 
a member of the father's clan only. Dr. Sims informed me 
from Stanley Pool on the Congo that among the Bateke the 
1 Maine, Dissertations on Early Law and Custom, p. 202. 
2 CJ. Lippert, Die Geschichte der Familie, pp. 5, 8, 9, &c. I have 
preserved the statement in the text from the earlier editions of this 
work, as it contains nothing about the time when the discovery of 
a connection between child-birth and sexual intercourse may 
have been made. If our early ancestors had a yearly pairing season, 
the regular appearance of pregnancy after a definite interval may, 
together with a natural association of child-birth with sexual inter- 
course on anatomical grounds, have led to the recognition of a 
causal connection between them. Mr. Heape asks (Sex Antagonism, 
p. 87) if it is to be believed that the women did not, year after 
year, associate impregnation, when confined to certain times, with 
the symptoms they experienced a month later ; and he thinks 
that even many of the lower animals anticipate the birth of their 
young and many nesting birds the result of mating. In his opinion 
" it is undoubtedly much simpler to connect the two facts of sexual 
congress and the birth of young than it is to invent any other theory 
whatever " (ibid. p. 92). See also Prof. Carveth Read's article, 
' No Paternity,' in Jour. Roy. Anthr. Inst, xlviii. 146 sqq., which 
appeared after the present chapter was written. Vin CRITICISM OF HYPOTHESIS OF PROMISCUITY 
287 
function of both parents in generation is held alike important. 
Mr. A. J. Swann wrote that the Waguha of West Tanganyika 
also recognise the part taken by both; and the same was 
asserted by Archdeacon Hodgson concerning certain other 
tribes of Eastern Central Africa, among whom descent is 
matrilineal. Mr. Cousins, again, informed me that according 
to the idea of the Cis-Natalian Kafirs a child descends 
chiefly, though not exclusively, from the father. 
Certain customs found among a large number of savage 
peoples indicate plainly .that they are well aware of the 
broad fact that pregnancy or child-birth is caused by sexual 
intercourse. In some cases a marriage does not become 
definite until a child is born, in other cases sexual relations 
which lead to pregnancy or child-birth make marriage 
obligatory or render the seducer liable to punishment. The 
custom of the couvade, which compels the father to take 
certain precautions at the birth of the child, undoubtedly 
implies the idea that there is an intimate relationship 
between the two. There are also, in the ethnological 
literature, various statements which directly tell us that 
sexual intercourse is known to have something to do with the 
production of a child, however defective this knowledge 
often is. 
The tribes near Fort Johnston in British Central Africa 
believe that " the child is formed only from what the man 
introduces into the woman-the semen."1 In Fiji there is a 
widespread belief " that when a woman has been cohabiting 
with more than one man before conception the paternity of 
her child is shared equally by all her paramours."2 The 
Maori of the Tuhoe tribe, according to Mr. Best, " would 
appear to have grasped the idea of active and passive 
agents in generation," although they, like many other 
peoples, employed various methods to cause married 
women to conceive when they were apparently barren, 
such as the invocations of a priest or the use of the mana, 
or supernatural power, ascribed to certain trees or stones. 
1 Stannus, ' Notes on some Tribes of British Central Africa, 
in Jour. Roy. Anthr. Inst. xl. 310. 
2 Thomson, Fijians, p. 239. 288 
THE HISTORY OF HUMAN MARRIAGE chap. 
According to their ideas it is during a certain stage of 
menstruation and immediately after it that the sexual act is 
fruitful-a belief which may be connected with the view 
which an old native said was held by his ancestors and elders, 
that the marriage of man and wife was a matter of no moment, 
the moon being the real husband.1 Some North American 
Indians believed that children receive their soul from their 
father and their body from their mother.2 When a Brazilian 
Bakairi wants to inform a stranger that he is another person's 
father, he takes hold of the organ of procreation.3 The 
Coroados even are of opinion that a child is exclusively 
indebted to its father for its existence, the mother only 
preserving and taking care of it.4 A similar idea prevails 
among various Australian tribes ; Dr. Howitt repeatedly 
heard the remark that a woman is nothing but a nurse who 
takes care of a man's children for him.5 With reference to 
the natives of some part of Australia, not specially mentioned, 
Mr. Purcell states that after the third initiation into the 
bora ceremony " the youth is made to drink semen that is 
taken from six or as many young clean gins and blacks, 
as may be in the camp at the bora ground," and that the 
same is done to an old man who is dying. For " they hold, 
that as semen brought them into the world, it should keep 
them alive and from dying."6 
Very different views are held by many other Australian 
tribes. The Arunta believe that there were certain great 
1 Best, ' Lore of Whare-Kohanga,' in Jour. Polynesian Soo. 
xiv. 2ii, 214 ; xv. 4 sqq. 
2 Lahontan, New Voyages to North-America, p. 461 (Canadian 
Indians). Carver, Travels through the Interior Parts of North America, 
p. 378 (Dakota). 
3 von den Steinen, Unter den N aturvblkern Zentral-Brasiliens, 
p. 191. 
4 Teschauer, ' Die Caingang oder Coroados-Indianer im brasilia- 
nischen Staate Rio Grande do Sul,' in Anthropos, ix. 22. 
5 Howitt, ' Australian Group Relations,' in Smithsonian Report, 
1883, p. 813. Idem, Native Tribes of South-East Australia, pp. 195, 
198, 255, 263, 284. Cameron, ' Notes on some Tribes of New South 
Wales,' in Jour. Anthr. Soc. xiv. 352. 
0 Purcell, ' Rites and Customs of Australian Aborigines,' in 
Verhandl. Berliner Gesellsch. Anthrop. 1893, p. 288. vm CRITICISM OF HYPOTHESIS OF PROMISCUITY 
289 
men and women, leaders of the various groups in the far 
past times, called Alcheringa, who carried with them and 
deposited in certain places numbers of spirit children, and 
that it is these who now enter women and are bom in the 
shape of human beings, as well as the spirits of the great 
leaders themselves.1 The belief that children enter women in 
the form of minute spirits is widely spread over the Central, 
Northern, and Western parts of the continent, although 
there are variations in details.2 In North Queensland, 
according to Dr. Roth, it is believed that " infants may be 
fashioned by spirits and then inserted in the mother."3 
Some of the aborigines there, the Tully River blacks, say 
that a woman begets children because she has been sitting 
over the fire on which she has roasted a particular species of 
black bream, " which must ha've been given to her by the 
prospective father " ; or because she has purposely gone 
a hunting and caught a certain kind of bull-frog ; or because 
some man may have told her to be in an interesting con- 
dition ; or because she may dream of having had the child 
put inside her. But Dr. Roth adds that " although sexual 
connection as a cause of conception is not recognised among 
the Tully River blacks so far as they themselves are con- 
cerned, it is admitted as true for all animals:-indeed this 
idea confirms them in their belief of superiority over the 
brute creation."4 So also the Arunta and Loritja, according 
1 Spencer, Native Tribes of the Northern Territory of Australia, 
p. 264 sq. Spencer and Gillen, Northern Tribes of Central Australia, 
p. 150. 
2 Spencer, op. cit. pp. 23-25, 263 sq. Speacer and Gillen, Northern 
Tribes, p. 145 sqq. Schulze, ' Aborigines of the Upper and Middle 
Finke River,' in Trans, and Proceed. Roy. Soc. South Australia, 
xiv. 237. Strehlow, Die Aranda- und Loritja-Stamme in Zentral- 
Australien, i. 14 ; ii. 52 sqq. ; vol. iii. pt. i. 7 n. I. Mrs. Bates, 
quoted by Lang, ' Questiones Totemicae,' in Man, 1906, p. 180 
(West Australian natives). Mrs. Langloh Parker, Euahlayi Tribe, 
p. 50 sqq. Bischofs, ' Die Niol-Niol in Nordwest-Australien,' in 
Anthropos, iii. 35. See also Hartland, op. cit. i. 236 sqq. ; van 
Gennep, op. cit. p. xliv. sqq. 
3 Roth, North Queensland Ethnography : Bulletin No. 5. Super- 
stition, Magic, and Medicine, p. 23. 
4 Ibid. p. 22. 290 
THE HISTORY OF HUMAN' MARRIAGE chap. 
to Strehlow, are aware of the causal connection between 
copulation and procreation in the case of animals.1 And 
even in the case of human beings the ignorance of such a 
connection is not complete. Spencer an.d Gillen say that, 
according to the beliefs of the Arunta, Loritja, and Ilpirra 
tribes, and probably also of others, sexual intercourse 
" merely as it were prepares the mother for the reception 
and birth also of an already formed spirit child who inhabits 
one of the local totem centres."2 Strehlow agrees with them 
in this respect: the Arunta and Loritja believe that the 
sexual act puts the vagina in such a condition as to enable 
it to receive a ratapa, or spirit child, whereas without coitus 
it remains " closed."3 
Dr. Malinowski has found a similar belief among the 
natives of Kiriwina in Trobriand Islands (British New 
Guinea). Their state of knowledge, he says, " is just at 
the point where there is a vague idea as to some nexus 
between sexual connection and pregnancy, whereas there is 
no idea whatever concerning the man's contribution towards 
the new life which is being formed in the mother's body."4 
The real cause of pregnancy is always a baloma, or the soul 
of a dead person, who enters into the body of a woman or 
is inserted into it in the form of a waiwaia, or " spirit child," 
by another baloma.3 But a woman who is a virgin cannot 
conceive and give birth to a child, because nothing can 
enter or come, out of her. She must be opened up or pierced 
through ; hence a woman who has much intercourse will be 
easier for a spirit child to enter than one who is fairly con- 
tinent. Sexual intercourse is thus quite unnecessary except 
for its mechanical action, and in default of it any other means 
of widening the passage may be used to enable a baloma 
to enter the woman or to insert a waiwaia into her.6 In 
the case of animals, however, copulation is recognised 
as the cause of pregnancy, in Kiriwina as well as among 
1 Strehlow, op. cit. ii. 52 n. 7. 
2 Spencer and Gillen, Native Tribes, p. 265. 
3 Strehlow, op. cit. vol. hi. pt. i. p. x. sq. 
4 Malinowski, ' Baloma,' in Jour. Roy. Anthr. Inst. xlvi. 407. 
5 Ibid. p. 403 sqq. 11 Ibid. p. 412. vm CRITICISM OF HYPOTHESIS OF PROMISCUITY 
291 
Australian tribes ; Dr. Malinowski was told of the pigs, 
" They copulate, copulate, presently the female will give 
birth."1 Most of the Papuans of Kaiser Wilhelm Land are 
said to know that pregnancy is connected with sexual 
intercourse, although there are women among the Kai 
people who in earnest deny the existence of any such connec- 
tion. But the man is supposed to play only a secondary 
part in the matter. He merely liberates the child which is 
already formed in the woman's womb, by " tearing it off," 
just as when the rind is tom off a cocoa-nut.2 The western 
tribes of the Bakongo, according to Mr. Torday, explain 
the phenomenon of conception by saying that the soul of 
a deceased person enters the woman and is reincarnated 
when it is wetted with the semen virile.3 
We must not expect from savages too accurate explana- 
tions of a physiological process so complicated and hidden as 
that of conception. What do the uneducated classes among 
ourselves know about the subject, beyond the facts that 
sexual commerce is often followed by pregnancy, and that 
there is no pregnancy without previous sexual commerce ? 
Nor must we infer that people are ignorant of the ordinary 
course of events because they maintain that children may 
also be produced in other ways. Among the Northern 
Massim of British New Guinea, according to Mr. Bellamy, 
" intercourse is recognised as the cause of children, although 
single girls who become pregnant have a curious habit 
of blaming some or other portion of their diet " ; 4 and 
this may very well be something more than a bare excuse. 
Stories of " supernatural birth " are by no means incon- 
sistent with the belief that in normal cases children owe their 
origin to sexual relations between men and women, nor do 
they prove the former absence of such a belief. Generally 
1 Ibid. p. 411. 
2 Neuhauss, Deutsch Neu-Guinea, i. 149; iii. 26. 
3 Torday and Joyce, Les Bushongo, p. m. According to Mr, 
Weeks {Among the Primitive Bakongo, p. 115), the Bakongo believe 
that " the only new thing about an infant is its body," the spirit 
or soul of the child being thought to have belonged either to a 
deceased person or to a living person or to a water-spirit. 
4 Seligman, Melanesians of British New Guinea, p. 704. 292 
THE HISTORY OF HUMAN MARRIAGE chap. 
speaking, I do not think that existing ideas about reproduc- 
tion take us back to a period when man was completely 
ignorant of the connection between coitus and pregnancy. 
Dr. Hartland maintains that not only ancient stories but 
superstitious practices still very frequently resorted to even 
at higher stages of culture for the purpose of getting 
offspring justify the conclusion that " it was a widespread 
belief in early times that pregnancy was caused otherwise 
than by sexual intercourse " ; 1 but, so far as I can see, 
these practices-bathing, the drinking of water, the eating 
of eggs, &c?-only show that sexual intercourse alone is 
often considered insufficient to secure fecundity.2 
Dr. Malinowski, again, in his brilliant essay, attaches some 
importance to the fact that the natives of Kiriwina belong 
to the most advanced Papuo-Melanesians : their ignorance, 
he says, " seems to indicate a much wider range of distri- 
bution and a much greater permanence through the higher 
stages of development than could be assumed hitherto. ' '3 He 
explains himself, however, this ignorance by the fact that 
" for the native of Kiriwina sexual intercourse is almost 
as common an occurrence as eating, drinking, or sleeping. 
What is there to guide the native observation, to draw his 
attention to the nexus between a perfectly normal, everyday 
occurrence, on the one hand, and an exceptional, singular 
event on the other ? How is he to realise that the very 
act which a woman performs almost as often as eating or 
drinking will, once, twice or three times in her life, cause her 
to become pregnant ? " 4 This explanation might apply to 
1 Hartland, Primitive Paternity, ch. i. sq., particularly vol. i. 154 
2 Cf. supra, p. 16. 3 Malinowski, loc. cit. pp. 415, 418. 
4 Ibid. p. 417. A similar explanation has, as Dr. Malinowski 
points out, been previously put forward by some other authors. 
See especially Roth, op. cit. p. 23 ; Spencer, Native Tribes of the 
Northern Territory, p. 25 ; Hartland, Primitive Paternity, ii. 253 sqq. 
Dr. Roth says with reference to the North Queensland aborigines 
that " when it is remembered that as a rule in all these Northern 
tribes a little girl may be given to and will live with her spouse as 
wife long before she reaches the stage of puberty-the relationship 
of which to fecundity is not recognised-the idea of conception 
not being necessarily due to sexual connection becomes partly 
intelligible." vin CRITICISM OF HYPOTHESIS OF PROMISCUITY 
293 
savages like those of Kiriwina, among whom " the un- 
married girls from six upwards are generally supposed 
to practise licence well nigh every night." But then we 
must not expect that savages, even of a lower intellect, 
whose girls remain chaste until they marry-and this, 
as we have seen, is the case with the girls of many savage 
races of a very rude type-should be equally ignorant of the 
cause of pregnancy as are the Kiriwinians. And even the 
latter know, as we have seen, that pregnancy is not quite 
independent of sexual intercourse, nay for animal impreg- 
nation they have no other explanation at all. How is it 
that in the case of animals " the physiological aspect," as 
Dr. Malinowski says, " is well known " ; are their sows and 
bitches more continent than their girls ? " When you ask 
about the animals, you get the answer that it is necessary 
that the physiological conditions should exist," whereas in 
the case of man " copulation is quite unnecessary except for 
its mechanical action."1 The difference, then, is not merely, 
as Dr. Malinowski says, that with regard to animals " the 
other side, the real problem of how life is created in the 
womb, is simply ignored "-unless indeed the expression 
" physiological conditions " in the case of animals also 
implies such conditions as may be produced by mechanical 
action. I am not quite sure, after all, that the Kiriwinian 
theory of human reproduction may not have to some extent 
been influenced by superimposed animistic ideas, although 
Dr. Malinowski is of another opinion.2 Nor does Dr. Roth's 
statement that the Tully River blacks find their belief in 
their own superiority over the brute creation confirmed 
by their knowledge of sexual connection being a cause of 
conception in animals suggest that their views of human 
procreation are really primitive. Nay, it even makes me 
suspect that their ignorance of the man's share in procreation 
is not quite so absolute as it is said to be. 
I must confess, then, that I have some doubts as to the 
present existence of any savage tribe where child-birth is 
considered to be completely independent of sexual inter- 
course. But it is obvious that there are simple peoples who 
1 Malinowski, Iog. cit. p. 413. 2 Ibid. p. 416. 294 
THE HISTORY OF HUMAN MARRIAGE chap. 
regard the man's participation in reproduction as entirely 
subordinate, whereas among others he seems to be regarded 
as the chief or even the only parent. The question now 
arises how far ideas of this sort have influenced the method 
of counting descent. 
When Lahontan asked some Canadian Indians the 
reason for their mother-right, they replied that the children 
having received their soul from their father and their 
body from their mother, it was but reasonable that the 
maternal name should be perpetuated; and when he 
represented to them " that it was more reasonable to derive 
the. original of that custom from the certainty that they had 
of the mother beyond that of the father," they positively 
affirmed that this reason was absurd.1 We are also told 
that the mother-right of the Papuans of Kaiser Wilhelm 
Land is due to the paramount importance they ascribe to 
the mother's share in reproduction ; but it is not clear 
whether this is the native view of the subject or the observer's 
own inference.2 In any case it is necessary to receive with 
cautiousness the explanations which savages give of their 
social institutions. Speaking of the natives of Kiriwina, 
Dr. Malinowski observes that such knowledge as the natives 
have in the matter of reproduction is of no sociological 
importance and " does not influence the native ideas of 
kinship."3 Nor can I find any facts which show that among 
the Australian tribes the method of counting descent is 
dependent upon theories relating to pregnancy and child- 
birth.4 Of the tribes that consider the child to be derived 
from the father alone there are some which have female 
descent, and among which, in addition, the right of betrothal 
lies with the mother or mother's brother.5 So also the 
tribes near Fort Johnston in British Central Africa, who 
believe that the child is formed only from what the man 
introduces into the woman, reckon descent through the 
1 Lahontan, op. cit. p. 461 sq. 2 Neuhauss, op. cit. i. 149. 
3 Malinowski, loo. cit. p. 413. 
4 Cf. N. W. Thomas, op. cit. p. 23. 
5 Howitt, Native Tribes of South-East Australia, p. 195. Cameron, 
loc. cit. p. 352. vm CRITICISM OF HYPOTHESIS OF PROMISCUITY 
295 
mother.1 Suppose that the practice of counting descent 
originated at a time when absolutely no connection was 
known to exist between pregnancy and sexual inter- 
course, and that mother-right was the consequence of 
this ignorance-might we not expect that a change in 
the ideas relating to consanguinity would have been 
accompanied by a change in the method of counting 
descent ? Or why should the latter only in the beginning 
have depended upon ideas of consanguinity and then 
remained unchanged when the principle on which it was 
based was given up ? I do not deny that mother-right 
may have directly something to do with child-birth, with 
the fact that the child comes from the mother, whatever ideas 
be held about fatherhood. Yet I think that there can be 
little doubt that not only father-right, but mother-right as 
well, at least in a large measure depends on social conditions. 
However impossible it might be to explain its prevalence in 
each separate case, even if our knowledge of the lower races 
were far less incomplete than it is at present, we can, I be- 
lieve, with a considerable degree of probability indicate in a 
general way the chief causes, or at least some of them, 
which have led to the system of mother-right, apart from any 
consideration of blood-relationship. 
Among savages in particular the tie between mother and 
child is much stronger than that which binds the child to 
its father.2 The savage mother is for a long time seen carry- 
ing the child at her breast, the sucking period lasting two, 
three, or four years or even more. In cases of separation, 
which occur frequently at lower stages of civilisation, the 
infant children always follow the mother, and so, very often, 
do the children more advanced in years. In polygynous 
families it is a customary arrangement that each wife has 
a hut for herself, where she lives with her children; and 
even where this is not the case, mother and children naturally 
keep together as a little sub-family. All over Equatorial 
Africa, says Mr. Connolly, the wives never eat with their 
husbands, but always with the children. With special 
1 Stannus, in Jour. Roy. Anthr. Inst. xl. 310. 
2 Of. Avebury, op. cit. p. 121. 296 
THE HISTORY OF HUMAN MARRIAGE - chap. 
reference to the Fantis, he adds that polygyny has not been 
without effect on succession to property, the rightful heir 
being the eldest nephew of the eldest sister.1 Indeed, 
speaking of certain negro tribes, Winterbottom suggested 
long ago that the prevalence of the female line was to be 
explained by the practice of polygyny, " which makes it 
easier among a number of children of the same family to 
distinguish them by their mothers' names, than if each 
had his father's prefixed " ; 2 and Dr. Starcke has more 
recently directed attention to the same point.3 The Rev. D. 
Macdonald likewise remarks, in his account of the Efatese 
of the New Hebrides, that the idea that children are more 
closely related to the mother than to the father is an idea 
perfectly natural among a polygynous people.4 Concerning 
the Basuto, a Bechuana tribe, Casalis observes that the 
authority of the eldest maternal uncle preponderates to 
excess especially in polygynous families, where the children 
have no strong affection for their father.5 
Among many peoples in different parts of the world it is 
the custom for a man on marrying to quit his home and go 
to live with his wife's people. In nearly all these cases 
mother-right prevails,6 and, as Tylor has pointed out, it 
generally does so more fully among peoples whose custom 
is for the husband not to take his wife to his own home.7 It 
seems therefore that mother-right in various cases is the 
result of matrilocal marriage. This conclusion is corro- 
borated by the fact that when matrilocal and patrilocal 
marriages occur side by side among the same people, descent 
1 Connolly, ' Social Life in Fanti-Land,' in Jour. Anthr. Inst. 
xxvi. 145 sq. 
2 Winterbottom, Account of the Native Africans in the Neighbour- 
hood of Sierra Leone, i. 151 sq. 
3 Starcke, Primitive Family, pp. 27, 28, 35, 36, 40, 41, &c. 
4 Macdonald, Oceania, pp. 184, 192 sq. 
6 Casalis, Basutos, p. 181. 
6 Hobhouse, Wheeler, and Ginsberg (op. cit. p. 151) have only 
found two cases in which patrilineal descent is combined with 
matrilocal marriage. To these may be added the Yukaghir 
(Jochelson, Yukaghir, pp. 92, 112 sq.). 
7 Tylor, ' On a Method of investigating the Development of 
Institutions,' in Jour. Anthr. Inst, xviii. 258. vni CRITICISM OF HYPOTHESIS OF PROMISCUITY 
297 
is-at least in some instances-reckoned through the 
mother if the husband goes to live with his wife's people 
and through the father if he takes her to his own home.1 
In Japan, should there be only daughters in the family, a 
husband is procured for the eldest, who enters his wife's 
family and at the same time takes its name.2 In this case 
the family wants to acquire a new member; but generally 
matrilocal marriage is undoubtedly, as Dr. Starcke remarks, 
due to the unwillingness of the woman's family to part with 
any of its members.3 This form of marriage seems particu- 
larly to occur among agricultural peoples, and is in such 
cases most probably connected with the fact that among 
simple peoples agriculture very largely devolves on the 
women.4 Among the Australian tribes, who are hunters 
and food-collectors, there is but the very slenderest evidence 
of matrilocal marriage, even where descent is reckoned 
through the mother.5 And pastoral tribes, as we have 
seen, are mostly patrilineal, and, of course, patrilocal; 
which well agrees with the fact that cattle-rearing is largely 
a masculine pursuit6 and consequently makes the woman 
less valuable and more readily dispensed with. 
I have now dealt with the chief facts which have been 
adduced as evidence of the hypothesis of promiscuity. They 
have been found to be no evidence at all. Not one of the 
customs alleged as relics of an ancient state of indiscriminate 
cohabitation of the sexes, or " communal marriage," pre- 
supposes the former existence of such a state. The numerous 
facts put forward in support of the hypothesis do not entitle 
us to assume that promiscuity has ever been the prevailing 
1 Robertson Smith, Kinship and Marriage in Early Arabia, p. 
74 sq. Marsden, History of Sumatra, p. 225. Cf. Forbes, Eleven 
Years in Ceylon, i. 333. 
2 Kiichler, ' Marriage in Japan,' in Trans. Asiatic Soc. Japan, 
xiii. 115. 
3 Starcke, op. cit. p. 79 sq. 
4 Westermarck, Origin and Development of the Moral Ideas, i. 
634- 
5 N. W. Thomas, op. cit. p. 16 sq. 
0 Westermarck, op. cit. i. 634. 298 
THE HISTORY OF HUMAN MARRIAGE ch. viii 
form of sexual relations among a single people, far less that 
it has constituted a general stage in the social development 
of man, and least of all, that such a stage formed the 
starting-point of all human history. 
It may seem to the reader that this question has received 
more attention than it deserves. But I have discussed 
some of the problems connected with it more fully than 
might have been strictly necessary for my purpose, partly 
because a criticism carries more weight if it is not merely 
destructive but substitutes other explanations for those it 
refutes, and partly because I have here found the most 
convenient place for discussing various customs and problems 
which cannot be ignored in a history of human marriage. 
And even now my criticism has not come to an end. Having 
shown that the hypothesis of promiscuity is unsupported 
by any evidence, I shall in the next chapter endeavour to 
demonstrate that it is contrary to certain facts which may 
be assumed to have been positive obstacles to a state of 
promiscuity in any period of the history of mankind. CHAPTER IX 
A CRITICISM OF THE HYPOTHESIS OF PROMISCUITY : 
MASCULINE JEALOUSY 
Darwin remarks that from what we know of the jealousy 
of all male quadrupeds, armed, as many of them arc, with 
special weapons for battling with their rivals, promiscuous 
intercourse is utterly unlikely to prevail in a state of nature. 
" Therefore," he continues, " looking far enough back in 
the stream of time, and judging from the social habits of 
man as he now exists, the most probable view is that he 
originally lived in small communities, each with a single 
wife, or if powerful with several, whom he jealously guarded 
against all other men." Yet it seemed certain to Darwin, 
from the lines of evidence afforded by Morgan, McLennan, 
and Lubbock, that almost promiscuous intercourse at a 
later time was extremely common throughout the world; 1 
and a similar view is held by some other writers.2 
If the facts adduced as evidence of former promiscuity 
really had proved that it was general at some stage of human 
development, we should, of course, be compelled to admit 
that jealousy on the part of the men could have been no 
1 Darwin, Descent of Man, ii. 394 sq. Already before Darwin, 
Virey argued {De la femme, p. 148) that promiscuity would have 
caused perpetual fighting between the men. 
2 Le Bon, L'homme et les sociMs, ii. 289 sq. Kautsky, ' Die 
Entstehung dcr Ehc,' in Kosmos, xii. 262. 
299 300 
THE HISTORY OF HUMAN MARRIAGE chap. 
serious hindrance to it. But the case is entirely different 
if we have no reason whatever to believe that there was a 
stage of promiscuity at any time. Then the prevalence of 
male jealousy both among the anthropoid apes and the 
existing races of men constitutes a strong prima-facie 
evidence of its prevalence in mankind in earlier ages as well. 
And this would have made general promiscuity exceedingly 
improbable, to say the least. 
It has, however, been asserted by certain writers, such as 
Giraud-Teulon1 and Dr. Le Bon,2 that jealousy is almost un- 
known among nearly all uncivilised peoples, and recently 
Dr. Hartland has produced a long list of savages who are said 
to be little addicted to it.3 According to him, the sense of 
ownership has been the seed-plot of jealousy, and in conse- 
quence this feeling operates only feebly in conditions where 
the sense of ownership is undeveloped or imperfect.4 This 
is to my mind a more than doubtful proposition. It is 
true that a savage often regards his wife as a kind of property 
and an adulterer as a thief.5 The Kiwai Papuans of British 
New Guinea,6 like the Western Islanders of Torres Straits,7 
say of a man who commits adultery with another man's 
wife that he " steals " her. The Arunta call such a man 
atna nylkna, which means vaginae fur, or inbitja, which means 
" thief."8 In some parts of Africa he is punished as a thief, 
1 Giraud-Teulon, Les origines de la famille, p. 79 note. 
2 Le Bon, op. cit. ii. 293. 
3 Hartland, Primitive Paternity, ii. 103 sqq. 4 Ibid. ii. 102 sq. 
5 See Casalis, Basutos, p. 225 ; Junod, Life of a South African 
Tribe, i. 194 ; Rochon, * Voyage to Madagascar,' in Pinkerton, 
Collection of Voyages and Travels, xvi. 747 ; Weeks, Among Congo 
Cannibals, p. 181 (Bolobi) ; Lumholtz, Among Cannibals, p. 126 
(natives of Northern Queensland). Guevara, ' Folklore Araucano,' 
in Anales de la Universidad de Chile, cxxvii. 626; Letourneau, 
Devolution du mariage, p. 258 sq. ; Westermarck, Origin and De- 
velopment of the Moral Ideas, ii. 449 n. 5. 
6 Landtman, Nya Guinea fdrden, p. 82. 7 Supra, p. 147. 
8 Spencer and Gillen, Native Tribes of Central Australia, p. 99. 
Strehlow, Die Aranda- und Loritja-Stamme, vol. iv. pt. i. 92. These 
epithets are given if the woman belongs to the class within which 
the adulterer is allowed to marry ; otherwise a much more oppro- 
brious term is applied to him. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
301 
having his hands, or one of them, ent oft'.1 In the ' Laws 
of Manu ' it is said that " seed must not be sown by any 
man on that which belongs to another."2 But if jealousy 
has anything to do with the sense of ownership, the reason 
is that it is primarily connected with the desire of posses- 
sion, which is something different. 
Ownership implies that a certain person or certain persons 
are recognised as having a right to the exclusive disposal of 
a certain thing. It owes its origin to the desire of an indi- 
vidual to keep and dispose of what he has appropriated or 
produced, but it is by no means identical with mere posses- 
sion.3 The male animal jealously keeps for himself the 
female he has appropriated and is enraged if his possession 
of her is interfered with, but we cannot say that his jealousy 
depends on a " sense of ownership." Sexual jealousy, as 
Mr. Shand remarks, springs from sexual love. " But sexual 
love cannot be separated from self-love, with which • it 
constantly interacts; and it is due to the desire of self-love 
to possess certain things exclusively for self, such as women, 
power, and reputation, that jealousy principally arises. 
Thus La Rochefoucauld observes : ' Il y a dans la jalousie 
plus d'amour propre que d'amour.' "4 The jealousy of a 
man, particularly a civilised man, differs from that of a 
male animal, apart from any feeling of injured rights- 
ownership or any other right. It is coloured by the nature 
of his love. It is accompanied with humiliation, because 
" the loss of possession to which jealousy refers, or the 
failure to obtain it, is of such a nature as carries with it a 
lowering of a man's self-valuation."5 There may also be 
fear of another man's offspring being born into the family.6 
1 Reade, Savage Africa, p. 61 (Fernando Po). Waitz, Anthro- 
pologic der Naturvblker, ii. 472 (Fulah). 
2 Ldws of Manu, ix. 42. 
3 See Westermarck, Origin and Development of the Moral Ideas, 
ii. 1, 51 sqq. 
4 Shand, Foundations of Character, p. 258. 5 Ibid. p. 258. 
6 It was a doctrine of the Roman jurists that adultery is a crime 
in the wife, and in the wife only, on account of the danger of intro- 
ducing strange children to the husband (Hunter, Exposition oj 
Roman Law, p. 1071) 302 
THE HISTORY OF HUMAN MARRIAGE chap. 
But there is one characteristic.common to sexual jealousy in 
all its forms, namely, that it is an angry feeling aroused by 
the loss, or the fear of the loss, of the exclusive possession of 
an individual who is the object of one's sexual desire. And 
this dominant feature of sexual jealousy, which is found in 
brute and man, savage and civilised alike, is strongly 
opposed to the idea of promiscuous intercourse. The 
variations of jealousy are therefore of no moment in the 
present connection. In polemising against me, Mr. Finck 
complains that " language should be so crude as to use the 
same word jealousy to denote three such entirely different 
things as rage at a rival, revenge for stolen property, and 
anguish at the knowledge or suspicion of violated chastity 
and outraged conjugal affection."1 But language has coined 
a word which excellently suits our purpose, and its different 
shades of meaning do not concern us. 
We shall now see how far male jealousy can be said to 
be prevalent among the lower races. Let us start with 
the lowest. Nearly all authorities on the Pygmy peoples 
agree that they are very faithful in their marriages, and 
that adultery, if it does occur among them, is mostly 
punished with death.2 As regards the Bushmen, Dr. 
Hartland quotes a statement by Alexander, indicating a 
considerable lack of jealousy,3 but other writers picture 
them in a very different light. Speaking of both men and 
women, Theal says that " their passionate tempers prevented 
the presence of rivals in the same abode."4 According to 
von Fran?ois, the Bushman is very jealous of the exclusive 
possession of his wife, and avenges an encroachment on 
his marital rights in the Italian manner.5 Among the Auin 
the seducer of another man's wife is killed, and the woman 
1 Finck, Primitive Love and Love-Stories, p. 88 n. i. 
2 Schmidt, Die Stellung der Pygmaenvblker in der Entwick- 
lungsgeschichte des Menschen, p. 162. Cf, however, Johnston, 
George Grenfell and the Congo, ii. 674. 
3 Hartland, op. cit. ii. 214. 
4 Theal, Yellow and Dark-skinned People of Africa south of the 
Zambesi, p. 47. 
'■ v. Francois, Nama und Damara Deutsch-Siid-West-Afrika, 
P- 237. ix CRITICISM OF HYPOTHESIS OF PROMISCUITY 
303 
is beaten ; 1 and among the Namib Bushmen adultery is 
said to be very rare.2 Among the wild tribes of the Malpy 
Peninsula Dr. Martin saw obvious signs of jealousy in the 
men.3 The Vcddas of Ceylon are represented by various 
writers as exceedingly jealous of their wives,4 and infidelity 
seems to be hardly known among them ;5 " nothing short of 
murder," says Nevill, " would content the injured party." 
The Australian aborigines-in spite of the prevalence 
among them of customs such as the lending and exchange 
of wives and, in some tribes, of group-relations-are by no 
means lacking in jealousy, as appears from the statements 
of a great many of our authorities.7 Summing up the 
evidence, Dr. Malinowski observes that if by jealousy is 
meant a natural aversion of an individual towards an 
encroachment on his sexual rights, it is, on the contrary, a 
feeling which is very strongly developed among them.8 
Messrs. Spencer and Gillen are much less positive. They 
say that among those Australian natives with vhem they 
1 Kaufmann, ' Die Auin,' in Mittheil. Deutsch. Schutzgeb. xxiii. 156. 
2 Trenk, ' Die Buschleute der Namib,' in Mittheil. Deutsch. 
Schutzgeb. xxiii. 169. 
3 Martin, Die Inlandstamme der Malayischen Halbinsel, p. 874. 
4 Bailey, ' Account of the Wild Tribes of the Veddahs of Ceylon,' 
in Trans. Ethn. Soc. N.S. ii. 292. Nevill, ' Vaeddas of Ceylon,' in 
Taprobanian, i. 178. Sarasin, Ergebnisse naturwissenschaftlicher 
Forschungen auf Ceylon, iii. 462 sq. 
5 Bailey, loc. cit. ii. 291 sq. Hartshorne, ' Weddas,' in Indian 
Antiquary, viii. 320. 6 Nevill, loc. cit. p. 178. 
7 Curr, The Australian Race, i. 100, 109. Waitz-Gerland, Anthro- 
pologic der Naturvblker, vi. 774. Wilkes, Narrative of the United 
States Exploring Expedition, ii. 195. Salvado, Memoires historiques 
sur I'Australie, p. 280. Grey, Journals of Two Expeditions of Dis- 
covery in North-West and Western Australia, ii. 252. Withwell, 
Customs and Traditions of the Natives of North Western Australia, 
p. 17. Mathew, ' Australian Aborigines,' in Jour, and Proceed. Roy. 
Soc. N. S. Wales, xxiii. 404. Breton, Excursions in New South 
Wales, &c., p. 231. Schuermann, ' Aboriginal Tribes of Port Lincoln,' 
in Woods, Native Tribes of South Australia, p. 223. Eylmann, 
Die Eingeborenen der Kolonie Sudaustralien. pp. 130, 138. Strehlow, 
Die Aranda- und Loritja-Stamme, vol. iv. pt. i. 94. 
8 Malinowski, The Family among the Australian Aborigines, 
p. 125. See ibid. p. 124 sqq. 304 
THE HISTORY OF HUMAN MARRIAGE chap. 
have come in contact " the feeling of sexual jealousy is 
not developed to anything like the extent to which it would 
appear to be in many other savage tribes." Now and 
again it will come into play, " but as a general rule this is 
a feeling which is undoubtedly subservient to that of the 
influence of tribal custom, so far as the latter renders it 
obligatory for a man to allow other men, at certain times, 
to have free access to his wife, or so far as it directs him to 
lend his wife to some other individual as a mark of personal 
favour to the latter."1 Mr. Gason states that among the 
Dieri it is set down by the law of the tribe, under pain of 
strangulation, that no jealousy shall be shown or exist 
between persons who have entered into group-relations with 
each other; but he adds that in spite of this strict law he 
is positive from his own observations that nearly all their 
quarrels spring just from those relations.2 
We are told that the Fuegians " are exceedingly jealous 
of their women and will not allow any one, if they can help 
it, to enter their huts, particularly boys."3 The Botocudos, 
who are known to change wives very frequently, are never- 
theless much addicted to jealousy ; they have a horror of 
adultery and punish severely any woman found guilty of 
it.4 Concerning the Coroados, v. Spix and v. Martius say 
that revenge and jealousy are the only passions that can 
rouse their stunted soul from its moody indifference.5 
Castelnau wrote of the Guatos, " La jalousie est la passion 
qui agite le plus fortement 1'ame de ces sauvages, et la garde 
de leurs nombreuses femmes parait les absorber entiere- 
ment."6 The Guarayos of South-West Amazonia " are 
1 Spencer and Gillen, Native Tribes of Central Australia, p. 99. 
2 Gason, ' Of the Tribes, Dieyerie, &c.,' in Jour. Anthr. Inst. xxiv. 169. 
3 Wilkes, op. cit. i. 125. Cf. Skottsberg, * Observations on the 
Natives of the Patagonian Channel Region,' in American Anthro- 
pologist, N.S. xv. 596 (West Patagonians). 
4 v. Martius, Beitrage zur Ethnographic Amerika's, i. 322. Keane, 
' On the Botocudos,' in Jour. Anthr. Inst. xiii. 206. de Lodi, ' Ex- 
trait d'une lettre,' in Annales de la propagation de la Foi, xvii. 415. 
5 v. Spix and v. Martius, Travels in Brazil, ii. 241. 
0 Castelnau, Expedition dans les parties centrales de V AmMque 
du Sud, iii. 13. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
305 
extremely jealous " and punish adultery by death of both 
man and woman.1 The Jibaros and Napo Indians of 
Ecuador are likewise said to be very jealous of their women ; 
among the former, the women are very frequently not 
allowed to talk to strangers and, " with Oriental exclusive- 
ness, not even to show themselves."2 Of another tribe in 
Ecuador, the Zaparos, it is stated that " they are not at 
all jealous, but allow their women great liberty," although 
this is " entirely contrary to other neighbouring tribes."3 
The Indians of Peru4 and the Arawaks5 are said to commit 
horrible crimes out of jealousy. 
Concerning the Huichol of Mexico Dr. Lumholtz writes, 
" That jealousy is highly developed among them is shown by 
their strenuous resentment of matrimonial indiscretion."6 
Of the Sia, one of the Pueblo peoples, we are told that 
although it is common among them both for the married 
and the unmarried to live promiscuously with one another, 
they nevertheless " have their share of latent jealousy," 
which is evident from the secrecy observed on the part of a 
married man or woman to prevent the anger of the spouse.7 
Among the Creeks " it was formerly reckoned adultery if a 
man took a pitcher of water off a married woman's head, 
and drank of it."8 The Illinois, according to La Salle, " are 
very jealous, and punish the infidelity of their wives with 
a great severity."9 The Alibamu, says Bossu, " are jealous 
to excess."10 In his description of " a vast country between 
New France and New Mexico " Hennepin observes that 
1 Church, Aborigines of South America, p. 114 sq. 
2 Simson, Travels in the Wilds of Ecuador, pp. 90, 158. Rivet, 
Les Indiens Jibaros,' in L' Anthropologic, xviii. 605. 
3 Simson, op. cit. p. 173. 
4 v. Schiitz-Holzhausen, Der Amazonas, p. 70. 
6 v. Martins, op. cit. i. 693. 
G Lumholtz, Unknown Mexico, ii. 91. 
7 Mrs. Stevenson, ' Sia,' in Ann. Rep. Bur. Ethnol. xi. 20. 
8 Adair, History of the American Indians, p. 143. 
9 La Salle, ' Last Expedition and Discoveries in North America,' 
in Collections of the New York Historical Society, for the Year 1814, 
ii. 238. 
10 Bossu, Travels through Louisiana, i. 231. 306 
THE HISTORY OF HUMAN MARRIAGE chap. 
" the men of the hot country are more jealous of their wives 
than those of the north ; the first are so jealous in this 
matter, that they wound themselves, and sometimes kill 
themselves in a blind passion of love, which prompts them 
to this fury."1 Speaking of the Hurons, Charlevoix remarks 
that " what most commonly disturbs domestic peace among 
the people of Canada is jealousy, which is equal on both 
sides. The Iroquois boast that they are never troubled 
with it : but those who are most acquainted with them, 
affirm that they are jealous to excess."2 Lahontan, on the 
other hand, says of Canadian Indians that the men are cold 
and incapable of jealousy- but this hardly agrees with his 
statement that the married women " had as good be dead 
as be guilty of adultery."3 Of some Californian Indians we 
are told that if a married woman is seen even walking in 
the forest with another man than her husband, she is 
chastised by him, whilst a repetition of the offence is gener- 
ally punished with speedy death.4 Among the Indians of 
San Diego in Upper California, according to a Mission 
record, " the men pursue one another with death for 
jealousies and other vexations."5 A considerable degree of 
jealousy was found among the Cree by Richardson,6 among 
the Takulli by Harmon,7 among the Haida by Dixon,8 
among the Kutchin by Richardson and Hardisty,9 among the 
1 Hennepin, New Discovery of a Vast Country between New France 
and Neiv Mexico, p. 483. Cf. Perrin du Lac, Travels through the 
Two Louisianas, p. 80. 
2 Charlevoix, Voyage to North-America, ii. 37 sq. 
3 Lahontan, New Voyages to North-America, pp. 453, 460. 
4 Powers, Tribes of California, p. 412. 
5 Kroeber, ' Mission Record of the California Indians,' in Uni- 
versity of California Publications in American Archaeology and 
Ethnology, viii. 6. 
6 Franklin, Journey to the Shores of the Polar Sea, p. 67. 
7 Harmon, Journal of Voyages and Travels in the Interior of North 
America, p. 293. 
8 Dixon, Voyage round the World, p. 225 sq. Cf. Swanton, Haida, 
P- 5\ 
9 Richardson, Arctic Searching Expedition, i. 383. Hardisty, 
' Loucheux Indians,' in Smithsonian Report, 1866, p. 312. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
307 
Atkha Aleut by Father Yakof.1 The Indians on the 
eastern side of the Rocky Mountains visited by Harmon, 
in their fits of jealousy, " often cut off all the hair from the 
heads of their wives, and, not unfrequently, cut off their 
noses also ; and should they not in the moment of passion 
have a knife at hand, they will snap it off at one bite, with 
their teeth. . . . The man is satisfied in thus revenging a 
supposed injury ; and having destroyed the beauty of his 
wife, he concludes that he has secured her against all future 
solicitations to offend."2 According to a Tlingit myth, the 
jealousy of man is older than the world itself. There was 
an age, it is supposed, when men groped in the dark in 
search of the world. At that time a Tlingit lived who had 
a wife and a sister ; and he was so jealous of his wife that he 
killed all his sister's children because they looked at her.3 
Among the Eskimo the jealousy of the men seems to be 
feebler than among most other natives of America ; but it 
is not absent. Speaking of the curious amusements of the 
Greenlanders, Dalager says that the men are too jealous 
to allow their wives to take part in any debauchery in case 
they have borne to them children.4 Cranz, who had no 
high opinion of the sexual morality of those natives, admits 
that if any infidelity occurs in the wife, the injured husband 
revenges himself in a similar way when an opportunity occurs, 
and that the disagreement " frequently costs the wife a black 
eye, which is rather surprising, as the Greenlanders are not 
at all quarrelsome or addicted to blows."5 Dr. Boas knows 
of instances of quarrels arising from jealousy among the 
1 Yakof, quoted by Petroff, ' Report on the Population, &c. of 
Alaska,' in Tenth Census of the United States, p. 158. For other 
instances of jealousy among North American Indians see Waitz, 
op. cit. iii. 328 ; Hearne, Journey from Prince of Wales's Fort to 
the Northern Ocean, p. 310 ; Mackenzie, Voyages from Montreal to 
the Frozen and Pacific Oceans, p. 147. 
2 Harmon, op. cit. p. 343. 
3 Holmberg, ' Ethnographische Skizzen iiber die Volker des 
russischen Amerika,' in Acta Soc. Scient. Fennicae, iv. 332 sq. Dall, 
Alaska, p. 421. 4 Dalager, Gr^nlandske Relationer, p. 67 sq. 
5 Cranz. History of Greenland, i. 147. 308 
THE HISTORY OF HUMAN MARRIAGE chap. 
Eskimo of Davis Strait and Cumberland Sound, although 
Lyon states that this passion is unknown among the 
Iglulirmiut of Baffin Land.1 Among the Eskimo of Hudson's 
Bay jealousy, resulting from laxity of morals, produces so 
much disagreement that one or the other of the parties 
usually leaves with little ceremony.2 Among the Eskimo 
about Bering Strait a man who finds that his wife is un- 
faithful may beat her or send her away ; but he rarely 
revenges himself on the adulterer.3 
From Siberia and other parts of the former Russian 
empire, inhabited by more or less uncivilised tribes, we hear 
little about jealousy but a great deal about disregard of 
chastity in married as well as unmarried women, which, as 
we have seen, is partly at least the result of Russian influence. 
Among the Koryak, who have been comparatively little 
subject to this influence, adultery is, according to Dr. 
Jochelson, less frequently met with than among civilised 
peoples. The injured husband does not touch the adulterer, 
but casts out or thrashes the faithless wife without mercy, 
and formerly many wives were killed by passionate husbands. 
It is said that a married woman had to go in a dirty dress 
and with unwashed face, so as not to attract the attention of 
strange men.4 Captain Arnesen noticed great jealousy 
among the Samoyed.5 Dr. A. O. Heikel informs me that a 
Tartar may repudiate his wife if he sees her shaking hands 
with another man. On the other hand, we are told that 
the Burnt6 and Ostyak7 know little of jealousy, and that 
among the Kirghiz husbands even encourage their friends to 
close intimacy with their wives.8 
1 Boas, * Central Eskimo,' in Ann. Rep. Bur. Ethn. vi. 579. 
2 Turner, ' Ethnology of the Ungava District,' in Ann. Rep. 
Bur. Ethn. xi. 188 sq. See also ibid. p. 178. 
3 Nelson, * Eskimo about Bering Strait,' in Ann. Rep. Bur. Ethn. 
xviii. 292. 
4 Jochelson, Koryak, p. 756. Georgi, Beschreibung alter Nationen 
des Russischen Reichs, p. 348 sq. 
■' Arnesen, ' Fran Gyda-viken till Obdorsk,' in Ymer, iii. 144. 
0 Valikhanof, &c., Russians in Central Asia, p. 85. 
7 Miss Czaplicka, Aboriginal Siberia, p. 128. 
8 Valikhanof, &c., op. cit. p. 85. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
309 
Among the savages of Formosa " unfaithfulness on the 
part of a wife is unheard of."1 Among the Sakai of 
Sumatra the men are said to be addicted to jealousy,2 
and the same is the case among various other tribes in the 
Malay Archipelago,3 as also in Yap (one of the Pelew Islands),4 
the Shortland Islands,6 New Caledonia,6 and New Guinea.7 
In Mailu, British New Guinea, according to Dr. Malinowski, 
" sexual jealousy is very pronounced, and this sentiment 
is not based only upon the sense of ownership established 
by marriage, since it exists very markedly in respect to the 
Ui'ui relationships "-that is, preconjugal intrigues without 
intention to marry.8 Among the Southern Massim Dr. 
Seligman repeatedly heard that in the old days a man might 
be killed for even talking to another man's wife if he met 
her alone in the bush or on the sea-shore, and that no married 
woman would speak to a stranger unless other women were 
present.9 In the Santa Cruz Islands, says Dr. Speiser, 
" masculine jealousy seems to have reached its climax . . ., 
for no man from another village even dares look at a woman."10 
In the Sandwich Islands, where there seems to have been 
some kind of sexual communism, jealousy was nevertheless, 
1 Davidson, Island of Formosa, p. 583. 
2 Moszkowski, Auf neuen Wegen durch Sumatra, p. 105. ' 
3 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, 
pp. 5, 335, 448. Modigliani, Un viaggio a Nias, p. 471. 
4 Kramer, ' Studienreise nach den Zentral- und Westkarolinen,' 
in Mittheil. Deutsch. Schutzgeb. xxi. 178. 
5 Ribbe, Zwei Jahre unter den Kannibalen der Salomo-Inseln, 
p. 142. 
6 Moncelon, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. ix. 368. 
Brainne, La Nouvelle-Caledonie, p. 251. Lambert, Mceurs et 
Superstitions des Nto-Caledoniens, p. 96. 
7 Vetter, ' Bericht liber papuanische Rechtsverhaltnisse,' in 
N achrichten uber Kaiser Wilhelms-Land, 1897, p. 87 (natives of Sim- 
bang). Moszkowski Die Volkerstamme am Mamberamo in Hol- 
landisch-Neuguinea,' in Zeitschr. f. Ethnol. xliii. ,339. 
8 Malinowski, ' Natives of Mailu,' in Trans. Roy. Soc. South 
Australia, xxxix. 572. 
9 Seligman, Melanesians of British New Guinea, p. 567 sq. 
10 Speiser, Two Years with the Natives in the Western Pacific, 
p. 284. 310 
THE HISTORY OF HUMAN MARRIAGE chap. 
according to Lisiansky, extremely prevalent j1 and in Nuka- 
hiva, where polyandry was practised, the men punished 
their wives with severity upon the least suspicion of in- 
fidelity.2 The Areois of Tahiti, too, although given to 
every kind of licentiousness, are described by Ellis as 
utterly jealous.3 The Maori, who treated incontinence 
lightly in the case of a young girl, looked upon it as a serious 
offence in a married woman, who was sometimes punished 
with death for adultery.4 
With reference to the Negroes of Fida, Bosman states that 
a rich man will not suffer any other man to enter the houses 
where his wives reside, and on the least suspicion will sell 
them to Europeans. The natives of the Gold Coast, says 
Cruickshank, " watch over the objects of their affection with 
a. jealous eye " ; and he speaks of " the spirit of jealousy, 
which in the African takes the form of a strong instinct." 6 
In the Nile countries and various other parts of Africa 
it is customary for the men to preserve the fidelity of their 
wives in a way not unlike a method used in the age of the 
Crusades.7 Among the Beni Mzab a man who speaks in 
the street to a married woman of quality is punished with 
a fine of two hundred francs and banishment for four years.8 
That jealousy is a powerful agent in the social life of 
civilised nations is a fact which it is unnecessary to dwell 
upon. In the Muhammadan world it is, strictly speaking, 
unlawful for a man to see the faces of any other women than 
those whom he is forbidden to marry and his own wives and 
female slaves ;9 but the feminine custom of wearing a veil is 
1 Lisiansky, Voyage round the World, p. 128. 2 Ibid. p. 82. 
3 Ellis, Polynesian Researches, i. 239. 
4 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 51 sqq. For the jealousy of South Sea 
Islanders see also Macdonald, Oceania, p. 194. 
5 Bosman, ' New Description of the Coast of Guinea,' in Pinker- 
ton, op. cit. xvi. 479. 
0 Cruickshank, Eighteen Years on the Gold Coast of Africa, ii. 
195, 212. 
7 Bastian, Die Rechtsverhdltnisse bei verschiedenen Vblkern der 
Erde, p. xx. Waitz, op. cit. ii. 516. 
8 Chavanne, Die Sahara, p. 315. 
9 Lane, Manners and Customs of the Modern Egyptians, i. 138. ix CRITICISM OF HYPOTHESIS OF PROMISCUITY 
311 
by no means universally observed there,1 and as I have 
pointed out elsewhere, may not only be due to masculine 
jealousy, but serve the object of protecting the woman against 
the evil eye.2 A man who penetrates into the harem of 
another man may easily lose his life ; and Dr. Polak states that 
in Persia a European physician cannot, without being con- 
sidered indecent, even ask about the health of a Muhamma- 
dan's wife and daughter although they are ill.3 In Japan, 
as I am told by a native of that country, it was customary for 
women, when getting married, to have their eyebrows shaved 
off, because thick and beautiful eyebrows are considered 
one of a woman's greatest attractions. This reminds us 
of the widespread practice of depriving a woman of her 
ornaments as soon as she is married. 
It is not, however, statements directly referring to 
jealousy that are our chief source of information with 
regard to its prevalence in mankind. It is mentioned only 
when it, for some reason or other, attracts the attention of 
the traveller ; and if it is said to be lacking, this statement 
seems generally to be an inference from the fact that a man 
under certain conditions permits, or is by custom compelled 
to permit, some other man to have intercourse with his 
wife. The real evidence of the general prevalence of 
masculine jealousy is found in the customs or laws concerning 
adultery, which make the seducer or the unfaithful wife or 
both liable to punishment of some sort, inflicted either by the 
injured husband himself or by the society of which he is 
a member, or at all events give him power to divorce the 
wife. In a monogamous or polygynous marriage the 
husband has, in ordinary circumstances at least, an exclusive 
right of access to his wife, and in the case of polyandry or 
1 It is not strictly observed among the lower classes in Arabia 
(Palgrave, Journey through Central and Eastern Arabia, i. 271 sq.) nor 
by the Muhammadans of Africa (Munzinger, Ostajrikanische Studien, 
p. 511; d'Escayrac de Lauture, Die afrihanische Wiiste, p. 63; 
Chavanne, Die Sahara, p. 349). Among the country people of 
Morocco the veil is used in some tribes but not in others. 
2 Westermarck, ' Magic Origin of Moorish Designs,' in Jour. 
Anthr. Inst, xxxiv. 211. 
3 Polak, Persien, i. 224. 312 
THE HISTORY OF HUMAN MARRIAGE chap. 
group-relations a similar right belongs to the husbands or 
recognised cicisbeos jointly. I doubt whether there really 
are any exceptions to this rule, although we are told that 
among a few peoples adultery is not held to be wrong.1 
Dr. Rivers suspects that among the Todas, instead of 
adultery being considered wrong, immorality is rather 
attached to the man who grudges his wife to another ;2 
but according to his own statement the consent of the 
husband or husbands is required if a woman is to become the 
formal mistress of a man who is not her husband.3 The 
disapproval of adultery varies greatly in degree among 
different peoples. Some savages are said to regard it as 
a venial offence, others as a most heinous crime. Among 
many the seducer escapes by paying a fine to the injured 
husband ; among others he is beaten or has his head shaved, 
his ears cut off, one of his eyes destroyed, or his legs speared. 
Sometimes he is merely paid in his own coin, or the punish- 
ment falls on his wife, who in that case seems to be looked 
upon as the real cause of her husband's unfaithfulness.4 
Very frequently he has to pay with his life. Among the 
Mishmis in Assam adultery committed against the consent 
of the husband was punished with death, although all other 
crimes, including murder, were punished only by fines.5 
The Baganda also, in former days, punished adultery more 
severely than murder.6 In the south-eastern extremity of 
British New Guinea there is a much stronger public feeling 
against adultery than against homicide, and the white man's 
law which provides that the former shall be punished by a 
fine or imprisonment instead of exposing the perpetrator to 
being killed by the aggrieved husband, as was formerly the 
case, seems to the native inadequate.7 
1 Westermarck, Origin and Development of the Moral Ideas, ii. 
447 n. i. 
2 Rivers, Todas, p. 529 sq. 3 Ibid. p. 527. 
4 Moncelon, in Bull. Soc. d'Anthr. Paris, ser. iii. vol. viii. 361 
(New Caledonians). 
5 Griffith, ' Journal of a Visit to the Mishmee Hills in Assam,' 
in Jour. Asiatic Soc. Bengal, vi. 332. 
6 Wilson and Felkin, Uganda, i. 201. 
7 Seligman, op. cit. p. 567 sq. Cf. Waitz-Gerland, op. cit. vi. 661. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
313 
Winwoqd Reade remarks that among savages generally 
it is the seducer who suffers, not the unfaithful wife.1 This 
is more or less true of many uncivilised peoples in different 
parts of the world, in Africa in particular,2 but it can certainly 
not be laid down as a general rule. More commonly the 
unfaithful wife is also treated as an offender, being discarded, 
beaten, or ill-treated in some way or other, and not infre- 
quently killed ; and among some peoples she is the only one 
who has to suffer.3 Often, too, she is disfigured by her 
enraged husband in such a way as to be deprived of her 
attractions : he bites or cuts off her nose, or cuts off one or 
both of her ears or her hair, or shaves her head.4 A similar 
1 Reade, Savage Africa, p. 61. 
2 Delafosse, ' Les Agni,' in L' Anthropologic, iv. 429 sq. Clozel 
and Villamur, Les coutumes indigenes de la C6te d'Ivoire, pp. 102, 
103, 319- Klose, Togo unter deutscher Flagge, p. 256. Ward, 
A Voice from the Congo, p. 253 (if the act is committed outside 
the village boundary). Weeks, Among Congo Cannibals, p. 127 sq. 
(Bolobi on the Upper Congo). Monteiro, Angola and the River 
Congo, i. 243 (Mussurongo, Ambriz, and Mushicongo). Kohler, 
' Das Recht der Herero,' in Zeitschr. f. vergl. Rechtswiss. xiv. 309. 
Magyar, Reisen in Siid-Afrika, p. 281 (Kimbunda). Junod, op. cit, 
i. 194 (Thonga). Gouldsbury and Sheane, Northern Rhodesia, p. 57 
(Awemba). Johnston, Uganda Protectorate, p. 689. Roscoe, Northern 
Bantu, p. 279 (Nilotic Kavirondo). O'Sullivan, ' Dinka Laws and 
Customs,' in Jour. Roy. Anthr. Inst. xl. 188. Scaramucci and Giglioli, 
' Notizie sui Danakil,' in Archivio per Vantrapologia e la etnologia, 
xiv. 26. Macpherson, Memorials of Service in India, p. 133 (Kandhs). 
Watt, ' Aboriginal Tribes of Manipur,' in Jour. Anthr. Inst. xvi. 355 
(Kaupuis). Moore, Marriage Customs, p. 187 (Ladrone Islanders). 
Batchelor, Ainuandtheir Folk-Lore, p. 285. Powers, Tribes of California, 
PP- 75» 24$. 27°- Schoolcraft, Information respecting the History, 
&c. of the Indian Tribes of the United States, ii. 132 (Comanche), 
de Herrera, General History of the West Indies, iv. 140 (ancient in- 
habitants of Honduras). Azara, Voyages dans V Amerique meri- 
dionale, ii. 95 (Guanas). Falkner, Description of Patagonia, p. 126. 
3 Morgan, League of the Iroquois, p. 331. James, Account of an 
Expedition from Pittsburgh to the Rocky Mountains, i. 233 (Omaha). 
Jochelson, Koryak, p. 756. 
4 Schoolcraft, op. cit. i. 236, ii. 132, v. 167, 683, 684, 686 ; Adair, 
op. cit. p. 144 sq. ; Carver, Travels through the Interior Parts of North 
America, p. 375 ; Hennepin, op. cit. p. 482 ; Bradbury, Travels 
in the Interior of America, p. 168 sq.; Bancroft, Native Races of the 
Pacific States of North America, i. 514 ; Kohler, ' Die Rechte der 314 
THE HISTORY OF HUMAN MARRIAGE chap. 
punishment has figured even in European law-books. 
According to a law of Cnut, an adulteress shall have her 
nose and ears cut off j1 whilst in the ' Uplands-lag,' an old 
Swedish provincial law, it is prescribed that an adulteress 
who cannot pay the fine of forty marks shall lose her hair, 
ears, and nose.2 This is, in a manner, a punishment in- 
flicted on " the offending member " ; and so is in a still 
higher degree the custom prevalent among some North 
American Indians of compelling an adulteress to " pass 
through the meadow," that is, to submit to the embraces of 
a large number of men successively.3 
Adultery may be condemned as wrong and be subject to 
punishment for various reasons. It is sometimes supposed 
to mar the fertility of the earth and to blight the crops.4 
But this belief, like the kindred belief which attributes the 
same injurious effects to incest and fornication, undoubtedly 
springs from the fact that it is illicit love and therefore 
regarded as polluting.5 Such a belief naturally increases 
the gravity of the offence in the eyes of those who hold 
Urvolker Nordamarikas,' in Zeitschr. f. vergl. Rechtswiss. xii. 386 
sq. (various North American tribes). James, op. cit. i. 233 (Omaha). 
Marston, ' Letter to Rev. Dr. J. Morse,' in Blair, Indian Tribes oj 
the Upper Mississippi Valley, &c., ii. 167 (Potawatomi). Forsyth, 
' Manners and Customs of the Sauk and Fox Nations,' ibid. ii. 214 
sq. Bossu, op. cit. i. 234 (Alibamu). ten Kate, Reizen en onder- 
zoekingen in Noord-Amerika, p. 175 (Apache). Waitz, op. cit. iv. 
367 (Chibchas). Thomas, Anthropological Report on Sierra Leone, 
i. 100 (Timne). Bowdich, Mission from Cape Coast Castle to Ashantee, 
p. 302 (Ashantees). Wilson, Western Africa, p. 182. Fiilleborn, 
Das Deutsche Njassa- ztnd Ruwuma-Gebiet, p. 307 (Konde people) 
Lobo, ' Voyage to Abyssinia,' in Pinkerton, op. cit. xv. 25 sq. For. 
syth, Highlands of Central India, p. 149 (Gonds and Korkus). Watt, 
in Jour. Anthv. Inst. xvi. 358 (Kolyas). Smith, Five Years' Resi- 
dence at Nepaul, i. 153. Pennell, Wild Tribes of the Afghan Frontier, 
p. 193. Krauss, Sitte ztnd Brauch dcr Sitdslaven, p. 569 sq. (South 
Slavonians). 
1 Laws of Cnut, ii. 54. 
2 Uplands-Lagen, Aerfdae Balkaer, ch. 6. 
3 Bossu, op. cit. i. 308 sq. (Chactaw). James, op. cit. i. 243 
(Omaha). 
4 Frazer, Magic Art, ii. 107 sqq. Idem, Psyche's Task, p. 44 sqq. 
■' See Westermarck, Origin and Development of the Moral Ideas, 
ii. 417, 747 ; infra, i. 407. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
315 
it, but it presupposes that there is some primary reason for 
censuring adultery. 
The Southern Bambala in Congo maintain that adultery 
is generally the cause of the deaths of infants. Mr. Torday 
says that it is only when a child is expected that the marriage 
of a couple seems to become definite and conjugal fidelity 
becomes obligatory on both parties, as the child is otherwise 
supposed to die. Nay, even previous intercourse with 
a lover may kill a child born in wedlock ; when a married 
woman perceives that she is pregnant, she must confess all 
her former lovers to her husband, and if she forgets the name 
of a single one it is believed that the child will die.1 The 
Thonga in South-Eastern Africa fear terrible complications 
at the birth of a child when adultery has had something to 
do with it; a protracted and difficult birth at once proves 
that the child is not legitimate. But adultery may also, in 
their opinion, be dangerous to the husband. The whole 
punishment, says M. Junod, falls on the man, first, because 
he is a thief, and secondly, because of the so-called mathdana. 
" Tlulana means to jump over each other, to compete with 
each other. In the sexual domain it is said of two men 
who have relations with the same woman, ' They have 
met together in one life through the blood of that woman ; 
they have drunk from the same pool.' This establishes 
between them a most curious mutual dependence : should 
one of them be ill, the other must not visit him ; the patient 
might die. If he runs a thorn into his foot, the other must 
not help him to extract it. It is taboo. The wound would 
not heal. If he dies, his rival must not assist at his mourning 
or he would die himself. ... It is a frightful taboo ! "2 
Senor Guevara mentions the prevalence of a very similar 
belief among the Araucanians of Chili. They look upon 
adultery as a most heinous offence, both because it is theft 
and because it makes the husband's existence in a mysterious 
manner dependent upon that of the adulterer through the 
medium of the woman : if the adulterer falls ill or dies, the 
same may easily happen to the husband.3 
1 Torday, Camp and Tramp in African Wilds, p. 203. 
2 Junod, op. cit. i. 194 sq. 
3 Guevara, in Anales de la Universidad de Chile, cxxvii. 626 316 
THE HISTORY OF HUMAN MARRIAGE chap. 
We do not know how widely spread ideas of this sort are, 
but in the prohibition of adultery they could only have played 
a subsidiary part. Like the belief that adultery has an 
injurious effect on the harvest, so the belief that it has an 
injurious effect on the offspring could in fact never have 
arisen unless the act had been disapproved of. There can 
be no doubt that the customs and laws which prohibit 
the seduction of another man's wife and unfaithfulness on 
her part are in the first place due to the facts that a man 
desires to keep for himself a woman whom he has got 
possession of, and that the society of which he is a member 
sympathises with him and consequently feels resentment on 
his behalf if his possession is interfered with.1 In other 
words, the prohibition of adultery is mainly due to the 
jealousy of the injured husband and to the society's readiness 
to sympathise with it. Its readiness to do so depends, of 
course, upon the disposition for jealousy in its men in 
general-you cannot sympathise with a feeling which you 
are incapable of experiencing; and thus customs or laws 
forbidding adultery may be taken as evidence of the general 
prevalence of male jealousy in the societies where they are 
found. I do not say that this is the only factor which 
determines the society's attitude towards adultery ; the 
faithfulness which a wife owes her husband is also very 
closely connected with the power he possesses over her. 
But without jealousy there would be no crime of adultery 
at all. It may be contended that adultery is a crime against 
property, an offence against a man's honour, a breach of 
faith. But it would be neither the one nor the other if 
there were not that desire of exclusive possession which is 
the root of all jealousy. 
There may even be jealousy with reference to the past. 
We have seen that among many peoples a husband not only 
requires chastity from his wife, but demands that the 
woman whom he marries shall be a virgin ; and there can 
be little doubt that this demand partly owes its origin to the 
1 In my work on The Origin and Development of the Moral Ideas 
1 have endeavoured to show that this is the usual way in which 
individual rights originate. Sec especially vo'. i. ch. v. sq. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
317 
same powerful feeling as keeps watch over the purity of 
the marriage-bed. And a husband's claims may go farther 
still. He often demands that the woman he chooses for 
his wife shall belong to him, not during his lifetime only, 
but after his death. So strong is the idea of a wife belonging 
exclusively to her husband that among several peoples 
she has to die with him. He may want her in the other 
world ; or, in any case, she shall not be able to become the 
wife of another man.1 
Thus among the Comanche, when a man died his favourite 
wife was killed at the same time.2 In certain Californian 
tribes widows were sacrificed on the pyre with their deceased 
husbands; 3 and Mackenzie was told that this practice 
sometimes occurred among the Cree.4 When one of the 
Incas died, says Acosta, the woman whom he loved best, 
and also his servants and officers, were put to death, 
" that they might serve him in the other life."5 Among 
the Guarani widows were not killed, but they " used to throw 
themselves from some eminence high enough to cripple them 
for the remainder of their lives."6 In various African 
tribes widows were buried alive with their husbands or 
otherwise put to death7-according to Dr. Nassau, either 
that the dead might not be without companionship in the 
spirit world or as a punishment for not having cared better 
for him in the preservation of his life.8 In Dahomey, on 
the death of a king, a large number of his widows were 
put into the tomb living, with broken legs, and cautioned 
to treat their lord well.9 In Uganda there is a tradition 
that at one time some of the widows of kings and hereditary 
1 See Westermarck, Origin and Development of the Moral Ideas, 
i. 472. 2 Schoolcraft, op. cit. ii. 133. 
3 Ibid. iv. 226; v. 217. 4 Mackenzie, op. cit. p. xcviii. 
5 Acosta, Natural and Moral History of the Indies, ii. 313. 
0 Charlevoix, History of Paraguay, i. 204. 
7 Dennett, Folklore of the Fjort {French Congo), p. 20 sq. Poupon, 
' £tude ethnographique des Baya de la circonscription du M'Bimou,' 
in L' Anthropologic, xxvi. 128. Reade, Savage Africa, p. 359. Waitz, 
op. cit. ii. 192, 193, 419. 
8 Nassau, Fetichism in West Africa, p. 9. 
u Labarthe, quoted by Ling Roth, Great Benin, p. 41 n. 1. 318 
THE HISTORY OF HUMAN MARRIAGE chap. 
chiefs were tied up in the enclosures round their husbands' 
graves and starved to death.1 In various Polynesian and, 
especially, Melanesian islands wives often died with their 
husbands.2 In Fiji, for instance, they were either buried 
alive or strangled, often at their own desire, because they 
believed that in this way alone could they reach the realms 
of bliss, and that she who met her death with the greatest 
devotedness would become the favourite wife in the abode 
of spirits. On the other hand, a widow who did not permit 
herself to be killed was considered an adulteress.3 In 
Pentecost it has until recently been the custom to strangle 
all the widows of a chief at the death-feast. This custom 
is now dying out owing to the influence of the planters and 
missionaries ; but " strange to say, the women were not 
altogether pleased with this change, many desiring to die, 
for fear they might be haunted by the offended spirit of their 
husband. ' '4 We are told that in the New Hebrides a wife was 
strangled even when her husband was long absent from home.5 
Among the Tartars, according to Navarette, on a man's 
death one of his wives hanged herself " to bear him company 
in that journey."6 In China the funeral sacrifice of human 
beings was abolished even for members of the imperial 
family in the fourteenth century, but still maintains itself 
in a modified shape : widows often take their lives in order 
1 Felkin, ' Notes on the Waganda Tribe,' in Proceed. Roy. Soc. 
Edinburgh, xiii. 745. 
2 Codrington, Melanesians, p. 289 (natives of Aurora Island, 
New Hebrides). Ribbe, op. cit. p. 273 (natives of Rubiana, Solomon 
Islands). Polack, Manners and Customs of the Neiv Zealanders, 
i. 156 sqq. Best, ' Maori Marriage Customs,' in Trans, and Proceed. 
New Zealand Institute, xxxvi. 62. Waitz-Gerland, op. cit. vi. 130, 
640 sq. 
3 Wilkes, op. cit. iii. 96. Zimmermann, Die Inseln des indischen 
mid stillen Meeres, i. 359, 377. Seemann, Viti, pp. 192, 398. Wil- 
liams, Missionary Enterprises in the South Sea Islands, p. 557. 
Pritchard, Polynesian Reminiscences, p. 372. 
4 Speiser, op. cit. p. 118. 
5 Inglis, 1 Missionary Tour in the New Hebrides,' in Jour. Ethn. 
Soc. London,' iii. 63. 
G Navarette. ' Account of the Empire of China,' in Churchill, 
Collection of Voyages, i. 77. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
319 
to follow their husbands into the next world, and although 
it has been enacted that no official distinctions shall be 
awarded to such suttees, the old custom still meets with 
the same applause as ever, and many a woman is no doubt 
prevailed upon, or even compelled, by her own relations 
to become a suttee.1 In India widows were sacrificed 
until recently on the funeral piles of their husbands. 
It has been argued, it is true, that the burning of widows 
begins rather late in India.2 Yet, although the modern 
ordinance of suttee-burning be a corrupt departure from 
the early Brahmanic ritual, the practice seems to be, not 
a new invention by the later Hindu priesthood, but the 
revival of an ancient rite belonging originally to a period 
even earlier than the Veda.3 In the Vedic period the widow, 
in some cases at any rate, married her husband's younger 
brother ;4 but in the Atharva-Veda the practice of burning 
the widow on the funeral pile of her dead husband is referred 
to as an " ancient custom,"5 and in the Vedic ritual there 
are ceremonies which indicate the previous existence of 
such a rite.6 From Greece we have the instances of Evadne 
throwing herself into the funeral pile of her husband,7 and 
of the suicide of the three Messenian widows mentioned by 
Pausanias.8 Sacrifice of widows occurred, as it seems as 
a regular custom, among the Scandinavians,9 Heruli,10 and 
Slavonians.11 " The fact," says Mr. Ralston, " that in 
1 de Groot.. Religious System of China (vol. ii. book) i. 724, 735, 748, 
754. 2 Hopkins, Religions of India, p. 274. 
3 Zimmer, Altindisches Leben, p. 331. Tylor, Primitive Culture, 
i. 465 sqq. Risley, People of India, p. 182 sq. 
4 Zimmer, op. cit. p. 329. Macdonell and Keith, Vedic Index 
of Names and Subjects, i. 476 sq. 
5 Athavva- Veda, xviii. 3, 1, quoted by Zimmer, op. cit. p. 331. 
0 Rig-Veda, x. 18. 8 sq. Macdonell, Vedic Mythology, p. 165. 
Hillebrandt, ' Ei ne Miscelle aus dem Vedaritual,' in Zeitschr. d. 
Deutschen Morgenldnd. Gesellsch. xl. 711. Oldenberg, Religion des 
Veda, p. 587. > Euripides, Supplices, 1000 sqq. 
8 Pausanias, Descriptio Graeciae, iv. 2. 7. 
9 Grimm, Deutsche Rechtsalterthumer, p. 451. 
,,J Procopius, ii. 14. 
11 Dithmar of Merseburg, Chronicon, viii. 2 (Pertz, Monumenta 
Germaniae historica, v. 861). Zimmer, op. cit. p. 330. 320 
THE HISTORY OF HUMAN MARRIAGE chap. 
Slavonic lands, a thousand years ago, widows used to 
destroy themselves in order to accompany their dead 
husbands to the world of spirits, seems to rest on incon- 
testable evidence."1 
In other cases the dead husband is less exacting. Among 
the Takulli the widow is compelled by the kinsfolk of the 
deceased to lie on the funeral pile where the body of her 
husband is placed, whilst the fire is lighting, until the 
heat becomes unbearable. Then, after the body is con- 
sumed, she is obliged to collect the ashes and deposit them 
in a small basket, which she must always carry about with 
her for two or three years, during which time she is not at 
liberty to marry again.2 Among the Kutchin the widow, 
or widows, are bound to remain near the body for a year 
to protect it from animals, &c. ; and only when it is quite 
decayed and merely the bones remain are they permitted 
to remarry, " to dress their hair, and put on beads and 
other ornaments to attract admirers."3 Among the Mos- 
quito Indians " the widow was bound to supply the grave 
of her husband with provisions for a year, after which 
she took up the bones and carried them with her for another 
year, at last placing them upon the roof of her house, and 
then only was she allowed to marry again."4 Among the 
Pima Indians of Arizona, during the time of mourning- 
which rarely was observed for the full period of four years 
-the widow was compelled to remain at home, to refrain 
from washing her hair, and to cry aloud the name of the 
deceased every morning at daybreak.5 Among the Minas 
on the Slave Coast the widows are shut up for six months 
in the room where their husband is buried.6 Among the 
Nsakara north of the Mubangi, where the marriage tie is 
very loose and married women are most unchaste, chastity 
is expected of the widow-and even of the daughters and 
1 Ralston, Songs of the Russian People, p. 327 sq. 
2 Wilkes, op. cit. iv. 453. Cf. Richardson, op. cit. ii. 31. 
3 Hardisty, in Smithsonian Report, 1866, p. 319. 
4 Bancroft, op. cit. i. 731. 
5 Russell, ' Pima Indians,' in Ann. Rep. Bur. Ethn. xxvi. 195. 
0 Bouche, La C6te des Esclaves, p. 218. Cf. Ellis, Ewe-speaking 
Peoples of the Slave Coast, p. 160. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
321 
sisters-of dead chiefs of importance ; and they have to 
spend their lives maintaining a perpetual fire on the dwelling- 
tomb of the deceased potentate.1 A somewhat similar 
custom prevailed formerly in Uganda.2 Among the Betsi- 
misaraka of Madagascar the widow was obliged to lie with 
her dead husband's body during the time it was exposed.3 
In Pentecost, in the case of ordinary people, who are buried 
in their own houses, the widow often had to sleep beside 
the decaying body of her deceased husband for one hundred 
days.4 Among the Papuans in the neighbourhood of 
Finschhafen, in Kaiser Wilhelm Land, she lies down beside his 
grave and is covered up with a mat, and a roof of branches 
is erected over her ; here she has to mourn for three months.5 
Among the Kukis of the Chittagong Hills a widow was 
compelled to remain for a year beside the tomb of her 
dead husband, her family bringing her food.6 
Among various peoples custom requires that a widow 
shall never remarry. This is the rule in Tikopia,7 Rotuma,8 
and the Marquesas Islands,9 among the gentry of the Line 
Islanders of the Gilbert group,10 the Biduanda Kallang in 
Johor (occupying the extremity of the Malay Peninsula),11 
and the savages of Formosa,12; and it seems formerly to 
have been the rule among the Ainu of Japan.13 Among the 
Masai, too, widows may not remarry, but they are allowed 
1 Johnston, George Grenfell and the Congo, ii. 674. 
2 Ibid. ii. 674 n. 2. 
3 Grandidier, Ethnographic de Madagascar, ii. 240. 
4 Speiser, op. cit. p. 118 sq. 
5 Schellong, ' Weitere Mitteilungen iiber die Papuas (Jabim) 
der Gegend des Finschhafens,' in Zeitschr. f. Ethnol. xxxvii. 618, 
Cf. Neuhauss, op. cit. i. 168. 
G Rennel, quoted by Lewin, Wild Races of South-Eastern India, 
p. 280. 
7 Rivers, History of Melanesian Society, i. 310. 
8 Waitz-Gerland, op. cit. vol. v. pt. ii. 191. 9 Ibid. vi. 130. 
10 Tutuila, ' Line Islanders,' in Jour. Polynesian Soc. i. 267. 
" Logan, ' Biduanda Kallang of the River Pulai in Johore,' in 
Jour. Indian Archipelago, i. 300. 
12 Davidson, op. cit. p. 583. Muller, ' Ueber die Wildenstamme 
der Insel Formosa,' in Zeitschr. f. Ethnol. xlii. 230. 
13 Batchelor, op. cit. p. 565. 
VOL. I Y 322 
THE HISTORY OF HUMAN MARRIAGE chap. 
to live together with men belonging to the age-class of their 
deceased husband.1 Among the Baganda, " if a widow 
had children she did not remarry, but devoted herself to 
her children, and continued to Hve at the grave of her late 
husband."2 In ancient Peru, according to Garcilasso 
de la Vega, very few widows who had no children ever 
married again, and even widows who had children con- 
tinued to live single ; "for this virtue was much commended 
in their laws and ordinances. "3 Nor is it in China considered 
proper for a widow to contract a second marriage, and in 
genteel families such an event rarely, if ever, occurs ; indeed, 
a lady of rank who married a second time would thereby 
expose herself to a penalty of eighty blows.4 " As a faithful 
minister does not serve two lords, neither may a faithful 
woman marry a second husband "-this is to the Chinese a 
principle of life, a maxim generally received as gospel.5 
Among the Hindus the widows of the highest castes are for- 
bidden to marry again, and this custom shows signs of ex- 
tending itself beyond its present limits. So far, however, 
it has only a comparatively limited currency. Inquiries 
made at the end of the last century showed that out of a 
population of forty millions of Hindus no more than twenty- 
four per cent, prohibited widow marriage, while seventy-six 
per cent, permitted or even encouraged the practice.6 But 
for a woman of the Brahman caste the bare mention of a 
second marriage would be considered the greatest of insults, 
and if she married again, " she would be hunted out of society, 
and no decent person would venture at any time to have 
the slightest intercourse with her." 7 In Greece 8 and Rome 9 
1 Hinde, The Last of the Masai, p. 72. Merker, Die Masai, 
p. 47. 2 Roscoe, Baganda, p. 96. 
3 Garcilasso de la Vega, First Part of the Royal Commentaries of 
the Yncas, i. 305. 4 Gray, China, i. 215. 
6 de Groot, op. cit. (vol. ii. book) i. 745. 
0 Risley, op. cit. pp. 183, 187. Crooke, North-Western Provinces 
of India, p. 228 sq-. Idem, Tribes and Castes of the North-Western 
Provinces and Oudh, i. p. cxcii. Gait, Census of India, 1911, vol. i. 
(India) Report, p. 246. 
7 Dubois, Description of the Character, &c. of the People of India, p. 99. 
8 Pausanias, ii. 21. 7. 9 Rossbach, Rdmische Ehe, p. 262. ix CRITICISM OF HYPOTHESIS OF PROMISCUITY 
323 
a widow's remarriage was regarded as an insult to her 
former husband; and so it is still looked upon among the 
Southern Slavs.1 The early Christians, especially the 
Montanists and Novatians, strongly disapproved of second 
marriages by persons of either sex ;2 such a marriage was 
described by them as a " kind of fornication,"3 or as 
" specious adultery."4 
Much more commonly, however, the prohibition of re- 
marrying refers only to a certain period after the husband's 
death. When Mr. Batchelor first came to the Ainu, widows 
and widowers were supposed to remain single for five years.5 
Among the Omaha a widow was obliged to wait from four 
to seven years after the death of her husband before marry- 
ing again.6 The Angoni, a Zulu tribe, do not allow a 
widow to remarry until four years have elapsed.7 The 
Creeks looked upon a widow as an adulteress if she spoke 
or made free with any man within four summers after the 
death of her husband.8 Among the Chickasaw in Northern 
Mississippi widows were obliged to live a chaste life for 
three years at the risk of the law of adultery being executed 
against the recusants.8 Among some of the Kukis a widow 
may marry again after three years have elapsed from the 
death of her husband, but only with the consent of his 
relatives.10 The same period of widowhood is required by 
1 Krauss, op. cit. p. .578. Cf. Ralston, op. cit. p. 115 (Bulgarians). 
2 Mayer, Die Rechte der Israeliten, Athene? und Romer, ii. 290. 
Bingham, Works, vi. 427 sq.; viii. 13 sq. 
3 Tertullian, De exhortatione castitatis, 9 (Migne, Patrologia 
cursus, ii. 924). 
4 Athenagoras, Legatio pro Christianis, 33 (Migne, op. cit. Ser. 
Graeca, vi. 967). 
6 Batchelor, op. cit. p. 565 sq. 
6 Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethn. iii. 267. 
' Wiese, ' Beitrage zur Geschichte der Zulu im Norden der 
Zambesi,' in Zeitschr. f. Ethnol. xxxii. 192. 
8 Schoolcraft, op. cit. v. 269. 0 Adair, op. cit. p. 186. 
10 Butler, Travels and Adventures in the Province of Assam, 
p. 85. See also Stewart, ' Notes on Northern Cachar,' in Jour. Asiatic 
Soc. Bengal, xxiv. 621 (Old Kukis). According to Soppitt (Short 
Account of the Kuki-Lushai Tribes, p. 16), however, a Kuki widow 
is at liberty to remarry at once. 324 
THE HISTORY OF HUMAN MARRIAGE chap. 
the Gold, one of the Amoor tribes,1 and the Kunama in 
North-East Africa.2 The natives of Sarae,3 those of Bukoba 
in East Africa,4 and the Algonkin of Canada5 have fixed 
that period at two years; and among the last-mentioned 
people the new husband had to be approved of by the 
mother-in-law. In Madagascar the new marriage- was 
to be authorised by the parents-in-law, and usually they did 
not give their consent until at least a year had passed.6 
Among many uncivilised peoples the widow must lead 
a single life for a year at least after her husband's 
death,7 and among others for six8 or five9 or three10 or 
1 Laufer, ' Notes and Explorations among the Amoor Tribes,' 
in American Anthropologist, N.S. ii. 322. 
2 Munzinger, Ostafrikanische Studien, p. 488. 3 Ibid. p. 387. 
4 Richter, ' Der Bezirk Bukoba,' in Mittheil. Deutsch. Schutzgeb. 
xii. 86. 
5 Perrot, ' Memoir on the Manners, &c. of the Savages of North 
America,' in Blair, op. cit. i. 70. 
6 Grandidier, op. cit. ii. 239. Cf. Sibree, The Great African 
Island, p. 255. 
7 Hill Tout, ' Report on the Ethnology of the Stlatlumh of British 
Columbia,' in Jour. Anthr. Inst. xxxv. 133. Loskiel, History of the 
Mission of the United Brethren among the Indians in North America, 
i. 64 (Iroquois, &c.). Schoolcraft, op. cit. iii. 238 (Dakota). 
Powers, op. cit. p. 383 (Yokut of California). Falkner, op. cit. 
p. 119 (Patagonians). Poupon, in L'Anthropologic, xxvi. 128 
(Baya). Clozel and Villamur, op. cit. p. 104 (Baoul6 of the Ivory 
Coast). Meyer, quoted by Kohler, in Zeitschr. f. vergl. Rechtswiss. 
xiv. 305 (Herero). Minnie Martin, Basutoland, p. 88. Munzinger, op. 
cit. pp. 208 (Takue), 241 (Marea). Best, in Trans, and Proceed. 
New Zealand Institute, xxxvi. 63 (Maori, probably, if the husband 
was a person of importance). Finsch, Neu-Guinea, p. 82. Jenks, 
Bontoc Igorot, p. 70. Shakespear, Lushei Kuki Clans, p. 53 (Lushais, 
formerly), v. Siebold, Ethnologische Studien uber die Aino auf 
der Insel Yesso. p. 34; cf. Bird, Unbeaten Tracks in Japan, ii. 100, 
and Batchelor, op. cit. p. 565 sq. 
8 Lyon, Private Journal during the Voyage of Discovery under 
Captain Parry, p. 369 (Eskimo of Iglulik and Winter Island). 
Lahontan, p-p. cit. p. 459 (Canadian Indians). Ashe, Travels in 
America, p. 250 (Shawnee). Heriot, Travels through the Canadas, 
p. 325 (Californian Indians). Trenk, in Mittheil. Deutsch. Schutzgeb. 
xxiii. 168 (Namib Bushmen) ; cf. Kaufmann, ibid, xxiii. 157 (Auin). 
Maass, ' Durch Zentral-Sumatra,' in Zeitschr. f. Ethnol. xli. 160. 
9 Dunbar, 'Pawnee Indians,' in Magazine of American. History, 
viii. 735. 10 Shakespear, op. cit. p. 53 (Lushais. nowadays). IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
325 
two1 months. Various American tribes required that the 
head of the widow should be shaved or her hair cut off, 
and that she should remain unmarried until her locks 
regained their wonted length.2 
We know that among some peoples, as the Chinese and 
Hindus, the prohibition of remarriage has succeeded an 
earlier custom which demanded that the widow should die 
with her husband ; but such a prohibition is by itself no 
evidence of former widow sacrifice. Nor must we assume 
that the interdict against remarriage for a certain period 
is generally a survival of compulsory widowhood for life, 
although we know that it sometimes is so3 and that in other 
cases the period in question has been shortened.4 The 
same cause may in some instances lead to an absolute pro- 
hibition of remarrying and in other instances to a prohibition 
for a limited period only. 
We are told that among the Omaha a widow was obliged 
to wait for several years before she could remarry, partly 
in order to show the proper respect to the memory of her 
deceased husband, but partly also " to enable her to wean 
her infant, if she had one by him, before she became enceinte 
by her next husband " ;5 and another explanation, referring 
to compulsory widowhood for a shorter time, may be that 
the prohibition serves the object of removing all apprehen- 
sions as to pregnancy. But this explanation is, at any 
rate, insufficient, even though the widowhood last only for 
a few months. In Ruanda, where a widow is strictly prohi- 
bited from sexual intercourse during her two months of 
mourning, it is believed that if she does not observe this 
1 Schumacher, ' Das Eherecht in Ruanda,' in Anthropos, vii. 
8. Parkinson, Dreissig Jahre in dev Sudsee, p. 396 (Moanus of the 
Admiralty Islands). 
2 Lo d, The Naturalist in Vancouver Island and British Columbia. 
ii. 235. Catlin, Illustrations of the Manners, &c. of the North American 
Indians, i. 95 (Mandan), de Bry, Narrative of Le Moyne, Descrip- 
tion of the Illustrations, p. 8 (Florida Indians). Schomburgk, 
Reisen in Britisch-Guiana, i. 227. 
3 Batchelor, op. cit. p. 565 (Ainu). 
4 Dorsey, loc. cit. p. 267 sq. (Omaha). Shakespear, op. cit. p. 53 
(Lushais). ' 
6 Dorsey, loc. ci . p. 267 sq. 326 
THE HISTORY OF HUMAN MARRIAGE chap. 
rule her fingers and nose will fall off j1 and the Namib 
Bushmen have the idea that a widow who marries soon 
after her husband's death will also die.2 There can be no 
doubt that the rule which among many uncivilised peoples 
compels a widow to remain single either for the rest of her 
life or for a certain time is chiefly founded on superstition.3 
She is held to be defiled. The Thonga believe, according 
to M. Junod, that " when a man dies all his relatives are 
contaminated by the defilement of death. . . . There are 
concentric circles around him, some people being more 
affected than others. The wives form the first circle, 
especially the first wife."4 But the* widow is not only pol- 
luted by the contagion of death :5 she is haunted by her 
dead husband, who jealously watches her behaviour, ready 
to punish her if she swerves from the duty he still thinks 
she owes him. Among the Thompson River Indians of 
British Columbia a widow sometimes wore a breech cloth 
made of dry bunch-grass for several days, that the ghost of 
the husband should not have connection with her.6 The 
Iroquois and other Indians connected with them, who 
prohibited a widow from remarrying within a year after 
the death of her husband, said that he did not forsake her 
before that time, but that then his soul went to the mansions 
of departed spirits.7 The Igorot of Bontoc in Northern 
Luzon believe that if either a widow or a widower marries 
before a year has elapsed, she or he will be killed "by an 
anito whose business it is to punish such sacrilege " ;8 and 
we may guess that the anito referred to originally was the 
dead person's soul. Among the Maori " widows were not 
permitted to marry until their dead husband's bones were 
1 Schumacher, in Anthropos, vii. 8. 
2 Trenk, in Mittheil. Deutsch. Schutzgeb. xxiii. 168. 
3 Cf. R^vesz, Das Trauerjahr der Witwe, passim. 
4 Junod, op. cit. i. 197. 
5 On the contagion of death see Crawley, Mystic Rose, p. 95 
sqq. ; Westermarck, Origin and Development of the Moral Ideas, 
ii. 536 sqq. 
0 Teit, ' Thompson Indians of British Columbia,' in Memoirs of 
the American Museum of Natural History, Anthropology, i. 333. 
7 Loskiel, op. cit. i. 64. 8 Jenks, op. cit. p. 70. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
327 
taken to their final resting-place."1 This rule evidently 
owes its origin to an idea similar to that which prevents a 
widow among the Syntengs of Assam from remarrying 
during the time she keeps her dead husband's bones : they 
believe that " as long as the bones remain in the widow's 
keeping, the spirit of her husband is still with her. ' '2 Among 
some of the Nagas marriage with the deceased husband's 
brother is- not permitted till the expiry of the general 
funeral rite following after the death of the husband ; and 
as the purpose of this rite is in part to lay to rest the ghosts 
of those who have died in the year, the reason for the 
restriction is obvious.3 A Dyak widow " is regarded as 
belonging to her deceased husband until she is formally 
freed from him by the feast of the Sungkup."^ In Aus- 
tralian tribes a widow has to perform many formalities 
before she can remarry ;5 and " the spirit of the dead man 
was supposed to have been watching all these proceedings 
as he lay at the bottom of the grave."6 
The remarriage of a widow may, indeed, be dangerous 
not only to her but to the new husband. Among the 
Baganda, " when a man wished to marry a widow, he first 
paid the deceased husband a barkcloth and a fowl, which 
he put into the little shrine at the grave; in this way he 
imagine^ he could pacify the ghost."7 Of the Stlatlumh 
in British Columbia we are told that " should a widow 
marry shortly after the death of her former husband without 
going through a course of ceremonial cleansing, it was 
believed that her second or subsequent husband's life would 
be very short."8 Among the Kamchadal nobody would 
marry a widow before an outsider had had sexual inter- 
1 Thomson, Story of New Zealand, i. 178. 
2 Gurdon, Khasis, p. 76 sq. 
3 Hodson, Naga Tribes of Manipur, p. 95 sq. 
4 Brooke Low, quoted by Ling Roth, Natives of Sarawak and 
British North Borneo, i. 130. 
6 Spencer and Gillen, Native Tribes of Central Australia, p. 500 
sqq. Howitt, Native Tribes of South-East Australia, p. 248. 
Spencer and Gillen, op. cit. p. 507. 
7 Roscoe, Baganda, p. 97. 
8 Hill Tout, in Jour. Anthr. Inst. xxxv. 139. 328 
THE HISTORY OF HUMAN MARRIAGE chap. 
course with her, which was called " removing the sin from 
her," since otherwise it was thought that the new husband 
would also die. As Dr. Jochelson remarks, she was 
thereby evidently freed from the old union, and her 
new husband could take her to his own family hearth 
without incurring vengeance on the part of her first hus- 
band's spirit. This explanation is favoured by the fact 
that a man might take his deceased brother's widow without 
any ceremonies. The person who would undertake to 
" remove the sin " from the woman was paid for this service ; 
and prior to the coming of the Cossacks it was difficult to 
find among the Kamchadal men who would volunteer for 
this undertaking, which, according to their belief, was 
fraught with danger, and was looked upon as " very dis- 
honourable for the man."1 So also among the Akamba, 
east of the Victoria Nyanza, " in the event of a widow not 
going to her brother-in-law, but to a stranger, she must 
first go through a ritual coitus with another elderly man 
(mutumid), otherwise her prospective husband's earlier 
wives will become barren, or her children will die."2 Similar 
customs have been found among the tribes near Fort 
Johnston in British Central Africa3 and the Baronga, farther 
south.4 Among the Maori the brother of the dead husband 
had to " free from tapu " the latter's bed, for both bed and 
widow would be tapu ; and the new husband often assumed 
the name of the dead brother, discarding his former one. 
After the widow had thus married her brother-in-law, she 
might be divorced from him by the priest if she took a 
dislike to him, and could then marry any one she pleased, 
1 Steller, Beschreibung von dem Lande Kamtschatka, p. 346 sq. 
Krasheninnikoff, History of Kamschatka, p. 214 sq. Jochelson, 
Koryak, p. 752. 
2 Lindblom, Akamba, p. 82. See also Hildebrandt, ' Ethno- 
graphische Notizen uber Wakamba und ihre Nachbaren,' in 
Zeitschr. f. Ethnol. x. 402. 
3 Stannus, ' Notes on some Tribes of British Central Africa,' in 
Jour. Roy. Anthr. Inst. xl. 316. 
4 Junod, ' Les conceptions physiologiques des bantou sud- 
africains et leurs tabous,' in Revue d'ethnographie et de sociologie, 
i. 163. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
329 
because she had become free from tapu by first marrying 
the brother of her former husband.1 
That the prohibition of the remarriage of a widow even 
for a limited period has little, if anything, to do with the 
question of pregnancy is indicated by the fact that a widower 
is likewise very frequently compelled to live single for a 
certain length of time.2 Among some peoples he is supposed 
to do so for the rest of his life.8 The period of compulsory 
celibacy is in many cases the same for widowers and widows, 
but in other cases it is shorter for the former. Thus in 
Bukoba, in East Africa, it lasts one year for a widower, but 
two for a widow.4 Butler states that among the Kukis 
the widower can marry again in the second year after the 
wife's death, but the widow only when three years have 
elapsed after the death of her husband.5 According to 
Mr. Soppitt, however, another rule prevails among the 
Kukis, or some of them, namely, that a widower is not 
1 Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 63. 
2 Cranz, op. cit. i. 148 (Greenlanders). Lyon, op. cit. p. 369 
(Eskimo of Iglulik and Winter Island). Bancroft, op. cit. i. 93 
n. 133 (Aleut). Schoolcraft, op. cit. iii. 238 (Dakota); v. 655 
(Indians of Oregon). Teit, loc. cit. p. 334 (Thompson River 
Indians). Lahontan, op. cit. p. 459 ; Perrot, loc. cit. i. 73 (Canadian 
Indians). Powers, op. cit. p. 383 (Yokut). Dorsey, loc. cit. p. 268 
(Omaha). Ashe, op. cit. p. 250 (Shawnee). Waitz, op. cit. iv. 367 
(Chibchas). v. Martius, op. cit. i. 649 ; Appun, ' Die Indian er von 
Britisch-Guayana,' in Das Ausland, xliv. 447 (Macusis). Meyer, 
quoted by Kohler, in Zeitschr. vergl. Rechtswiss. xiv. 305 (Herero). 
Trenk, in Mittheil. Deutsch. Schutzgeb. xxiii. 168 (Namib Bushmen). 
Neuhauss, op. cit. i. 168 (Papuans of Kaiser Wilhelm Land). Blumen- 
tritt, Versuch einer Ethnographic der Philippinen, p. 28 ; Meyer, ' Die 
Igorroten von Luzon,' in Verhandl. Berliner Gesellsch. Anthrop. 
1883, p. 385 ; Jenks, op. cit. p. 70 (Igorot of Luzon). Stewart, in 
Jour. Asiatic Soc. Bengal, xxiv. 620 (Old Kukis). Shakespear, 
op. cit. p. 53 (Lushais). Laufer, in American Anthropologist, N.S. 
ii. 322 (Gold, an Amoor tribe), v. Siebold, op. cit. p. 34; 
Bickmore, ' Notes on the Ainos,' in Trans. Ethn. Soc. N.S. 
vii. 20 ; Batchelor, op. cit. p. 566 ; Bird, op. cit. ii. 100 (Ainu). 
3 Logan, in Jour. Indian Archipelago, i. 300 (Biduanda Kallang 
in Johor). Muller, in Zeitschr. J. Ethnol. xlii. 230 (savages of For- 
mosa). Batchelor, op. cit. p. 565 (Ainu, formerly). 
4 Richter, in Mittheil. Deutsch. Schutzgeb. xii. 86. 
6 Butler, op. cit. p. 85. 330 
THE HISTORY OF HUMAN MARRIAGE chap. 
permitted to remarry within the period of three years, 
whereas a widow is at liberty to remarry at once.1 
The prohibitions relating to a widower have obviously 
a similar origin to those, relating to a widow. He is haunted 
by his dead wife, and he is contaminated with the defilement 
of death. Among the Lillooet Indians in the south-western 
interior of British Columbia both widowers and widows 
had to perform many purificatory ceremonies, and a number 
of restrictions were placed on them; and if they did not 
keep all the observances, they were supposed to become 
lame or crippled in some part of the body.2 Among the 
Thompson River Indians the widower, on the fourth day 
after the death of his wife, tied buckskin thongs round the 
right ankle, knee, and wrist, and round the neck-evidently 
in order to keep away the ghost of his wife-and he 
should not marry until the thongs had fallen off.3 Among 
the Stlatlumh a young widower must'be careful to refrain 
from sexual intercourse for a year, the more particularly 
if he possessed esoteric or mystery powers ; and it was not 
unusual for such a person to go apart into the forest by 
himself for that period and purify himself from the death 
defilement and seek mystery powers.4 Moreover, as the 
remarriage of a widow may be dangerous to her new husband, 
so the remarriage of a widower may also be dangerous to 
his new wife. Among the tribes near Fort Johnston, 
" after the death of a wife the husband mourns five or six 
months and is then given medicine, after which he may 
marry again, and without which, should he marry, his new 
wife would die."5 Among the Hindus of the Punjab there 
is a belief that the jealousy of the spirit of the first wife 
is instrumental in causing the death of the subsequent 
wives. Hence when a widower has to marry a second time, 
a miniature picture of the first wife, either cased in silver 
1 Soppitt, op. cit. p. x6. 
2 Teit, ' Lillooet Indians,' in Publications of the Jesup North Pacific 
Expedition, ii. 271. 
3 Idem, ' Thompson Indians,' in Mem. American Mus. Nat. 
History, Anthropology, i. 333 sq. 
4 Hill Tout, in Jour. Anthr. Inst. xxxv. 139. 
6 Stannus, ibid. xl. 315 sq. ix CRITICISM OF HYPOTHESIS OF PROMISCUITY 
331 
or gold or engraved on a silver or gold plate, is hung round the 
neck of the bride at the wedding ceremonies. When a 
picture cannot be obtained or engraved, the name of the 
deceased wife is substituted for the picture. "The idea seems 
to be to humour the spirit of the first wife, by proving the 
fidelity of the husband, who in marrying the second wife 
pretends to really marry the picture or name of the deceased 
wife, thus identifying the second wife with the first."1 
It is worth noticing that the prohibition of the remarriage 
of a widower, either for the rest of his life or for some 
definite period, is chiefly found among peoples who are 
strictly monogamous or practise polygyny only as a rare 
exception. 
There are certainly savage peoples who have been said 
to be almost devoid of jealousy. This has often been 
inferred from the practice of lending, exchanging, or prosti- 
tuting wives. But such an inference cannot be accepted 
if sexual jealousy simply means, as I take it to mean, the 
angry feeling which is aroused by the loss, or the fear of the 
loss, of the exclusive possession of an individual who is the 
object of one's sexual desire. First, who can exactly tell 
what a savage feels when he, for reasons stated above 
or simply in compliance with the customs of his people, 
lends his wife to a visitor ? And secondly, a person may 
have a tendency, and even a strong tendency, to experience 
a certain feeling although it on some special occasions fails 
to make its voice heard. I have nowhere received more 
profuse, and apparently more ungrudging, hospitality 
than in Jewish homes in Morocco, but I should be greatly 
mistaken if I took this for evidence of any lack of frugality, 
or even parsimony, on the part of my hosts. A desire may 
be silenced by another desire, which for the moment at 
least is the stronger of the two. A man may, generally 
1 Pandit Harikishan Kaul, Census of India, 1911, vol. xiv. (Pun- 
jab) Report, p. 283 sq. Instances of haunted widows and widowers 
are also given by Hartland, in his essay ' The Haunted Widow ' 
(in Ritual and Belief, p. 194$^.), and by Frazer, Folk-Lore in the Old 
Testament, i. 523 sqq. See also infra on Marriage Rites. 332 
THE HISTORY OF HUMAN MARRIAGE chap. 
speaking, be quite jealous of his wife and yet, under the 
influence of love of money, prostitute her to a stranger or, 
as negro husbands sometimes do 1 use her for entrapping 
other men and making them pay a heavy fine. Moreover, 
sexual jealousy presupposes some degree of sexual love, 
and neither a savage nor a civilised man always loves his 
wife. As we have seen, one reason why a man exchanges 
his wife for that of another man is that he has got thoroughly 
tired of her. 
By all this I certainly do not mean to deny that there 
are peoples who are remarkably little addicted to jealousy. 
This is said of many who practise polyandry.2 But these 
peoples cannot on that account be quoted in support of the 
theory of promiscuity, since polyandry, as will be seen, 
owes its origin to specific causes which never would have 
produced general communism in women. It is also in the 
present connection important to note that polyandry has 
been mainly found, not among savages of the lowest type, 
but among peoples who have flocks and herds or who practise 
agriculture and some of whom cannot be called savages 
at all. These facts do not speak in favour of McLennan's 
statement that " polyandry must be regarded as a modifi- 
cation of and advance from promiscuity."3 Nor could the 
practice of prostituting wives in any case be taken as 
evidence of primitive lack of jealousy. It has certainly not 
the flavour of primitiveness, and among many peoples it is 
known to have arisen through their contact with a " higher 
culture." In a book on the Tasmanians Mr. Bonwick 
observes that " husbands, after the degradation of a pseudo- 
civilisation, are sometimes found ready to barter the virtue 
of a wife for a piece of tobacco, a morsel of bread, or a 
silver sixpence." 4 Curr observes that among the Aus- 
tralian natives " husbands display much less jealousy of 
'white men than of those of their own colour," and that 
1 Reade, Savage Africa, p. 44. ' Negersitten,' in Das Ausland, 
liv. 1028. Torday and Joyce, Les Bushongo, p. m sq. 
2 Infra, on Polyandry. 
3 McLennan, Studies in Ancient History, p. 93 sq. 
4 Bonwick, The Last of the Tasmanians, p. 308. IX CRITICISM OF HYPOTHESIS OF PROMISCUITY 
333 
they will more commonly prostitute their wives to strangers 
visiting the tribe than to their own people.1 " Under no 
circumstances," says Sir George Grey, "is a strange native 
allowed to approach the fire of a married man."2 According 
to Bosman, the Negroes of Benin were very jealous of their 
wives with their own countrymen, though not in the least 
with European foreigners ;3 and Lisiansky states exactly 
the same as regards the Sandwich Islanders.4 In California, 
says Powers, " since the advent of the Americans the 
husband often traffics in his wife's honour for gain, and even 
forces her to infamy when unwilling ; though in early days 
he would have slain her without pity and without remorse 
for the same offence."5 The like is true of the Columbians 
about Puget Sound.6 Georgi remarks that the nomadic 
Koryak torment their wives by their jealousy, sometimes 
even killing them from this passion ; whereas those Koryak 
who lead a stationary life, being far more advanced in 
civilisation, are so little addicted to it, that they even have 
a relish for seeing foreigners make love to their wives, whom 
they dress accordingly.7 Indeed, if the hypothesis of an 
animal pairing time in the infancy of mankind holds good, 
we may assume that jealousy at that stage was no less 
intense than among other mammals. 
Dr. Hartland maintains that the view taken by a large 
number of savages in different parts of the world of what 
we should call serious offences against virtue must lead the 
student seriously to ask whether the masculine passion of 
jealousy can be as fundamental and primitive as it is some- 
times asserted to be. He believes that the answer must 
be in the negative, and that certain hypothetical recon- 
structions of the history of marriage will in consequence 
1 Curr, Australian Race, i. no. Cf. Lumholtz, Among Cannibals, 
P- 345 sq. 
2 Grey, op. cit. ii. 252 sq. 
3 Bosman, loc. cit. p. 525. 1 Lisiansky, op. cit. p. 128. 
5 Powers, op. cit. p. 413. 6 Bancroft, op. cit. i. 218. 
' Georgi, op. cit. p. 349. See also Sarytschew, ' Account of 
a Voyage of Discovery to the North-East of Siberia,' in Collection 
of Modern and Contemporary Voyages and Travels, vi. 76 sq. (Aleut) ; 
Bogoras, Chukchee, p. 607. 334 
THE HISTORY OF HUMAN MARRIAGE chap. 
need reconsideration.1 In his long chapter on marital 
jealousy, however, I find nothing to support this view, 
whereas we have an overwhelming array of facts which 
seem to me to be entirely at variance with it. Nor can I 
in the least agree with the suggestion sometimes made that, 
however jealous man originally may have been, his gregarious 
way of living made promiscuity a necessity. The men of a 
group, it is argued, must either have quarrelled about 
their women and separated, splitting the horde into hostile 
sections, or indulged in promiscuous intercourse. But why 
should tribal organisation in olden times have prevented 
a man from having his special wife as it does not do so 
among savages still existing ? There must always have been 
in the group stronger or more influential men and weaker or 
less influential men, and the former-whether their superi- 
ority depended upon their physical strength or their skill 
in the magic art or any other quality-would naturally 
have appropriated the most comely women and kept them 
for themselves. So it is nowadays, and I can see no reason 
to suppose that it ever was otherwise. 
I even venture to go a step further and suggest that 
masculine jealousy has not only been a powerful obstacle 
to promiscuity, but that its peculiar violence may possibly 
serve the very purpose of preventing it. The murderous 
rage of the male which tolerates no rivals may be the result 
of natural selection if promiscuous intercourse is, as it is 
often supposed to be, unfavourable to fecundity. It is a 
well-known fact that prostitutes very seldom have children, 
except in the beginning of their career.2 This has been 
attributed to various causes-such as ablutions, injections, 
abortions, and venereal disease ; but an explanation which, 
however hypothetical, is of interest in the present con- 
nection is that excessive sexual intercourse leads to a 
pathological state of the vagina which is injurious to 
1 Hartland, op. cit. ii. 242 sq. 
2 See, e.g., Parent-Duchatelet, La prostitution dans la ville de Paris, 
i. 218 ; Mireur, La prostitution a Marseille, p. 177 sqq. ; Reuss, 
La prostitution au point du vue de I'hygi&ne et de I'administration, 
p. 221 sq. ix CRITICISM OF HYPOTHESIS OF PROMISCUITY 
335 
fecundation.1 Moreover, it is conceivable that the sper- 
matozoa of different individuals have a counteracting effect 
upon each other; this, as I am told by my Mend 
Dr. Harry Federley, has been found to be the case 
with the spermatozoa of males belonging to different 
animal species, whereas I am not aware of any experi- 
ments made with males of the same species. It has 
been noticed that prostitutes who give up their profession 
and marry very often become mothers.2 And Dr. Carpenter 
refers to the efforts of the American planters to form the 
negroes into families, as the promiscuity into which they 
were liable to fall produced infertility, and fertility had 
become important to the slave-owners through the pro- 
hibition of the slave-trade.8 The saying, "Il ne pousse 
pas d'herbe dans les chemins ou tout le monde passe,"4 
may thus perhaps express a general physiological truth 
which helps us to understand the peculiar violence of 
masculine jealousy. 
It may possibly be argued that if there is any truth in 
these suppositions the same evil effects as are ascribed to 
promiscuity ought also to result from polyandry. Now 
the two cases are not quite similar. Among many poly- 
androus peoples, at least, the various husbands live or 
cohabit with their common wife in turn, or more than one 
of them are seldom at home at the same time. Neverthe- 
less various statements relating to polyandrous peoples 
rather support the view that intercourse between one 
woman and several men is unfavourable to reproduction. 
In Tibet, where polyandry is common, the families are, 
according to Rockhill, generally small, containing three or 
four children, but more generally two.5 Mr. Sherring 
makes a very similar statement, and adds that " it has been 
noticed in our own hills that where polyandry has existed 
1 For an original suggestion of this kind, see Mireur, op. cit. 
p. 182 sqq. 2 Reuss, op. cit. p. 222. 
3 Carpenter, quoted by Maine, Dissertations on Early Law and 
Custom, p. 204 sq. note. 
4 Bertillon, quoted by Witkowski, La qMration humaine, p. 218. 
6 Rockhill, Land of the Lamas, pp. 213, 212. 336 
THE HISTORY OF HUMAN MARRIAGE ch. ix. 
the result has been small families."1 Drew2 and Knight3 
observe that in Ladakh polyandry has a striking tendency 
to keep down the population in more than one way : " not 
only are fewer families founded than would be otherwise, 
but the families are smaller," and the children are far be- 
tween. Baierlein noticed the great scarcity of children 
among the Todas.4 Concerning the Herero, among whom 
two men enter into a peculiar relation which gives each 
of them the right of access to the other one's wives, Dr. 
Brincker states that this custom has led to the result that 
the pagan Herero women bear comparatively few children.5 
Dr. Tautain expresses the opinion that a cause of the low 
birth-rate among the Marquesas Islanders is "la liberte 
extreme des mceurs pouss^e au point qu'elle ressemble a 
l'£tat de communaut^ des femmes."6 Among the Ahirs in 
the North-Western Provinces of India, on the other hand, 
Mr. Crooke was struck by the fact that the women who 
practised fraternal polyandry gave birth to as many children 
as those who had but a single husband ; but he was informed 
that " the visits of the brothers were not so frequent as to 
produce any effect of this kind."7 
It is not, of course, impossible that among some peoples 
the intercourse between the sexes may have been almost 
promiscuous. But the hypothesis according to which pro- 
miscuity has formed a general stage in the social history 
of mankind, instead of belonging-as Giraud-Teulon puts 
it-to the class of hypotheses which are scientifically per- 
missible,8 is in my opinion one of the most unscientific 
ever set forth within the whole domain of sociological 
speculation. 
1 Sherring, Western Tibet and the British Borderland, p. 88. 
2 Drew, The Jummoo and Kashmir Territories, p. 251. 
3 Knight, Where Three Empires meet, p. 140. 
4 Baierlein, Nach und aus Indien, p. 250. 
6 Brincker, ' Character, Sitten und Gebrauche speciell der Bantu 
Deutsch-Sudwestafrikas,' in Mittheil. d. Seminars f. orientalische 
Sprachen, iii. 3. 86. 
6 Tautain, ' £tude sur la depopulation de 1'archipel des Marquises,' 
in L' Anthropologic, ix. 420, 421, 423. 
7 Crooke, Tribes and Castes of the North-Western Provinces, ii. 445. 
8 Giraud-Teulon, Les origines du mariage et de la famiUe, p. 70. CHAPTER X 
THE FREQUENCY OF MARRIAGE AND THE MARRIAGE AGE 
Among the uncivilised races of men marriage not only 
exists but is much more frequent than among ourselves. 
As a general rule, nearly every man endeavours to marry 
when he has reached the age of puberty, and practically 
every woman gets married. Among some savages priests 
or magicians have to remain single ; and there are inverts, 
male and female, who avoid sexual intercourse with the 
opposite sex. With the former I shall deal in the next 
chapter, of the latter I have spoken in another work.1 I 
shall now give some instances of what, with variations in 
details, undoubtedly may be said to be the rule among the 
vast majority of savage tribes. 
Mr. Bridges wrote to me that among the Yahgans of 
Tierra del Fuego none but mutes and imbeciles remained 
single, except some lads of vigour who did so from choice, 
influenced by licentiousness. But " no woman remained 
unmarried ; almost immediately on her husband's death 
the widow found another husband." So also among the 
Lengua Indians of the Paraguayan Chaco there is no such 
thing as a spinster, and even a widow rarely remains single, 
unless she be old ; " their mode of life and state of society 
are such that every girl must have a recognised protector." 2 
Dr. Nordenskiold never heard of the existence of any elderly 
unmarried women among the Choroti or in any other 
Chaco tribe.3 The Charruas of Paraguay, according to 
1 Origin and Development of the Moral Ideas, vol. ii. ch. xliii., 
' Homosexual Love.' 
2 Grubb, An Unknown People in an Unknown Land, p. 214 sq. 
8 Nordenskiold, Indianliv i El Gran Chaco (Syd-Amerika), p. 88. 
337 338 
THE HISTORY OF HUMAN MARRIAGE chap. 
Azara, " ne restent jamais dans le celibat, et ils se marient 
aussitot qu'ils sentent le besoin de cette union." 1 Among 
the Mosquito Indians on the western shores of the Caribbean 
Sea " there are very rarely unmarried girls. If there are 
any spare girls, the principal men take two more wives, so 
that the supply is never in excess of the requirements, and 
all are happy. ... To be without a wife is not only an 
ignominious, but a most distressing plight for an Indian."2 
Mr. Russell writes of the Pima Indians of Arizona :-" Occa- 
sionally a man possessed such a character that no woman 
would marry him, and more rarely a woman would remain 
unmarried. There is one such at Casa Blanca and one at 
Blackwater at the present time, the latter being regarded 
as possessed of supernatural powers because of her spinster- 
hood."3 Harmon found that among the Blackfeet, Cree, 
Chipewyan, and other tribes on the eastern side of the 
Rocky Mountains celibacy was a rare exception;4 and 
Ashe noticed the same among the Shawnee of the Mississippi.5 
Prescott says of the Dakota : -"I do not know of a bachelor 
among them. They have a little more respect for the women 
and themselves, than to live a single life."6 According to 
Adair, many Indian women thought virginity and widow- 
hood the same as death.7 Among the Eskimo round Repulse 
Bay " every woman is married as soon as she arrives at a 
marriageable age, and whenever a man dies his wife is 
taken by some one else." 8 Lieutenant Holm found only 
one unmarried woman among the Eastern Greenlanders 
visited by him, and she was the mother of two children.9 
1 Azara, Voyages dans l'Amerique meridionale, ii. 21. 
2 Bell, Tangweera, p. 261 sq. 
3 Russell, ' Pima Indians,' in Ann. Rep. Bur. Ethnol. xxvi. 184. 
4 Harmon, Journal of Voyages and Travels in the Interior of North 
America, p. 339. 
5 Ashe, Travels in America, p. 250. 
6 Prescott, in Schoolcraft, Information respecting the History, &c. 
of the Indian Tribes of the United States, iii. 238. 
7 Adair, History of the American Indians, p. 187. 
®, Gilder, Schwatka's Search, p. 246. 
9 Holm, ' Konebaads-Expeditionen til Gr0nlands 0stkyst 1883-85/ 
in Geografisk Tidskrift, viii. 91. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
339 
Among the Reindeer Chukchee in the north-western 
extremity of Asia, according to Dr. Bogoras, no man can 
live a tolerable life without a separate house of his own and 
a woman to take care of it; and another motive for marriage 
is the idea of the necessity of continuing the family line. 
On the other hand, among the Maritime portion of the tribe 
marriage is not so indispensable, and the unmarried state 
is consequently somewhat more common.1 
Among the Kafirs of the Hindu-Kush " a young woman 
who remains unmarried must be a hopelessly bad character."2 
Mr. Sherring observes that among the Bhotias of British 
Garhwal, in the Himalayas, everybody, man and woman, is 
married, whereas among those of Darma Pargana many 
unmarried persons are to be found in every village, "the 
reason being that there marriage depends upon the will of 
the parties, who are always of mature age at the time of the 
marriage contract ; and instances are not uncommon of 
men and women who have remained unmarried all their 
lives, because nobody would marry them." 3 Among the 
Santals of Bengal " nobody but an idiot remains a celibate." 4 
Among the Toungtha of the Chittagong Hills it is unheard 
of for a man or woman to be unmarried after the age of 
thirty, and among the Chukmas living in the same region a 
bachelor twenty-five years old is rarely seen.5 Concerning 
the Todas of the Nilgiri Hills Mr. Marshall writes :-" Every 
man and every woman, every lad and every girl is somebody's 
husband or wife ; tied at the earliest possible age. . . . 
With the exception of a cripple girl, and of those women who, 
past the child-bearing age, were widows, I did not meet 
with a single instance of unmarried adult females." 6 The 
Kachins of Burma, according to Father Gilhodes, seem to 
have no idea of voluntary celibacy. It is an honour for 
1 Bogoras, Chukchee, p. 569 sq. 
2 Scott Robertson, Kafirs of the Hindu-Kush, p. 534. 
3 Sherring, ' Notes on the Bhotias of Almora and British Garhwal, ' 
in Memoirs Asiatic Soc. Bengal, i. 98, 106. 
4 Macphail, ' Cycle of the Seasons in a Santal Village,' in Calcutta 
Review, N.S. i. 159. 
5 Lewin, Wild Races of South-Eastern India, pp. 193, 175. 
a Marshall. A Phrenologist amongst the Todas, p. 220. 340 
THE HISTORY OF HUMAN MARRIAGE chap. 
every one of them to marry and have children, and a shame 
to die without offspring. Yet there are among them a few 
bachelors and old maids, mostly semi-idiots or persons of 
an impossible character.1 
Among the Hill and Sea Dyaks of Borneo old maids and 
bachelors are said to be unknown or almost so.2 Marsden 
writes of the inhabitants of the districts in Sumatra which 
were under his charge that he does not conceive it would be 
possible to find among a population amounting to about 
eight thousand persons even ten instances of men of the 
age of thirty who were unmarried.3 Hagen observes that 
among the Orang Kubu, an aboriginal tribe of the same 
island, old maids are as absent as among the Malays ; even 
the ugliest woman finds a husband because there are more 
men than women among them.4 In Java Crawfurd " never 
saw a woman of two-and-twenty that was not, or had not 
been, married." 5 Mr. Jenks states that among the Igorot 
of Bontoc, in Northern Luzon, there were only two men 
who had never married, and both were deaf and dumb.6 
Of various Papuans of New Guinea it is said that nobody 
remains unmarried, that a girl is married as soon as she has 
reached the proper age, and that a young widow gets another 
husband immediately after the period of mourning has 
come to an end.7 In New Ireland (New Mecklenburg) it is 
the rule that every man and every girl marries.8 In Buin, 
in the island of Bougainville of the Solomon Group, prac- 
1 Gilhodes, ' Mariage et Condition de la Femme chez les Katchins,' 
in Anthropos, viii. 374. 
2 Wallace, Malay Archipelago, i. 141. Gomes, Seventeen Years 
among the Sea Dyaks of Borneo, p. 127. 
3 Marsden, History of Sumatra, p. 256 sq. 
4 Hagen, Die Orang Kubu auf Sumatra, p. 130. 
5 Crawfurd, History of the Indian Archipelago, i. 86. 
8 Jenks, Bontoc Igorot, p. 69. 
7 Vetter, ' Bericht uber papuanische Rechtsverhaltnisse, nament- 
lich bei den Jabim,' in Nachrichten uber Kaiser Wilhelms-Land, 
1897, PP- 89, 93. Schellong, ' Familienleben und Gebrauche der 
Papuas der Umgebung von Finschhafen,' in Zeitschr. f. Ethnol. 
xxi. 17. Landtman, Nya Guinea far den, p. 81 (Kiwai Papuans). 
8 Stephan and Graebner, Neu-Mecklenburg (Bismarck-Archip el), 
p. 109. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
341 
tically every woman from fifteen years of age upwards is 
married.1 In East Mallicolo, of the New Hebrides, " a 
woman without a husband is rare." 2 In New Caledonia 
there are in every tribe a certain number of women who 
have to live single, namely such as in their childhood were 
betrothed to a great chief while still an infant, and after- 
wards when he became a man were refused by him. Rochas 
assures us that among the men, on the other hand, celibacy 
is unknown ;3 but according to Brainne it sometimes happens 
that a man can find no woman who is willing to marry him 
and consequently has to remain a bachelor all his life.4 
Mariner states that in Tonga there were but a few women 
who, from whim or some accidental cause, remained single 
till their death.5 Among the Australian aborigines girls 
are as a rule, it seems, betrothed in infancy,6 and marriage 
takes place very early. The betrothed girls are said to be 
removed from their parents when they are between eight 
and fourteen years of age and most generally at the age of 
about ten or eleven.7 Curr never heard of a woman over 
sixteen who, prior to the breakdown of aboriginal customs 
after the coming of the whites, had not a husband.8 
Passing to African peoples, we notice that among the 
Namib Bushmen most men and girls many as soon as they 
have arrived at puberty.9 In Ondonga all men and women 
are said to do so.10 At Latuka, in Bechuanaland, Campbell 
was assured that there was not an unmarried man, young 
1 Thurnwald, Forschungen auf den Salomo-Inseln und dem 
Bismarck-Archipel, iii. 81. 
2 Paton, quoted by Serbelov, ' Social Position of Men and Women 
among the Natives of East Malekula, New Hebrides,' in American 
Anthropologist, N.S. xv. 279. 
3 de Rochas, La Nouvelle Caledonie, p. 229 sq. 
4 Brainne, La Nouvelle Caledonie, p. 251. 
6 Mariner, Account of the Natives of the Tonga Islands, ii. 168. 
8 Malinowski, The Family among the Australian Aborigines, 
p. 35 sqq., particularly p. 48. 
7 Ibid. p. 257 sq. 
8 Curr, The Australian Race, i. no. 
9 Trenk, ' Die Buschleute der Namib,' in Mittheil. Deutsch 
Schutzgeb. xxiii. 169. 
10 Schinz, Deutsch-Sud-West-Afrika, p. 311. 342 
THE HISTORY OF HUMAN MARRIAGE chap. 
or old, in the whole town.1 " Such a thing as an ' old maid ' 
is almost unknown amongst the heathen Basuto, nor are 
widows allowed to remain as such for any length of time." 2 
Among the natives of South Africa generally, according to 
Burchell, neither men nor women ever pass their lives in a 
state of celibacy.3 In his description of the Thonga tribe 
of Mozambique M. Junod observes :-" The kind of indivi- 
dual called bachelor • does not abound amongst the Bantu. 
The wretched, the invalids, the weak-minded only, are 
deprived of the legal marriage which for the black man is 
and remains the one object in life. It is through his wife 
and children that he becomes somebody in the society." 
And so also every girl in the primitive Bantu tribe gets 
married, some girls, however, sooner than others.4 Among 
the natives of Northern Rhodesia widows, unless old women, 
very seldom remain bereft of a husband for more than a 
year, when they are inherited by the nearest male relative ; 
and although the divorced state is very common, young 
women rarely continue in it for long.5 Speaking of the 
Bagesu, a Bantu tribe living on the slopes of Mount Elgon, 
on the eastern boundary of the Uganda Protectorate, Mr. 
Roscoe remarks that " an unmarried and childless woman 
finds no place in social life among primitive people, for 
woman's principal function is child-bearing and in the 
second place that of making a home for some man."6 
Among the Bantu Kavirondo "it is highly improbable that 
any woman goes to her death unmarried ; for if no suitor 
asks for her in the ordinary way, she will single out a man 
and offer herself to him at a ' reduced price.' "7 Among 
the Mandingo of Senegambia, Caillie met with no instance of 
a young woman, pretty or plain, who had not a husband.8 
1 Campbell, Second Journey in the Interior of South Africa, p. 214. 
2 Minnie Martin, Basutoland, p. 88. 
3 Burchell, Travels in the Interior of Southern Africa, ii. 58. 
Cf. ibid. ii. 565. 
4 Junod, Life of a South African Tribe, i. 125, 183. See also ibid. i. 
101, 184. 
Gouldsbury and Sheane, Northern Rhodesia, p. 171. 
0 Roscoe, Northern Bantu, p. 172. 
7 Johnston, Uganda Protectorate, p. 746. 
8 Caillie, Travels through Central Africa, i. 348. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
343 
In various other tribes in the western parts of Africa celibacy 
is said to be rare or almost unknown.1 Barth tells us that 
the Western Touareg found no other fault with him than 
that he was unmarried ; they could not understand how 
this was possible.2 
Very frequently we are told that a person who does not 
marry is looked upon as an unnatural being or is an object 
of contempt or ridicule. Among the Araucanians of Chili 
" old bachelors are called, by way of contempt, vuchiapra, 
and old maids cudepra, that is, old, idle, good for nothing." 3 
In the Tupi tribes of Brazil no man was allowed to partake 
in the drinking-feast while he remained single.4 Among 
the Chukchee " a man fullgrown and unmarried is despised 
by the people, and in reality is looked upon as a good-for- 
nothing, a lounger, a tramp, idly wandering from camp to 
camp."5 The Tipperahs of the Chittagong Hills do not 
consider a man a person of any importance until he is 
married.6 Among the Santals a man who remains single 
"is at once despised by both sexes, and is classed next to a 
thief, or a witch: they term the unhappy wretch ' No man.' "7 
Among the Kafirs a bachelor has no voice in the kraal,8 
among the Baganda he was not allowed to possess land.9 
The Batamba in the Uganda Protectorate have a saying 
which means that an unmarried woman is death itself ; 
she is a dead quantity, of no use whatever to anyone and 
therefore might as well be dead.10 So also the Todas consider 
1 Delafosse, ' Le peuple Si6na ou S6noufo,' in Revue des etudes 
ethnographiques et sociologiques, i. 483. Schuster, ' Die sozialen 
Verhaltnisse des Banjange-Stammes (Kamerun),' in Anthropos, 
ix. 951. Bosman, ' New Description of the Coast of Guinea,' in 
Pinkerton, Collection of Voyages and Travels, xvi. 424. Torday and 
Joyce, * Notes on the Ethnography of the Ba-Mbala,' in Jour. 
Anthr. Inst. xxxv. 410 sq. 
2 Barth, Reisen in Nord- und Central-Afrika, i. 489. 
3 Molina, History of Chili, ii. 115. 
4 Southey, History of Brazil, i. 240. 5 Bogoras, op. cit. p. 569. 
6 Dalton, Descriptive Ethnology of Bengal, p. no. 
7 Man, Sonthalia and the Sonthals, p. 101. 
8 v. Weber, Vier Jahre in Afrika, ii. 215. 
9 Decle, Three Years in Savage Africa, p. 446. 
10 Condon, ' Contribution to the Ethnography of the Easoga 
Batamba,' in Anthropos, vi. 372. 344 
THE HISTORY OF HUMAN MARRIAGE chap. 
it a disgrace for a woman not to get married, and if a girl 
among them does not succeed in securing a husband by the 
natural process of sexual selection, her father bribes a man 
to marry her by the present of a buffalo.1 On the coast of 
Malabar, if anyone died a maid, some poor man of the family 
was asked to deflower her ; he was well paid for his service, 
but as soon as he had done it stones were thrown at him till 
he ran out of reach.2 The Kachins of Burma perform 
derisory funeral rites on the death of bachelors and old 
maids.3 The natives of Futuna, in the Western Pacific, 
maintained that it was necessary to be married in order to 
hold a part in the happy future life, and that all celibates, 
both men and women, had to submit to a chastisement of 
their own before entering the fale-mate, or " home of the 
dead."4 According to Fijian beliefs, he who died wifeless 
was stopped by the god Nangganangga on the road to 
Paradise, and smashed to atoms.5 
The frequency of marriage depends of course largely 
on the age when people generally marry. Although in the 
latter respect as well as in the former statistical data are 
entirely lacking so far as savages are concerned, I think we 
may safely say that among all of them the girls marry at an 
earlier age than among the peoples of Western civilisation, 
and the same is probably in most cases true of the men. 
The following statements seem to justify these conclusions, 
1 Thurston, ' Todas of the Nilgiris,' in the Madras Government 
Museum's Bulletin, i. 158. 
2 Faria y Sousa, Asia Portoguesa, ii. 4.6.3, vol. ii. 708. 
3 Gilhodes, in Anthropos, viii. 374. 
4 Percy Smith, ' Futuna,' in Jour. Polynesian Soc. i. 39 sq. 
5 Pritchard, Polynesian Reminiscences, pp. 368, 372. Seemann, 
Viti, p. 399 sq. Fison, ' Fijian Burial Customs,' in Jour. Anthr. 
Inst. x. 139. Williams and Calvert, Fiji, p. 206. For other 
instances see Armstrong, Discovery oj the North-West Passage, 
p. 192 (Eskimo) ; Lansdell, Through Siberia, ii. 226 (Gilyak) ; Wilken, 
' Over de primitieve vormen van het huwelijk en den oorsprong van 
het gezin,' in De Indische Gids, 1880, vol. ii. 633 n. 2 (natives of 
the Malay Archipelago) ; Klose, Togo unter deutscher Flagge, p. 508 
(Bassari) ; Hutter, Wanderungen und Forschungen im Nord- 
Hinterland von Kamerun, p. 376 (Bali) ; Torday and Joyce, Les 
Bushongo, p. 114 (Bangongo). x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
345 
even though many or most of them can only be regarded as 
approximately accurate.1 
Starting with the American aborigines, we notice that 
the betrothal of girls, or of both boys and girls, in their early 
childhood occurs more or less frequently among a large 
number of them.2 Even though the betrothal be not strictly 
binding, we may assume that as a ru'e the marriage is 
consummated when the persons thus betrothed reach the 
age of puberty, if not before. In South America. the custom 
of parents contracting marriages for their children when 
infants is particularly prevalent among the Arawaks and 
Macusis of British Guiana,3 and in North America among 
1 See also Ploss-Bartels, Das Weib, i. 668 sqq. 
2 Falkner, Description of Patagonia, p. 124 ; King and Fitzroy, 
Narrative of the Voyages of the Adventure and Beagle, ii. 152 sq. 
(Patagonians). Stade, Captivity in A.D. 1547-1555 among the Wild 
Tribes of Eastern Brazil, p. 143 (Tupinambase). Tocantins, ' Estudos 
sobre o tribu " Mundurucu," ' in Revista trimensal do Instituto 
Historico Geographico e Ethnographico do Brazil, vol. xl. pt. ii. 113. 
v. Martius, Beitrage zur Ethnographic Amerika's, i. 322 (Bo'tocudos). 
Whifien, North-West Amazons, p. 162 (Witoto and Boro). Coudreau, 
Chez nos Indiens, p. 127 (Roucouyennes of French Guiana). Bell, 
op. cit. p. 262 (Mosquito Indians). Stephen, ' Navajo,' in American 
Anthropologist, vi. 356. Lewis and Clarke, Travels to the Source 
of the Missouri River, p. 307 (Shoshoni). James, Account of 
an Expedition from Pittsburgh to the Rocky Mountains, i. 231 
(Omaha). Loskiel, History of the Mission of the United Brethren 
among the Indians in North America, i. 56 (Iroquois). Richardson, 
Arctic Searching Expedition, ii. 23 ; Mackenzie, Voyages from Montreal 
to the Frozen and Pacific Oceans, p. cxxiii. (Chipewyan). Dixon, 
' Notes on the Achomawi and Atsugewi Indians of Northern 
California,' in American Anthropologist, N.S. x. 217. Sapir, ' Notes 
on the Takelma Indians of Southwestern Oregon,' ibid. N.S. ix. 
274. Bancroft, Native Races of the Pacific States of North America, 
i. 276 sq. (Inland Columbians). Mayne, Four Years in British 
Columbia, p. 276 (Nootka). Swanton, Haida, p. 50. Kohler, ' Die 
Rechte der Urvolker Nordamerikas,' in Zeitschr. f. vergl. Rechtswiss. 
xii. 378. 
3 Appun,. ' Die Indianer von Britisch-Guayana,' in Das Ausland, 
xliv. 124, 446. Bemau, Missionary Labours in British Guiana, 
p. 59. Schomburgk, Reisen in Britisch Guiana, ii. 460. Brett, Indian 
Tribes of Guiana, p. 99 sq. v. Martius, op. cit. i. 645. Im Thurn, 
Among the Indians of Guiana, p. 221. 346 
THE HISTORY OF HUMAN MARRIAGE chap. 
the Eskimo,1 except in Greenland, where the practice seems 
always to have been rare, though not unknown.2 Dalager, 
who wrote in the eighteenth century, says that the young 
Greenlanders marry as soon as they feel the desire and are 
able to support a wife, and that they choose a woman of 
their own age ; but he also states that there are many men 
who remain bachelors till the age of twenty-seven or twenty- 
eight.3 According to Holm, a Greenlander on the east 
coast often marries before he is grown-up, chiefly in order' 
-to get a woman to attend to his fishing implements and to 
prepare what he catches ; hence it may be that a young 
lad has a wife who is old enough to be his mother.4 Among 
the Thompson River Indians of British Columbia girls 
were often betrothed while mere infants to men sometimes 
twenty years their seniors. " They were considered 
marriageable only after they had finished the ceremonies 
attendant upon reaching the age of puberty. This was 
approximately in the seventeenth or eighteenth year, but 
sometimes the ceremonies were continued until the twenty- 
third year. Most of the men married from three to seven 
years after finishing the puberty ceremonials, and it may 
be said that most of them married between the ages of 
twenty-two and twenty-five." 5 Among the coast Indians 
in Washington Territory Mr. Swan noticed as a general 
thing that the young men married women much older than 
themselves, while the young girls were married to men much 
their seniors ; the reason they gave for this was that if the 
1 Franklin, Narrative of a Journey to the Shores of the Polar Sea, 
p. 263. Hall, Arctic Researches, p. 567. Klutschak, Ais Eskimo 
unter den Eskimos, p. 233. Ross, Narrative of a Second Voyage in 
Search of a North-West Passage, p. 269. Kumlien, Contributions 
to the Natural History of Arctic America, p. 16. Gilder, op. cit. 
p. 249 $#. Boas, 'Central Eskimo,' in Ann. Rep. Bur. Ethnol. vi. 578. 
Murdoch, ' Ethnological Results of the Point Barrow Expedition,' 
ibid. ix. 410. Turner, ' Ethnology of the Ungava District, Hudson 
Bay,' ibid. xi. 188. 
2 Cranz, History of Greenland, i. 146. 
3 Dalager, Grfnlandske Relationer, pp. 7, 67. 
1 Holm, in Geografisk Tidskrift, viii. 91. 
5 Teit, ' Thompson River Indians of British Columbia,' in 
Memoirs American Museum Natural History, Anthropology, i. 321. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
347 
young men marry young girls, both are so foolish that they 
do not know how to take care of each other.1 Among the 
Omaha, a tribe of the Missouri valley in the State of Nebraska, 
men formerly did not marry till they were twenty-five or 
thirty years old and women not till they were twenty ; but 
nowadays it is customary for girls to be married at the age 
of fifteen, sixteen, or seventeen.2 So also among the 
Natchez it was rare for young men to marry before they 
were twenty-five, as till then they were looked upon as too 
weak and devoid of understanding and experience.3 But 
of a large number of other North and also Central American 
tribes we are told that marriages take place at an early age.4 
Among the Sauk and Foxes, for instance, most men married 
from sixteen to twenty and the girls from fourteen to 
eighteen, but generally before sixteen.5 The Havasupai 
girls of Arizona marry as soon as possible after attaining 
puberty, mostly between the ages of thirteen and fourteen ;6 
and almost the same is said of the girls of the Pawnee.7 
Concerning the Tarasco of Mexico, Dr. Lumholtz states 
that " if a man is bearded, though he may be only twenty 
and not yet married, he generally has to content himself 
with a widow, for the girls are suspicious lest something 
prevented him from marrying when it was time for him to 
1 Swan, Northwest Coast, p. 170. 
2 Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethnol. iii. 259. 
3 Le Page du Pratz, History of Louisiana, p. 344. 
4 Ross, ' Eastern Tinneh,' in Smithsonian Report, 1866, p. 305. 
Powers, Tribes of California, p. 413. Catlin, Illustrations of the 
Manners, &c. of the North American Indians, i. 121 (Mandan). 
Schoolcraft, Information Respecting the History, &c. of the Indian 
Tribes of the United States, ii. 132 ; ten Kate, Reizen en onderzoekingen 
in N oord-Amerika, p. 390 (Comanche). Hrdlicka, ' Notes on the 
Indians of Sonora, Mexico,' in American Anthropologist, N.S. vi. 69 
(Yaqui). Bancroft, op. cit. i. 632 (Central Mexican tribes). Bovallius, 
Resa i Central-Amerika, i. 248 (Talamanca). 
5 Marston, ' Letter,' in Blair, Indian Tribes of the Upper Missis- 
sippi Valley, ii. 165. Forsyth, ' Account of the Manners and Customs 
of the Sauk and Fox Nations of Indians,' ibid. ii. 216. 
8 James, Indians of the Painted Desert Region, p. 227. 
' Dunbar, ' Pawnee Indians,' in Magazine of American History, 
iv. 264. 348 
THE HISTORY OF HUMAN MARRIAGE chap. 
do so."1 In some Central American tribes the parents try 
to get a wife for their son when he is nine or ten years old.2 
Among the Guaraunos on the Orinoco, according to M. 
Chaffanjon, the men marry at the age of fourteen or fifteen 
and the women at the age of ten or twelve.3 Among the 
natives of British Guiana marriages are concluded by boys 
of fifteen or sixteen and by girls of twelve or thirteen.4 
Among Kayuas of the Gran Chaco the men seldom marry 
before they are seventeen or eighteen years old, but the 
girls marry when they are eleven or twelve.5 Among the 
Coroados on the Rio Xipoto, according to v. Spix and v. 
Martius, a man generally marries at the age of from fifteen 
to eighteen and a woman from ten to twelve.6 Dr. Krause 
states that the Kayapo, on the Araguaya, marry early, 
between the ages of about fifteen and twenty, and that 
among the Karaya, living on the same river, the men marry 
apparently when they are from seventeen to twenty and the 
women between fourteen and sixteen.7 But he adds that 
among the latter there are men who put off their marriage 
as long as possible so as not to be burdened with marital 
duties, and that they are enticed to do so by a number of 
elderly bachelors, who extol the advantages of a single 
life ; and widowers also prefer remaining unmarried if they 
have no children who are ip need of maternal care.8 Among 
the Guaycurus of the Gran Chaco even the majority of the 
men were said by an old writer to refrain from marrying 
until they were between twenty-five and thirty years old 
in order to be able to lead a free and easy life.9 Father 
Baucke, who lived in Paraguay in the middle of the eighteenth 
century, states that among the Mocobis in the southern 
part of. the Gran Chaco the man is generally past twenty 
1 Lumholtz, Unknown Mexico, ii. 416. 
2 Morelet, Reisen in Central-Amerika, p. 257. 
3 Chaffanjon, L'Orenoque et le Caura, p. 11. 
4 Bemau, op. cit. p. 59. 
5 v. Koenigswald, ' Die Cayuds,' in Globus, xciii. 381. 
6 v. Spix and v. Martius, Travels in Brazil, ii. 248. 
7 Krause, In den Wildnissen Brasiliens, pp. 401, 324. 
8 Ibid. pp. 160, 190, 325 sq. 
9 Sdnchez Labrador, El Paraguay Cat6lico, ii. 26. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
349 
when he enters into the married state.1 But the Guarani 
of the Plata, according to Azara, marry at an earlier age, and 
among the Guanas of the Gran Chaco the girl " qui se marie 
le plus tard, se marie A neuf ans."2 In Tierra del Fuego, as 
we are told by Lieutenant Bove, a girl looks about for a 
husband when twelve or thirteen years old, and a youth 
marries at the age of from fourteen to sixteen.3 It is a 
widespread custom in South America that older men marry 
young girls and young men elderly women ; when in the 
latter case the wife becomes too old, the husband can take 
another one, either keeping or sending away the first.4 _ Thus 
among the Jibaros of Ecuador men of thirty or forty often 
marry girls of from six to twelve.6 Among the Witoto and 
Boro of the North-West Amazons the general disparity of 
age is between five and fifteen years. A man will choose 
an undeveloped girl, perhaps nine or ten years old, and 
hand her over to the women of his own family. Mr. Whiffen 
says that he thereby " seeks to gain affection by association. 
The girl lives with him and his people, they become to all 
intents and purposes her people ; she is trained by custom 
to their habits of life."6 
As among the American Eskimo so also among the Chuk- 
chee on the other side of Bering Strait there are betrothals 
of infants.7 Dr. Bogoras states that families who are 
friendly to each other, though not connected by ties of 
blood, sometimes even agree to a marriage between their 
children before these are born ; but more frequently such 
agreements are concluded between friendly families on the 
basis of the exchange of one woman for another.8 He also 
1 Kohler, Pater Florian Baucke, ein Jesuit in Paraguay, p. 311. 
2 Azara, op. cit. ii. 60, 61, 94. 
3 Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 132. 
4 Ehrenreich, Beitrdge zur V olkerkunde Brasiliens, p. 27. Krause, 
op. cit. p. 325 (KarayA). Gumilla, El Orinoco ilustrado, i. 197 sq. 
(Otomacos). Im Thum, op. cit. p. 223 (Macusis). 
5 Rivet, ' Les Indiens Jibaros,' in L'Anthropologic, xviii. 607 sq. 
6 Whifien, op. cit. p. 162. 
7 Hooper, Ten Months among the Tents of the Tuski, p. 209. 
Bogoras, op. cit. p. 571. 
8 Bogoras, op. cit. p. 577 sq. 350 
THE HISTORY OF HUMAN MARRIAGE chap. 
says that the young men marry early, and that not infre- 
quently very young girls bear children.1 Among the 
Gilyak betrothals are made in childhood ;2 among those of 
Sakhalin the marriage is said to take place at the age of 
thirteen or fourteen in the case of men and at the age of 
twelve or thirteen in the case of women.3 Among the Ainu 
it seems to have been an ancient custom, though not general, 
to betroth children, but the young man and his fiancee were 
not absolutely bound to marry. They consider their 
daughters to be marriageable at about sixteen or seventeen 
years of age, whilst the men marry when about nineteen or 
twenty.4 .Among the Koryak girls generally marry when 
twenty years old or even older ; but in former times marriages 
with minors were more common.5 The Buryat usually 
betroth their children in infancy,6 but the custom by which 
the parents of two families where there are sons and 
daughters exchange brides for their sons may lead to a 
great difference of age between husband and wife-a young 
boy may be married to an elderly woman and an old man to 
a young girl.7 Infant-betrothals are practised by the 
Samoyed 8 and Yakut,9 and according to Vamb&y by all 
peoples of the Turkish stock.10 Among the Lake Dwellers 
of Lob-Nor girls enter into matrimony at the age of fourteen 
or fifteen, and men at the same age or a little later.11 
Among the Kafirs of the Hindu-Kush it is comparatively 
rare to find a girl of twelve who is unmarried ; infants in 
arms are sometimes married, or at least affianced, to grown 
men, and full-grown or even middle-aged women are on the 
other hand sometimes married to boys, because a wife is a 
1 Ibid. pp. 37, 573. 2 Miss Czaplicka, Aboriginal Siberia, p. 100. 
3 Labbd, ' L'ile de Sakhaline,' in Bull. Soc. de geographic commercials 
de Paris, xxiii. 124. 
1 Batchelor, Ainu and their Folk-Lore, p. 225 sqq. 
5 Jochelson, Koryak, p. 753. 6 Miss Czaplicka, op. cit. p. 118. 
7 Melnikow, 'Die Burjaten des Irkutskischen Gouvernements,' 
in Verhandl. Berliner Gesellsch. Anthrop. 1899, p. 441 sq. 
s Arnesen, ' Fran Gyda-viken till Obdorsk,' in Ymer, iii. 144. 
•s Miss Czaplicka, op. cit. p. 108. 
10 Vambdry, Das Turkenvolk, p. 109. 
11 Prejevalsky, From Kulja to Lob-nor, p. ill sq. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
351 
field-slave and an orphan lad who is the owner of fields 
requires one to get his land cultivated.1 So also there is 
often a very great disproportion of age in the marriages^ of 
the Todas of Southern India. The woman may be older, 
owing to the custom of marrying a cousin, but more com- 
monly the man is much the older, which is partly due to the 
practice of infant marriage, a child being often married 
when only two or three years of age.2 Among the Kotas, 
another tribe of the Nilgiri Hills, when a boy is from fifteen 
to twenty years old, his parents ask in marriage for him some 
girl of six or eight.3 Among the Kasubas of the Nilgiris 
" marriage is either infant or adult; the former is more 
common. Adult marriage is, strange to say amongst this 
forest tribe, looked down upon with something akin to 
contempt." 4 In Tamil land, when the daughters are from 
six to eight years old it is time to marry them ;5 whilst 
among the Tamils in ancient days girls were considered 
marriageable at twelve and boys at sixteen.6 A Kandh boy 
marries when he reaches his tenth or twelfth year, his wife 
being usually about four years older.7 Among the Oraons, 
a Dravidian tribe in Chota Nagpur, the average age of the 
bridegroom is sixteen and that of the bride fourteen or 
fifteen.8 Of the Mundas, another Chota Nagpur tribe, we 
are told that early marriages are not infrequent nowadays 
among the. well-to-do people, but the days are still remem- 
bered when no young Munda could marry before he was 
able to construct a plough with his own hands, and no girl 
was given away in marriage before she could weave mats 
with palm-leaves and spin cotton.9 Among the Bodo and 
1 Scott Robertson, op. cit. p. 534. 2 Rivers, Todas, pp. 504, 523. 
3 Breeks, Account of the Primitive Tribes of the Nilagiris, p. 46. 
4 Hayavadana Rao, ' Kasubas,' in Anthropos, iv. 179. 
6 Baierlein, Land of the Tamulians, p. 35. 
6 Kanakasabhai, Tamils Eighteen Hundred Years ago, p. 124. 
7 Hunter, Rural Bengal, iii. 82. 
8 Dehon, ' Religion and Customs of the Uraons,' in Memoirs 
Asiatic Soc. Bengal, i. 161. 
9 Sarat Chandra Roy, Mundas, p. 436. See also Jellinghaus, 
' Sagen, Sitten und Gebrauche der Munda-Kohls in Chota Nagpore,' 
in Zeitschr. f. Ethnol. iii. 366 sq. 352 
THE HISTORY OF HUMAN MARRIAGE chap. 
Dhimals " marriage takes place at maturity, the male 
being usually from twenty to twenty-five years of age, and 
the female from fifteen to twenty." 1 A Santal lad marries, 
as a rule, about the age of sixteen or seventeen, and a girl at 
that of fifteen.2 Among the Mikirs, a Tibeto-Burman 
people inhabiting the Province of Assam, child-marriage is 
unknown, the age for marriage being from fourteen to 
twenty-five, but most usually eighteen or nineteen, for men, 
and ten to fifteen, but most usually fifteen, for girls.3 
Generally speaking, among the aboriginal tribes of India 
marriage is almost as universal as among the Hindus, but 
the age at which it is contracted is higher and infant-marriage 
rare.4 The aborigines of the Malay Peninsula, as a rule, 
marry early in fife ; among the Besisi, for instance, boys of 
fourteen marry girls of thirteen,5 but sometimes a girl is 
married when a little child, in which case the husband 
carries her to his house and brings her up.6 The Veddas of 
Ceylon marry young ;7 whereas in the Andaman Islands a 
marriage rarely takes place before the man is twenty-six.8 
Child-marriage is practised in various tribes of the 
Philippine Islands.9 Among the Subanu, a mountain folk 
of Mindanao, the women usually marry upon arriving-at the 
age of puberty, but the men often after they have passed that 
age.10 Among the Bagobo in Southern Mindanao marriage 
is said to take place much later than is common among most 
Philippine tribes, the couple being eighteen or twenty years 
of age.11 Betrothals of children are also frequent in various 
1 Hodgson, ' K6cch, Bodo and Dhimal People,' in Jour. Asiatic 
Soc. Bengal, vol. xviii. pt. ii. 734. 
2 Hunter, Rural Bengal, i. 205. Cf. Man, Sonthalia, p. 20. 
3 Stack, Mikirs, p. 17. 
4 Gait, Census of India, 1911, vol. i.(India) Report, pp. 263, 266, 270. 
5 Martin, Die Inlandstamme der Malayischen Halbinsel, p. 865 sq. 
8 Logan, ' Five Days in Naning,' in Jour. Indian Archipelago, 
iii. 490. 7 Nevill, ' Vaeddas of Ceylon,' in Taprobanian, i. 178. 
8 Portman, History of Our Relations with the Andamanese, i. 29. 
9 Reed, Negritos of Zambales, p. 57 ; Blumentritt, Versuch einer 
Ethnographic der Philippinen, p. 7 (Negritos). Worcester, Philippine 
Islands, p. 108 (Tagbanuas of Palauwan). Jenks, Bontoc Igorot, 
p. 68. 10 Finley and Churchill, Subanu, p. 39. 
11 Cole,' Wild Tribes of Davao District, ' in Field Museum of Natural 
History, Anthropological Series, xii. 101. See also ibid. p. 143 (Bila-an). x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
353 
other parts of the Malay Archipelago.1 Among the Sea 
Dyaks of Borneo the young men marry when about twenty 
years of age and the girls at sixteen or seventeen; some- 
times, however, marriage is postponed till later, though it 
is very unusual to meet a bachelor above the age of twenty- 
five.2 Among the Kayans of the same island a young man 
" is usually not much over twenty years of age when he 
becomes accepted as the future husband of a girl some years 
his junior." 3 The Kubus of Sumatra marry at a very early 
age.4 Among the Battas the men generally marry when 
they are eighteen years old and the girls when they are 
fifteen or still younger, sometimes already in their ninth 
year.5 In the kingdom of Acheh, on the east coast of 
Sumatra, a young man usually marries for the first time at 
the age of from sixteen to twenty years, whilst girls marry 
earlier in fife than perhaps in any other Muhammadan 
country of the Eastern Archipelago ; girls of eight or ten, 
nay, even of seven years of age are actually handed over to 
their husbands, though the latter may be grown-up or 
elderly men.6 Mr. Bickmore says that among the Malays 
the boys usually marry for the first time when about sixteen 
and the girls at the age of thiiteen or fourteen, and occasion- 
ally still earlier.7 
Among the Pacific Islanders betrothals very frequently 
1 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. 161 sqq. ; Schmidt, Liebe und Ehe im alten und modernen 
Indien, p. 307 sq. ; Hickson, A Naturalist in North Celebes, p. 270 
(various peoples in the Malay Archipelago). ' Aus dem Wanderbuche 
eines Weltreisenden,' in Das Ausland, liv. 569 (Javanese). Riedel, 
De sluik- en kroesharige rassen tusschen Selebes en Papua, p. 21 ; 
Martin, Reisen in den Molukken, &c., p. 289 (natives of Burn ; 
children are even betrothed before their birth). 
2 Gomes, op. cit. p. 127. See also Ling Roth, Natives of Sarawak, 
i- IX5- 
3 Hose and McDougall, Pagan Tribes of Borneo, 11. 170. 
4 Hagen, Die Orang Kubu auf Sumatra, p. 130. 
5 Junghuhn, Die Bai aldnder auf Sumatra, ii. 133. See also v. 
Brenner, Besuch bei den Kannibalen Sumatras, p. 248. 
6 Snouck Hurgronje, Achehnese, i. 295 sqq. 
7 Bickmore, Travels in the East Indian Archipelago, p. 278. 
VOL. I A A 354 
THE HISTORY OF HUMAN MARRIAGE chap. 
take place in childhood or infancy or even before birth.1 
Mariner' supposed that in Tonga about one-third of the 
married women had been thus betrothed.2 As a rule, it 
seems, the tender girl is promised to a boy more or less of 
her own age, but in some places she is not infrequently 
affianced to a grown-up man 3 or even to a man who is old 
enough to be her grandfather.4 She may remain with her 
1 Finsch, Neu-Guinea, pp. 102, 116. Guillemard, Cruise of the 
Marchesa, p. 389. Neuhauss, Deutsch Neu-Guinea, i. 160 sq. 
Vetter, in Nachrichten uber Kaiser Wilhelms-Land, 1897, p. 89. 
Seligman, Melanesians of British New Guinea, p. 745 (natives of 
Mukaua and Bogaboga). Williamson, Mafulu Mountain People of 
British New Guinea, p. 170. Moszkowski, ' Die Volkerstamme am 
Mamberamo in Hollandisch-Neuguinea,' in Zeitschr. f. Ethnol. 
xliii. 323 (natives of Dorey). Hahl, 'UeberdieRechtsanschauungen 
der Eigeborenen eines Theiles der Blanchebucht und des Innem 
der Gazelle Halbinsel,' in Nachrichten uber Kaiser Wilhelms-Land, 
1897, p. 78 ; Burger, Die Kusten- und Bergvblker der Gazellehalbinsel, 
p. 23 ; Parkinson, Dreissig fahre in der Sudsee, p. 63 (natives of 
Gazelle Peninsula of New Britain). Powell, Wanderings in a Wild 
Country, p. 85 (natives of New Britain). Stephan und Graebner 
Neu-Mecklenburg, p. 107. Brown, ' Notes on the Duke of York 
Group,' in Jour. Roy. Geograph. Soc. xlvii. 148. Thumwald, For- 
schungen auf den Salomo-Inseln, iii. 12 (natives of Buin). Codrington, 
Melanesians, pp. 237, 241. Robertson, Erromanga, p. 395. de Rochas, 
op. cit. p. 231; Lambert, Moeurs et Superstitions desNeo-Caledoniens, 
p. 91 (New Caledonians). Thomson, Fijians, p. 201. Wilkes, 
op. cit. iii. 92 (Fijians); v. 102 (Gilbert Islanders). Senfft, ' Die 
Insel Nauru,' in Mittheil. Deutsch. Schutzgeb. ix. 106; Kohler, 
' Das .Recht der Marschallinsulaner,' in Zeitschr. f. vergl. Rechtswiss. 
xiv. 417 sq. Tautain, ' £tude sur le mariage chez les Polynesians 
des ties Marquises,' in L'Anthropologic, vi. 644 sq. Waitz-Gerland, 
Anthropologic der Naturuolker, vi. 127 (Nukahivans). Jarves, 
History of the Hawaiian Islands, p. 43. Stair, Old Samoa, p. 171. 
Ellis, Polynesian Researches, i. 267, 270 (Tahitians). Geiseler, Die 
Oster-Insel, p. 27. Gisborne, Colony of New Zealand, p. 27 ; Polack, 
Manners and Customs of the New Zealanders, i. 135 sq.; Tregear, 
Maori Race, p. 285 ; Best, ' Maori Marriage Customs,' in Trans, and 
Proceed. New Zealand Institute, xxxvi. 32, 42 sq. 
2 Mariner, op. cit. ii. 167. 
8 See, e.g., Neuhauss, op. cit. i. 161 (natives of the part of New 
Guinea which formerly belonged to Germany) ; Stephan and 
Graebner, op. cit. p. 107 (natives of New Ireland) ; Thomson, Story 
of New Zealand, i. 176. 
4 Thomson, Fijians, p. 201. Stair, Old Samoa, p. 171. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
355 
parents until she is marriageable, or she may be at once 
transferred to the family of the fiance. In the latter case 
the betrothed persons may nevertheless be kept away from 
each other and even strictly prohibited from speaking to 
one another ; nay, it sometimes happens that whilst the 
girl is taken to the boy's family the boy is taken to the girl's 
people to be educated by them.1 As regards the age when 
marriages are concluded or consummated we are told that 
the Papuans of the Maclay Coast of New Guinea marry soon 
after they are circumcised, at the age of between thirteen 
and fifteen;2 whereas among the inland tribes of Dutch New 
Guinea the men are said to marry comparatively late, 
living together as bachelors in the men's house and indulging 
there in homosexual orgies.3 In the Mekeo district of British 
New Guinea a boy may marry at the age of about thirteen 
and a girl at ten, but as a rule the youngest ages at which 
marriages are concluded would be more like eighteen for a 
boy and sixteen for a girl.4 Among the Mafulu mountain 
people of British New Guinea a boy is regarded as having 
reached a marriageable age at about sixteen, seventeen, or 
eighteen, and a girl a few years earlier; but they do not as 
a rule marry before they have received their perineal bands, 
although there does not appear to be any definite custom 
against their doing so.6 At Port Moresby " few men over 
twenty years of age remain single." 6 Among the natives 
of Bogadjim, in the part of New Guinea which formerly 
belonged to Germany, marriages are seldom contracted 
before the age of twenty.7 
In the groups of islands to which the Germans gave the 
collective name Bismarck Archipelago the earliest age for 
marrying is in the case of boys fourteen or fifteen and in the 
1 Moszkowski, in Zeitschr. f. Ethnol. xliii. 323 (Papuans of Borey). 
2 Miklucho-Maclay, * Anthropologische Bemerkungen uber die 
Papuas der Maclay-Kiiste in Neu-Guinea,' in Natuurkundig 
Tijdschrift voor Nederlandsch Indie, xxxiii. 245. 
3 Moszkowski, in Zeitschr. f. Ethnol. xliii. 339. 
4 Williamson, Ways of the South Sea Savage, p. 129. 
5 Idem, Mafulu Mountain People of British New Guinea, p. 168. 
0 Stone, ' Port Moresby and Neighbourhood,' in Jour. Roy. 
Geograph. Soc. xlvi. 55- 7 Hagen, Unter den Papua's, p. 241. 
A A 2 356 
THE HISTORY OF HUMAN MARRIAGE chap. 
case of girls nine or ten, but it seldom happens that people 
marry as early as that; few girls, however, seem to remain 
unmarried after the age of fifteen.1 In the Buin district of 
Bougainville, one of the Solomon Islands, where practically 
all women in the child-bearing age, from about fifteen years 
upwards, are married, the men do not marry till they are 
between twenty and twenty-five, and formerly they had to 
wait till they were about from'twenty-five to thirty.2 In 
East Mallicolo, of the New Hebrides, the girls are usually 
married at the age of twelve.3 In New Caledonia it is very 
common for men and women to marry when they have 
attained the age of from eighteen to twenty.4 In Fiji 
" young men of the lower orders married rather late in life 
for a primitive race, rarely, it appears, before the age of 
twenty-five." 5 In Easter Island the ancient custom, which 
still prevails there to a considerable extent, is that when a 
boy arrives at the age of, say, twelve years, his father tries 
to find a suitable wife for him, the girl being perhaps of the 
same age or more likely younger than the boy;6 according 
to some authorities, the girls were generally ten years old 
when they married.7 In Niue Island, as in fact among all 
Polynesians, the women married at an early age.8 " Early 
marriages," says Mr. Best, " seem to have been common 
among the Maori, although the elders appear to have 
believed them to be harmful." 9 According to Mr. Tregear, 
however, such marriages were not the rule among them ; the 
men, especially, often reached mature age before they took 
wives.10 
The custom of betrothing females in infancy or sometimes 
1 Pfeil, Studien und Beobachtungen aus der Sudsee, p. 26 sq. 
2 Thumwald, op. cit. iii. 81. 
3 Serbelov, in American Anthropologist, N.S. xv. 278. 
4 Brainne, op. cit. p. 251. 5 Thomson, Fijians, p. 172 sq. 
6 Cooke, ' Te Pito Te Henna,' in Smithsonian Report, 1897, vol. 
i. 716. 
7 Gana, quoted in Jour. Anthr. Inst. v. 112. Tregear, ' Easter 
Island,' in Jour. Polynesian Soc. i. 99. 
8 Percy Smith, ' Niue Island,' in Jour. Polynesian Soc. xi. 206. 
9 Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 32. 
10 Tregear, The Maori Race, p. 284. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
357 
before they are born is extremely widespread in Australia.1 
The girl may be promised to a boy of a friendly family, or 
to a grown-up man, who even may be married before and 
old enough to be her father. She generally remains with 
her parents till she attains puberty, when she is claimed by 
her husband ; but sometimes he claims her or has inter- 
course with her while she is still a mere child. Among the 
Arunta and Southern Loritja, according to Strehlow, when 
a girl is betrothed to a boy of her own age, the marriage does 
not take place until the latter gets a beard or even till the 
first grey hairs make their appearance in it.2 A man may 
have to wait a long time before he gets a wife. Mr. Curr 
1 Howitt, Native Tribes of South-East Australia, pp. 60, 61, 194, 196, 
197, 216, 217, 219, 222, 232, 236, 237, 241, 245, 251, 260, 262 sq. 
Ridley, Kamilaroi, Dippil, and Turrubul, p. 157. Holden, in Taplin, 
Folklore, &c. of the South Australian Aborigines, p. 17 (Maroura tribe, 
Lower Darling). Bonney, ' Customs of the Aborigines of the River 
Darling,' in Jour. Anthr. Inst. xiii. 129, 301. Cameron, ' Notes on 
some Tribes of New South Wales,' ibid. xiv. 352. Newland, 
' Parkengees, or Aboriginal Tribes on the Darling River,' in 
Proceed. Roy. Geograph. Soc. Australasia : South Australian Branch, 
ii. 21. Curr, Recollections of Squatting in Victoria, pp. 129, 249 sq. 
Mathews, Ethnological Notes on the Aboriginal Tribes of N. S. Wales 
and Victoria, pp. 96, 97, 100 sq. Fraser, Aborigines of New South 
Viales, p. 26. Wilhelmi, ' Manners and Customs of the Australian 
Natives, in particular of the Port Lincoln District,' in Trans. Roy. 
Soc. Victoria, v. 179. Westgarth, Australia Felix, p. 69 (aborigines 
of Port Phillip, N. S. Wales). Schulze, 'Aborigines of the Upper 
and Middle Finke River,' in Trans, and Proceed. Roy. Soc. South 
Australia, xiv. 236. Gillen, ' Notes on some Manners and Customs 
of the Aborigines of the McDonnell Ranges belonging to the Arunta 
Tribe,' in Report on the Work of the Horn Expedition to Central 
Australia, iv. 165. Spencer and Gillen, Native Tribes of Central 
Australia, p. 558 sqq. Sturt, Narrative of an Expedition into Central 
Australia, ii. 284 sq. Mrs. Langloh Parker, Euahlayi Tribe, p. 55 sq. 
Purcell, ' Aborigines of Australia,' in Trans. Roy. Geograph. Soc. 
Australasia (Victoria Branch), xi. 19 (natives of Queensland and the 
Northern Territory). Brown, ' Three Tribes of Western Australia,' 
in Jour. Roy. Anthr. Inst, xliii. 156. Waitz-Gerland, Anthropologic 
der Naturvolker, vi. 772. Malinowski, The Family among the Australian 
Aborigines, p. 118. 
2 Strehlow, Die Aranda- und Loritja-Stamme in Zentr al-Ausir alien, 
vol. iv. pt. i. 89, 100 sq. 358 
THE HISTORY OF HUMAN MARRIAGE chap. 
even goes so far. as to say that as a rule wives are not 
obtained by the men until they are at least th'rty years of 
age ;1 and when young men are allowed to marry they 
generally get old lubras as wives.2 Mr. Taplin states that 
among the Narrinyeri the men rarely marry before they are 
eighteen or twenty.3 
Infant- or child-betrothals are also practised, more or 
less frequently and sometimes almost as exclusive custom, 
among a large number of African tribes.4 The fiance may 
be a small boy, a youth, a man already married, or even a 
greybeard. In many cases two families agree that their 
1 Curr, The Australian Race, i. no. 
2 See infra, on Group-marriage and other Group-relations. 
3 Taplin, op. cit. p. 35. 
4 Ruelle, ' Notes sur quelques populations noires du 2e territoire 
militaire de l'Afrique Occidentale fran^aise,' in L'Anthropologic, 
xv. 684 (Mossi of the French Sudan). Henry, Les Bambara, p. 199. 
Sarbah, Fanti Customary Laws, p. 45. Hartter, ' Sitten und 
Gebrauche der Angloer (Ober-Guinea),' in Zeitschr. f. Ethnol. 
xxxviii. 44. Cruickshank, Eighteen Years on the Gold Coast of Africa, 
ii. 192 ; Bowman, loc. cit. p. 424 (natives of the Gold Coast). Bowdich, 
Mission to Ashantee, p. 302 ; Beecham, Ashantee and the Gold Coast, 
p. 126; Wilson, Western Africa, pp. 181 (Ashanti), 113 (Kru 
people of the Grain Coast). Ellis, Ewe-speaking Peoples of the Slave 
Coast of West Africa, p. 201 ; Spieth, Die Ewe-Stdmme, p. 182 
(Ewhe of Ho) ; Klose, Togo unter deutscher Flagge, p. 252 (F.whe of 
the interior). Eliis, Yoruba-speaking Peoples of the Slave Coast of 
West Africa, p. 183 ; Dennett, Nigerian Studies, p. 165 (Yoruba). 
Mockler-Ferryman, British Nigeria, p. 232. Thomas, Anthropological 
Report on the Edo-speaking Peoples of Nigeria, i. 47. Idem, Anthropo- 
logical Report on Ibo-speaking Peoples of Nigeria, iv. 61 (Asabe). 
Landolphe, Memoires, ii. 50 (people of Benin). Dennett, At the Back 
of the Black Man's Mind, p. 198 sq. (Bini). Partridge, Cross River 
Natives, p. 254. Zoller, Forschungsreisen in der deutschen 'Colonie 
Kamerun, ii. 58 (Dualla). Poupon, ' £tude des Baya,' in L'Anthro- 
pologic, xxvi. 126. Torday and Joyce, Les Bushongo, p. 115 (Ban- 
gongo). Torday, Camp and Tramp in African Wilds, p. 94 (Bambala). 
Johnston, George Grenfell and the Congo, ii. 677 (Mongwandi of the 
Upper Mongala region). Weeks, Among Congo Cannibals, p. 122 
(Boloki of the Upper Congo). Hahn, * Die Ovahererd,' in Zeitschr. 
d. Gesellsch. f. Erdkunde zu Berlin, iv. 490 ; v. Francois, Nama und 
Damara Deutsch-Sud-West-Afrika, p. 195 ; Schinz, Deutsch-Sud- 
West-Afrika, p. 172 (Herero). Gottschling, ' Bawenda,' in Jour. 
Anthr. Inst. xxxv. 373. Kidd. The Essential Kafir, p. 211. ' Theal, x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
359 
children shall intermarry. Among the Yao of the former 
German East Africa, for instance, it is said to be very common 
for a woman who has just had a child to say to a neighbour 
who is expecting one,- " I have a son-if you will have a 
daughter, let him marry her " ; and this, in due course, is 
done.1 In British Nyasaland the matrimonial future of a 
girl is nearly always settled within a few months of her birth, 
the fiance being generally a boy of tender years.2 Among 
some African peoples the betrothal is considered binding 
on the family of the girl,3 among others it is not.4 But 
before the man can take the girl to his home and consummate 
the marriage the bride price must be paid and the girl, as 
a rule, must have attained the age of from ten to twelve 
years at least. 
The age of about twelve or thirteen seems to be a very 
common average age for the marriage of girls in Africa.5 
Yellpiv and Dark-skinned People of Africa south of the Zambesi, 
pp. 159 (Barotse), 219 sq. (Makaranga). Holub, Seven Years in 
South Africa, ii. 314 ; Decle, Three Years in Savage Africa, p. 78 
(Barotse). Gouldsbury and Sheane, Northern Rhodesia, p. 156 
(Awemba). Miss Werner, Natives of British Central Africa, p. 129. 
Junod, Life of a South African Tribe, i. 99 (Pedi). Dahlgriin, 
' Heiratsgebrauche der Schambaa,' in Mittheil. Deutsch. Schutzgeb. 
xvi. 224. Fiilleborn, Das Deutsche Njassa- und Ruwuma-Gebiet, 
p. 345 sqq. (Konde people). Duff, Nyasaland tinder the Foreign 
Office, p. 311. Roscoe, Northern Bantu, pp. 38, 39 (Banyoro), 118 
(Bahima), 262 sq. (Bateso). Johnston, Uganda Protectorate, p. 747 
(Bantu Kavirondo). Hobley, Eastern Uganda, pp. 17 (Bantu 
Kavirondo), 28 (Nilotic tribes of Kavirondo). Gregory, Great Rift 
Valley, p. 343 (Wapokomo). Merker, Die Masai, p. 44. Munzinger, 
Ueber die Sitten und das Recht der Bogos, p. 61 sq. Grandidier, 
Ethnographic de Madagascar, ii. 163 sqq. 
1 Weule, Native Life in East Africa, p. 306. 2 Duff, op. cit. p. 311. 
3 Sarbah, op. cit. p. 45 sq. (Fanti). Cruickshank, op. cit. ii. 192 
(Gold Coast natives). Ellis, E^e-speaking Peoples of the Slave Coast, 
p. 201. 
4 Torday, op. cit. p. 94 (Bambala). Decle, op. cit. p. 490 (Akamba). 
5 Cruickshank, op. cit. ii. 193 (Gold Coast natives). Landolphe, 
op. cit. ii. 50 (people of Benin). Johnston, George Grenfell, ii. 678 
(Bakongo). Le Vaillant, Travels from the Cape of Good-Hope, into the 
Interior Parts of Africa, ii. 68 (Hottentots). Magyar, Reisen in Siid- 
Afrika, p. 283 (Kimbunda). Decle, op. cit. p. 490 (Akamba). Roscoe, 
Baganda, p. 87. Grandidier, op. cit. ii. 164 (Malagasy). 360 
THE HISTORY OF HUMAN MARRIAGE chap. 
Among the Hereto the girls are married when they are from 
twelve to fifteen years old,1 in the tribes near Fort Johnston 
in British Central Africa when they are about fifteen.2 In 
some Kafir tribes a girl is married as early as thirteen, but 
in others not until seventeen ; "if an average has to be 
struck," says Mr. Kidd, " it would be best to place sixteen 
as the average age for marriage."3 Among the Auin 
Bushmen the girls marry between about thirteen and 
sixteen.4 Among the Bassari of Togoland they do not 
marry, or are not allowed to marry, before fifteen or sixteen 
years of age,5 among the Akikuyu of British East Africa not 
before sixteen or seventeen,6 among the Wadshagga of 
Mount Kilimanjaro not before seventeen,7 among the 
Baganda according to a recent enactment of the native 
council of Uganda not before seventeen or eighteen.8 The 
age at which an African male enters into the married state 
depends upon his ability to pay the bride price, provided 
that he has attained the age of puberty. Among the 
Bassari in Togoland a man is not allowed to marry until he 
is seventeen years old,9 but elsewhere in West Africa he 
can do so as early as fifteen or sixteen.10 According to the 
recent enactment of the native council of Uganda, just 
referred to, a youth must not get married " under the 
age of twenty English years " ;11 but formerly young 
Baganda of about sixteen married, as there was no pro- 
hibition of marrying at any age after attaining puberty if 
the husband had enough property to pay the dowry.12 
1 Hahn, in Zeitschr. d. Gesellsch. f. Erdhunde zu Berlin, iv. 490. 
v. Francois, op. cit. p. 195. 
2 Stannus, ' Notes on some Tribes of British Central Africa,' in 
Jour. Roy. Anthr. Inst. xl. 310. 3 Kidd, op. cit. p. 211. 
4 Kaufmann, ' Die Auin,'in Mittheil. Deutsch. Schutzgeb. xxih. 156. 
6 Klose, op. cit. p. 508. 
• Routledge, With a Prehistoric People, p. 124. 
7 Volkens, Der Kilimandscharo, p. 251. 
8 Cunningham, Uganda and its Peoples, p. 164. 
9 Klose, op. cit. p. 508. 
10 Nassau, Fetichism in West Africa, p. 12. 
11 Cunningham, op. cit. p. 164. 
19 Felkin, 'Notes on the Waganda Tribe of Central Africa,'in 
Proceed. Roy. Soc. Edinburgh, xiii. 744. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
361 
According to Fanti customary law "it is the duty of the 
father to engage a wife for his son as soon after he reaches 
the age of puberty as possible." 1 Among the Herero the 
men marry from sixteen years of age upwards,2 among the 
Auin Bushmen between about sixteen and eighteen,3 in the 
tribes near Fort Johnston in British Central Africa4 and 
among the Central African Fors at about seventeen,5 among 
the Wanyamwezi to the south of the Victoria Nyanza from 
seventeen to twenty,6 among the Wafipa in the former 
German East Africa from eighteen to twenty,7 among the 
Wadshagga generally at twenty-four.8 Nowhere in Africa, 
so far as I know, have the men been reported to marry at an 
earlier age than in Madagascar, where the majority of them, 
according to M. Grandidier, did so when about fourteen 
years old.9 
To most savages marriage seems absolutely indispensable. 
Among very many uncivilised peoples, as we have seen, a 
man cannot easily, or cannot at all, gratify his sexual desire 
but in the married state or in relations which normally lead 
to marriage. On the other hand, where no such restrictions 
exist we sometimes hear of young men preferring to lead a 
bachelor's life. We have noticed instances of this among 
certain South Ameiican Indians ; and with reference to the 
1 Sarbah, op. cit. p. 45. 2 Hahn, loc. cit. p. 490. 
3 Kaufmann, loo. cit. p. 156. 
4 Stannus, in Jour. Roy. Anthr. Inst. xl. 310. 
0 Felkin, ' Notes on the For Tribe of Central Africa,' in Proceed, 
Roy. Soc. Edinburgh, xiii. 207. 
6 Decle, op. cit. p. 348. 
7 Fromm,' Ufipa-Land und Leute,' in Mitiheil. Deutsch. Schutzgeb 
xxv. 96. 
8 Volkens, op. cit. p. 251. 
9 Grandidier, op. cit. ii. 164.-For additional instances of early 
marriages among African peoples see Ward, A Voice from the Congo, 
p. 251 ; Torday and Joyce, Les Bushongo, p. 113 (" On pent voir 
des femmes mariees dont la poitrine est a peine form6e, et des peres 
famille qui sont encore des jeunes gardens ") ; Wilson and Felkin, 
Uganda and the Egyptian Soudan, ii. 145 sq. ; Felkin, ' Notes on the 
Madi or Moru Tribe of Central Africa,' in Proceed. Roy. Soc. Edin- 
burgh, xii. 323 ; Petherick, Egypt, the Soudan and Central Africa, 
p. 396 (Djour tribes on the White Nile) ; Munzinger, Ostafrikamsche 
Studien, p. 324 (Beni-Amer) ; Parkyns, Life in Abyssinia, ii. 41. 362 
THE HISTORY OF HUMAN MARRIAGE chap. 
Dravidian tribes of India Mr. Crooke remarks that the 
prenubial laxity of their girls " enables the men to avoid 
marriage till they are well advanced in life, and desire to 
found homes for their old age." 1 Yet even where the man 
can freely indulge in extra-matrimonial intercourse, marriage 
will sooner or later become a necessity for him. He must 
have a female companion who takes care of his house : she 
procures wood and water, lights and attends to the fire, 
prepares the food, dresses skins, makes clothes, gathers roots 
and berries, and among agricultural peoples very frequently 
cultivates the soil. He must have a woman who bears to 
him children, and nurses and looks after them ; for a man 
without offspring is an unfortunate being under savage con- 
ditions of life where individual safety and welfare depend 
upon family ties, and the old have to be supported by the 
young. The childless man may even have to suffer after 
his death for lack of offspring. The Eskimo about Bering 
Strait " appear to have great dread of dying without being 
assured that their shades will be remembered during the 
festivals, fearing if neglected that they would thereby suffer 
destitution in the future life " ; hence a pair of childless 
Eskimo frequently adopt a child, so that when they 
die there will be some one left whose duty it will be to make 
the customary feast and offerings to their shades at the 
festival of the dead.2 The Rev. J. Roscoe says that among 
the Bagesu of British East Africa marriage is " a matter of 
expediency rather than love, because children are the means 
of assisting the ghost after death." 3 Among the Ainu of 
Japan, according to Mr. Batchelor, there are various reasons 
why the men wish for at least one boy:-that he may act 
as family priest when the father dies ; that he may inherit, 
preserve, and hand down to posterity the principal heirlooms 
and family treasures; that he may act as the head of the 
family, and take the place of the father to the younger 
1 Crooke, Tribes and Castes of the North-Western Provinces and 
Oudh, i. p. cxcvii. 
2 Nelson, ' Eskimo about Bering Strait,' in Ann. Rep. Bur. 
Ethnol. xviii. 290. 
3 Roscoe, Northern Bantu, p. 172. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
363 
members thereof, should there be any ; and that he may keep 
the father in his old age.1 
Moreover, the young savage will generally find little 
difficulty in supporting • a family. Speaking of some 
aboriginal tribes in Farther India, Lieutenant-Colonel Low 
observes :-" In the infancy of society and where a youth 
of activity has equal means with the older persons of the 
tribe of subsisting himself by the chase or by fishing, he 
finds nothing to prevent his marrying early. In fact it 
becomes a matter requiring little thought. He soon per- 
ceives it to be his real interest to enter into that state." 2 
The same is true of pastoral and agricultural tribes; Bick- 
more's statement with reference to the Malays, that the 
difficulty in supporting a family is unknown to them,3 has 
a much wider application. Indeed, far from being a burden 
to a man, wife and children are frequently a source of 
prosperity. In his description of some Indians of Canada 
Heriot remarks that children form the wealth of savage 
tribes.4 
There are, however, even in savage life circumstances 
which may compel persons to live unmarried for a longer or 
shorter time. When the wife has to be bought, it may be 
difficult for a young man to provide the necessary price to 
be paid for her. Thus, with regard to the Zulus Mr. Eyles 
wrote to me that " young men who are without cattle have 
often to wait many years before getting married." When 
Campbell asked some Kandhs why they remained single, 
they replied that they did so because wives were too expen- 
sive.5 In the island of Pentecost, one of the Northern 
New Hebrides, according to Dr. Speiser, " young men, as a 
rule, either cannot marry, being too poor to buy a wife, or, 
at best, can only afford to pay for an old widow, a low-priced 
article. The young, pretty girls are generally bought by 
1 Batchelor, Ainu and their Folk-Lore, p. 231 sq. 
2 Low, * Karean Tribes or Aborigines of Martaban and Tavai,' in 
Jour. Indian Archipelago, iv. 418. See also Hunter, Rural Bengal, 
i. 205. 
3 Bickmore, op. cit. p. 278. 
4 Heriot, Travels through the Canadas, p. 337. 
5 Campbell, Wild Tribes of Khondistan, p. 143. 364 
THE HISTORY OF HUMAN MARRIAGE chap. 
old men." 1 In fact, of many simple peoples we are told 
that the necessity of paying a bride price is a more or less 
frequent obstacle to early marriages or even a cause of lasting 
celibacy.2 It affects of course in the first place the men, 
but may also increase the number of unmarried women. 
It is said that among the Munda Kols and Hos, in consequence 
of the high prices of brides, are to be found " what are 
probably not known to exist in other parts of India, respect- 
able elderly maidens " ;3 and in the Solomon Islands a 
chief's daughter seldom marries young on account of the 
exorbitant price demanded for her hand.4 Yet, however 
frequent it be, we must not exaggerate the importance of 
this obstacle to marriage. When a young man is not able 
to buy a wife for himself, he may in many cases acquire 
her by working for some time with her parents, or by eloping 
with her. Moreover, as Lord Avebury remarks, the price 
of a wife is generally regulated by the circumstances of the 
tribe, so that nearly every industrious young man is enabled 
1 Speiser, op. cit. p. 234. 
2 Richardson, Arctic Searching Expedition, i. 383 (Kutchin). 
Bogoras, op. cit. p. 570 (Tungus). Melnikow, ' Die Burjaten des 
Irkutskischen Gouvernements,' in Verhandl. Berliner Gesellsch. f. 
Anthrop. 1899, p. 441. Ahlqvist, Unter Wogulen und Ostjaken,' 
in Acta Societatis Scient. Fennicae, xiv. 291 (Ostyak). Idem, Die 
Kulturworter der westfinnischen Sprachen, p. 203 sq. (Tartars). 
Finley and Churchill, op. cit. p. 39 (Subanu of Mindanao). v. 
Brenner, op. cit. p. 248 (Battas). Zollinger, ' Lampong Districts,' 
in Jour. Indian Archipelago, v. 697. Romilly, ' Islands of the New 
Britain Group,' in Proceed. Roy. Geograph. Soc. N.S. ix. 8. Pfeil, 
op. cit. p. 26 sq. (natives of the Bismarck Archipelago). Thumwald, 
op. cit. iii. 81 (natives of Buin in the Solomon Islands, especially in 
former times). Penny, Ten Years in Melanesia, p. 93 (Solomon 
Islanders). Waitz-Gerland, op. cit. vi. 126 (Tahitians). Tessmann, 
Die Pangwe, ii. 262. Zoller, op. cit. ii. 58 (Dualla). Chavanne, 
Reisen im Kongostaate, p. 399 (Bafi6te tribes). Torday and 
Joyce, ' Notes on the Ethnography of the Ba-Mbala,' in Jour. 
Anthr. Inst. xxxv. 410. lidem, Les Bushongo, p. 114 (Bangongo). 
v. Weber, Vier Jahre in AJrika, ii. 216 (Kafirs). Petherick, op. cit. 
p. 393 (certain Dinka). 
3 Watson and Kaye, People oj India, i. no. 18. Dalton, op. cit. 
p. 192. 
4 Penny, op. cit. p. 93. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
365 
to get one.1 Speaking of the Sumatrans, Marsden observes 
that the necessity of purchasing does not prove such an 
obstacle to matrimony as is supposed, for there are few 
families who are not in possession of some small substance, 
and the purchase-money of the daughters serves also to 
provide wives for the sons.2 
A poor man may find it particularly difficult, if not 
impossible, to procure a wife where the males outnumber 
the females, as is the case in many savage communities.3 
When such a disproportion between the sexes is combined 
with the practice of polyandry it need not of course lead 
to celibacy among the men, but when it occurs among a 
non-polyandrous people there must inevitably be a certain 
number of men who are unable to marry. This must, for 
instance, be the case in many of the South Sea Islands,4 
where there is often a striking excess of males over females.5 
Concerning the Easter Islanders we have the statement that 
bachelors were very frequent among them owing to the 
great preponderance of males.6 With regard to the late age 
at which the Solomon Islanders of Buin formerly married, 
Dr. Thurnwald observes that it was due partly to the high 
bride price and partly to the smaller number of females 
compared with that of males. He adds that in spite of this 
polygyny exists there;7 and Hagen holds polygyny respon- 
sible for the large number of unmarried men in other parts 
of the Solomon Islands.8 On the Gazelle Peninsula of New 
1 Avebury, Origin of Civilisation, p. 106. Cf. Bosman, loc. cit. 
pp. 419, 424 (Negroes of the Gold Coast). 
2 Marsden, op. cit. p. 256 sq. 
3 See infra, on Monogamy and Polygyny. 
4 In Micronesia it is common for the poorer class and the slaves 
to be doomed to perpetual celibacy (Wilkes, Narrative of the United 
States Exploring Expedition, v. 74 ; Romilly, Western Pacific, p. 69 
sq. ; Waitz-Gerland, op. cit. vol. v. pt. ii. 125). In Tikopia many 
people never marry (Rivers, History of Melanesian Society, i. 346). 
5 See infra, on Monogamy and Polygyny. 
6 Geiseler, op. cit. p. 29. See also Cooke, in Smithsonian Report, 
1897, p. 716. 
7 Thurnwald, op. cit. iii. 81. 
8 Hagen, ' Les indigenes des ties Salomon,' in L' Anthropologic, 
iv. 4. See also Parkinson, Dreissig Jahre in der Sildsee, p. 481 sq. 366 
THE HISTORY OF HUMAN MARRIAGE chap. 
Britain (New Pomerania) the males are in the majority, 
polygyny is extensively practised by those who can afford 
it, and many a poor young man can get no1 wife at all.1 
Among the Australian aborigines there are likewise, as a 
rule, more males than females, the old men appropriate to 
themselves a plurality of women, especially young ones, 
and in consequence the younger men have generally to wait 
a long time until they can marry.2 
Polygyny, in connection with an unequal distribution of 
property, is very frequently reported to cause celibacy 
among the poorer and younger men. I have received state- 
ments to this effect from several correspondents living 
among different savage peoples-from Mr. Hadfield with 
reference to the natives of Lifu, from Mr. Cousins with 
reference to the Cis-Natalian Kafirs, from Mr. Ingham with 
reference to the Bakongo, from Dr. Sims with reference to 
the Bateke. Mr. Weeks writes of the Boloki on the Upper 
Congo that the effect of polygyny among them was to tie 
up the women to a comparatively small number of men who 
had either inherited them or got the wealth with which to 
pay their marriage money. " There was a constant com- 
plaint amongst the young and vigorous men of the middle 
and lower orders that it was almost impossible for them to 
procure wives. Thus we found a small number of men 
possessing nearly all the women in a town, having from 
four or five up to twenty-five and thirty each, and a large 
number of young men who could not secure wives." 3 So 
also, according to Wilson, among the maritime tribes of 
Southern Guinea, where polygyny flourishes, " a very large 
proportion of the younger male members of society have 
no wives." 4 In Makin, one of the Gilbert Islands, a great 
number of young men were unmarried owing to the majority 
of the women being monopolised by the wealthy and 
1 Burger, op. cit. p. 28 sq. 
2 See infra, on Group-marriage and other Group-relations. 
3 Weeks, Among Congo Cannibals, p. 135. See also Idem, ' Anthro- 
pological Notes on the Bangala of the Upper Congo River,' in Jour. 
Roy. Anthr. Inst. xl. 418 sq. 
1 Wilson, Western Africa, p. 266. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
367 
powerful.1 Among the Kutchin Indians, according to 
Hardisty, there are but few young men who have wives- 
unless they can content themselves with some old cast-off 
widow-on account of all the chiefs, medicine-men, and 
those who possess rank acquired by property having two, 
three, or more wives.2 Polygyny may be a cause of male 
celibacy even where there are many more females than 
males. Dr. Felkin states that among the Baganda, owing 
to the frequency of polygyny, a large number of the poorer 
men were unable to marry although the proportion of 
females to males was as 31 to 1.3 So also among the Konde 
people in the former German East Africa, who have a striking 
excess of girls, many young men are compelled to lead a 
single life because the old and well-to-do men indulge in a 
plurality of wives.4 On the other hand, it must not be 
supposed that wherever polygyny occurs, it necessarily or 
even generally leads to compulsory celibacy for a considerable 
number of men. Among most peoples practising polygyny 
it is restricted to a small minority of the population, and 
very often it is combined with a surplus of females, which 
makes it possible for practically every man to procure a 
wife although there are some who have more than one. 
Among the polygynous Waguha of Lake Tanganyika, for 
instance, as I am informed by Mr. A. J. Swann, unmarried 
grown-up men do not exist, the females being more numerous 
than the males. It should also be noticed that polygyny 
could hardly have been a potent cause of celibacy at the 
lowest stages of civilisation, where it was little practised ; 
nor is it likely that men had to remain unmarried on account 
of poverty in those primitive conditions where there was no 
accumulated property worth speaking of. 
Whilst polygyny may lead to celibacy among the men, its 
effect upon the women is exactly the opposite. As Junod 
1 Wilkes, op. cit. v. 102. 
2 Hardisty, ' Loucheux Indians,' in Smithsonian Report, 1866, 
p. 312. 
8 Felkin, in Proceed. Roy. Soc. Edinburgh, xiii. 744. See also 
Wilson and Felkin, op. cit. i. 224. 
4 Zache, quoted by Fiillebom, op. cit. p. 345. 368 
THE HISTORY OF HUMAN MARRIAGE chap. 
observes in his description of the life of a South African 
tribe, " every girl finds a husband in the land where polygamy 
flourishes." 1 And it seems that in countries unaffected by 
European civilisation polygyny is practised whenever the 
women are in the majority ; hence the almost complete 
absence of old spinsters among savage races. For it must 
not be supposed that polyandry in some countries leads to 
that which polygyny prevents in others. Apart from the 
fact that among peoples practising polyandry the males 
are, often at least, more numerous than the females, it is 
among many of them quite an exceptional form of marriage ; 
and in no case is it known to be the exclusive form, mono- 
gamy, polygyny, and sometimes group-marriage being 
practised side by side with it. 
Among the obstacles barring the way to matrimony for 
many a young savage should also be mentioned the rule of 
exogamy. Dr. Bogoras remarks that unmarried men are 
more frequently met with among the Tungus than among 
the Chukchee, not only on account of the considerable price 
to be paid for a bride, but " because marriages are strictly 
exogamic, and a bride may be taken only from another 
clan than that of the bridegroom." 2 Concerning the natives 
of Pentecost Dr. Speiser observes, " All marriages between 
near relations being impossible, the chances to marry at 
all are considerably diminished, so that nowadays, with the 
decreased population, a man very often cannot find a wife 
even though surrounded by any number of girls." 3 To the 
causes already referred to, which partly account for the 
large number of bachelors among the Australian aborigines, 
should be added their peculiar marriage rules, which only 
allow marriages between members of certain small groups 
of people.4 
Sometimes a military life is a hindrance to marriage 
1 Junod, op. cit. i. 184. See also ibid. i. 273. 
2 Bogoras, op. cit. p. 570. 3 Speiser, op. cit. p. 234. 
4 Cf. Schulze, in ' Aborigines of the Upper and Middle Finke 
River,' in Trans, and Proceed. Roy. Soc. South Australia, xiv. 
p. 224 ; Strehlow, Die Aranda- und Loritja-Stamme, vol. iv. pt. i. 
98. See infra, on Exogamy. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
369 
among savages as well as among civilised nations. Old 
writers tell us that among the Iroquois and neighbouring 
peoples young warriors seldom married or had anything to 
do with women until they became .thirty years of age, 
because, as they said, commerce with women exhausted 
their knees, and rendered them heavy in the course. Those 
who married before that age were looked upon as unfit for 
war or hunting and were despised as effeminate persons.1 
Among the Masai of Eastern Africa the warriors are abso- 
lutely prohibited from marrying-but, as we have noticed 
before, not from having extra-matrimonial connections with 
the other sex.2 The younger men, or warriors, do not dwell 
with the rest of the tribe, but have separate kraals where 
they five with the immature unmarried girls ; and they are 
not allowed by the elders of the tribe to marry until they 
have reached the age of about thirty and have accumulated 
a fair amount of property, or else have so distinguished 
themselves by their bravery as to merit an early retirement. 
The warrior then becomes himself an elder (pLmoruo). 
According to their own traditions, the practice of living with 
the girls was permitted to the warriors because it was 
found that the girls, if kept apart from their own warriors, 
allowed themselves to be courted by the enemy and betrayed 
the interests of the tribe.3 " In the Zulu-country," says 
Shooter, " bachelors require the king's permission to take 
wives-a permission sometimes not given until they are 
thirty or thirty-five years old. The reason of its being so 
long withheld is that unmarried men are thought to make 
better soldiers than those who have wives and families to 
attach them to life. Tshaka permitted very few of his 
people to marry, but his successors have found it politic to 
be more indulgent. This unnatural practice, introduced 
by Tshaka, seems to have been based on an existing institu- 
1 Hennepin, New Discovery of a Vast Country in America, p. 483. 
Lahontan, New Voyages to North-America, p. 451. 
2 Supra, p. 134. 
3 Eliot, East Africa Protectorate, p. 136 sq. Johnston, Uganda 
Protectorate, p. 822. Hinde, The Last of the Masai, p. 105. Merker, 
op. cit. pp. 45, 334. Hollis, Masai, p. 292. 370 
THE HISTORY OF HUMAN MARRIAGE chap. 
tion."1 In this connection it may be noticed that in 
ancient Rome the milites gregales, who were Roman citizens, 
were prohibited from marrying. The women with whom 
they cohabited-commonly, as it appears, called focariae- 
were regarded as concubines, even when they had been the 
regular wives of the soldiers before the date of enrolment. 
These restrictions, however, seem to have been removed by 
Septimius Severus ; in any case, it is certain that the milites 
gregales in the fourth century possessed the right to marry 
during service.2 In a subsequent chapter I shall deal with 
the Nayars of Malabar and the Zaporog Cossacks, who were 
also forbidden to contract actual marriages though allowed 
to have mistresses.3 In the Sandwich Islands most of the 
people about court remained unmarried, but homosexual 
practices and fornication were common among them.4 
The difficulties experienced by the men in procuring wives, 
or young wives, may be one cause of the practice of infant- 
or child-betrothal, which is so widely spread among the 
uncivilised races. Its apparently universal prevalence 
among the Australian aborigines is no doubt connected with 
the great demand for women in their tribes. Of the natives 
of Eastern Victoria we are told that more than one girl is 
usually betrothed to the same youth, as the girl may die, 
and that on the other hand one girl may also be betrothed 
to several young men; if the man to whom she was first 
betrothed dies before he is old enough to claim her, then she 
becomes the wife of one of the others.6 To acquire a girl 
in her childhood may also be a matter of economy. Among 
the Mongwandi of the Upper Mongala region on the Congo 
the youth selects by preference a girl child of six or seven 
years because she can be bought cheaper at that age than 
when she grows older.6 So also the child-betrothals among 
the Negritos of Northern Zambales in Luzon are attributed 
by Mr. Reed to a desire of the boy's parents to get the girl 
cheaper than they could by waiting until she was of marriage- 
1 Shooter, Kafirs of Natal and the Zulu Country, p. 47. 
2 Meyer, Der romische Konkubinat, p. 106. « 
3 Infra, on Polyandry. * Malo, Hawaiian Antiquities, p. 93. 
5 Mathews, Ethnological Notes on the Aboriginal Tribes of N. S. 
Wales and Victoria, p. 100 sq. 6 Johnston, George Grenfell, ii. 677. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
371 
able age, as well as to a desire on the part of the girl to come 
into early possession of the things which are paid for her; 
and in support of this view he mentions the fact that the 
practice is not met with in Southern Zambales and Bataan, 
where marriage does not seem to partake so much of the 
nature of a sale as in the other district.1 With special 
reference to child-marriage in the Malay Archipelago it has 
been observed that where elopement or the abduction of 
girls is a recognised custom parents are anxious to marry 
their daughter as early as possible in order to prevent an 
undesirable match.2 The child-betrothal may further be 
a means of preserving the virginity of the girl, which among 
so many peoples is a highly valued quality in a bride. 
Among the Yoruba of the Slave Coast, for instance, " non- 
virginity in a bride is only a valid ground for repudiation 
when the girl has been betrothed at a tender age ; for 
unbetrothed girls can bestow their favours upon whom they 
please."3 Finally, a very frequent motive for infant- or 
child-betrothals among uncivilised peoples is the wish of 
two families to be drawn together or to cement and per- 
petuate their friendship, or a desire to become connected 
with a family of importance.4 In many cases such betrothals 
are said to be particularly prevalent among people of conse- 
quence or wealth or among chiefs, and may thus be alliances 
of political moment.5 
1 Reed, op. cit. p. 57. 
2 Schmidt, Liebe und Ehe im alien und modernen Indien, p. 308. 
3 Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 184. 
4 Dennett, Nigerian Studies, p. 165 (Yoruba). Cruickshank, 
op. cit. ii. 192 (natives of the Gold Coast). Bowdich, op. cit. p. 302 
(Ashanti). Zoller, op. cit. ii. 58 (Dualla). Munzinger, Ueber die 
Sitten und das Recht der Pogos, p. 61 sq. Mathews, op. cit. p. 96 sq. 
(natives of Eastern Victoria). Williamson, Mafulu Mountain People of 
British New Guinea, p. 170. Gisborne, op. cit. p. 27 ; Polack, op. cit. i. 
*35 Yb • Best, in Trans, and Proceed. New Zealand Institute, xxxvi. 43 
(Maori). Tautain, in L'Anthropologic, vi. 644 (Marquesas Islanders). 
Stair, Old Samoa, p. 171. Geiseler, op. cit. p. 29 (Easter Islanders). 
6 Ellis, Yoruba-speaking Peoples, p. 183. Dennett, At the Back of 
the Black Man's Mind, p. 198 (Bini of Southern Nigeria). Codrington, 
Melanesians, pp. 237, 241. Best, loc. cit. p. 42 sq. (Maori). Tautain, 
loo. cit. p. 645 (Marquesas Islanders). Jenks, op. cit. p. 68 (Bontoc 
Igorot). 372 
THE HISTORY OF HUMAN MARRIAGE chap. 
Among many uncivilised peoples younger sons1 or 
daughters,2 or both younger sons and daughters,3 must not 
marry before the elder ones as long as there is a chance of 
the latter getting married ; or it is at any rate considered 
improper for them to do so. Such customs are also found 
in China, where children are married according to seniority,4 
and among Semitic and Aryan peoples. We read in Genesis : 
" Laban said, It must not be so done in our country, to give 
the younger before the firstborn."5 In modern Egypt a 
father very often objects to marrying a younger daughter 
before an elder.6 Among the Vedic people it was considered 
proper that the younger brothers and sisters should not 
1 Dugmore, in Maclean, Compendium of Kafir Laws and Customs, 
p. 45 ; Kidd, op. cit. p. 211 (Kafirs). Wilken, in Bijdragen tot de taal-, 
land- en volkenkzmde van Nederlandsch-Indie, ser. v. vol. i. 143 (Battas 
of Sumatra). Carey and Tuck, Chin Hills, i. 189 (Siyin tribe). 
2 Casalis, Basutos, p. 184 (Bechuanas). Junod, Life of a South 
African Tribe, i. 235 (Thonga). Heese, ' Sitte und Branch der Sango,' 
in Archiv f. Anthrop. N.S. xii. 134 (Sango of East Africa). Riedel, 
' Galela und Toberloresen,' in Zeitschr. f. Ethnol. xvii. 76 (natives of 
Halmahera, to the west of New Guinea). Snouck Hurgronje, op. cit. 
i. 301 (Achehnese of Eastern Sumatra and other natives of the Malay 
Archipelago), v. Rosenberg, Der malayische Archipel, p. 155 (natives 
of Nias). Among certain tribes in the Malay Archipelago a man who 
marries a younger sister before an elder must pay the tatter a fine 
(Frazer, Folk-Lore in the Old Testament, ii. 291). Among the Salish 
of British Columbia the eldest daughter is always the first to be 
disposed of in marriage, and it is customary'- for her husband to marry 
her younger sisters afterwards (Hill Tout, ' Report on the Ethnology 
of the Stlatlumh of British Columbia,' in Jour. Anthr. Inst. xxxv. 
133). Among the Meitheis (Hodson, Meitheis, p. 76 n. *) and Garos 
(Playfair, Garos, p. 69) of Assam a man may marry two sisters only 
on condition that he marries the elder before the younger. Many 
other races allow or even encourage a man to marry all his wife's 
sisters " provided that he marries them one after the other in order 
of seniority, beginning with the eldest and working his way down 
to the youngest " (Frazer, op. cit. ii. 264). 
3 Nassau, op. cit. p. 9 (West Africans). Thumwald, ' Ermittlungcn 
fiber Eingeborenenrechte der Sfidsee. A. Buin auf Bougainville 
(Deutsche Salomo-Inseln),' in Zeitschr. f. vergl. Rechtswiss. xxiii. 335. 
Sorge, ' Nissan-Inseln im Bismarck-Archipel,' in Steinmetz, Rechts- 
verhaltnisse von eingeborenen Vblkern in Afrika und Ozeanien, p. 407. 
4 Gray, China, i. 190. 6 Genesis, xxix. 26. 
6 Lane, Manners and Customs of the Modern Egyptians, p. 172. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
373 
anticipate their elders by marrying before them ;1 and in 
the ancient law-books of India a transgression of this rule of 
marrying in order of seniority is punished with damnation, 
or, at any rate, has to be atoned for by the performance of 
a penance.2 In the Punjab Muhammadans and Sikhs, as 
well as Hindus, consider it a disgrace to marry the younger 
son or daughter before the elder one.3 In Serbia and Bulgaria 
custom requires that the eldest son should marry before his 
younger brothers ; and the daughters always precede their 
brothers in marriage, except when one of the sisters is 
married and the other still is a child, in which case the 
brother is not obliged to wait till his younger sister is 
marriageable.4 In modern Greece, too, it is considered 
wrong for the sons to marry until all the daughters have 
been disposed of, and the latter must marry in order of 
seniority.5 In Ireland it was likewise usual that girls 
should be married in order of age, beginning with the eldest;6 
and reminiscences of the same rule survive in English and 
Scotch marriage ceremonies. Brand quotes a statement 
to the effect that " if the youngest daughter in a family 
should chance to be married before her elder sisters, they 
must all dance at her wedding without shoes. This, it is 
held, will counteract their ill-luck, and procure them 
husbands."7 This custom, which is also alluded to by 
Shakespeare,8 is still observed or remembered in Shropshire 8 
and the north of England.10 In Worcestershire an elder 
1 Macdonell and Keith, Vedic Index of Names and Subjects, i. 476. 
2 Laws of Manu, iii. 172. Sacred Laws of the Aryas, ii. 103 
(Vasishtha), 217 (Baudhayana). Institutes of Vishnu, liv. 16. 
3 Pandit Harikishan Kaul, Census of India, 1911, vol. xiv. 
(Punjab) Report, p. 268. 
4 Demelic, quoted by Frazer, Folk-Lore in the Old Testament, 
ii. 287 sq. Piprek, Slaivische Brautwerbungs- und Hochzeitsgebrauche, 
p. 137 (Bulgaria). 
8 Rodd, Customs and Lore of Modern Greece, p. 92. 
6 Joyce, Social History of Ancient Ireland,.ii. 6. 
7 Brand, Observations on Popular Antiquities, p. 398. 
8 Shakespeare, Taming of the Shrew, ii. 1. 33. 
9 Miss Burne, Shropshire Folk-Lore, p. 290. 
10 Henderson, Notes on the Folk-Lore of the Northern Counties of 
England and the Borders, p. 41. 374 
THE HISTORY OF HUMAN MARRIAGE chap. 
sister has to dance barefooted or to jump over a pig trough 
at the wedding of a younger sister,1 and in Gloucestershire 
she has to carry a broom on the same occasion as a penalty 
for allowing the younger sister .to get engaged and married 
before her.2 In Wales, " if the youngest of a family was 
married before the eldest, the seniors had to dance shoeless 
for penance to the company " 3-which seems to imply that 
elder brothers also had to dance without shoes at the 
weddings of their younger brothers.4 In the neighbourhood 
of Balmoral, when a younger sister marries before the elder, 
the latter is forcibly made to wear green garters at the 
wedding, and any young man who takes them off is destined 
to be her future husband.5 In the north-east of Scotland 
the bride had to give her unmarried elder sister green garters;6 
whilst among the mining folk of Fife green garters were 
surreptitiously pinned on to the clothing of the elder un- 
married brother or sister of the bride, and afterwards, when 
discovered, tied round the left arm and worn for the rest of 
the evening.7 It is obvious that all these customs are 
ultimately based on the idea that men and women should 
marry as soon as they arrive at the proper age, and that it 
is unnatural for an elder brother or sister to remain unmarried 
when a younger one becomes marriageable. 
Among peoples of archaic culture, as among most 
uncivilised races, celibacy is a great exception and marriage 
is regarded as a duty. In ancient Mexico, when a young 
man became between twenty and twenty-two years of age, 
a wife was singled out for him, unless he was going to be a 
priest ; and for girls the customary marriage age was from 
eleven, or, according to Clavigero, sixteen, to eighteen. In 
Tlascala, we are told, the unmarried state was so despised 
that a grown-up man who would not marry had his hair 
1 Notes and Queries, ser. vi. vol. ix. 135. 
2 Ibid. ser. vi. vol. viii. 147. 
3 Marie Trevelyan, Folk-Lore and Folk-Stories of Wales, p. 274. 
4 Cf. Frazer, op. cit. ii. 288. 
3 Miss Burne, in Folk-Lore, xix. 339. 
0 Gregor, Notes on the Folk-Lore of the North-East of Scotland, p. 90. 
7 Simpkins, County Folk-Lore, vol. vii. Fife, p. 393. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
375 
cut off for shame.1 Among the ancient Peruvians, every 
year or every two years each governor in his district had to 
arrange for the marriage of all the young men at the age of 
twenty-four and upwards, and of all the girls from eighteen 
to twenty.2 
" Almost all Chinese," says Dr. Gray, " robust or infirm, 
well-formed or deformed, are called upon by their parents 
to marry as soon as they have attained the age of puberty. 
Were a grown-up son or daughter to die unmarried, the 
parents would regard it as most deplorable." Hence a 
young man of marriageable age, whom consumption or 
any other lingering disease had marked for its own, would 
be compelled by his parents or guardians to marry at 
once.3 So indispensable is marriage considered by the 
Chinese that even the dead are married, the spirits of all 
males who die in infancy or in boyhood being in due time 
married to the spirits of females who have been cut off at 
a like early age.4 To die without leaving a son to perpetuate 
the family cult is considered one of the greatest misfortunes 
that could befall a man, and at the same time an offence 
against the whole line of ancestors. For it would doom 
father, mother, and all the ancestry in the Nether-world to 
a pitiable existence without descendants enough to serve 
them properly, to worship at the ancestral tombs, to take 
care of the ancestral tablets, and duly to perform all rites 
and ceremonies connected with the departed dead. Accord- 
ing to Confucius no crime is greater than filial impiety ; and 
Mencius said that " there are three things which are unfilial, 
and to have no posterity is the greatest of them." Conse- 
quently there can be no greater crime that a man can commit 
than to remain single. And if his wife has reached her 
fortieth year without bringing him a son, it is an imperative 
duty for him to take a concubine.5 
1 Clavigero, History of Mexico, i. 319. Bancroft, op. cit. ii. 251 sq. 
2 Garcilasso de la Vega, First Part of the Royal Commentaries of the 
Yncas, i. 306 sq. 3 Gray, op. cit. i. 186. 4 Ibid. i. 216 sq. 
5 Hozumi, Ancestor-Worship and Japanese Law, p. 129 sqq. de 
Groot, Religiotts System of China (vol. ii.) book i. 617. Giles. 
Strange Stories from a Chinese Studio, i. 64 n. 10. Indo-Chinese 
Gleaner, iii. 58. 376 
THE HISTORY OF HUMAN MARRIAGE chap. 
The doctrine of the Chinese moralists on marriage was 
also taught in Japan for more than a thousand years. The 
consequence of it was that, although celibacy was not posi- 
tively forbidden by law, it was denounced by public opinion ; 
indeed, according to Professor Hozumi, the obligation to 
marry was so effectively insisted upon by the latter that 
there was no need of enforcing it by legislation. But he 
adds that the customary prohibition of celibacy only extended 
to the present or future head of a family. As to the other 
members of the family, on the contrary, celibacy was, as a 
rule, obligatory. Before the Restoration of 1868 only the 
house-head, his eldest son, who was the presumptive heir, and 
his eldest grandson, who would become the presumptive heir 
after the eldest son, were allowed-or obliged-to marry, but 
younger sons could not lawfully contract marriage. There 
was no need for the latter to marry and have children, 
because they had no apparent hope of ever becoming 
house-heads and continuators of the cult. This rule was 
strictly followed among the samurai, or military class ; for 
permission was not given by their feudal lords for the 
marriage of younger sons. Among merchants, artisans, and 
farmers, again, there were sometimes exceptional cases in 
certain localities, but in such cases the newly married 
couple generally established a " new house." 1 At the time 
of the Taiho-ryo (a.d. 701-1192) a male was allowed to 
marry at the age of fifteen and a female at thirteen ;2 but 
the present Civil Code requires the age of seventeen for the 
male and fifteen for the female.3 
In Corea the men usually marry at the age of fifteen or 
sixteen, and it is considered correct for them to marry girls 
a year or two older than themselves; but it frequently 
happens that a girl of twelve or thirteen marries a boy of 
ten or less.4 Very often the marriage is decided upon when 
1 Hozumi, op. cit. pp. 131, 136 sqq. Cf. Kikuchi, Japanese 
Education, p. 264. 
2 Nakajima, 'Marriage (Japanese and Korean),' in Hastings, 
Encyclopedia of Religion and Ethics, viii. 459. 
3 Civil Code of Japan, § 765, p. 201. 
4 Saunderson, ' Notes on Corea and its People,' in Jour. Anthr. 
Inst. xxiv. 305. Nakajima, loc. cit. p. 460, x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
377 
the future bride and bridegroom are in their early infancy.1 
It is said that " the male human being who is unmarried is 
never called a ' man,' whatever his age, but goes by the 
name of ' yatow,' a name given by the Chinese to un- 
marriageable young girls ; and the ' man ' of thirteen or 
fourteen has a perfect right to strike, abuse, order about the 
' yatow ' of thirty, who dares not as much as open his 
lips to complain." 2 
Among the Semites we meet with the idea that a dead 
man who has no children will miss something in Sheol 
through not receiving that kind of worship which ancestors 
in early times appear to have received.3 The Hebrews 
looked upon marriage as a religious duty.4 According to the 
Shulman 'Arukh, the recognised Jewish code, he who abstains 
from marrying is guilty of bloodshed, diminishes the image 
of God, and causes the divine presence to withdraw from 
Israel; hence a single man past twenty may be compelled 
by the court to take a wife.5 The Mishna regards eighteen 
as the normal age of mai riage for a man, whilst girls were 
treated as marriageable from the beginning of their 
thirteenth year, when they attained their majority.6 But 
at various times very youthful marriages have prevailed 
among the Jews : by the thirteenth century a large pro- 
portion of Jewish girls were married during their minority, 
and in the second half of the seventeenth century the 
bridegroom was frequently not more than ten years old 
and the bride was younger still.' 
Although Islam considers marriage a civil contract, it 
1 Saunderson, loc. cit. p. 305. 8 Ross, History of Corea, p. 313. 
3 Cheyne, ' Harlot,' in Cheyne and Black, Encyclopedia Biblica, 
ii. 1964. 
4 Mayer, Rechte der Israeliten, pp. 286, 353. Lichtschein, Ehe 
nach mosaisch-talmudischer Aztffassung, p. 5 sqq. Klugmann-, Die 
Frau im Talmud, p. 39 sq. 
6 Shulhan 'Arukh, iv. (Ebhen ha-'ezer) i. 1,3. See also YebhamSth. 
fol. 63 b. sq., quoted by Margolis, ' Celibacy,' in Jewish Encyclopedia, 
iii. 636. 
6 Abrahams, Jewish Life in the Middle Ages, p.. 166 sq. Idem. 
Marriage (Jewish),' in Hastings, op. cit. viii. 461. 
7 Idem, Jewish Life in the Middle Ages, p. 167 sq.. 378 
THE HISTORY OF HUMAN MARRIAGE chap. 
nevertheless enjoins it as a religious duty " incumbent on 
all who possess the ability." 1 " When a servant of God 
marries, verily he perfects half his religion." 2 It is related 
in the Traditions that the Prophet once asked a man if 
he was married, and being answered in the negative, said, 
" Art thou sound and healthy ? " When the man replied 
that he was, the Prophet said, " Then thou art one of the 
brothers of the devil." 3 In Morocco I was told that a married 
man is blessed in this life and goes to Paradise after death, 
whereas a grown-up man who dies a bachelor does not find 
the road to Paradise, but will rise again with the devil.4 
In Muhammadan countries it is certainly considered im- 
proper and even disreputable for a man to abstain from 
marrying when he has attained a sufficient age and when 
there is no just impediment;5 and, as Niebuhr remarked, 
" nothing is more rarely to be met with in the East, than a 
woman unmarried after a certain time of life." She will 
rather marry a poor man, or become second wife to a man 
already married, than remain in a state of celibacy.6 Among 
the Persians, according to Dr. Polak, almost every girl of 
good repute is married before her twenty-first year, and 
old bachelors are unknown.7 Elphinstone states that in 
Afghanistan the common age for marriage is twenty for 
the men and fifteen or sixteen for the women. Men unable 
to pay the price of a wife are often unmarried till forty, 
and women are sometimes single till twenty-five ; but on 
the other hand the rich sometimes marry before the age of 
puberty.8 
1 Sayings of Muhammad, edited by Abdullah al-Mamun al- 
Suhrawardy, p. 55. 
2 Quoted by Lane, Arabian Society in the Middle Ages, p. 221. 
3 Mishhat, book xiii. ch. i., quoted by Hughes, Dictionary of 
Islam, p. 313. 
4 See Westermarck, Marriage Ceremonies in Morocco, p. 359. 
5 See, 5.g.,Lane, Manners and Customs of the Modern Egyptians, 
i. 213. 
6 Niebuhr, ' Travels in Arabia,' in Pinkerton, op. cit. x. 151. 
Cf. Burckhardt, Notes on the Bedouins and, Wahabys, p. 64. 
7 Polak, Persien, i. 205. Cf. Wilson, Persian Life and Customs, 
p. 261. 
8 Elphinstone, Accoimt of the Kingdom of Kaubul, i. 236 sq. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
379 
The so-called Aryan nations in ancient times, as M. 
Fustel de Coulanges and others have pointed out, regarded 
celibacy as an impiety and a misfortune : "an impiety, 
because one who did not marry put the happiness of the 
manes of the family in peril; a misfortune, because he 
himself would receive no worship after his death." A 
man's happiness in the next world depended upon his 
having a continuous line of male descendants, whose duty 
it would be to make the periodical offerings for the repose 
of his soul.1 It was no doubt possible to adopt a son ; but 
in the Rig-Veda at least that custom is plainly viewed as 
unsatisfactory.2 The old idea still survives in India: "a 
Hindu man must marry and beget children to perform his 
funeral rites, lest his spirit wander uneasily in the waste 
places of the earth."3 Marriage is a duty which every 
parent must perform for his children ; otherwise they owe 
him no reverence. But although a man who is unmarried 
is looked upon as an almost useless member of the com- 
munity and, indeed, as beyond the pale of nature, he does not 
disgrace his family. On the other hand, there is no greater 
reproach for a high-class Hindu than to have a daughter 
unmarried at the age of puberty. A family with such a 
daughter is supposed to labour under the displeasure of 
the gods; on a strict rendering of certain texts her 
unmarried state entails retrospective damnation on three 
generations of ancestors.4 Of the unmarried girls in India 
in general only one in every fourteen has turned her 
fifteenth year, and of the bachelors' only one in twenty-four 
is over thirty ; and among the Hindus marriage is con- 
tracted at an earlier age than in the population as a whole. 
1 Fustcl de Coulanges, La cite antique, p. 54 sq. Hearn, The 
Aryan Household, pp. 69, 71* Mayne, Treatise on Hindu Law and 
Usage, p. 83. 
2 Rig-Veda, vii. 4.7.8. Macdonell and Keith, op. cit. i. 486. 
3 Risley, People of India, p. 154. See also Monier-Williams, 
Brahmanism and Hinduism, p. 243 sq.; Gait, op. cit. p. 263. 
4 Sleeman, Rambles and Recollections of an Indian Official, i. 50. 
Dubois, Description of the Character, &c. of the People of India, 
p. 96 sq. Risley, op. cit. pp. 154, 191. Forster, Sketches of the 
Mythology, &c. of the Hindoos, p. 58. 380 
THE HISTORY OF HUMAN MARRIAGE chap. 
At the higher ages practically no one is left unmarried, except 
persons suffering from some infirmity or disfigurement, 
beggars, prostitutes, concubines, religious devotees and 
mendicants, and a few members of certain hypergamous 
groups who have been unable to effect alliances of the kind 
which alone are permitted to them by the rules of their 
community.1 
A considerable number of Hindus of both sexes are married 
already as infants, namely, 10 males and 18 females in 
every thousand children in the age-period 0-5, 48 males and 
132 females in the age-period 5-10, and 159 males and 488 
females in the age-period 10-15. Infant-marriage prevails 
chiefly in Bihar and Orissa, Bombay, Baroda, the Central 
India Agency, and Hyderabad, that is, in certain central 
tracts, whereas it is rare in the east, west, and south of 
India.2 The Hindus as a body are more addicted to it than 
any other religious community ; it is much rarer among 
the Muhammadans and much rarer still among Christians 
and Animists, whilst among the Buddhists marriage below 
the age of ten is practically unknown and marriage below 
the age of fifteen extremely rare.3 It should be noticed 
that the Hindu child-marriage does not as a rule imply 
immediate cohabitation. Yet in some parts of the country 
cohabitation often takes place before the child-wife has 
reached the age of puberty, and it does so, at the latest, 
immediately after her first menstruation.4 
These early betrothals have been attributed to various 
causes. First of all it should be noted that they cannot be 
regarded as survivals of a primitive custom, either 
Dravidian or Aryan ; for most of the Dravidian tribes 
have adult marriage,5 whilst in the early Vedic texts 
marriage appears essentially as a union of two persons of 
full development and child-wives first occur regularly in 
the Sutra period.6 Many of the law-books fix a definite age 
for the marriage of girls; and the later the treatise, the 
1 Gait, op. cit. pp. 262, 263, 266. 2 Ibid. p. 267. 
3 Ibid. p. 267 sq. 4 Ibid. p. 267. Risley, op. cit. p. 193 sq. 
6 Gait, op. cit. p. 270. Risley, op. cit. p. 187. 
6 Macdonell and Keith, op. cit. i. 474 sq. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
381 
earlier is the age which it prescribes. According to the 
' Laws of Manu ' a man of thirty should marry a girl of 
twelve, and a man of twenty-four a girl of eight ;1 whilst 
later books fix the higher limit of age in such cases at ten 
or eight years and reduce the lower limit to seven, six, or 
even four years.2 It has been surmised that one cause of 
the child-marriage was the growth of the patriarchal power 
of the head of the family, who must have been adverse to 
any assertion of independence on the part of its female 
members, and more especially to their exercising the right 
of choosing their husbands for themselves, which might 
prove detrimental to the interests of the family. Moreover, 
the gradual lowering of the position of women from the 
high standard of Vedic times and the distrust of their virtue 
induced by the example of pre-matrimonial licence set by 
the Dravidian races may have operated in the same 
direction ; it is hinted in the literature of the subject that 
a girl would be married as a child in order to avert the 
possibility of her causing scandal later on.3 It has further 
been said that child-marriage was introduced on account 
of the frequent invasions of foreigners and the insecurity 
of the times.4 As the most potent motive, however, has 
been represented the desire of parents to secure appropriate 
bridegrooms for their daughters.5 The custom of hypergamy, 
which prohibits a woman from marrying a man belonging 
to a caste lower than her own, limits the number of possible 
husbands for the girls of the higher classes and makes 
fathers anxious to get their daughters married as soon as an 
acceptable match offers. And, says Sir Herbert Risley, 
" when the custom of infant marriage had once been started, 
under pressure of social necessity, by the fa.milics of the 
highest group, who had the largest surplus of marriageable 
daughters, a sort of fashion would have been set and would 
be blindly followed through all the grades." 6 
1 Laws of Manu, ix. 94. 2 Risley, op. oil. p. 189. 
3 Ibid. p. 189 sq. 
4 Kanjilal, ' Hindu Early Marriage,' in Calcutta Review, cxxviii. 
168. 
6 Forbes, Oriental Memoirs, iii. 302. 6 Risley, op. cit. p. 190. 382 
THE HISTORY OF HUMAN MARRIAGE chap. 
This opinion has been contradicted by Mr. Gait, who 
observes that the practice of hypergamy, although it in 
some cases leads to early marriages, often, on the contrary, 
leads to the postponement of marriage. " A poor man with 
many daughters," he writes, " finds it extremely difficult 
to pay the bridegroom price ; and it often happens in 
consequence that his girls remain unmarried until long after 
the age of puberty. So frequently is this the case that, in 
various castes, the hypergamous sections no longer penalise 
a man for failing to give his daughters in marriage before 
they attain puberty. The Rajputs, who are much addicted 
to hypergamy, are by no means in the front rank as regards 
infant marriage. On the whole, therefore, it cannot be said 
that hypergamy leads to early marriage. It seems rather to 
be the case that infant marriage is most common where the 
difficulty of obtaining a husband is small and the marriage 
ceremony inexpensive." 1 His own view is that the practice 
had its origin neither in the pure Aryan nor in the pure 
Dravidian culture, but was the result of their impact. He 
remarks that though the non-Hinduised Dravidian tribes 
do not ordinarily give their girls in marriage before puberty, 
they allow them great sexual freedom so long as they are 
spinsters. When such tribes come under the influence of 
Hinduism, this premarital communism falls into disrepute, 
and the simplest method of putting an end to it would 
obviously be by providing the girls with husbands before 
the promptings of nature could lead them astray. And 
in the same way the new-born desire to get virgin wives for 
their sons would lead parents to select girls who are so young 
that there can be little fear of their having already lost their 
virginity. Mr. Gait thinks that this hypothesis is the one 
which, on the whole, seems to fit most closely into the 
facts. It explains how it is that while the non-Hinduised 
tribes have adult marriage, those that have become 
Hinduised are ordinarily more addicted to infant-marriage 
than any other section of the community. The statistics 
show that the practice is least common in the north-west 
of India, where the Aryan element is strongest, and that 
1 Gait, op. cit. p. 271. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
383 
elsewhere it is often most prevalent among the lower rather 
than the higher castes, that is, among the communities of 
Dravidian origin.1 In this fact he finds an argument 
against the theory that infant-marriage originated with high 
caste Hindus and spread gradually from them to the lower 
castes.2 But his argument does not seem convincing as it 
is quite possible that a certain custom, which has originated 
in one class of society, is not only adopted but carried to an 
extreme by a lower class. Mr. Gait thinks that a further 
cause tending to encourage the marriage of very young girls, 
where that of widows is forbidden, is the fact that girls are 
wanted as wives by widowers as well as bachelors ;3 but 
this suggestion seems of little importance considering that 
the very castes who practise infant-marriage as a general 
rule allow their widows to marry again.4 Infant-marriage 
in India may no doubt be due to various causes, but it 
appears that the most potent of them, at all events, are the 
dread of future celibacy and the preference given to virgin 
brides-causes which, as we have seen, have led to the 
same result in many other parts of the world. 
In Burma-as well as in Assam, the North-West Frontier 
Province, Cochin, Travancore, and Mysore-marriage before 
the age of ten is practically non-existent.6 Most Burmese 
lads are said to get married when they are eighteen or 
nineteen, and very few, not in the sacred order, are 
unmarried when they are twenty; whilst thirteen or 
fourteen is a common age for the girls.6 So also in Siam 
marriage takes place early-about fourteen in the case of 
girls and about twenty in the case of men ;7 and. " every 
girl gets married sooner or later, so that old maids do not 
exist." 8 
In the Zoroastrian books, as in the sacred books of India, 
we meet with the idea that a man should marry and get 
progeny. Ahura Mazda said to Zoroaster :-" The man 
who has a wife is far above him who lives in continence ; 
1 Ibid. p. 26g sq. 2 Ibid. pp. 264, 269. 3 Ibid. p. 271. * Ibid. p. 269. 
6 Ibid. p. 267. 8 Shway Yoe, The Burman, pp. 55, 64. 
7 Young, Kingdom of the Yellow Robe, p. 85. Bowring, Kingdom 
and People of Siam, i. 118. 8 Young, op. oil. p. 85. 384 
THE HISTORY OF HUMAN MARRIAGE chap. 
he who keeps a house is far above him who has none ; he 
who has children is far above the childless man."1 The 
greatest misfortune which could befall an ancient Persian 
was to be childless.2 To him who has no child the bridge 
of Paradise shall be barred ; the first question the angels 
there will ask him is, whether he has left in this world a 
substitute for himself, and if the answer be " No " they 
will pass by and he will stay at the head of the bridge, full 
of grief. The primitive meaning of this is plain : the man 
without a son cannot enter Paradise because there is nobody 
to pay him the family worship.3 Ashi Vanguhi, a feminine 
impersonification of piety, and the source of all the good 
and riches that are connected with piety, rejects the 
offerings of barren people-old men, courtesans, and 
children.4 It is said in the Yasts, " This is the worst deed 
that men and tyrants do, namely, when they deprive maids 
that have been barren for a long time of marrying and 
bringing forth children."6 And in the eyes of all good Parsis 
of the present day, as in the time of King Darius and the 
contemporaries of Herodotus, the two greatest merits of a 
citizen are the begetting and rearing of a numerous family, 
and the fruitful tilling of the soil.6 It is by no means fitiusual, 
therefore, for Parsis to enjoin by will or by a trust that a 
certain amount of their wealth be expended in aiding poor 
brides to marry, and certain institutions provide special 
funds for this purpose.7 
The ancient Greeks regarded marriage as a matter both 
of public and private importance.8 In various places 
1 Vendiddd, iv. 47. 
2 Rawlinson, in his translation of Herodotus, i. 262 n. 1. Cf. 
Herodotus, i. 133, 136 ; Dind-t Maindg-t Khirad, xxxv. 19. 
3 Darmesteter, in Sacred Books of the East, iv. 47. Cf. Idem, 
Ormazd et Ahriman, p. 294. 
4 Yasts, xvii. 54. 5 Ibid. xvii. 59. 
6 Darmesteter, in Sacred Books of the East, iv. p. Ixii. 
7 Jivanji Jamshedji Modi, ' Marriage (Iranian).-1. Zoroastrian,' 
in Hastings, op. cit. viii. 455. 
8 Muller, History and Antiquities of the Doric Race, ii. 300 sq. 
Fustel de Coulanges, op. cit. p. 55. Hearn, op. cit. p. 72. Dollinger, 
The Gentile and the Jew, ii. 234 sq. Kessler, Plutarchs Leben des 
Lykurgos, p. 64. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
385 
criminal proceedings might be taken against celibates.1 
Plato remarks that every individual is bound to provide 
for a continuance of representatives to succeed himself as 
ministers of the Divinity;2 and Isaeus says, " All those 
who think their end approaching look forward with a 
prudent care that their houses may not become desolate, 
but that there may be some person to attend to their 
funeral rites and to perform the legal ceremonies at their 
tombs." 3 So also the conviction that the founding of a 
house and the begetting of children constituted a moral 
necessity and a public duty had a deep hold of the Roman 
mind in early times.4 Cicero's treatise ' De Legibus '- 
which generally reproduces in a philosophical form the 
ancient laws of Rome-contains a law according to which 
the Censors had to impose a tax upon unmarried men.5 
But in later periods, when sexual morality reached a very 
low ebb in Rome, celibacy-as to which grave complaints 
were made as early as 520 b.c.-naturally increased in 
proportion, especially among the upper classes. Among 
these marriage came to be regarded as a burden which 
people took upon themselves at the best in the public 
interest. Indeed, how it fared with marriage and the 
rearing of children is shown by the Gracchan agrarian 
laws, which first placed a premium thereon;6 and later 
the Lex Julia et Papia Poppaea imposed various penalties 
on those who lived in a state of celibacy after a certain age,7 
though with little or no result.8 According to the later 
1 Pollux, Onomasticum, iii. 48. 
2 Plato, Leges, vi. 773. 
3 Isaeus, Oratio de Apollodori hereditate, 30, p. 66. Rohde observes 
(Psyche, p. 228), however, that such a belief did not exist in the 
Homeric age, when the departed souls in Hades were supposed to be 
in no way dependent upon the survivors. 
4 Mommsen, History of Rome, i. 74. 
5 Cicero, De legibus, iii. 3. Fustel de Coulanges, op. cit. p. 55. 
6 Mommsen, op. cit. iii. 121 ; iv. 186 sq. For the growing unpopu- 
larity of marriage in Rome see also Fowler, Social Life at Rome in 
the Age of Cicero, p. 149 sq. ; Meyer, Der rbmische Konkubinat, p. 17. 
7 Rossbach, Rbmische Ehe, p. 418. 
8 Mackenzie, Studies in Roman Law, p. 104. 386 
THE Lt I STORY OF HUMAN MARRIAGE chap. 
Roman law a man could marry from the age of fourteen 
and a woman from the age of twelve.1 
The ancient Teutons as described by Caesar, on the other 
hand, accounted it in the highest degree scandalous to have 
intercourse with the opposite sex before the twentieth year.2 
Tacitus also asserts that the young men married late, and 
that the maidens did not hurry into marriage ;3 and the 
accuracy of this statement is confirmed by what we know 
about Teutonic custom in the Middle Ages.4 It seems 
probable, however, that at a mature age celibacy was almost 
unknown among the Teutons, except in the case of women 
with a bad reputation, for whom neither beauty nor riches 
could procure a husband.5 
Among the Russian peasantry celibacy is unheard of : 
when a youth reaches the age of eighteen he is informed 
by his parents that he ought to marry at once.6 Formerly 
the father was even anxious to marry his son at an earlier 
age in order to secure an additional female labourer, and 
he cohabited himself with the wife during the son's 
minority.7 Professor Giorgevitch of Belgrade informs me 
that such marriages are occasionally found to this day in 
Serbia ; in former times they were more common there, 
but the younger generation were unwilling to submit to the 
old custom, and this contributed to the breaking up of 
the zadruga, or joint-family institution. Infant-betrothal 
occurred both in Serbia and Montenegro,8 and in High 
Albania it is still in vogue. According to Miss Durham, most 
of the children there are betrothed in infancy or in early 
childhood, and some even before birth. A man, as soon as 
1 Dernburg, Pandekten, iii. 16. 2 Caesar, De bello Gallico, vi. 21. 
3 Tacitus, Germania, ch. 20. 
4 Hoops, Reallexikon der Germanischen Alter tumskunde, ii. 486. 
5 Tacitus, op. cit. ch. 19. 6 Mackenzie Wallace, Russia, i. 138. 
7 v. Haxthausen, Transcaucasia, p. 403 note. Le Bon, L'homme 
et les societes, ii. 295. Bogoras, Chukchee, p. 578 n. 4. A Serbian 
writer, Sava Tekelija, found this custom, in the government of 
Kursk, when travelling there in 1786 (Sava Tekelija, ' Auto- 
biografija,' in Letopis Matice Srpske, cxix. 39 sq. ; I am indebted to 
Professor Giorgevitch for this reference). 
8 Piprek, op. cit. p. 119. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
387 
a son is born to him, seeks a suitable family with which to 
be allied, and should there be no daughter available, 
bespeaks the next one born. The girl may afterwards 
refuse to marry the man, but only on condition that she 
swears before witnesses to remain virgin all her life, and 
should she break this vow there would be endless bloodshed 
in consequence. Girls are married as young as thirteen 
and boys at fifteen or even fourteen, but the age is rising.1 
The laws of all Christian countries fix the lowest ages at 
which marriage is allowed for males and females. The 
stipulation of the Roman law, according to which a man may 
marry at the age of fourteen and a woman at the age of 
twelve, was adopted by the Church, and is, under the 
influence of Canon Law, still preserved in Great Britain, 
in some of the United States, and in Spain, Portugal, 
Greece, Mexico, Chili, and several other Roman Catholic 
States in America. In Austria both men and women 
become capable of marrying at the age of fourteen; in 
Serbia and Caucasia, men at fifteen and women at thirteen ; 
in Brazil, North Carolina, and Texas, men at sixteen and 
women at fourteen; in Alabama, Arkansas, Georgia, and 
Illinois, men at seventeen and women at fourteen; in 
Peru, men at eighteen and women at fourteen ; in France, 
Belgium, Italy, Rumania, California, Minnesota, New 
Mexico, Oregon, and Wisconsin, men at eighteen and 
women at fifteen; in Russia, Holland, Indiana, Michigan, 
Montana, Nebraska, Nevada, Ohio, and Wyoming, 
men at eighteen and women at sixteen ; in Switzerland, 
men at twenty and women at eighteen; in Germany, men 
at twenty-one and women at sixteen; in Sweden, men at 
twenty-one and women at eighteen. In many countries, 
however, where the canonic age-limit has not been pre- 
served, the obstacle to marrying at an earlier age than that 
which the law admits may be removed by dispensation. 
The German Civil Code allows such dispensation only with 
regard to women, but a man is nevertheless in special cases 
1 Miss Durham, ' High Albania and its Customs,' in Jour. Roy. 
Anthv. Inst. xl. 459 sq. Idem, High Albania, p. 35 sq. 388 
THE HISTORY OF HUMAN MARRIAGE chap. 
able to marry before twenty-one, namely, if the Court of 
Wardship declares him to be of full age, which the Court 
has power to do at any time after he has reached the age 
of eighteen. The general tendency of modern legislation 
has been to raise the age-limit; and it is therefore quite 
possible that some of the figures given above have become 
antiquated in the course of the last few years.1 
Besides the age-limit below which marriage is prohibited 
there is frequently another, higher limit below which a 
marriage can only be contracted with the consent of parents, 
guardians, or other persons having control of the parties. 
This subject will be dealt with in a subsequent chapter.2 We 
shall now proceed to consider the frequency of marriage 
and the age at which people actually marry in Europe. 
Generally speaking, modern civilisation has proved 
unfavourable to the frequency of marriage, and has raised 
the average age at which marriages are concluded. In 
Europe more than a third of the male and female population 
beyond the age of fifteen live in a state of voluntary or in- 
voluntary celibacy ; 3 and among them there are many who 
never marry. In the middle of the nineteenth century 
Wappaus found that in Saxony 14-6 per cent, of the un- 
married adult population died single, in Sweden 14-9 per 
cent., in the Netherlands 17-2 per cent., and in France 20-6 
per cent.4 The marriage-rate varies greatly in different 
European countries, as appears from the following figures, 
1 Roguin, Traite. de droit civil compare. Le mariage, p. 48 sqq. 
Winroth, Aktenskapshindren, p. 100 sq. (some statements found in 
these works are already antiquated). Wright, Report on Marriage 
and Divorce in the United States, p. 29. Code civil, art. 144. Das 
allgemeine biirgerliche Gesetzbuch filr das Kaisertum Oesterreich, §§21, 
49. Schweizerisches Zivilgestzbuch vom 10 Dezember 1907, art. 96. 
Burgerliches Gesetzbuch (German Civil Code), § 1303. Lag om 
dktenskaps ingdende och upplbsning av den 12 november 1915, 
ii. 1 (Sweden). As a curious exception to the general tendency to 
raise the age-limit may be mentioned the law of the grand-duchy 
of Hesse before the year 1875, which prohibited a man from 
marrying before the age of twenty-five (Roguin, op. cit. p. 49 sq.). 
2 Infra, on Consent. 
3 v. Oettingen, Die Moralstat istik, p. 140 n. 1. 
1 Wappaus Allgemeine Bevblkerungsstatistik, ii. 267. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
389 
which give the annual number of marriages per 10,000 
marriageable persons, that is, males 18 years of age or over 
and females 15 years or over who are either single, widowed, 
or divorced. It was in Serbia (1896-1905) 1386, in Bulgaria 
(1910-1911) 1223, in Russia-not including Finland and 
Poland-(1896-1897) 921, in Rumania (1896-1903) 873, in 
Hungary (1906-1915) 778, in Germany (1907-1914) 569, in 
France (1910-1911) 539, in Austria (1908-1913) 536, in Eng- 
land and Wales (1907-1914) 507, in Norway (the same period) 
418, in Scotland (the same period) 411, in Finland (1906-1915) 
398, in Sweden (1908-1913) 367, in Iceland (1906-1915) 
335, and in Ireland (1909-1912) 254.1 The average age of 
the bachelors and the spinsters who enter into matrimony 
was in Serbia (1896-1900) 2r8 and 197 respectively, in 
Italy (1911-1914) 27'2 and 23'6, in Germany (the same 
period) 27-4 and 247, in England (1906-1914) 27'4 and 
257, in Scotland (the same period) 27'8 and 25-8, in France 
(1906-1910) 28-0 and 237, and in Sweden (1906-1913) 
28'8 and 26'4.2 
In recent times the proportion of unmarried people is 
known to have increased in various European countries 3 
and the marriage age to have risen. In England and Wales 
the annual number of marriages per 10,000 marriageable 
persons was, in 1876-1885, 568 ; in 1886-1895, 529 ; in 
1896-1905, 531 ; in 1907-1914, as already said, 507 ;4 but 
in 1911-1914, a little higher than in 1906-1910.5 In Hungary 
it was, in 1876-1885, 1038 ; in 1886-1895, 999; in 1896- 
1905, 813. In Finland it was, in 1876-1885, 536; in 
Sweden, during the same period, 417; in Norway, in 1881- 
1885, 458; in Germany, in 1901, 603.6 In England and 
Wales the average age of bachelor-bridegrooms and of 
spinster-brides was, in 1876-1885, 25-9 and 24-4 respectively ; 
1 Annuaire international de statistique, p. n sqq. 2 Ibid. p. 28 sq. 
3 Wappaus, op. cit. ii. 229. v. Oettingen, op. cit. p. 120. Haus- 
hofer, Lehr- und Handbuch der Statistik, p. 396. Mayo-Smith, 
Science of Statistics, i. 96. 
♦ Annuaire international de statistique, p. 12. 
Baines, ' Recent Trend of Population in England and Wales, * 
in Jour. Roy. Statistical Soc. Ixxix. p. 401. 
6 Annuaire international de statistique, p. 11 sqq. 390 
THE HISTORY OF HUMAN MARRIAGE chap. 
in 1886-1895, 26-4 and 24-9 ; in 1896-1905, 26-8 and 25-3 ; in 
1906-1910, 27-2 and 25-6? The tendency towards increasing 
age at marriage in England may also be demonstrated by 
taking the percentages of marriages contracted at ages under 
25. These were, in 1896-1900, 48-2 for bachelors and 64-4 for 
spinsters ; in 1901-1905, 45-4 and 6r6 respectively; in 
1906-1910, 42'5 and 59'1; and in 1911-1915, 40'4 and 
56-9.2 " 
A very important cause of the decline of the marriage-rate 
and the rise of the age for marriage in Europe is the 
difficulty of supporting a family in modern society. The 
importance of the economic factor has been much em- 
phasised by statisticians. From "the days of Siissmilch till 
the middle of the last century it was regarded almost as a 
statistical axiom that the number of marriages varies 
inversely with the price of corn. In Germany this tendency 
of the number of marriages to decrease when the price of 
rye was high and to increase when the price was low was 
observed until about i860 ; but since then Germany has 
become so industrial and commercial that the price of food 
is only one element in the economic well-being of the masses.3 
In the English records, at any rate so far back as 1820, no 
such relation is noticeable between the number of marriages 
and the price of corn ; on the contrary, the marriage-rate 
has often been found to vary not inversely but directly 
with the price of wheat.4 In explanation of this fact Dr. 
Ogle points out that increased exports and imports put up 
freights and thus increase the price of corn, while at the 
same time men marry in greater numbers when trade is 
brisk and the value of exports increases ; the amount of 
exports gauges the opportunity for employment, and is 
therefore an index of economic prosperity.5 But the corre- 
spondence between the marriage-rate and the general 
1 Ibid. p. 28. 
2 Seventy-eighth Annual Report of the Registrar-General, p. xii. 
3 Mayr, Bevblkerungsstatistik, p. 385. Mayo-Smith, op. cit. i. 190. 
« Ogle, ' On Marriage-Rates and Marriage-Ages,' in Jour. Roy. 
Statistical Soc. liii. 258. 
6 Ibid. p. 260 sqq. X FREQUENCY OF MARRIAGE AND MARRIAGE AGE 
391 
prosperity of the country, as indicated by the course of 
trade, is certainly not absolute : the curves show in fact 
that it is only the oscillations which are correlated. At the 
beginning of the present century Hooker observed that over 
the preceding forty years as a whole the trade curve had 
risen in England, whereas the marriage-rate had fallen.1 
So also Hermann's formula that " the number of marriages 
in any period expresses the expectation of economic 
prosperity prevailing at that time "2 has only a relative 
value.3 
That in spite of the general economic progress people 
have become more and more unwilling to venture upon 
marriage is no doubt to some extent due to the ever- 
increasing standard of comfort among all classes, which 
has led either to a retardation or to an abandonment of 
marriages.4 In this respect there is a difference between 
different social classes. Professor Fahlbeck has found that 
among the Swedish nobility 43'17 per cent, of the marriage- 
1 Hooker, ' Correlation of the Marriage-Rate with Trade,' in 
Jour. Roy. Statistical Soc. Ixiv. (1901) 485 sq. See also Ogle, loc. cit. 
p. 265. 
8 Hermann, quoted by Mayo-Smith, op. cit. i. 190. 
• 3 As for the earlier observation that the effect of war is to bring 
down the marriage-rate (Haushofer, op. cit. p. 401 ; Ogle, loc. cit. 
p. 255 sq. ; Mayo-Smith, op. cit. i. 99 sq.) it is interesting to note that 
during the second, third, and fourth quarters of 1915 and the first 
quarter of 1916 there was, on the contrary, a sudden and phenomenal 
rise in the marriages in England which was directly due to the war. 
Sir Bernard Mallet (' Vital Statistics as Affected by the War,' in 
Jour. Roy. Statistical Soc. Ixxxi. 7) observes that whilst war marriages 
are of perfectly natural occurrence at the outbreak of war, they 
were in the present case enormously stimulated by the very generous 
provision of separation allowance and pensions, as also by the policy 
which prevailed in the first period of the war of taking single men 
first. Crudely stated the war resulted in 200,000 persons in England 
being married between August 1914 and June 1917 who in the 
ordinary course of events would not have been married (ibid. p. 3). 
On the other hand, taking an average of marriages in Hungary, 
the effect of the war, up till May 1917, was that over 600,000 who 
in ordinary circumstances would have married did not do so 
(ibid. p. 6). 
4 CJ. Ogle, loc. cit. p. 265, 392 
THE HISTORY OF HUMAN MARRIAGE chap. 
able men and 46'15 of the marriageable women are un- 
married, while the figures for the whole population of Sweden 
are only 31'42 and 31'54 respectively; and the number of 
unmarried persons has greatly increased in recent times.1 
Generally speaking, the average age for marriage is more 
advanced among the upper classes than among the lower 
ones.2 A " gentleman " before marrying thinks it 
necessary to have an income of which a mere fraction would 
suffice for a married workman. He has to offer his wife a 
home in accordance with her social position and his own; 
and unless she brings him some fortune, she contributes 
but little to the support of the family. Moreover, from the 
economic point of view there is hardly any reason to put off 
marrying for men who earn in youth almost as much as in 
later life, as is the case with miners, tailors, shoemakers, 
artisans, and others, who have been found to marry earlier 
than men of the professional class.3 It may also be said that 
a man requires more time to gain his living by intellectual 
than by material work. 
When we consider the differences in the marriage-rate 
and the age for marriage in the different countries of 
Europe, we notice that the tendency to marry is greatest 
in the East of Europe with its more primitive civilisation. 
In Russia there is an extraordinarily large number of 
bridegrooms under the age of twenty, and more than one- 
half of the brides are under that age ;4 and the reason for 
this is that an enormous proportion of the population 
consists of small agriculturists, who are in the habit of 
arranging for the marriage of their sons at as early an age as 
possible in order to secure an additional female labourer.5 
The exceptionally low marriage-rate in Northern Europe is, 
partly at least, due to emigration. And the emigration of 
young men and women is undoubtedly the cause of Ireland 
1 Fahlbeck, Sveriges add, ii. 118. 
2 Mayo-Smith, op. cit. i. 103 sq. Dumont, ' L'age au mariage,' in 
Bull, et mem. Soc. d'Anthr. Paris, ser. v. vol. iii. 260. 
3 Forty-ninth Annual Report of the Registrar-General, p. viii. 
4 Mayo-Smith, op. cit. i. 105. 
5 Semenow, quoted in Forty-sixth Ann. Rep. Reg.-Gen. p. ix. 
Cf. supra, p. 386 ; Leroy-Beaulieu, Empire of the Tsars, i. 495. x FREQUENCY OF MARRIAGE AND MARRIAGE AGE, 
393 
having a lower marriage-rate than any other country in 
Europe.1 
An economic cause of the declining tendency to marry 
which still remains to be mentioned is the increasing 
economic independence of women. At the lower stages of 
civilisation a woman is a helpless being who depends on the 
support of a man, whereas modern civilisation provides her 
with means of earning her living by her own efforts alone. 
But besides factors of an economic character there are 
others, equally important, which have made men and 
women less inclined to enter into matrimony. 
A modern writer justly points out that " by the general 
diffusion of education and culture, by the new inventions 
and discoveries of the age, by the increase of commerce and 
intercourse and wealth, the tastes of men and women have 
become widened, their desires multiplied, new gratifications 
and pleasures have been supplied to them. By this increase 
of the gratifications of existence the relative share of them 
which married life affords has become just so much less. 
The domestic circle does not fill so large a place in life as 
formerly. It is really less important to either man or 
woman. Married life has lost in some measuie its advantage 
over a single life. There are so many pleasures, now, that 
can be enjoyed as well or even better in celibacy." More- 
over, " by the diffusion of a finer culture throughout the 
community, men and women can less easily find any one 
whom they are willing to take as a partner for life ; their 
requirements are more exacting ; their standards of ex- 
cellence higher ; they are less able to find any who can 
satisfy their own ideal, and less able to satisfy anybody 
else's ideal. Men and women have, too, a livelier sense of 
the serious and sacred character of the marriage union, and 
of the high motives from which alone it should be formed. 
They are less willing to contract it from any lower motives." 2 
The marriage and divorce laws of Christian countries are 
also responsible for a certain number of persons remaining 
1 Mayo-Smith, op. cit. i. 96 sq. 
2 ' Why is Single Life becoming more General ? ' in Nation 
(New York), vi. 190 sq. 394 
THE HISTORY OF HUMAN MARRIAGE chap, x 
celibates. We may assume that if the union could be more 
easily dissolved it would be more readily entered into. And 
the law of monogamy is necessarily a cause of celibacy 
where the adult women outnumber the adult men. If we 
reckon the age for marriage from twenty to fifty years, a 
hundred men may in Europe choose amongst a hundred and 
three or four women, so that about three or four women per 
cent, are in normal circumstances doomed to a single life 
on account of our obligatory monogamy.1 
1 v. Oettingen, op. cit. p. 60, CHAPTER XI 
CELIBACY 
Side by side with the idea that marriage is highly desirable 
or obligatory for all ordinary men and women we find the 
opposite notion, according to which celibacy in extra- 
ordinary cases commands more respect than marriage or is 
even looked upon as a stringent duty. 
The Tuhoe, a Maori tribe, had the custom of making the 
first-born daughter of a chief, but nobody else, a puhi. This 
implied that she was rendered tapu-that " she was not 
allowed to have sexual connection with any man, nor to 
perform any work except such as the weaving of the better- 
class garments. . . . The idea was to make her an important 
person in the tribe, a lady of rank, to be treated with respect 
and looked up to. If a puhi were detected in illicit inter- 
course with any man she was degraded and the tapu taken 
off her."1 
We are told that the Shawnee Indian had a great respect 
for certain persons who observed celibacy.2 Among other 
North American tribes certain effeminate men who assumed 
the dress and habits of women were looked upon as wizards 
or supernatural beings and held in repute.3 Among many 
peoples persons whose function it is to perform religious 
or magical rites must be celibates.4 The Tlingit believe 
that if a shaman does not observe continuous chastity his 
1 Best, * Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 34. 
2 Ashe, Travels in America, p. 250. 
3 See Westermarck, Origin and Development of the Moral Ideas, 
ii. 472 sq. 
4 Some instances of this are stated by Landtman, Origin of 
Priesthood, p. 156 sq. 
395 396 
THE HISTORY OF HUMAN MARRIAGE chap. 
own guardian spirits will kill him.1 In Patagonia the male 
wizards were not allowed to marry.2 In some tribes of the 
Guarani of Paraguay " the female Payes were bound to 
chastity, or they no longer obtained credit."3 Celibacy was 
compulsory on the priests of the Chibchas in Bogota.4 The 
Tohil priests in Guatemala were vowed to perpetual con- 
tinence.5 In Ichcatlan the high-priest was obliged to live 
constantly within the temple, and to abstain from commerce 
with any woman whatsoever; and if he failed in this duty 
he was cut in pieces, and the bloody limbs were given as a 
warning to his successor.6 Of the women who held positions 
in the temples of ancient Mexico we are told that their 
chastity was most zealously guarded ; during the perform- 
ance of their duties they were required to keep at a proper 
distance from the male assistants, at whom they did not even 
dare to glance. The punishment to be inflicted upon those 
who violated their vow of chastity was death; whilst, if 
their trespass remained entirely secret, they endeavoured to 
appease the anger of the gods by fasting and austerity of life, 
dreading that in punishment of their crime their flesh would 
rot.7 In Yucatan there was, connected with the worship 
of the sun, an order of vestals the members of which generally 
enrolled themselves for a certain time, but were afterwards 
allowed to leave and enter the married state. Some of 
them, however, remained for ever in the service of the 
temple and were apotheosised. Their duty was to attend 
to the sacred fire, and to keep strictly chaste, those who 
broke their vows being shot to death with arrows.8 In 
1 Veniaminof, quoted by Landtman, op. cit. p. 156. 
2 Falkner, Description of Patagonia, p. 117. 
3 Southey, History of Brazil, ii. 371. 
4 Simon, quoted by Dorman, Origin of Primitive Superstitions, 
p. 384. 
& Bancroft, Native Races of the Pacific States of North America, 
Hi. 489. 
6 Clavigero, History of Mexico, i. 274. 
7 Ibid. i. 275 sq. Torquemada, Monarchia Indiana, ii. 188 sqq. 
Bancroft, op. cit. iii. 435 sq. Cf. Acosta, History of the Indies, ii. 
333 sq. 
8 Bancroft, op. cit. iii. 473. Lopez Cogolludo, Historia de Yu- 
cathan, p. 198. XI CELIBACY 
397 
Peru there were likewise virgins dedicated to the sun, who 
lived in perpetual seclusion to the end of their lives, who 
preserved their virginity and were forbidden to converse 
or have sexual intercourse with or to see any man, or even 
any woman who was not one of themselves.1 And besides 
the virgins who thus professed perpetual virginity in the 
monasteries, there were other women, of the blood royal, 
who led the same life in their own houses, having taken a 
vow of continence. These women " were held in great 
veneration for their chastity and purity, and, as a mark of 
worship and respect, they were called Ocllo, which was a name 
held sacred in their idolatry " ; but if they lost their virtue, 
they were burnt alive or cast into " the lake of lions."2 
Among the Guanches of the Canary Islands there were 
virgins, called Magades or Harimagades, who presided over 
the cult under the direction of the high-priest, and there 
were other virgins, highly respected, whose function was to 
pour water over the heads of newborn children, and who could 
abandon their office and marry whenever they pleased.3 
The priestesses of the Tshi- and Ewhe-speaking peoples on 
the West Coast of Africa are forbidden to marry.4 In a 
wood near Cape Padron, in Lower Guinea, lives a priestly 
king who is allowed neither to leave his house nor to touch a 
woman.5 
Among the Todas of the Nilgiris, in Southern India, the 
" dairyman " or priest is bound to a celibate existence.6 
Among the Hindus, in spite of the great honour in which 
marriage is held, celibacy has always commanded respect in 
1 Garcilasso de la Vega, First Part of the Royal Commentaries of 
the Yncas, i. 291 sqq. 
2 Ibid. i. 305. 
3 Bory de St. Vincent, Essais sur les Islesv Fortunees, p. 96 sq. 
Nunez de la Pena, Conqvista y antigvedades de las islas de la Gran 
Canaria, p. 27. 
4 Ellis, Tshi-speaking Peoples, p. 121. Idem, Ewe-speaking 
Peoples, p. 142. Supra, p. 219 sq. 
5 Bastian, Die deutsche Expedition an der Loango-Kiiste, i. 287 sq. 
6 King, Aboriginal Tribes of the Nilgiri Hills, p. 19. Thurston, 
Anthropology of the Todas and Kotas,' in the Madras Govern- 
ment Museum's Bulletin, i. 169, 170, 193. Rivers, Todas, pp. 80, 
99, 236. 398 
THE HISTORY OF HUMAN MARRIAGE chap. 
instances of extraordinary sanctity.1 Those of the Sann- 
yasis who are known to lead their lives in perfect celibacy 
receive on that account marks of distinguished honour and 
respect.2 Already the time-honoured Indian institution of 
the four Asramas contained the germ of monastic celibacy, 
the Brahmacarin, or student, being obliged to observe 
absolute chastity during the whole course of his study.3 
The idea was further developed in Jainism and Buddhism. 
The Jain monk was to renounce all sexual pleasures, " either 
with gods, or men, or animals " ; not to give way to sen- 
suality ; not to discuss topics relating to women; not to 
contemplate the forms of women.4 Buddhism regards 
sensuality as altogether incompatible with wisdom and 
holiness ; it is said that " a wise man should avoid married 
life as if it were a burning pit of live coal." 5 According to 
the legend, Buddha's mother, who was the best and purest of 
the daughters of men, had no other sons, and her conception 
was due to supernatural causes.6 One of the fundamental 
duties of monastic life, by an infringement of which the guilty 
person brings about his inevitable expulsion from Buddha's 
order, is that " an ordained monk may not have sexual 
intercourse, not even with an animal." 7 In Tibet some 
sects of the Lamas are allowed to marry, but those who do not 
are considered more holy ; and in every sect the nuns must 
take a vow of absolute continence.8 At Taklakot in Western 
Tibet one boy in every family must invariably be set apart 
for the priesthood, and similarly of the daughters one or 
more are assigned to perpetual virginity.9 The Buddhist 
1 Monier-Williams, Buddhism, p. 88. 
2 Dubois, Description of the Character, Manners, and Customs of 
the People of India, p. 133. Cf. Monier-Williams, Brahmanism and 
Hinduism, p. 261. 
3 Kern, Manual of Indian Buddhism, p. 73. 
4 Hopkins, Religions of India, p. 294. 
5 Dhammika-Sutta, 21, quoted by Monier-Williams, Buddhism, 
p. 88. « Rhys Davids, Hibbert Lectures on Buddhism, p. 148. 
7 Oldenberg, Buddha, p. 350 sq. 
8 Wilson, Abode of Snow, p. 213. Waddell, 'Celibacy (Tibetan),' 
in Hastings, Encyclopaedia of Religion and Ethics, iii. 277. 
8 Sherring, Western Tibet, p. 188. XI CELIBACY 
399 
priests of Ceylon are totally debarred from women.1 Chinese 
law enjoins celibacy on all priests, Buddhist or Taouist.2 
And among the immortals of Taouism there are some women 
also, who have led an extraordinarily ascetic life.3 
In ancient Persia there were sun priestesses who were 
obliged to refrain from intercourse with men.4 The nine 
priestesses of the oracle of a Gallic deity in Sena were 
devoted to perpetual virginity.5 The Romans had their 
Vestal virgins, whose office, according to tradition, was 
instituted by Numa. They were compelled to continue 
unmarried during thirty years, which time they employed 
in offering sacrifices and performing other rites ordained by 
the law; and if they suffered themselves to be debauched 
they were delivered up to the most miserable death, being 
placed in a subterraneous cell, in their funeral attire, without 
any sepulchral column, funeral rites, or other customary 
solemnities.6 After the expiration of the term of thirty 
years they might marry on quitting the ensigns of their 
priesthood ; but we are told that very few did this, as those 
who did suffered calamities which were regarded as ominous 
by the rest, and induced them to remain virgins in the temple 
of the goddess till their death.7 in Greece priestesses were 
not infrequently required to be virgins, if not for their whole 
life, at any rate for the duration of their priesthood.8 Ter- 
tullian writes :-" To the Achaean Juno, at the town Aegium, 
a virgin is allotted ; and the priestesses who rave at Delphi 
know not marriage. We know that widows minister to the 
1 Percival, Account of the Island of Ceylon, p. 202. 
2 Ta Tsing Leu Lee, sec. cxiv. p. 118. Medhurst, 'Marriage in 
China,' in Trans. Roy. Asiatic Soc. China Branch, iv. 18. Davis, 
China, ii. 53. 
3 Reville, La religion Chinoise, p. 451 sq. 
4 Justin, quoted by Justi, ' Die Weltgeschichte des Tabari,' in 
Das Ausland, xlviii. 307. 
3 Pomponius Mela, Chorographia, iii. 6. 
6 Dionysius of Halicarnassus, Antiquitates Romana, ii. 64 sqq. 
Plutarch, Numa, x. 7 sqq. 
7 Dionysius of Halicarnassus, ii. 67. 
8 Strabo, xiv. 1; 23. Muller, Das sexuelle Leben dev alien Kultur- 
volker, p. 44 sqq. Bliimner, Home Life of the Ancient Greeks, p. 325. 
Gotte, Das Delphische Orakel, p. 78 sq. 400 
THE HISTORY OF HUMAN MARRIAGE chap. 
African Ceres ; they not only withdraw from their still living 
husbands, but they even introduce other wives to them in 
their own room, all contact with males, even as far as the 
kiss of their sons, being forbidden them. . . . We have 
heard, too, of continent men, and among others the priests 
of the famous Egyptian bull."1 There were eunuch priests 
connected with the cults of the Ephesian Artemis,2 the 
Phrygian Cybele,3 and the Syrian Astarte.4 
A small class of Hebrews held the idea that marriage is 
impure. The Essenes, says Josephus, " reject pleasure as 
an evil, but esteem continence and the conquest over our 
passions to be virtue. They neglect wedlock."5 This 
doctrine exercised no influence on Judaism, but perhaps 
much upon Christianity. St. Paul considered celibacy to be 
preferable to marriage. " He that give th her (his virgin) in 
marriage doeth well; but he that giveth her not in marriage 
doeth better." G " It is good for a man not to touch a woman. 
Nevertheless, to avoid fornication, let each man have his 
own wife, and let each woman have her own husband."7 
If the unmarried and widows cannot contain let them marry, 
" for it is better to marry than to bum."8 These and other 
passages9 in the New Testament inspired a general enthusiasm 
for virginity. Commenting on the words of the Apostle, 
Tertullian points out that what is better is not necessarily 
good. It is better to lose one eye than two, but neither is 
good; so also, though it is better to marry than to bum, it 
is far better neither to marry nor to burn.10 Marriage 
1 Tertullian, Ad uxorem, i. 6 (Migne, Patrologies cursus, i. 1284). 
Idem, De exhortatione castitatis, 13 (Migne, ii. 928 sq.). Cf. Idem, 
De monogamia, xq (Migne, ii. 953). 
2 Strabo, xiv. 1. 23. 
3 Arnobius, Adversus gentes, v. 7 (Migne, op. cit. v. 1095 sqq.). 
Farnell, ' Sociological Hypotheses concerning the Position of Women 
in Ancient Religion,' in Archiv f. Religionswiss. vii. 78. 
4 Lucian, De dea Syria, 15, 27, 50 sqq. 
6 Josephus, De bello Judaico, ii. 8. 2. See also Solinus, Collectanea 
rerum memorabilium, xxxv. 9 sq. 
6 1 Corinthians, vii. 38. 7 Ibid. vii. 1 sq. 8 Ibid. vii. 9. 
9 St. Matthew, xix. 12; Revelation, xiv. 4 ; &c. 
10 Tertullian, Ad uxorem, i. 3 (Migne, op. cit. i. 1278 sq.). Idem, 
De monogamia, 3 (Migne, ii. 932 sq.). XI CELIBACY 
401 
" consists of that which is the essence of fornication " j1 
whereas continence " is a means whereby a man will traffic 
in a mighty substance of sanctity."2 The body which our 
Lord wore and in which He carried on the conflict of life in 
this world He put on from a holy virgin; and John the 
Baptist, Paul, and all the others " whose names are in the 
book of life " 3 cherished and loved virginity.4 Virginity 
works miracles : Mary, the sister of Moses, leading the female 
band, passed on foot over the straits of the sea, and by the 
same grace Thecla was reverenced even by Hons, so that the 
unfed beasts, lying at the feet of their prey, underwent a holy 
fast, neither with wanton look nor sharp claw venturing to 
harm the virgin.5 Virginity is Hke a spring flower, always 
softly exhaling immortality from its white petals.6 The 
Lord himself opens the kingdoms of the heavens to eunuchs.7 
If Adam had preserved his obedience to the Creator he would 
have Hved for ever in a state of virgin purity, and some 
harmless mode of vegetation would have peopled paradise 
with a race of innocent and immortal beings.8 It is true 
that, though virginity is the shortest way to the camp of the 
faithful, the way of matrimony also arrives there, by a longer 
circuit.9 TertuUian himself opposed the Marcionites, who 
prohibited marriage among themselves and compelled those 
who were married to separate before they were received by 
1 Idem, De exhortatione castitatis, 9 (Migne, op. cit. ii. 925). 
2 Ibid. 10 (Migne, op. cit. ii. 925). 
3 Philippians, iv. 3. 
4 St. Clement of Rome, Epistola I. ad virgines, 6 (Migne, op. cit. 
Ser. Graeca, i. 392). 
5 St. Ambrose, Epistola LXIII. 34 (Migne, op. cit. xvi. 1198 sq.}. 
8 Methodius, Convivium decern virginum, vii. 1 (Migne, op. cit. 
Ser. Graeca, xviii. 125). 
7 Tertullian, De monogamia, 3 (Migne, op. cit. ii. 932). 
8 This opinion was held by Gregory of Nyssa and, in a later time, 
by John of Damascus. It was opposed by Thomas Aquinas, who 
maintained that the human race was from the beginning propagated 
by means of sexual intercourse, but that such intercourse was 
originally free from all carnal desire (von Eicken, Geschichte dev 
mittelalterlichen Weltanschauung, p. 437 sq.; see also Gibbon, History 
of the Decline and Fall of the Roman Empire, ii. 186). 
9 St. Ambrose, Epistola LXIII. 40 (Migne, op. cit. xvi. 1200). 
VOL. I D D 402 
THE HISTORY OF HUMAN MARRIAGE cttAt. 
baptism into the community.1 And in the earlier part of 
the fourth century the Council of Gangra expressly con- 
demned anyone who maintained that marriage prevented a 
Christian from entering the kingdom of God.2 But, at the 
end of the same century, a council also excommunicated the 
monk Jovinian because he denied that virginity was more 
meritorious than marriage.3 The use of marriage was per- 
mitted to man only as a necessary expedient for the con- 
tinuance of the human species, and as a restraint, however 
imperfect, on the natural licentiousness of desire.4 The 
procreation of children is the measure of a Christian's 
indulgence in appetite, just as the husbandman sowing the 
seed into the ground awaits the harvest, not sowing more 
upon it.5 
These opinions led by degrees to the obligatory celibacy 
of the secular and regular clergy. The conviction that a 
second marriage of a priest, or the marriage of a priest with 
a widow, is unlawful, seems to have existed from the earliest 
period of the Church;6 and as early as the beginning of 
the fourth century a synod held in Elvira in Spain insisted 
on the absolute continence of the higher ecclesiastics.7 The 
1 Tertullian, Adversus Marcionem, i. I, 29 ; iv. 11 ; &c. (Migne, 
op. cit. ii. 247, 280 sqq., 382). Idem, De monogamia, 1. 15 (Migne, 
ii. 931, 95°)- Cf. Irenaeus, Contra Hczreses, i. 28. 1 (Migne, op. cit. 
Ser. Graeca, vii. 690 sq.) ; Clement of Alexandria, Stromata, iii. 3 
(Migne, op. cit. Ser. Graeca, viii. 1113 sqq.\ 
* Concilium Gangrense, can. 1 (Labbe-Mansi, Sacrorum Concili- 
orum collectio, ii. 1106). 
3 Concilium Mediolanense, a.d. 390 (Labbe-Mansi, op. cit. iii. 
689 sq.}. 
4 St. Justin, Apologia I. pro Christianis, 29 (Migne, op. cit. Ser. 
Graeca, vi. 373). Clement of Alexandria, Stromata, ii. 23 (Migne, op- 
cit. Ser. Graeca, viii. 1089). Gibbon, op. cit. ii. 186. 
8 Athenagoras, Legatio pro Christianis, 33 (Migne, op. cit. Ser. 
Graeca, vi. 966). 
8 Lea, Sacerdotal Celibacy in the Christian Church, p. 37. Lecky, 
History of European Morals, ii. 328 sq. 
7 Concilium Eliberitanum, a.d. 305, ch. 33 (Labbe-Mansi, op. cit. 
ii. 11) :-" Placuit in totum prohiberi episcopis, presbyteris, et 
diaconibus, vel omnibus clericis positis in ministerio, abstinere se a 
conjugibus suis, et non generare filios : quicumque vero fecerit, 
ab honore clericatus exterminetur." XI CELIBACY 
403 
celibacy of the clergy in general was prescribed by 
Gregory VIL, who " looked with abhorrence on the con- 
tamination of the holy sacerdotal character, even in its 
lowest degree, by any sexual connection." But in many 
countries this prescription was so strenuously resisted, that 
it could not be carried through till late in the thirteenth 
century.1 
These rules and ideas relating to celibacy may be traced 
to various sources. In the first place, in a community where 
almost everybody is anxious to marry, a person who willingly 
refrains from doing so must appear queer, and such queemess 
may, under otherwise favourable circumstances, help to 
establish a reputation for sanctity. Yet this could only, 
at the most, have been a subsidiary cause of religious 
celibacy. 
In many cases the priestess is obviously regarded 
as married to the god whom she is serving, and is therefore 
forbidden to marry anybody else. In ancient Peru the Sun 
was the husband of the virgins dedicated to him.2 They were 
obliged to be of the same blood as their consort, that is to say, 
daughters of the Incas. "For though they imagined that 
the Sun had children, they considered that they ought not 
to be bastards, with mixed divine and human blood. So the 
virgins were of necessity legitimate and of the blood royal, 
which was the same as being of the family of the Sun." 3 
And the crime of violating the virgins dedicated to the Sun 
was the same and punished in the same severe manner 
as the crime of violating the women of the Inca.4 Con- 
cerning the priestesses of the Tshi-speaking peoples of the 
Gold Coast, Major Ellis remarks that the reason for their 
celibacy appears to be that " a priestess belongs to the god 
she serves, and therefore cannot become the property of a 
man, as would be the case if she married one." 5 So also 
the Ewhe-speaking peoples of the Slave Coast regard the 
1 Gieseler, Text-Book of Ecclesiastical History, ii. 275. Milman, 
History of Latin Christianity, ii. 150. 
2 Garcilasso de la Vega, op. cit. i. 297. 
3 Ibid. i. 292. 4 Ibid. i. 300. 
5 Ellis, Tshi-speaking Peoples, p. 121. 404 
THE HISTORY OF HUMAN MARRIAGE chap. 
women dedicated to a god as his wives.1 In the great temple 
of Jupiter Belus, we are told, a single woman used to sleep, 
whom the god had chosen for himself out of all the women 
of the land ; and it was believed that he came down in person 
to sleep with her. " This," Herodotus says, " is like the 
story told by the Egyptians of what takes place in their 
city of Thebes, where a woman always passes the night in the 
temple of the Theban Jupiter. In each case the woman is 
said to be debarred all intercourse with men."2 In the 
Egyptian texts there are frequent references to " the divine 
consort," neter hemt, a position which was generally occupied 
by the ruling queen, and the king was believed to be the 
offspring of such a union.3 As Plutarch states, the Egyptians 
thought it quite possible for a woman to be impregnated by 
the approach of some divine spirit, though they denied that 
a man could have corporeal intercourse with a goddess.4 
Nor was the idea of a nuptial relation between a woman and 
the deity foreign to the early Christians. St. Cyprian speaks 
of women who had no husband and lord but Christ, with 
whom they lived in a spiritual matrimony-who had 
" dedicated themselves to Christ, and, retiring from carnal 
lust, vowed themselves to God in flesh and spirit." 5 In 
the following words he condemns the cohabitation of such 
virgins with unmarried ecclesiastics, under the pretence of 
a purely spiritual connection" If a husband come and 
see his wife lying with another man, is he not indignant 
and maddened, and does he not in the violence of his jealousy 
perhaps even seize the sword ? What ? How indignant 
and angered then must Christ our Lord and Judge be, when 
He sees a virgin, dedicated to Himself, and consecrated to 
His holiness, lying with a man ! and what punishments does 
1 Idem, Ewe-speaking Peoples, pp. 140, 142. 
2 Herodotus, i. 181 sq. 
3 Wiedemann, Herodots zweites Buch, p. 268. Cf. Erman, Life in 
Ancient Egypt, p. 295 sq. 
4 Plutarch, Numa, iv. 5. Idem, Symposiaca problemata, viii. 
1. 6 sq. 
St. Cyprian, De habitu virginum, 4, 22 (Migne, op. cit. iv. 443, 
462). Cf. Methodius, Convivium decern virginum, vii. 1 (Migne 
op. cit. Ser. Graeca, xviii. 125). XI CELIBACY 
405 
He threaten against such impure connections. . . . She who 
has been guilty of this crime is an adulteress, not against a 
husband, but Christ."1 According to the gospel of Pseudo- 
Matthew, the Virgin Mary had in a similar manner dedicated 
herself as a virgin to God.2 The idea that the deity is jealous 
of the chastity of his or her servants may also perhaps be at 
the bottom of the Greek custom according to which the hiero- 
phant and the other priests of Demeter were restrained from 
conjugal intercourse and washed their bodies with hemlock- 
juice in order to kill their passions,3 as also of the rule which 
required the priests of certain goddesses to be eunuchs? 
Religious celibacy is further enjoined or commended as a 
means of self-mortification supposed to appease an angry god, 
or with a view to raising the spiritual nature of man by sup- 
pressing one of the strongest of all sensual appetites. Thus 
we find in various religions celibacy side by side with other 
ascetic observances practised for similar purposes. Among 
the early Christians those young women who took a vow of 
chastity " did not look upon virginity as any thing if it were 
not attended with great mortification, with silence, retire- 
ment, poverty, labour, fastings, watchings, and continual 
praying. They were not esteemed as virgins who would not 
deny themselves the common diversions of the world, even 
the most innocent."5 Tertullian enumerates virginity, 
widowhood, and the modest restraint in secret on the marriage- 
bed among those fragrant offerings acceptable to God which 
1 St. Cyprian, Epistola LXII., ad Pomponium de virginibus, 
3 sq. (Migne, op. cit. iv. 368 sqq.). See also Neander, General History 
of the Christian Religion and Church, i. 378. The Council of Elvira 
decreed that such fallen virgins, if they refused to return back to 
their former condition, should be denied communion even at the 
moment of death (Concilium Eliberitanum, a.d. 305, ch. 13 [Labbe- 
Mansi, op. cit. ii. 8]). 
2 Gospel of Pseudo-Matthew, 8 (Ante-Nicene Christian Library, 
xvi. 25). See also Gospel of the Nativity of Mary, 7 (ibid. xvi. 57 sq.f 
3 Wachsmuth, Hellenische Alterthumskunde, ii. 560. 
* Cf. Lactantius, Divines Institutiones, i. 17 (Migne, op. cit. vi. 
206) :-" Deum mater et amavit formosum adolescentem, et eumdem 
cum pellice deprehensum exsectis virilibus semivirum reddidit ; 
et ideo nunc sacra ejus a Gallis sacerdotibus celebrantur." 
6 Fleury, Manners and Behaviour of the Christians, p. 128 sq. 406 
THE HISTORY OF HUMAN MARRIAGE chap. 
the flesh performs to its own especial suffering.1 Moreover, 
it was argued that marriage prevents a person from serving 
God perfectly, because it induces him to occupy himself 
too much with worldly things.2 Though not contrary to the 
act of charity or the love of God, says Thomas Aquinas, it 
is nevertheless an obstacle to it.3 This was one, but certainly 
not the only, cause of the obligatory celibacy which the 
Christian Church imposed upon her clergy. 
A very important cause of religious celibacy is the idea that 
sexual intercourse is defiling and in certain circumstances 
a mysterious cause of evil.4 The Syryenians maintain that 
it develops a peculiar dangerous and contagious quality or 
substance " which has no definite shape or form " called 
pez; hence a man who has had intercourse with a woman 
must undergo a rite of purification.5 So also the Sulka in 
the central part of New Britain believe that such intercourse 
pollutes both men and women, married as well as unmarried 
-a pollution which they call a sle. Married persons can 
themselves cleanse it away, whereas unmarried ones have 
to ask somebody else to do it for them. The latter, as long 
as they are polluted, are dangerous both for other people and 
for themselves. Everybody avoids them and children are 
warned of them by their parents ; and if they are not purified 
by a particular ceremony, the sle with which they are affected 
will kill them. Even a person who overtakes a couple having 
sexual intercourse is considered to be defiled and in need of 
purification, though of a lighter kind.6 The Tahitians, 
according to Captain Cook, believed that if a man refrained 
from all connection with women some months before death, 
1 Tertullian, De resurrectione carnis, 8 (Migne, op. cit. ii. 806). 
2 Vincentius Bellovacensis, Speculum naturale, xxx. 43. See also 
von Eicken, op. cit. p. 445. 
3 Thomas Aquinas, Summa theologica, ii.-ii. 184. 3. 
4 The danger attributed to sexual intercourse has been much 
emphasised by Mr. Crawley in The Mystic Rose. See also Wester- 
marck, Origin and Development of the Moral Ideas, ii. 414 sqq. ; Idem, 
Marriage Ceremonies in Morocco, passim, and especially p. 334 sqq. 
6 Nalimov, Zur Frage nach den ursprunglichen Beziehungen der 
Geschlechter bei den Syrjanen, p. 2 sq. 
* Parkinson, Dreissig Jahre in der Siidsee, p. 179 sq. XI CELIBACY 
407 
he passed immediately into his eternal mansion without any 
purification.1 When Jellinghaus once asked some Munda 
Kols in Chota Nagpur if a dog can sin, the answer was, " If 
the dog did not sin, how could it breed? "2 Herodotus 
writes :-" As often as a Babylonian has had intercourse with 
his wife, he sits down before a censer of burning incense, and 
the woman sits opposite to him. At dawn of day they wash ; 
for till they are washed they will not touch any of their 
common vessels. This practice is also observed by the 
Arabs."3 Among the Hebrews both the man and woman 
had to bathe themselves in water, and were " unclean until 
the even." 4 The Nandi, in Eastern Africa, say that people 
are dirty (simwek) when they have had sexual intercourse, 
hence they must purify themselves by bathing.5 The belief 
in the defiling effect of such intercourse seems to underlie 
the idea prevalent among various peoples that incontinence, 
and especially illicit love, injures the harvest.6 In New 
Caledonia, " in the case of women engaged in planting, 
cohabitation is not allowed for some time before and after."7 
Among the Koita of British New Guinea " continence is 
insisted on when making a new garden. If a man so em- 
ployed approach his wife his yams will grow but poorly." 
And among the Motu, another tribe in the same colony, a 
similar restriction applies to the " master of a turtle or 
dugong fishing party, or to any one connected with making 
1 Cook, Voyage to the Pacific Ocean, ii. 164. 
2 Jellinghaus, ' Sagen, Sitten und Gebrauche der Munda-Kohls 
in Chota Nagpore,' in Zeitschr. f. Ethnol. iii. 367. 
3 Herodotus, i. 198. 4 Leviticus, scv. 18. 
5 Hollis, Nandi, p. 92. 
6 Frazer, Magic Art, ii. 105 sqq. This is in my opinion a more 
natural explanation than the one suggested by Sir James G. Frazer 
(ibid. ii. 117) namely, that uncivilised man imagines " that the vigour 
which he refuses to expend in reproducing his own kind, will form as 
it were a store of energy whereby other creatures, whether vegetable 
or animal, will somehow benefit in propagating their species." This 
theory entirely fails to account for the fact that illicit love, by 
preference, is supposed to mar the fertility of the earth and to blight 
the crops-a belief which is in full accordance with my own ex- 
planation, in so far as such love is considered particularly polluting. 
7 Atkinson. ' Natives of New Caledonia,' in Folk-Lore, xiv. 256. 408 
THE HISTORY OF HUMAN MARRIAGE chap. 
new turtle or dugong nets.1 Among the Moanus of the 
Admiralty Islands a man must observe continence five 
days before he goes out fishing with large nets, two or three 
days before he goes to war, and two days before he visits the 
bachelors' section of the men's house.2 Among the Nagas 
of Manipur, according to Mr. Hodson, the periods when a 
man is in special danger are marked by the prohibition of 
sexual intercourse; for example, when he is preparing for 
or returning from a raid.3 The Wagiriama of British East 
Africa believe that if men cohabit with their wives during 
war time " they will be unable to kill any of their enemies, 
and that if they themselves receive a trifling wound it will 
prove fatal."4 Their neighbours, the Wasania, say that if a 
man has intercourse with his wife during hunting, he will 
have bad luck and see no animals to kill.5 Among some 
other East African tribes-the Akamba, the Akikuyu, and 
the Atheraka-it was believed that if men and women 
cohabited during the hours in which the cattle -were out 
grazing this would cause the stock to die, and that it would 
bring ill-luck upon the village if a man had sexual intercourse 
with a woman when on a journey.6 Among the Masai, when 
poison is being made and when honey wine is being brewed, 
sexual intercourse is considered dangerous for the man 
who is making poison and for the man and woman who 
have been selected for brewing honey wine. Were they 
to have intercourse during the six days when the honey wine 
is brewing, the wine would be undrinkable and the bees which 
made the honey would fly away.7 Similar taboos are very 
widespread.8 
1 Seligman, Melanesians of British New Guinea, p. 140. 
2 Parkinson, op. cit. p. 395. 
3 Hodson, Naga Tribes of Manipur, p. 88. 
4 Barrett, ' Notes on the Customs and Beliefs of the Wa-Giriama, 
etc., British East Africa,' in four. Roy. Anthr. Inst. xli. 22. 
5 Ibid. p. 31. 
8 Dundas, ' Organization and Laws of some Bantu Tribes in 
East Africa,' in Jour. Roy Anthr. Inst. xlv. 274. 
7 Hollis, ' Note on the Masai System of Relationship and other 
matters connected therewith,' in Jour. Roy. Anthr. Inst. xl. 480 sq. 
8 See, e.g., Frazer, Totemism and Exogamy, iv. 226 sq. XI CELIBACY 
409 
The fact which chiefly interests us in the present connec- 
tion is that the idea of sexual defilement is particularly cdh- 
spicuous in connection with religious observances. It is 
a common rule that he who performs a sacred act or enters a 
holy place must be ceremoniallyclean/andno kind of unclean- 
ness is to be avoided more carefully than sexual pollution. 
Among the Chipewyan, " if a chief is anxious to know the 
disposition of his people towards him, or if he wishes to settle 
any difference between them, he announces his intention of 
opening his medicine-bag and smoking in his sacred stem. . . . 
No one can avoid attending on these occasions; but a 
person may attend and be excused from assisting at the 
ceremonies, by acknowledging that he has not undergone 
the necessary purification. The having cohabited with his 
wife, or any other woman, within twenty-four hours preced- 
ing the ceremony, renders him unclean, and, consequently, 
disqualifies him from performing any part of it" 2 Herodotus 
tells us that the Egyptians, like the Greeks, " made it a point 
of religion to have no converse with women in the sacred 
places, and not to enter them without washing, after such 
converse."3 This statement is corroborated by a passage 
in the ' Book of the Dead.' 4 In Greece 5 and India6 those 
who took part in certain religious festivals were obliged to 
be continent for some time previously. Before entering 
the sanctuary of Men Tyrannos, whose worship was extended 
over the whole of Asia Minor, the worshipper had to abstain 
from garlic, pork, and women, and had to wash his head.7 
Among the Hebrews it was a duty incumbent upon all to 
be ritually clean before entering the temple-to be free from 
sexual defilement,8 leprosy,8 and the pollution produced by 
1 See Westermarck, Origin and Development of the Moral Ideas, 
ii. 294, 295. 352 sq. 
2 Mackenzie, Voyages to the Frozen and Pacific Oceans, p. cii. sq. 
8 Herddotus, ii. 64. 
4 Wiedemann, Herodots zweites Buch, p. 269 sq. 
6 Wachsmuth, op. cit. ii. 560. 
• Oldenberg, Die Religion des Veda, p. 411. 
7 Foucart, Des associations religieuses chez les Grecs, pp. 119, 
123 sq. 
8 Leviticus, ch. xv. 8 Ibid. ch. xiii. sq. 410 
THE HISTORY OF HUMAN MARRIAGE chap. 
the association with corpses of human beings, of all animals 
not permitted for food, and of those permitted animals 
which had died a natural death or been killed by wild beasts j1 
and eating of the consecrated bread was interdicted to per- 
sons who had not been continent for some time previously.2 
A Muhammadan would remove any defiled garment before 
he commences his prayer, or otherwise abstain from praying 
altogether ; he would not dare to approach the sanctuary of 
a saint in a state of sexual uncleanness; and sexual inter- 
course is forbidden for those who make the pilgrimage to 
Mecca.3 The Christians prescribed strict continence as a 
preparation for baptism4 and the partaking of the Eucharist.5 
They further enjoined that no married persons should par- 
ticipate in any of the great festivals of the Church if the night 
before they had lain together;6 and in the ' Vision ' of 
Alberic, dating from the twelfth century, a special place of 
torture, consisting of a lake of mingled lead, pitch, and resin, 
is represented as existing in hell for the punishment of married 
people who have had intercourse on Sundays, church festivals, 
or fast-days.7 And they abstained from the marriage-bed at 
other times also, when they were disposed more freely to 
give themselves to prayer.8 
Holiness is a delicate quality which is easily destroyed if 
anything polluting is brought into contact with the holy 
object or person.9 In Efate namim, or uncleanness, sup- 
posed to be contracted in various emergencies, was especially 
avoided by the sacred men, because it was believed to destroy 
1 Ibid. xi. 24 sqq. ; xvii. 15. Numbers, xix. 14 sqq. Montefiore, 
Hibbert Lectures on the Religion of the Ancient Hebrews, p. 476. 
2 1 Samuel, xxi. 4 sq. 3 Koran, ii. 193. 
4 St. Augustine, De fide et operibus, vi. 8 (Migne, op. cit. xl. 202). 
5 St. Jerome, Epistola XLVIII. 15 (Migne, op. cit. xxii. 505 sq.). 
8 Lecky, History of European Morals, ii. 324. St. Gregory the 
Great, Dialogi, i. 10 (Migne, op. cit. Ixxvii. 200 sq.). 
7 Albericus, Visio, ch. 5, p. 17. Delepierre, L'enfer deer it par ceux 
qui I'ont vu, p. 57 sq. On this subject see also Muller, Das sexuelle 
Leben der christlichen Kulturvolker, pp. 52, 53, 120 sq. 
8 St. Jerome, Epistola XLVIII. 15 (Migne, op. cit. xxii. 505). 
Fleury, op. cit. p. 75. 
• On this subject I may refer the reader to my essay on The Moorish 
Conception of Holiness (Baraka), ch. iii. p. 121 sqq. XI CELIBACY 
411 
their sacredness.1 The priestly taboos, of which Sir James G. 
Frazer has given such an exhaustive account in The 
Golden Bough, have undoubtedly in a large measure* a 
similar origin. Nay, sexual uncleanness is not only 
injurious to holiness with which it comes into contact, but 
it may also injure holy persons or objects in a more 
positive manner. When the supreme pontiff in the 
kingdom of Congo left his residence to visit other 
places within his jurisdiction, all married people had to 
observe strict continence the whole time he was out, as 
it was believed that any act of incontinence would prove 
fatal to him.2 In self-defence, therefore, gods and holy 
persons try to prevent polluted individuals from approaching 
them, and their worshippers are naturally anxious to do the 
same. But apart from the resentment which the sacred 
being naturally feels against the defiler, it appears that holi- 
ness is supposed to react quite mechanically against pollu- 
tion, to the destruction or discomfort of the polluted indivi- 
dual. I shall illustrate these various effects supposed to 
result from contact between sexual uncleanness and holiness 
by some ideas I found prevalent in Morocco.3 
If a person who has been polluted by any discharge of 
sexual matter enters a holy place, a mosque or a shrine, 
before he has washed himself, it is believed that he will suffer 
some misfortune : he will become blind or lame or mad, or 
he or some member of his family will become ill or die, or 
he will lose some of his animals, or his corn crop will be bad. 
I was told that if a person who is not sexually clean visits the 
tomb of the Aglu saint Sidi Daud, which is situated on an 
island in Southern Morocco, he will find that the water in 
the sea has suddenly risen to such a height that he cannot 
go back to the mainland, but has to wait till it has gone 
down. Sexual cleanness is required of those who have any- 
thing to do with the com-the ploughman, the women who 
clear 'the crops of weeds in the spring, the reapers, the 
1 Macdonald, Oceania, p. 181; 
2 Labat, Relation historique de I'llthiopie occidentale, i. 259 sq. 
3 See Westermarck, Marriage Ceremonies in Morocco, p. 334 sqq. ; 
Idem, Moorish Conception of Holiness (Baraka'), p. 124 sqq. 412 
THE HISTORY OF HUMAN MARRIAGE chap. 
threshers, the women who grind com ; such persons are 
otherwise supposed to pollute its bar aka, or holiness, and also, 
in many cases, to do injury to themselves. So also if an 
unclean person goes into a granary, it is believed not only 
that the grain will lose its baraka, but that he himself will 
fall ill; a Berber from the Ait Warain told me that he once 
got bad boils because he entered a granary in a state of 
sexual uncleanness. Nor must an unclean individual enter 
the vegetable garden, as such a visit would do harm both to 
the garden and to the person who went there.1 If an unclean 
person goes among the sheep they will die because they are 
holy animals, or the person himself will suffer some mis- 
fortune. If a person who is sexually unclean rides a horse, 
another holy animal, some evil will happen either to the 
horse or the rider or both; the horse will get sores on its 
back, or fall down with its rider, and the latter will have 
boils, or become ill or die, or be late in arriving at his destina- 
tion, or not succeed in his business. So also the person who 
robs the bees of their honey must be clean, lest they should 
leave the place or die, or the person himself be stung by them ; 
for there is baraka in the bees. Sexual intercourse, moreover, 
destroys the magic efficacy of a charm if it is not removed 
before the act, and I have also heard that in such a case the 
person who wears it may become ill; but some people main- 
tain that a charm enclosed in a metal case is not in danger of 
being defiled. 
It should also be noticed that, owing to the injurious effect 
of pollution upon holiness, an act generally regarded as 
sacred would, if performed by an unclean individual, lack 
that magic efficacy which otherwise would be ascribed to it. 
Muhammad described the ablution which is a necessary pre- 
paration for prayer as " the half of faith and the key of 
prayer."2 The Moors say that a scribe is afraid of evil 
spirits only when he is sexually unclean, because then his 
reciting of passages of the Koran-the most powerful weapon 
against such spirits-would be of no avail. The Syrian 
1 Idem, Ceremonies and Beliefs connected with Agriculture, &c. in 
Morocco, pp. 17, 22, 23, 28, 46, 47, 54. 
2 Hughes, Dictionary of Islam, p. 3. XI CELIBACY 
413 
philosopher Jamblichus speaks of the belief that " the gods 
do not hear him who invokes them, if he is impure from 
venereal connections."1 A similar notion prevailed among 
the early Christians ; with reference to a passage in the First 
Epistle of the Corinthians,2 Tertullian remarks that the 
Apostle added the recommendation of a temporary absti- 
nence for the sake of adding an efficacy to prayers.3 To 
the same class of beliefs belongs the notion that a sacrificial 
victim should be clean and without blemish.4 The Chibchas 
of Bogota considered that the most valuable sacrifice they 
could offer was that of a youth who had never had inter- 
course with a woman.5 
If ceremonial cleanliness is required even of the ordinary 
worshipper it is all the more indispensable in the case of a 
priest ;6 and of all kinds of uncleanness none is to be more 
carefully avoided than sexual pollution. Sometimes ad- 
mission into the priesthood is to be preceded by a period 
of continence.7 In the Marquesas Islands no one could 
become a priest without having lived chastely for several 
years previously.8 Among the Tshi-speaking peoples of 
the Gold Coast men and women, in order to become members 
of the priesthood, have to pass through a long novitiate, 
generally from two to three years, during which they live 
in retirement and are instructed by the priests in the secrets 
of the craft; and " the people believe that, during this 
period of retirement and study, the novices must keep their 
bodies pure, and refrain from all commerce with the other 
sex." 9 The Huichol of Mexico, again, are of opinion that a 
man who wishes to become a shaman must be faithful to 
his wife for five years, and that, if he violates this rule, he 
1 Jamblichus, De mysteriis, iv. n. 2 i Corinthians, vii. 5. 
3 Tertullian, De exhortations castitatis, 10 (Migne, op. cit. ii. 926). 
4 See Westermarck, Origin and Development of the Moral Ideas, 
ii. 295 sq. 
5 Simon, quoted by Waitz, Anthropologie der Naturvolker, iv. 363. 
6 Cf. Westermarck, Origin and Development of the Moral Ideas, 
ii- 352 sq. 
7 Cf. Landtman, op. cit. p. 118 sqq. 
8 Waitz-Gerland, Anthropologie der Naturvolker, vi. 387. 
9 Ellis, Tshi-speaking Peoples, p. 120. 414 
THE HISTORY OF HUMAN MARRIAGE chap. 
is sure to be taken ill and will lose the power of healing.1 
In ancient Mexico the priests, all the time that they were 
employed in the service of the temple, abstained from all 
other women but their wives, and " even affected so much 
modesty and reserve, that when they met a woman they 
fixed their eyes on the ground that they might not see her. 
Any incontinence amongst the priests was severely punished. 
The priest who, at Teohuacan, was convicted of having 
violated his chastity, was delivered up by the priests to the 
people, who at night killed him by the bastinado."2 Among 
the Kotas of the Nilgiri Hills the priests-who, unlike the 
" dairymen " of their Toda neighbours, are not celibates- 
are at the great festival in honour of Kamataraya forbidden 
to live or hold intercourse with their wives for fear of pollu- 
tion, and are then even obliged to cook their meals them- 
selves.3 It seems that, according to the Anatolian religion, 
married hieroi had to separate from their wives during the 
period they were serving at the temple.4 The Hebrew priest 
should avoid all unchastity ; he was not allowed to marry a 
harlot, or a profane, or a divorced wife,5 and the high-priest 
was also forbidden to marry a widow.6 Nay, even in a 
priest's daughter unchastity was punished with excessive 
severity, because she had profaned her father; she was to 
be burned.7 
Carried further, the idea underlying all these rules and 
practices led to the notions that celibacy is more pleasing 
to God than marriage,8 and that it is a religious duty for those 
members of the community whose special office is to attend 
to the sacred cult. For a nation like the Jews, whose am- 
bition was to live and to multiply, celibacy could never 
become an ideal; whereas the Christians, who professed the 
most perfect indifference to all earthly matters, found no 
difficulty in glorifying a state which, however opposed it 
1 Lumholtz, Unknown Mexico, ii. 236. 2 Clavigero, op. Cit. i. 274. 
3 Thurston, in the Madras Government Museum's Bulletin, i. 193. 
4 Ramsay, Cities and Bishoprics of Phrygia, i. 136, 137, 150 sq. 
5 Leviticus, xxi. 7. 8 Ibid. xxi. 14. 7 Ibid. xxi. 9. 
8 Cf. Westermarck, Origin and Development of the Moral Ideas, 
ii- 358- Xi CELIBACY 
415 
was to the interests of the race and the nation, made men pre- 
eminently fit to approach their god. Indeed, far from being 
a benefit to the kingdom of God by propagating the species, 
sexual intercourse was on the contrary detrimental to it by 
being the great transmitter of the sin of our first parents. 
This argument, however, was of a comparatively late origin. 
Pelagius himself almost rivalled St. Augustine in his praise 
of virginity, which he considered the great test of that 
strength of free-will which he asserted to be at most only 
weakened by the fall of Adam.1 
An important question now calls for an answer:-Why 
is sexual intercourse looked upon as unclean and defiling, 
or in other words, as a mysterious source of danger ? That 
the danger is supposed to be particularly alarming in the 
case of contact between the polluted individual and anything 
holy is merely an instance of the general belief that holiness 
is exceedingly sensitive to, and readily reacts against, 
external influences; indeed it is not only exceptionally 
susceptible to influences that are, or are supposed to be, in- 
jurious also in ordinary cases, but it is affected or influenced 
by various acts or omissions which are otherwise considered 
perfectly harmless.2 Sexual uncleanness, as we have seen, 
belongs to the former kind of influences, even though its 
dangerous character chiefly displays itself when it comes 
into contact with holiness. Sometimes it is said to cause 
effeminacy either in the unclean man himself or in other 
men, who are contaminated by his company; this is the 
professed reason why among the Moanus of the Admiralty 
Islands a man must be continent for a couple of days before 
he goes to war or enters the bachelors' room.3 Here we have 
evidently to do with a belief in the transmission of qualities 
-intercourse with a woman makes the man womanly, or 
makes him a medium by which womanliness is transferred to 
others ; and such a belief may be supported, if not suggested, 
by the feeling of weakness or depression following upon sexual 
intercourse.4 
1 Milman, op. cit. i. 151, 153. 
2 See Westermarck, Moorish Conception of Holiness, p. 137 sqq. 
3 Parkinson, op. cit. p. 395. 1 Cf. Crawley, op. cit. p. 187. 416 
THE HISTORY OF HUMAN MARRIAGE chap. 
This explanation, however, is by no means sufficient to 
account for all the various evils attributed to sexual unclean- 
ness. It should be remembered that woman is frequently 
regarded as an unclean being ; hence* intimate contact with 
her may be considered polluting. Particularly during men- 
struation, or when with child, or at child-birth, she is 
supposed to be charged with mysterious baneful energy- 
no doubt on account of the marvellous nature of these pro- 
cesses and especially the appearance of blood. And it is 
probably such frequent temporary defilement of a specifi- 
cally feminine character that has led to the notion of the 
permanent uncleanness of the female sex.1 It is true that 
not only connection with a woman but any discharge of 
sexual matter is considered to pollute a man. The Nandi 
say that a man is " dirty " not only after sexual intercourse, 
but after an involuntary seminal emission.2 A similar 
opinion is held by Muhammadans ; once when I was staying 
in a country place in Morocco where the water supply had 
become extremely scarce, my Berber teacher, who had always 
before most regularly said his daily prayers, refrained from 
doing so for a couple of days because his clothes had been 
defiled by a pollution. And in Christian penitentials it is 
prescribed that a man who has a nightly pollution shall get 
up and sing a certain number of psalms, which varies accord- 
ing as he was polluted " ex cogitatione," or " non voluntate."3 
As woman, however, is the normal centre of sexual life, the 
notion of her uncleanness may have influenced the concep- 
tion of its various manifestations. But at the same time I 
strongly believe that there are other reasons for the impurity 
1 See Westermarck, Origin and, Development of the Moral Ideas, i. 
663 sqq. ; Idem, Marriage Ceremonies in Morocco, p. 337 sqq. 
2 Hollis, Nandi, p. 92. 
3 Gregory III., Judicia congrua poenitentibus, ch. 24 (Labbe-Mansi, 
op. cit. xii. 293) :-" In somno peccans, si ex cogitatione pollutus, 
viginti duos psalmos cantet: si in somno peccans sine cogitatione, 
duodecim psalmos cantet." Poenitentiale Pseudo-Theodori, xxviii. 25 
(Wasserschleben, Bussordnungen der abendlandischen Kirche, p. 600):-- 
" Qui in somno, non voluntate, pollutus sit, surgat, cantetque vii. 
psalmos poenitentiales." Cf. ibid, xxviii. 6, 33 (Wasserschleben, 
P- 559 sqf. XI CELIBACY 
417 
ascribed to them. The sexual matter and the sexual organs 
are the seats of mysterious propensities, and the whole 
process of generation is a mystery. Among the Maori, says 
Mr. Best, " the organs of generation were deeply imbued 
with tapu (sacredness, sometimes ' uncleanness ') and mana 
(influence, prestige, supernatural power), both of an active 
and of a passive form. This may have sprung from an 
observance of the mystery of sex. ... When a person 
repeated a magic spell, he would place his hand on his genital 
organs in order to give force, supernatural power (mana) 
to his incantation. This is quite Oriental. Observe sundry 
passages in the Bible, where a man, when making a solemn 
promise, is said to have placed his hand ' in the hollow of his 
thigh.' "1 A similar practice occurs in Morocco in connec- 
tion with swearing. In Tanna, one of the New Hebrides, 
narak, or bewitching force, is ascribed to a man's penis, 
even the sight of which is looked upon as dangerous.2 And 
the widespread belief in the sorcery of indecent gestures 
indicates how commonly the organs of generation have been 
held to be the seat of evil-working power.3 Moreover, there 
is the secrecy drawn over the sexual functions and the 
feeling of sexual shame, which give them the appearance of 
something illicit and therefore, it seems, has together with 
other factors led to the idea that they are defiling, or even 
sinful. In the next chapter we shall examine the nature and 
origin of this feeling. 
1 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 208. See also Idem, 'Maori Beliefs con- 
cerning the Human Organs of Generation,' in Man, xiv. 132 sqq. 
2 Sommerville, ' Ethnological Notes on New Hebrides,' in Jour. 
Anthr. Inst, xxiii. 368. 
3 Hirn, Origins of Art, p. 217. CHAPTER XII 
SEXUAL MODESTY 
Sexual shame may be defined as the shame caused by the 
idea of the sexual function or anything which is apt to lead to 
such an idea ; and sexual modesty is the fear of, or the ten- 
dency to avoid, anything which would arouse sexual shame. 
It leads to the concealment of the sexual function itself, 
or of any part of the body the exposure of which may too 
openly call forth a thought of it, or to the avoidance of any 
gesture or word that may have a similar effect. I think there 
can be no doubt that sexual shame thus centres round the 
sexual function. It is felt at the exposure of certain parts of 
the body only when such exposure directs the thought to 
this function. As regards reticences of speech Dr. Havelock 
Ellis argues that they are not adequately accounted for by 
the statement that modesty tends to irradiate from the action 
to the words describing the action, because " there is a ten- 
dency for modesty to be more deeply rooted in the words 
than in the actions " j1 and in support of this view he 
quotes Kleinpaul's remark that modest women have a 
much greater horror of saying indecent things than of 
doing them, believing that " fig-leaves were especially made 
for the mouth."2 So also Stendhal observes that " a woman 
of feeling betrays her sentiments for her lover sooner by 
deed than by word."3 This, however, is no evidence against 
1 Ellis, Studies in the Psychology of Sex, vol. i. Evolution of 
Modesty, p. 66. 
2 Kleinpaul, Sprache ohne Worte, p. 309. 
3 Stendhal, De I'amour, p. ^7 sa. 
418 ch. xii SEXUAL MODESTY 
419 
the view T am advocating, considering that the deed implies 
the gratification of a strong desire, and therefore may more 
easily overcome the resistance of modesty than the word, 
which possesses no such inducement. We must not, of 
course, compare words said in public with deeds committed 
in secret. Persons who indulge in indecent talk would 
certainly shrink at performing openly the acts which they 
speak of ; this, as we shall see, is the case among savages 
as well as among ourselves.1 That sexual modesty is funda- 
mentally concerned with the sexual function is, in my opinion, 
a fact of the greatest importance for a right understanding of 
its genesis. 
Some degree of sexual modesty is in all probability 
universal in mankind. There are peoples among whom men 
or women, or both men and women, go absolutely naked 
or cover themselves in a manner which could have nothing 
to do with the feeling of shame. But, as we shall see in a 
subsequent chapter, this by no means proves the absence 
of sexual modesty. Cook states that in Tahiti it was cus- 
tomary to " gratify every appetite and passion before wit- 
nesses," and that the principal topic in the conversation of 
the people was always that which was also the principal 
source of their pleasure, everything being mentioned without 
any restraint or emotion and in the most direct terms by both 
sexes.2 The Nagas in the southern mountains of Assam, 
according to a Persian historian of the seventeenth century, 
" go about naked like beasts, and do not mind to copulate 
with their women in the streets and the bazaars, before the 
people and the chiefs' '; yet the women covered their breasts.3 
Charlevoix tells us that among the Guaycurus of Paraguay 
those actions which require most privacy were performed in 
1 Dr. Ellis remarks himself (op. cit. i. 68) that in modern Europe, 
as seems fairly evident from the early realistic dramatic literature of 
various countries, no special horror of speaking plainly regarding the 
sacro-pubic regions and their functions existed among the general 
population until the seventeenth century. 
1 Hawkesworth, Account of Voyages in the Southern Hemisphere, 
ii. 196. 
3 Blochmann, * Koch Bihar, Koch Hajo, and A'sam, in the 16th 
and 17th Centuries,' in Jour. Asiatic Soc. Bengal, vol. xli, pt. i. 84. 420 
THE HISTORY OF HUMAN MARRIAGE chap. 
the face of the whole world.1 It may be fairly doubted, 
however, that these and similar statements represent the 
ordinary habits of the peoples concerned; they may refer 
either to individual cases or to obscenities practised at cer- 
tain feasts or in connection with certain rites-occasions 
when the common rules of decency are frequently suspended. 
In no case must they be regarded as typical expressions of 
savage behaviour. 
Cook says of the Maori that in their "carriage and conversa- 
tion there was as much modest reserve and decorum with 
respect to actions, which yet in their opinion were not 
criminal, as are to be found among the politest people in 
Europe " ;2 and from a later time we have the statement 
that between unmarried persons of both sexes among them 
" there is much delicacy of intercourse in the presence of 
others,'' although their songs, dances, and conversation are 
occasionally indelicate.3 With reference to some tribes in 
South-East Australia, Ridley observes that " in their own 
way, within the Emits of traditional rule, the Aborigines are 
very strict in the observance of the dictates of natural 
modesty."4 Among the natives of Northern Queensland 
" a woman will rarely join in any filthy songs or stories in- 
dulged in by the men, and seldom converses upon any sexual 
matter except with other females. On the other hand, in 
times of quarrel, &c., she will give way to language of the 
most obscene character-the only method she can employ 
of showing her contempt-for the benefit of the man or men 
she may be angered with." 5 Like many other peoples,6 
these natives have both a decent and an indecent vocabulary 
for the sexual parts.7 The savages of the Bismarck Archi- 
1 Charlevoix, History of Paraguay, i. 92. Baron Nordenskiold 
says (Indianliv i El Gran Chaco, p. 82) that among the Ashluslay 
of the Gran Chaco several couples lie together after their dances, and 
that " spectators are not unusual." 2 Hawkesworth, op. cit. iii. 450. 
3 Thomson, Story of New Zealand, i. 177. 
* Ridley, Kamilaroi, Dippil, and Turrubul, p. 157. 
6 Roth, North Queensland Ethnography : Bulletin No. 8. Notes 
on Government, &c., p. 7. • Havelock Ellis, op. cit. i. 67. 
7 Roth, op. cit. p. 7. Idem, Ethnological Studies among the North- 
West-Central Queensland Aborigines, p. 184. XII SEXUAL MODESTY 
421 
pelago, who are anything but chaste in their habits, make use 
of euphemistic expressions in speaking of those parts j1 
and we are told that in the district of Laur in New Ireland 
men and women, boys and girls, sometimes commit suicide 
when an indecent word is shouted to them as an insult.2 
Of some of the Marshall Islanders, Chamisso wrote:- 
" Unmarried women enjoy their freedom, under the pro- 
tection of decorum. The girl requires presents from the 
man ; but the veil of modesty is drawn over all the connec- 
tions which unite the two sexes."3 So also among the 
Mortlock Islanders, although a girl is at liberty to have inter- 
course with men belonging to other tribes, " decorum is 
strictly observed in every case."4 In all the south-eastern 
parts of British New Guinea visited by Dr. Seligman, in spite 
of the great freedom granted to the unmarried of both sexes, 
the people " were absolutely modest in their behaviour and 
nowhere was an indecent gesture seen. The men carry their 
physical modesty further than do Europeans, for no one takes 
off his perineal bandage when bathing, and a native would 
be almost as much ashamed to remove this before a single 
kinsman as he would be to stalk naked through his hamlet. 
Women are equally particular and are said never to strip 
before one another." 5 Of the Koita-among whom " con- 
tinence, except when ceremonially imposed on the man, is 
practically unknown, and the girls are habitually unchaste "- 
we are told that " the greatest decorum is. however, observed 
between the sexes, while in public they usually avoid each 
other."6 Among the Mailu, whose culture is essentially of 
the same type as that of the Koita, much secrecy and 
reticence seem to obtain between man and woman in sex 
matters.7 So also there is modesty in the relations between 
1 Pfeil, Studien und Beobachtungen aus der Siidsee, p. 74. 
a Stephan and Graebner, Neu-Mecklenburg (Bismarck-Archipel}, 
p. no. 8 Kotzebue, Voyage of Discovery into the South Sea, iii. 172. 
4 Kubary, ' Die Bewohner der Mortlock Inseln,' in Mittheil. 
Geograph. Gesellsch. Hamburg, 1878-79, p. 252. 
5 Seligman, Melanesians of British New Guinea, p. 568. 
6 Ibid. p. 134. 
7 Malinowski, * Natives of Mailu,* in Trans. Roy. Soc. South 
Australia, xxxix. 564. 422 
THE HISTORY OF HUMAN MARRIAGE chap. 
the sexes among the natives of Kaiser Wilhelm Land.1 
Their notions of decency require that a married man shall 
never be seen publicly in the company of his wife, nor take 
the slightest notice of her in the presence of others ; if he 
has been long absent from his village, he, on his return, 
passes her without even looking at her.2 According to 
Seemann, it was quite contrary to Fijian ideas of delicacy 
that husband and wife should sleep under th- same roof ; 
thus a man spent the day with his family, but absented 
himself on the approach of night, and rendezvous between the 
couple were arranged in the depths of the forest, unknown 
to anybody else.3 A later writer, however, says that it is a 
mere fiction that the marriages of the Fijians were consum- 
mated in the bush.4 But in various islands belonging to the 
Malay Archipelago, according to Riedel, sexual intercourse 
takes place in the forest, not in the house.5 He says that 
in Ceram all natural functions, especially that of coitus, 
are performed in secret, by preference in the forest, and that 
two persons are forbidden to speak of sexual matters in the 
presence of a third.6 Among the Veddas of Ceylon indecent 
language, so common in Sinhalese and Tamil villages, is 
said to be unknown ; this, in the opinion of Mr. Nevill, " is 
probably due to the restraint felt in the presence of women 
known to be virtuous." 7 Among the Andamanese " coarse 
conversation appears to be of rare occurrence and to meet 
with little or no encouragement."8 
1 Vetter, ' Bericht uber papuanische Rechtsverhaltnisse, nament- 
lich bei den Jabim,' in Nachrichten uber Kaiser Wilhelms-Land, 1897, 
pp. 87, 93. Schellong, ' Ueber Familienleben und Gebrauche der 
Papuas der Umgebung von Finschhafen,' in Zeitschr. f. Ethnol. 
xxi. 18. 2 Neuhauss, Deutsch Neu-Guinea, i. 161 sq. 
3 Seemann, Viti, pp. 110,191. 4 Thomson, Fijians, p. 202. 
5 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, 
pp. 260 (Aru Islands), 448 (Wetter). For a similar custom among the 
Kaya-Kaya of Dutch New Guinea see Poch, ' Vierter Bericht uber 
meine Reise nach Neuguinea,' in Sitzungsberichte der mathematisch- 
naturwissenschaftlichen Klasse der Kaiserl. Akademie der Wissen- 
schaften, vol. cxv. pt. i. 902. 8 Riedel, op. cit. p. 96. 
7 Nevill, ' Vaeddas of Ceylon,' in Taprobanian, i. 178. 
8 Man, 'Aboriginal Inhabitants of the Andaman Islands,' in 
Jour. Anthr. Inst. xii. 94 n. 2. XII SEXUAL MODESTY 
423 
Among the Kalmucks " lads and girls allow themselves 
no frivolous jokes. Even young men are ashamed to speak of 
things which offer the European youth materials for daily 
entertainment."1 The Siberian Yukaghir, in spite of the 
sexual licence prevailing among them, are generally very 
bashful and modest in their speech.2 Tessmann says that 
if you discuss anything sexual with a West African Pangwe 
negro, you will hear him repeatedly utter the word oson, 
which means " shame."3 The Negroes of the Bali countries 
are not ashamed to speak of sexual matters even in the 
presence of women, but they have sexual intercourse only in 
private.4 Of the Negroes of Accra, Monrad wrote that they, 
in spite of their licentiousness and obscene dances, otherwise 
observed in their relations to the fair sex a decency which is 
often lacking even among civilised Europeans.5 
Among the Witoto and Boro of the North-West Amazons 
" the women are extraordinarily modest in their behaviour" ; 
but at the same time we are told that " sexual matters ap- 
peared to be discussed freely and lewdly by both sexes, and 
even by young children."6 On the other hand, Koch- 
Griinberg relates that when talking to some Desana Indians 
on sexual subjects, the conversation was stopped by them 
till the women were sent away, and only after their depar- 
ture the men began to talk freely and broadly.7 Among the 
Chiriguanos of the Gran Chaco, who allow the greatest free- 
dom in the relations between the sexes, Father Chome was 
struck by the propriety of their conversation and their 
behaviour in general; he never heard an indecent word 
among them.8 But at present they are anything but 
modest in their speech.8 Hans Stade, who in the middle of 
1 Bergmann, Nomadische Streifereien unter den Kalmiiken, iii. 288. 
2 Jochelson, Yukaghir, p. 62 sqq. 
3 Tessmann, Die Pangwe, ii. 256. 
4 Hutter, Wanderungen und Forschungen im Nord-Hinterland von 
Kamerun, pp. 331, 374, 376. 
6 Monrad, Bidrag til en Skildring af Guinea-Kysten, p. 56 sq. 
6 Whiffen, North-West Amazons, p. 262. 
7 Koch-Griinberg, quoted by Whiffen, op. cit. p. 262. 
8 Chomd, ' Dritter Brief,' in Stoecklein, Der neue Welt-Bolt, 
vol. iv. pt. xxix. 72. 8 Nordenskibld, op. cit. p. 205. 424 
THE HISTORY OF HUMAN MARRIAGE chap. 
the sixteenth century was a captive among the Tupi- 
nambase on the coast of Brazil, remarks that men and 
women " conduct themselves decently, and sleep with one 
another privately."1 The Indians of British Guiana consider 
it extremely indecent if a married couple show any signs of 
mutual affection in the presence of others.2 Speaking of 
certain North American Indians, Lafitau observes, " Ils 
n'osent aller dans les Cabanes particulieres ou habitent leurs 
Epouses, que durant 1'obscurite de la nuit; . . . ce seroit une 
action extraordinaire de s'y presenter de jour."3 Among 
various other peoples it is considered indecent or is even 
prohibited to have sexual intercourse in the daytime/' The 
night is also the time for courtship.5 Lahontan wrote of 
some Indians of Canada :-" Nothing of intrigue or court- 
ship must be mention'd to the savage ladies in the day time, 
for they will not hear it ; they'll tell you the night-time is the 
most proper season for that. . . . This is a general rule, 
that whoever designs to win the affection of a girl, must speak 
to her in the day-time, of things that" He remote from the 
intrigues of love."6 
Sexual modesty shows itself in various ways in the 
preliminaries to, or the conclusion of, a marriage, not only 
on the part of the woman but on the part of the man. 
1 Captivity of Hans Stade of Hesse, in A.D. 1547-1555, among the 
Wild Tribes of Eastern Brazil, p. 144. 
2 Appun, ' Die Indianer von Britisch-Guayana,' in Das Ausland, 
xliv. 832. 
8 Lafitau, Mceurs des sauvages ameriquains, i. 576. 
4 Neuhauss, op. cit. i. 149 (people of Tami, Kaiser Wilhelm Land). 
Tessmann, op. cit. ii. 253 (Pangwe). Vergette, Certain Marriage 
Customs of some of the Tribes in the Protectorate of Sierra Leone, p. 8. 
Penna di Billi, Breve notizia del regno del Thibet, p. 70. 
5 Long, Voyages and Travels of an Indian Interpreter, p. 174 
(Chippewa). Harmon, fournal of Voyages and Travels in the Interior 
of North America, p. 340 (Indians on the east side of the Rocky Moun- 
tains). Gomes, Seventeen Years among the Sea Dyaks of Borneo, 
p. 120 sq. ; Ling Roth, Natives of Sarawak, i. 109 sq. (Dyaks). 
Seligman, Melanesians of British New Guinea, pp. 76, 502, 708. 
a Lahontan, New Voyages to North-America, p. 453. See also 
Perrot, ' Memoir on the Manners, &c. of the Savages of North 
America,' in Blair, Indian Tribes, i. 69 (Algonkin) ; Carver, Travels 
through the Interior Parts of North America, p. 241 (Dakota). xn SEXUAL MODESTY 
425 
Very frequently he avoids directly asking the parents of 
the girl for her hand. In Futuna, north of the Fiji group, 
the suitor, without saying anything, left a hog at the house 
of the parents and then retired; if they retained the animal 
it was a sign of consent to the marriage, whereas their 
returning it was a sign of refusal.1 In Ugi and San 
Christoval, of the Solomons, where the marriage of a girl 
is arranged by her parents and friends, " if a man wants 
a wife, he cooks a dish full of yams and cocoa nuts and 
carries it to the house of his bride elect, whence he returns 
without uttering a word. The next morning he returns 
to take the dish away. If the food has not been eaten, 
he is not accepted, and he takes this as an insult; if, on 
the other hand, the bowl is empty and a couple of fathoms 
of money left instead of it, he is not accepted either, but 
the family wishes to keep on friendly terms with the suitor. 
Finally, if the dish be entirely empty he is accepted."2 In 
Samoa, among the middle and lower classes, if a young 
man desired to marry a girl, " he either made known his 
wishes by means of a friend, soa, or else, preparing an oven 
of food as a present, he posted off with it himself to the 
family of the young girl. If it was refused, it was an 
intimation that his attentions were discouraged; but if 
accepted, it was understood that his visits would be approved 
by the family, and the marriage was soon arranged. . . . 
Among the higher ranks, however, proposals for marriage 
were always made through the medium of a third party, 
or soa, literally a companion."3 Among the Koryak it 
frequently happens that a young man does not tell anybody 
of his intentions. " He goes to the house in which the 
girl lives whom he desires to marry, and, without saying 
a word, remains there, performing all house-work becoming 
to a man. The house-owner receives the suitor's services 
with the same silence as he renders them." This is particu- 
larly practised when parents disapprove of their son's choice 
1 Gr6zel, Dictionnaire Futunien-Fran^ais, p. 232. 
2 Elton, * Notes on Natives of the Solomon Islands,' in Jour. 
Anthr. Inst. xvii. 94. 
3 Stair, Old Samoa,Ip. 171 sq. 426 
THE HISTORY OF HUMAN MARRIAGE chap. 
and he does not want to submit to their disapproval; 
otherwise the parents or other elder relatives of the young 
man go to the bride's parents as match-makers for him.1 
The practice of employing one or more go-betweens for 
arranging a marriage is extremely widespread both among 
savages2 and among more civilised nations.3 It is common 
in the Muhammadan world,4 as it was among the heathen 
1 Jochelson, Koryak, p. 739. 
2 Prichard, Through the Heart of Patagonia, p. 9} (Tehuelches). 
Candelier, Rio-H ccha et les Indiens Goajires, p. 210. Russell, 
' Pima Indians,' in Ann. Rep. Bur. Ethnol. xxvi. 183. La Flesche, 
' Osage Marriage Customs,' in American Anthropologist, xiv. 129. 
Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethnol. iii. 259. 
Teit, ' Thompson Indians of British Columbia,' in Memoirs American 
Museum Nat. Hist., Anthropology, i. 322. Jochelson, Yukaghir, 
p. 88. Scott Robertson, Kafirs of the Hindu-Kush, p. 533. Hutchin- 
son, Account of the Chittagong Hill Tracts, p. 96 (Chukmas). Gurdon, 
Khasis, p. 128. Hodson, Naga Tribes of Manipur, p. 87. Rose and 
Brown, ' Lisu Tribes of the Burma-China Frontier,' in Memoirs 
Asiatic Soc. Bengal, iii. 262. Davidson, Island of Formosa, p. 573 
(Tsalisens). Worcester, Philippine Islands, p. 492 (Tagbanuas of 
Culion and Busuanga). Perelaer, Ethnographische beschrijving der 
Dajaks, p. 47. Meinicke, Die Inseln des Stillen Oceans, ii. 407 
(certain Ladrone Islanders). Hahl, ' Ueber die Rechtsanschauungen 
der Eingeborenen des Innern der Gazelle Halbinsel,' in Nachrichten 
uber Kaiser Wilhelms-Land und den Bismarck-Archipel, 1897, p. 78. 
Junod, Life of a South African Tribe, i. 103 (Thonga). Volkens, 
Der Kilimandscharo, p. 251 (Wadshagga). Baumann, Usambara, 
p. 45 (natives of the Tanga coast). Ling Roth, Great Benin, 
p. 38. Thomas, Anthropological 'Report on Ibo-speaking Peoples of 
Nigeria, i. 65.-Among various peoples the proposal is, often at 
least, made by the young man's father or parents (Goddard, Life 
and Culture of the Hupa, p. 55 ; Hutton, Among the Eskimos of 
Labrador, p. 76; Bogoras, Chukchee, p. 584 ; Jochelson, Koryak, 
p. 739 ; Batchelor, Ainu and their Folk-Lore, p. 224 ; Stack, Mikirs, 
p. 17 ; Proust de la Girondidre, Twenty Years in the Philippines, 
p. 271 [Aetas] ; Walen, * Sakalava ' [Madagascar], in Antananarivo 
Annual, no. viii. 52 ; Kropf, Das Volk der Xosa-Kaffern im bstlichen 
Sud-Afrika, p. 133 ; Weule, Native Life in East Africa, p. 306 
[Wayao] ; Cunningham, Uganda, p. 101 [Bakoki] ; Ellis, Yoruba- 
speaking Peoples of the Slave Coast of West Africa, p. 153). 
3 Bowring, Kingdom and People of Siam, i. 118 ; Young, Kingdom 
of the Yellow Robe, p. 89 (Siamese). Saunderson, ' Notes on Corea 
and its People,' in Jour. Anthr. Inst. xxiv. 305. 
* Westermarck, Marriage Ceremonies in Morocco, p. 19. XII SEXUAL MODESTY 
427 
Arabs;1 it prevailed in ancient India ;2 it has existed, or 
still exists, in European countries.3 
The bashfulness of the young man, as well as of the 
girl, may last after the match has been arranged. In the 
Banks Islands, according to Dr. Codrington, " boys and 
girls, and young people generally, who are engaged are 
verv shv about it, and will hardly look at one another ; 
but as the time for marriage draws on it is correct for the 
youth to make little presents and otherwise shew attention."4 
At Vanua Lava, in Port Patteson, says the same authority, 
" boys and girls who are engaged generally avoid one 
another, but through shyness, not by rule." 5 Among the 
Bororo Indians in Matto Grosso, if a man has accepted the 
proposal of marriage made by a girl, he still remains in the 
bahito (the men's house) one or two, or even ten or fifteen, 
days, because he is ashamed to be seen entering the house of 
his bride. Occasionally his prospective father-in-law fetches 
him late at night so that he may not be hurt by the gibes and 
mockings of the other men in the bahito. " This sense of 
shame is accentuated if neither of the two have had sexual 
intercourse before." 6 The bashfulness is particularly marked 
or symbolically accentuated at the wedding.7 Among the 
Andamanese, " when the elders of the Sept are aware that a 
young couple are anxious to be married, the bride is taken to 
1 Frey tag, Einleitung in das Studium dev Arabischen Sprache, 
p. 202. Wellhausen, Reste des arabischen Heidentums, p. 433 n. 1. 
2 Haas, ' Die Heirathsgebrauche der alten Inder,' in Weber's 
Indische Studien, v. 181, 236, 276, 288, 291-293, 380, 411. Winter- 
nitz, ' Das altindische Hochzeitsrituell,' in Denkschriften dev Kaiserl. 
Akademie der Wissenschaften, Philosophisch-historische Classe, xl. 21, 
39 sq. Macdonell and Keith, Vedic Index of Names and Subjects, 
i. 482. 
3 Weinhold, Die deutschen Frauen in dem Mittelalter, i. 316 sqq. 
v. Schroeder, Die Hochzeitsgebrduche der Esten und einiger andrer 
finnisch-ugrischer Vblkerschaften, in Vergleichung mit denen der 
indogetmanischen Volker, p. 32 sqq. Sartori, Sitte und Brauch, i. 52. 
4 Codrington, Melanesians, p. 240. 5 Ibid. p. 43. 
6 Eric and Radin, ' Contributions to the Study of the Bororo 
Indians,' in Jour. Anthr. Inst, xxxvi. 390. 
7 See Parsons, ' The Reluctant Bridegroom,' in Anthropos, x.-xi. 
65 sq. 428 
THE HISTORY OF HUMAN MARRIAGE chap. 
a newly made empty hut and made to sit down in it. The 
bridegroom runs away into the jungle, but, after some 
struggling and pretence at hesitation, is brought in by force 
and made to sit down on the bride's lap. This is the whole of 
the ceremony. The newly married couple have little to say to, 
and are very shy of, each other, for at least a month after 
marriage, when they gradually settle down together." 1 
Among the Roro of British New Guinea, when the bridegroom 
sees the bridal procession coming he hides in the men's club- 
house, but his companions soon drag him forth, and then, dis- 
regarding his complaints, paint and decorate him, and 
bring him to his father's house, to which the bride has 
already come. Here he is made to sit down near her, while 
the onlookers call out, " So-and-so is married to so-and-so." 
But the bridegroom and bride take no notice of each other, 
the former only talking to his friends; and in the evening 
he goes to sleep in the clubhouse, while the bride sleeps in 
the house of her father-in-law.2 In New Britain the bride 
has to stay five days alone in the bridegroom's house, while 
he himself is hiding away in a place in the forest or the high 
grass known only to the men.3 Among the Abkhasians of 
the Western Caucasus the bridegroom runs away and hides 
on his wedding night.4 On the third night of an Apache 
wedding both bride and bridegroom suddenly disappear- 
until then they have not even been permitted to speak 
with each other, and when they come back the young bride 
assumes the air and pretences of extraordinary modesty.5 
In Morocco the bashfulness of bride and bridegroom is 
expressed or symbolised in their extremely reserved be- 
haviour throughout the wedding, in many of the abstinences 
which they have to observe, and in the covering up of their 
faces, the shutting up of the bride in a box, and other 
1 Portman, History of Our Relations with the Andamanese, i. 42. 
See also Man, in Jour. Authr. Inst. xii. 137. 
2 Seligman, op. cit. p. 269. See also Williamson, ' Some unrecorded 
Customs of the Mekeo People of British New Guinea,' in Jour. Roy. 
Anthr. Inst, xliii. 276. 
3 Parkinson, Im Bismarck-Archipel, p. 98. 
4 v. Seidlitz, ' Die Abchasen,' in Globus, Ixvi. 40. 
8 Cremony, Life among the Apaches, p. 248. XII SEXUAL MODESTY 
429 
ceremonies.1 The mock-capture of the bride and/ the 
resistance made by her will be dealt with later on. 
We now come to the interesting question how to explain 
the origin of sexual modesty. Already St. Augustine asked 
why even conjugal intercourse, legitimate and honourable 
though it be, seeks retirement from every eye ; but the 
only answer he could give was, " Because that which is 
by nature fitting and decent is so done as to be accompanied 
with a shame-begetting penalty of sin."2 Stendhal observes : 
-" People have noticed that birds of prey hide them- 
selves to drink; the reason being that, obliged to plunge 
their head in the water, they are at that moment 
defenceless. After a consideration of what happens at 
Tahiti, I see no other natural basis for modesty."3 Several 
other writers have traced modesty at least in part to a similar 
origin.4 Thus Tillier writes, " Il nous parait evident que ce 
sentiment a eu comme origine premiere la necessity de 
nous derober au danger pendant 1'accouplement: les 
individus accouples etant absolument incapables de songer 
a leur conservation et d'y veiller, pendant 1'union sexuelle, 
ont dii necessairement rechercher la solitude."6 It has 
been pointed out that this is done by certain animals ; 
elephants are said to retire to the densest part of the jungle, 
and camels to the loneliest solitude of the desert, before 
they copulate.6 And it is argued that if our domestic animals 
take no such precaution, the reason is that they are safe 
without it.7 But would not the same apply to savage man, 
at least when he is in his shelter or when he is surrounded 
by friends, and to civilised man on practically all occasions ? 
1 See Westermarck, Marriage Ceremonies in Morocco, p. 345, 
and General Index, s.v. ' Bashfulness.' 
2 St. Augustine, De civitate Dei, xiv. 17 (Migne, Patrologies cursus, 
xli. 426 sq.). 
8 Stendhal, De I'amour, p. 55 (English translation by Woolf, p. 81). 
* Mortimer, Chapters on Human Love, p. 40. Crawley, Mystic 
Rose, p. 134. Maupetit, ' La pudeur,' in Bull, et memoires Soc. 
d'Anthr. Paris, ser. vi. vol. v. 412. 
6 Tillier, L'instinct sexuel chez Vhomme et chez les animaux, p. 254. 
8 Lejeune, in Bull, et mem. Soc. d'Anthr. Paris, ser. vi. vol. v. 417. 
Mortimer, op. cit. p. 40. 7 Tillier, op. cit. p. 254. 430 
THE HISTORY OF HUMAN MARRIAGE chap. 
How is it, then, that an instinct which in the case of 
domesticated animals has been lost, when no longer needed, 
has been preserved, nay intensified, in mankind, though 
equally superfluous ? Moreover, if the need of privacy 
or " the dread of the approach of enemies " were at the 
bottom of sexual modesty, might we not expect to find a 
feeling of bashfulness also with regard to sleep, which im- 
plies a much greater degree of defencelessness and a much 
longer exposure to danger than does the act of procreation ? 
According to another theory, closely allied ta the one just 
discussed, the secrecy in sexual matters is the result of the 
fear of being disturbed by rivals. " The practice of steal- 
ing away to some solitary spot, where no rival could disturb 
the enjoyment of this pleasure," says Professor Lester 
Ward, " grew into a regular and permanent custom, and like 
all customs was continued after the cause which originated 
it and made it necessary ceased to exist. . . . Fear of 
being circumvented and prevented from consummating a 
pleasurable act otherwise entirely natural and proper-- 
this was the oiiginating cause of modesty. We see some- 
thing approaching it in animals. The rivalries of wild 
animals, could we fully comprehend their effects, would 
show us some of the first steps in the direction of establish- 
ing such a custom as that which grew up among men. . . . 
Where there are several cocks, they seek to separate their 
chosen mates from the rest in order to enjoy perfect 
immunity."1 The last-mentioned fact, instead of support- 
ing the theory, discloses its weakness. The cock avoids the 
presence of other cocks, but not the presence of other hens 
besides its mate. So also the fear of rivals might have 
induced men to avoid intercourse in the presence of other 
men, not in the presence of women. But how, then, could 
such fear have been the originating cause of sexual modesty, 
considering that men exhibit this feeling chiefly towards 
women, and that it is even more marked in women than 
in men, or at least in young virgins compared with young 
men of the same age ? 
1 Ward, Dynamic Sociology, i. 634 sq. Cf. Crawley, op. cit. 
pp. 150, 180 ; Mortimer, op. cit. p. 40. XII SEXUAL MODESTY 
431 
Sexual modesty is further supposed to be due, in a large 
measure, to fear of causing disgust.1 Dr. Havelock Ellis 
points out that even savages very low down in the scale 
feel repugnance to the public satisfaction of natural needs. 
Thus there are peoples who consider it immodest to eat in 
public.2 When Dr. von den Steinen, in the presence of 
some Bakairf, ate a piece of fish offered him, they hung 
their heads and were ashamed.3 But I strongly doubt that 
the custom of eating in solitude, where it exists, is really 
in the first place due to fear of causing disgust. As Mr. 
Crawley observes, it is a common belief that the presence of 
other persons may contaminate the food either involun- 
tarily or through their evil eye.4 At any rate, the rule of 
eating alone is by no means sufficiently widespread to justify 
the assumption that there is in man an instinctive dis- 
position to feel disgust at the sight of eating ; among 
ourselves there is not a trace of such a feeling, provided 
that the eating is conducted in the manner prescribed by 
the social etiquette. It is different with the satisfaction 
of certain other natural needs. But even in the case of the 
excretory functions it seems that publicity is naturally apt 
to cause disgust chiefly when the function is productive of 
an obnoxious smell, whereas the concealment of urination 
is largely a matter of sexual modesty. As for the 
sexual function, I fail to see on what ground it, 
in normal cases, could be a cause of disgust, unless 
it for some other reason were looked upon as some- 
thing to be concealed. Sexual indecency is no doubt 
disgusting, but I am inclined to believe that it is disgusting 
because it is indecent, rather than indecent because it is 
1 Ellis, op. cit. i. 46 sqq. Renouvier and Prat, La nouvelle mona- 
dologie, p. 221 : "La pudeur est d'abord une crainte qui nous avons 
de d6plaire, d'avoir a rougir de nos imperfections de nature.'' 
Crawley, op. cit. p. 139. 
2 Ellis, op. cit. i. 48 sq. For instances of eating in solitude see 
Crawley, op. cit. p. 150 sqq. 
3 von den Steinen, Unter den Naturvolkern Zentval-Brasiliens, 
p. 66. 
4 Crawley, op. cit. p. 156 sqq. See also Westermarck, Origin 
and Development of the Moral Ideas, i. 561. 432 
THE HISTORY OF HUMAN MARRIAGE chap. 
disgusting. If the sexual function were intrinsically dis- 
gusting to outsiders, how could the feelings centering round 
it be the favourite topic of polite literature and the stage, 
whereas the excretory functions are at most the source of 
a vulgar joke ? 
By this criticism I certainly do not mean to deny that 
there may have been in mankind an original desire to 
perform the sexual function in secret, apart from any feeling 
of shame. Fear of being attacked by an enemy, fear of being 
interrupted by a rival, fear of causing unsatisfied desire 
and unpleasant emotions in others-which Dr. Havelock 
Ellis and Mr. Crawley seem to include in the fear of causing 
disgust-may all have induced men to observe such secrecy. 
Indeed, the mere presence of a more or less curious onlooker 
may be sufficiently disturbing to be dreaded by a loving 
couple. Yet I do not believe that any of these factors can 
account for the essential characteristics of sexual modesty. 
One of these characteristics is that it is chiefly, though 
by no means exclusively, displayed in the presence of per- 
sons of the opposite sex. Another is that it is more developed, 
or at least more ostensible, in women, or rather young women, 
than in men of the same age. The opinion has, it is true, 
been expressed that men are really more modest than women.1 
But I consider that Dr. Havelock Ellis is indisputably 
right in his statement that modesty-which he defines as 
" an almost instinctive fear prompting to concealment and 
usually centering around the sexual processes "-while 
common to both sexes, is more peculiarly feminine, so that 
it may almost be regarded as the chief secondary sexual 
character of women on the psychical side. The woman, 
he adds, who is lacking in this kind of fear is lacking, also, 
in sexual attractiveness to the normal and average man. 
The organic sexual demands of women call for coyness in 
1 Online Renooz, Psychologic compares de I'homme et de la femme, 
p. 86 sq. Sergi and Viazzi, quoted by Ellis, op. cit. i. 3. M. Tillier 
observes (op. cit. p. 255), " Nous n'avons aucune raison de croire 
que le sentiment instinctif de la pudeur soit plus developpe chez la 
femme: la reserve de celle-ci est certainement plus grande, mais 
c'est la un ph6nomene different." xn SEXUAL MODESTY 
433 
courtship. Their modesty is not an artificial garment, 
which they throw off or on at will: it is organic, " but like 
the snail's shell, it sometimes forms an impenetrable cover- 
ing, and sometimes glides off almost altogether. A man's 
modesty is more rigid, with little tendency to deviate toward 
either extreme. Thus it is, that, when uninstructed, a man 
is apt to be impatient with a woman's reticences, and yet 
shocked at her abandonments."1 I think there can be no 
doubt that the coyness which the female animal shows 
when she is courted by the male is one root of sexual modesty 
in mankind. This accounts for the peculiarly feminine 
character of this feeling, for the attractiveness it possesses in 
the eyes of the men, for the readiness with which its behaviour 
becomes a cover of coquetry, and, partly, for the broad fact 
that it is particularly felt towards a person of the opposite 
sex. Even if sexual modesty had no other root than 
feminine coyness, some reserve would, among human beings, 
be displayed not only by women in the presence of men, 
but also by men in the presence of women. For both for 
prudential and altruistic motives the men would naturally 
avoid uselessly to hurt the feelings of the other sex and 
make the women blush for shame. And when once observed 
in the relations between the sexes, the reserve might be 
fossilised into general social customs in the shape of decency 
and thus in some measure also affect the men's behaviour 
towards persons of their own sex. 
This theory of the origin of sexual modesty is of 
course very incomplete in so far that it accepts female 
coyness as a fact without trying to explain it. But an 
explanation of female coyness can only be attempted in 
connection with a general theory of courtship, and this re- 
quires a chapter by itself. At the same time our explana- 
tion of sexual modesty suffers from another defect. It is 
also incomplete because this feeling undoubtedly has more 
than one root. There are certain facts which cannot be 
explained as either the direct or indirect results of feminine 
coyness, but indicate a close connection between sexual 
modesty and the aversion to incest, or, more strictly 
1 Ellis, op. cit. i. i, 4 sq. 434 
THE HISTORY OF HUMAN MARRIAGE chap. 
speaking, the sexual aversion which prevails between 
members of the same domestic circle, with the exception, 
of course, of husband and wife. 
In Morocco a young man is very shy of his parents in 
all sexual matters, even such relating to his marriage. This 
shyness shows itself from the moment when the marriage 
is first thought of. till the wedding has passed and even 
longer. There is no conversation on the subject between 
the youth and his parents, or at least his father, and he 
sometimes entirely avoids the latter from the day when the 
arrangements for his marriage commence. At Fez the 
father keeps away from his son's wedding ; on the great 
occasion when the bridegroom, after the arrival of the 
bride, comes to his parents' house to meet her, his father 
hides himself somewhere in the house or in another house, 
or stays in the street. Sometimes the bridegroom puts off 
his first intercourse with the bride till all the guests have 
gone away, or till the evening of that day, because, as I 
was told, he is shy of his parents. Among the Berbers of 
the tribe Aith Yusi, in the neighbourhood of Fez, if he lives in 
the same tent as his parents or one of them, he will, out of 
bashfulness, stay away during the day even for a whole 
month, taking his meals somewhere else and visiting his 
wife only at night; and here, as in many other tribes, he 
resumes his intercourse with them by a special ceremony, 
that of kissing their heads. Among the neighbouring tribe 
the Aith Nd er the young husband enters his parents' tent 
in the company of his " vizier," or best-man, and another 
bachelor, all three with the hoods of their cloaks pulled over 
their faces, and one after the other they kiss the parents' 
heads ; he is too bashful to appear alone, and for some time 
afterwards he still shows great shyness of his parents. 
Among the Ait Warain, south of Fez, it is the rule that the 
young husband does not show himself or speak to his parents 
till some three weeks have passed, when he kisses their 
hands ; and when his wife becomes a mother he never holds 
the child in his arms in the presence of his father. There 
is also bashfulness in the relations between the bride and her 
parents. In Andjra, in Northern Morocco, she does not see XII SEXUAL MODESTY 
435 
them as long as the wedding lasts. Among the Berbers of 
the Aith Sadden, and at Fez, the girl avoids her father at 
the time of the betrothal. Nowhere in Morocco does the 
father go to his daughter's wedding. In some places her 
mother also stays at home, and the same may even be the 
case with her grown-up brothers.1 
We are told that in some parts of Algeria2 and Tunis, 
also, a son avoids his father or both his father and mother 
for a longer or shorter time after his marriage. Among some 
natives of Southern Tunis, " le fiance est mis au courant 
huit jours avant la celebration du mariage, par des amis de 
son age, des dispositions arretees entre parents ; il quitte 
alors la maison patemelle, car des ce moment, par pudeur 
et par respect, il ne doit pas se montrer a ses parents, et 
cette interdiction dure encore pendant plusieurs mois apres 
son mariage. Il va se cacher dans les jardins, ou ses amis 
lui apportent a manger et viennent lui tenir compagnie."3 
Of some Arabs of the desert an old traveller, d'Arvieux, 
writes:-" All the relations assist at the wedding, except 
the bride's father, who leaves the house that evening out 
of an odd piece of niceness that will not suffer him to be at 
home whilst they are putting his daughter to bed with a 
man. The fathers make this a point of honour." 4 Among 
some other peoples, also, the parents of the bride do not 
attend their daughter's wedding.5 In some parts of 
1 For some further details see my Marriage Ceremonies in Morocco, 
p. 314 sq. 
2 Gaudefroy-Demombynes, Les ceremonies du mariage chez les 
indigenes de l'Algerie, p. 75 (Tlemcen). Idem, ' Coutumes de 
mariage,' in Revue des traditions populaires, xxii. 55 (QaW of the 
Beni Rached). Destaing, Etude sur le dialecte berb^re des Beni Snous, 
p. 290. 
3 Menouillard, ' Un mariage dans le Sud Tunisian (Matmata),' 
in Revue Tunisienne, ix. 372. 
4 d'Arvieux, Travels in Arabia the Desart, p. 235. 
6 Majerus, ' Brautwerbung und Hochzeit bei den Wabende 
(Deutsch-Ostafrika),' in Anthropos, vi. 897. Among the Zulus the 
bride's mother must not be present at her daughter's wedding, 
and the father's presence is also forbidden when the bride is his 
eldest daughter (Hartland, 'Travel Notes in South Africa,' in 
Folk-Lore, xvii. 478). 436 
THE HISTORY OF HUMAN MARRIAGE chap. 
England 1 and in the West of Ireland 2 the bride's mother 
refrains from doing so, and in Shropshire her father, too, 
rarely goes to the church.8 At Booking, in Essex, " the 
parents of the bride keep studiously out of the way at the 
time of the marriage ceremony." 4 
In Rajput families of Tirhut " it is considered contrary 
to etiquette for a young married couple to see each other 
by day so long as the husband's parents are alive, and in 
particular they must avoid being seen together by the 
husband's parents, and must not speak to one another in 
their presence."5 Among the South African Basuto, 
" when a youth wishes to marry, he does not go to his father 
and ask for a wife. Such a course would be most dis- 
respectful, and altogether wanting in etiquette. ... He 
gets up very early one morning and takes out his father's 
cattle to the pasture without milking the cows, and lets the 
calves run with their mothers and drink all the milk. No 
notice of this is taken on his return, and the same course is 
pursued by him for thirty days. All his companions leave 
him severely alone, and nickname him ' the silly one.' On 
the thirtieth day his father says, ' Surely my son wants to 
get married.' This remark is repeated to the youth, and 
the next day the cattle return to their former habits and 
are milked as usual." 6 On some islands off the south-west 
coast of New Britain, when a young man has been 
accepted by the girl he desires to marry, he asks his brother 
to inform his mother about it, and she then visits the girl's 
parents to propose for the hand of their daughter on behalf 
of her son.7 In Erromanga, of the New Hebrides, " the 
girl's father and his friends arranged the match, but, 
1 Miss Burne, Shropshire Folk-Lore, p. 291. Henderson, Notes 
on the Folk-Lore qf the Northern Counties of England and the Borders, 
p. 34 (the presence of the bride's mother in the wedding party is 
inauspicious). 
2 Blake, 'Matrimonial Customs in the West of Ireland,' in 
Folk-Lore, xviii. 81. 3 Miss Burne, op. cit. p. 291. 
4 Evelyn Martinengo-Cesaresco, ' American Songs and Games,' 
in Folk-Lore Journal, ii. 246. 
5 Risley, Tribes and Castes of Bengal. Ethnographic Glossary, 
ii. 189. 8 Minnie Martin, Basutoland, p. 80. 
7 Vogel, Eine Forschungsreise ini Bismarck-Archipel, p. 223. XII SEXUAL MODESTY 
437 
according to their etiquette, the friends, not the father, 
made the first move ; for he was supposed to be narumfrum, 
' ashamed,' to take much part in the matter."1 In the 
Mortlock Islands all frivolous talk is strictly prohibited 
in the presence of a woman, in case some other member of 
her " tribe " also happens to be present. Not even such 
words as those for stomach, girdle, and the covering of the 
hips must be mentioned ; and if a European, ignorant of 
native custom, transgresses this rule, the two persons in 
question will immediately blush for shame and turn away 
their faces.2 On the same islands a brother is allowed to 
play with his sister only till he reaches the age of about 
seven years ; after this age he is chiefly in male company 
or plays with girls belonging to another " tribe."3 
The custom of avoidance between brother and sister has 
been found in various islands in the Pacific Ocean and in 
some other parts of the world as well.4 It has been held to 
imply the habitual existence in the past of sexual relations 
between brothers and sisters ; but this is only one of those 
arbitrary interpretations of customs as survivals of earlier 
hypothetical conditions which unfortunately are still only 
too common in social anthropology. M. de Rochas sees in 
the custom " an unreasonable exaggeration of a natural 
sentiment, the horror of incest " ; 5 and if he by this means 
that it is due to exaggerated sexual modesty inside the family 
1 Robertson, Erromanga, p. 396. 
2 Kubary, in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, 
p. 252. Cf. Kohler, ' Das Recht der Marschallinsulaner,' in Zeitschr. 
vergl. Rechtswiss. xiv. 433. 
3 Kubary, loc. cit. p. 261. See also Arago, Narrative of a Voyage 
round the World, ii. 17 (Caroline Islanders). 
4 de Rochas, La Nouvelle Caledonie, p. 239; Atkinson, ' Natives 
of New Caledonia,' in Folk-Lore, xiv. 254. Fox, ' Social Organiza- 
tion in San Cristoval, Solomon Islands,' in Jour. Roy. Anthr. Inst. 
xlix. 117. Rivers, History of Melanesian Society, i. 255 (Guadal- 
canar in the Solomon Group), 290 sq. (Fiji). Frazer, Totemism 
and Exogamy, i. 542, 565 sq. (some Australian aborigines) ; ii. 
77 sq. (Lepers' Island in the New Hebrides), 131 (Central New 
Ireland), 344 (Gilyak of the island of Saghalin), 638 (BetsimisarakS 
of Madagascar) ; iii. 245 (Navaho of Arizona and New Mexico, and 
Arapaho, an Algonkin tribe), 362 (Tsetsaut of British Columbia). 
5 de Rochas, op. cit. p. 239. 438 
THE HISTORY OF HUMAN MARRIAGE chap. 
circle, I think he is right. In Penrhyn Island or Tongarewa 
" a sister and brother, on meeting after a long absence, 
cannot fondly rush into each other's arms, but must sit 
down facing each other, and nod their heads, one to one side, 
the other to the opposite " j1 this is obviously a rule of 
etiquette founded on sexual modesty, rather than a precau- 
tion against incest. And the same may be said of many of 
the other avoidance customs. On the Gazelle Peninsula in 
New Britain,2 and in Lifu,3 a married sister is not allowed to 
converse with her brother. Among the Arunta an elder 
sister may speak to her brother at all times, but a younger 
sister may not speak to or look at her brother after both are 
grown up ; and a young woman must not speak to her father 
as soon as her marriage has been consummated.4 There 
may be shyness even in the relations between brothers ; in 
Lepers' Island a man who is laughing with a crowd will cease 
to laugh on the arrival of his brother.5 Among the Malagasy 
men and women are particularly ashamed to expose certain 
parts of their body in the presence of very near relatives.6 
Dr. Jochelson says of the Yukaghir that " the sense of 
shame prevailing in a high degree among near relatives 
undoubtedly helped to prevent sexual intercourse among 
them." 7 In this connection the following statement made by 
Dr. Codrington with reference to the natives of the Northern 
New Hebrides deserves mentioning :-" A girl betrothed in 
childhood is taken to her father-in-law's house and brought 
up there ; the boy often thinks she is his sister, and is 
much ashamed when he comes to know the relation in which 
he stands."8 
1 Lamont, Wild Life among the Pacific Islanders, p. 136. 
2 Hahl, in Nachrichten uber Kaiser Wilhelms-Land und den Bis- 
marck-Archipel, 1897, p. 80. 
8 Ray, ' People and Language of Lifu, Loyalty Islands,' in Jour. 
Roy. Anthr. Inst, xlvii. 287. 
* Gillen, * Notes on some Manners and Customs of the Aborigines 
of the McDonnell Ranges belonging to the A.runta Tribe,' in Report 
on the Work of the Horn Scientific Expedition to Central Australia, 
iv. 164, 166. 5 Rivers, op. cit. ii. 154. 
6 Grandidier, Ethnographic de Madagascar, ii. 137. 
' Jochelson, Yukaghir, p. 86. 8 Codrington, Melanesians, p. 240 XXI SEXUAL MODESTY 
439 
Facts like these show that among peoples in different 
parts of the world sexual modesty is especially strongly 
developed in relations between members of the same family 
circle or kin, whether of the same sex or of opposite sexes. 
We experience the same fact among ourselves-quite contrary 
to what is the case with the shame felt in connection with, 
for- example, the excretory functions. A young man is 
certainly in sexual matters more shy of his parents than 
of strangers, and even between brothers there is commonly 
more reserve in such matters than between other friends of 
the same age. I have little doubt that the custom of going 
on a honeymoon is chiefly due to the same shyness of 
relatives. Whenever I have asked experienced persons on 
the subject, the answer, spontaneously given, has confirmed 
this view. 
' In the present connection mention should also be made 
of the taboos so frequently imposed on the intercourse 
between a married man and his parents-in-law or other 
members of his wife's family.1 These taboos may be more 
or less strict and more or less comprehensive, and they may 
last for a longer or shorter time ; but I think that they have 
all in the main the same foundation. I shall again give 
some instances from my experience in Morocco, where the 
avoidance is less radical than in many other countries, but 
for this reason not less instructive. Among the Igliwa, a 
Berber tribe in the Great Atlas Mountains, the bridegroom 
kisses his mother-in-law on the head when he first meets 
her after her arrival at his house, but shows much bashfulness 
in her presence. In the beginning he must not speak to 
her, and he can never eat with her, unless she comes and 
lives with her daughter in his house. So also he for some 
time refrains from speaking to his father- and brothers-in- 
law, and even later he never eats with them, nor is he seen 
1 For instances of such taboos see Tylor, Researches into the 
Early History of Mankind, p. 288 sqq.; Avebury, Origin of Civilisation, 
p. 8 sqq. ; Andree, Ethnographische Parallelen und Vergleiche, p. 159 
sqq.; Crawley, op. cit. p. 400 sqq. ; Frazer, Taboo and the Perils 
of the Soul, p. 84 sqq. ; Idem, Totemism and Exogamy, Index, s.v. 
' Avoidance ' ; Idem, Psyche's Task, p. 75 sqq. 440 
THE HISTORY OF HUMAN MARRIAGE chap. 
together with them in public ; should any of them come 
where he is sitting with other persons, he at once gets up 
and leaves the company. If a man meets a friend whom he 
has not seen for a while, he jokingly says, " Why have you 
been running away from me, am I your brother-in-law ? " 
Among the Arabic-speaking Ulad Bu'aziz in the province 
of Dukkala a man refrains from eating together with or 
in the presence of his father-, mother-, or grown-up brother- 
in-law, if they are of another village. If, on the other hand, 
his father- and brothers-in-law are of his own village, he 
can eat with them, but there must be no indecent talk in 
their presence ; in case anything indecent is said, he goes 
away at once, and if it has reference to himself he may even 
complain of it to his governor, with the result that the 
offender is compelled to pay a fine. Between a man and 
his mother-in-law there is considerable mutual shyness. 
They speak to each other, but not much, and only "sensible 
words ; if they are from different villages they avoid 
looking at each other, while if they are from the same village 
she covers her face and refrains from eating in his presence, 
though he may eat in hers. In the Hiaina, not far from Fez, 
a man does not speak to his parents-in-law if his own parents 
are alive, but otherwise he begins to converse with them 
when a month or so has passed. He is still shyer of his 
father-in-law than of his mother-in-law, and also avoids a 
grown-up brother-in-law.1 
I was told in Morocco that a man would naturally be shy 
of the persons with whose daughter or sister he has sexual 
intercourse, and this I take to be the true explanation of 
the restrictions in question. We hear of similar explana- 
tions given by the natives of other countries. Among the 
Baganda, according to the Rev. John Roscoe, " no man 
might see his mother-in-law, or speak face to face with her. 
. . . This was said to be because he had seen her daughter's 
nakedness.''2 Among the Bahuana of Congo a man may 
never enter the house of his parents-in-law, and if he meets 
1 For further details see my Marriage Ceremonies in Morocco, 
p. 310 sqq. 
2 Roscoe, Baganda, p. 129. XII SEXUAL MODESTY 
441 
them in a road he must turn aside into the bush to avoid 
them. " Repeated inquiries as to the reason of this avoid- 
ance on the part of a man of his parents-in-law," says 
Mr. Torday, " elicited the invariable reply ' that h6 was 
ashamed ' ; to a further inquiry of what he was ashamed, 
the answer would be ' of marrying their daughter.' No 
other reason could be obtained."1 The shyness is naturally 
apt to be mutual. " All the Indians on the east side of the 
rocky mountain," says Harmon, " think it very indecent 
for a father or mother in law, to speak to, or look in the face 
of a son or daughter in law ; and they never do either unless 
they are very much intoxicated. The reason which they 
give for this custom, when questioned on the subject, is, 
the peculiar intercourse which this person has had with 
their child."2 Dr. Jochelson speaks of " the natural 
modesty and bashfulness of the Yukaghir, characterised 
by the belief that the souls of the father-in-law and his 
son-in-law are ashamed when they look at each other."3 
The custom of avoidance between a man and members 
of his wife's family I thus take to be ultimately based on 
the sexual bashfulness felt inside the domestic circle. We 
shall see that the aversion to incest, through an association 
of ideas and feelings, gives rise to various prohibitions, and 
it is therefore all the more easy to understand that it has 
led to a custom like .tjiat which we are now discussing. It 
seems only natural that where this aversion is so exceedingly 
strong as it is among peoples at the lower stages of culture, 
a man should feel sexual shame in the presence of persons 
of his wife's family, who would not only at once come to 
think of his relations to their daughter or sister but feel 
aversion to the very idea of those relations, and that they, 
on their part, also should be anxious to avoid him. Among 
the West African Pangwe, whose sexual modesty is highly 
developed,4 a woman bashfully avoids even the kinsfolk 
of her son-in-law.5 Now it mav be asked whv. as is the 
1 Torday and Joyce, ' Notes on the Ethnography of the Ba-Huana,' 
in Jour. Anthr. Inst, xxxvi. 285 sq. 
2 Harmon, op. cit. p. 341. 3 Jochelson, Yukaghir, p. 81. 
4 Tessmann, op. cit. p. 256. 6 Ibid. p. 261. 442 
THE HISTORY OF HUMAN MARRIAGE chap. 
case in Morocco, a married man's avoidance of his own 
parents should be of so much shorter duration than his 
avoidance of his parents-in-law, although it is, in a way, 
a more direct expression of the feeling which is at the 
bottom of both customs. But the answer is not difficult 
to find. There is no feeling which is more easily influenced 
by habit than shame, and it would be difficult for the son 
to avoid his own parents for long, when he lives in their 
vicinity, if not under the same roof, and has many interests 
in common with them. As appears from some statements 
relating to Morocco, already quoted, there is much less 
restraint in a man's relations to his parents-in-law, also, if 
they live in his own than if they live in another village, and 
his avoidance of them soon ceases if his own parents are dead 
and there, consequently, is more inducement for him to attach 
himself to his wife's family than would otherwise be the case. 
These last-mentioned facts are at variance with Sir 
E. B. Tylor's conclusion that the avoidance between a man 
and his wife's family is intrinsically connected with the 
custom of the husband taking up his abode with the latter.1 
He argues as follows :-" If the customs of residence and the 
customs of avoidance were independent, or nearly so, we 
should expect to find their coincidence following the ordinary 
law of chance distribution. In the tribes where the husband 
permanently lives with his wife's family (sixty-five out of 
three hundred and fifty), we should estimate that cere- 
monial avoidance between him and them might appear in 
nine cases, whereas it actually appears in fourteen cases. 
On the other hand, peoples where the husband at marriage 
takes his wife to his home (one hundred and forty-one out 
of three hundred and fifty), would rateably correspond with 
avoidance between him and her family in eighteen cases, 
whereas it actually appears in nine cases only."2 This 
argument, however, suffers from a very serious defect. 
1 Mr. Kidd (The Essential Kafir, p. 241) also observes that the 
statement that in nearly all cases where a man avoids his wife's 
people and especially his mother-in-law he has entered the wife's 
family, is hardly borne out in South Africa. 
2 Tylor, ' On a Method of investigating the Development of 
Institutions,' in Jottr. Anthr. Inst, xviii. 247. XII SEXUAL MODESTY 
443 
The fact that the avoidance in question has been noticed in 
a comparatively large number of tribes where the custom is 
for a man to reside with his wife's family by no means proves 
that it really is less common in tribes whose habit it is for 
the husband to take her to his own home ; for it would 
naturally be much more conspicuous in the former cases 
than in the latter. This alone would be a sufficient reason 
for finding fault with Tylor's interpretation of the custom, 
when he says that " as the husband has intruded himself 
among a family which is not his own, and into a house 
where he has no right, it seems not difficult to understand 
their marking the difference between him and themselves 
by treating him formally as a stranger."1 Accepting 
Tylor's conclusion that the taboo by the wife's relations of 
her husband is the result of his residence with the wife's 
kin, Dr. Hartland goes still further and declares that this 
taboo " beyond reasonable doubt " is a ceremonial expres- 
sion of that secrecy in the connection between husband and 
wife which characterised the earliest stage of marital rela- 
tions.2 But we have no evidence whatever that there was 
such a " stage." 
Another explanation of the avoidance between parents- 
in-law and son-in-law has been given by Lord Avebury. To 
him it " seems to be a natural consequence of marriage by 
capture. When the capture was a reality, the indignation 
of the parents would also be real; when it became a mere 
symbol, the parental anger would be symbolised also, and 
would be continued even after its origin was forgotten."3 
1 Ibid. p. 248. 2 Hartland, Primitive Paternity, ii. 93. 
3 Avebury, op. cit. p. 103. Mr. Mathew (Eaglehawk and Crow, 
p. 115 ; Two Representative Tribes of Queensland, p. 163 sqq.} likewise 
thinks that the avoidance between mother-in-law and son-in-law 
among Australian tribes is a survival of marriage by capture. With 
reference to such avoidance among the Kurnai of Gippsland, Mr. 
Morgan (in his introduction to Fison and Howitt, Kamilaroi and 
Kurnai, p. 16 sq.) observes that it is, not unlikely, due to the hostile 
feelings aroused against the son-in-law in the mind of his mother-in- 
law in consequence of his elopement with her daughter. Mr. Kidd 
(The Essential Kafir, p. 241) is of opinion that the custom of avoidance 
among the Kafirs has undoubtedly deep connection with wife-capture, 
although this cannot account for all the features of the custom. 444 
THE HISTORY OF HUMAN MARRIAGE chap. 
Against this view it may be said that marriage by capture 
has played a much more prominent part in the imagination 
of anthropologists than in the life of savages. According 
to M. Mauss, again, the avoidance between son-in-law and 
mother-in-law is the consequence of exogamy ; they belong 
to different phratries, between which there is still a barrier 
although they have been brought nearer each other through 
the marriage. In the phratry with which the man has allied 
himself he has a right only over his wife, whereas her mother 
remains sacree for him.1 This hypothesis seems to presuppose 
that the avoidance in question has grown up in matrilineal 
exogamous societies, although as a matter of fact it exists 
in societies which are neither matrilineal nor exogamous. 
Dr. Rivers maintains that the Melanesian rules of avoid- 
ance-which include prohibitions on the use of the personal 
name, on conversation, and on passing one another,-when 
applying to relatives by marriage of different sexes, " carry 
with them definite implications of potential sexual inter- 
course," and are connected with ancient sexual communism ; 
whereas similar avoidances between persons of the same sex 
imply the idea of hostility and the possibility that one 
relative may injure the other, " evidently to be connected 
with the condition of hostility between the two moieties of 
the community."2 An hypothesis which traces exactly 
similar rules when observed in the relations between persons 
of the same sex and in those between persons of different 
sexes to so widely different causes as sexual communism 
in one case and conditions of hostility between the members 
of the two moieties of the community in the other case 
does not inspire much confidence, especially as rules of the 
same kind are of almost world-wide prevalence. I have 
myself observed in a previous work 3 that the avoidance 
between a son-in-law and his parents-in-law may have some- 
thing to do with the fact that he has deprived the latter of 
their daughter ; but at the same time it is obvious from 
1 Mauss, at the International Congress of Ethnology and Ethno- 
graphy at Neuchatel, quoted in L' Anthropologic, xxv. 370 sq. 
2 Rivers, op. cit. ii. 168 sqq. 
3 Marriage Ceremonies in Morocco, p. 316 sq. XU SEXUAL MODESTY 
445 
various facts that this avoidance is essentially connected 
with sexual shame. In Morocco all indecent talk is particu- 
larly prohibited in the presence of a man and his father- or 
brother-in-law.1 So also among the Konde people in " German 
East Africa" a son-in-law is not allowed to hear any improper 
talk in the presence of his father-in-law, and so particular 
are they in this respect that even the hair of the body is a 
tabooed subject.2 The hlonipa custom of the Kafirs is 
obviously closely connected with sexual bashfulness-the 
word hlonipa itself is derived from a root which means 
" shame." A woman has to hlonipa all her husband's 
relations in the ascending line : she must not hold inter- 
course with them, or mention the emphatic syllable in their 
names, or even pronounce their names mentally to herself. 
And she must be specially careful not to uncover in the 
presence of people whom she has to hlonipa any part of the 
body which is usually covered up ; to uncover any part of 
her body in the presence of her mother-in-law would be 
considered very shameful. Moreover, the hlonipa custom 
enjoins that the mother-in-law must avoid being seen by 
her son-in-law with her breasts exposed, " for it is not seemly 
that the breasts that suckled his wife should be seen by him."3 
Among the Baganda, " if a son-in-law accidentally saw his 
mother-in-law's breasts, he sent her a barkcloth in compensa- 
tion, to cover herself, lest some illness, such as tremor, should 
come upon him." 4 
The avoidance between relatives by marriage of different 
sexes has frequently been explained as a means of preventing 
unlawful intercourse.5 The most obvious weakness of this 
1 See supra, p. 440, and Westermarck, Marriage Ceremonies in 
Morocco, p. 311. 
2 Fiilleborn, Das Deutsche Njassa- und Ruwuma-Gebiet, p. 350 sq. 
3 Kidd, op. cit. pp. 236, 237, 241. Leslie, Among the Zulus and 
Amatongas,p. 141 sq. Maclean, Compendium of Kafir Laws and Customs, 
p. 95 sq. 4 Roscoe, op. cit. p. 129. 
5 Fison, in Fison and Howitt, op. cit. p. 103 sq. Macdonald, 
Oceania, p. 189 (Efatese). Weeks, Among Congo Cannibals, p. 134 
(Boloki). Mr. Kidd {op. cit. p. 243) thinks that one reason for the 
practice of hlonipa among the Kafirs was " to render the temptations 
to incest rare." 446 
THE HISTORY OF HUMAN MARRIAGE chap. 
explanation has already been touched upon, namely, that 
it must be supplemented by another one for the avoidance 
between relatives by marriage who are of the same sex. 
Dr. Jochelson believes that among the Yukaghir mother- 
and son-in-law, father- and daughter-in-law, and the elder 
brother and the wite of the younger one, are on terms of 
avoidance " probably for the prevention of sexual cohabita- 
tion between them " ; whereas the avoidance between 
father- and son-in-law and the elder brother and the husband 
of his younger sister are on such terms on account of the 
feeling of bashfulness. But he seems to maintain that in 
the former cases also this feeling has been a co-operative 
cause of the avoidance.1 To Sir James G. Frazer the rules 
of avoidance likewise appear to have been adopted in order to 
prevent incest between persons who were deemed too nearly 
related by marriage ; and he argues that the avoidance 
between persons of opposite sexes were extended " by false 
analogy to include avoidance between persons of the same 
sex who stood in a similar relation to each other."2 I 
cannot believe, however, that any people could have been 
so utterly unreasonable as to apply rules established for the 
express purpose of preventing sexual intercourse between 
certain persons to other persons between whom such inter- 
course is entirely out of the question ; and least of all can 
I conceive that a very large number of peoples, many of 
whom are not even savages, could be guilty of a like 
absurdity. 
But even apart from the awkward necessity of resorting 
to different explanations of similar rules among the same 
people, I find it very hard to believe that the intention of 
preventing incestuous intercourse is at the bottom of the 
avoidances between relatives by marriage of opposite sexes 
-even if these avoidances were better suited for the purpose 
than many of them actually are. Provided that sexual 
intercourse between the persons of different sexes who are 
1 Jochelson, Yukaghir, p. 81 sq. 
2 Frazer, Totemism and Exogamy, iii. 112 sq. ; iv. 108 sqq. Idem, 
Taboo and the Perils of the Soul, p. 85 n. 6. Idem, Psyche's Task, 
p. 93 sqq.. XII SEXUAL MODESTY 
447 
subject to these taboos is really looked upon as incestuous- 
which is by no means certain-one may ask why not similar 
taboos have been applied to the other persons between whom 
sexual relations are condemned as incestuous. May we 
suppose that the mother-in-law-who is a particularly 
conspicuous figure in the rules of avoidance-is more than 
other forbidden women an object of sexual attraction for a 
man who probably in most cases is young enough to be her 
son ? Rather might we suspect that a father-in-law is in 
danger of falling in love with the wife of his son. Now there 
are among many peoples rules of avoidance between them 
also-and there are such rules even between a married 
woman and her mother-in-law. These rules are quite 
analogous to those relating to a man and his relatives by 
marriage, and I believe that they have a similar origin. 
Among the Bushongo, according to Mr. Torday, " le mari 
et sa femme doivent tous deux eviter de se rencontrer avec 
leurs beaux-parents respectifs, et la raison alleguee en faveur 
de cette coutume est qu'on est ' honteux d'avoir marie son 
enfant.' "x But the avoidance between a daughter-in-law 
and her father-in-law seems to be less strict or less frequent 
than that between a son-in-law and his mother-in-law,2 
although we might rather have expected the reverse if the pur- 
pose of the avoidances really had been to prevent incestuous 
intercourse. In Morocco there is no special avoidance at 
all noticeable in the relations between a wife and her parents- 
in-law, although-or, as I should say, probably because- 
they live in very near contact with each other. Moreover, 
the avoidance of either kind is in many cases only temporary ; 
it may cease or become less strict when a child is born.3 
1 Torday and Joyce, Les Bushongo, p. 116. 
2 Frazer, Totemism and Exogamy, iii. 113. Crawley, op. cit. p. 407. 
3 Tylor, in Jour. Anthr. Inst, xviii. 249. Frazer, Psyche's Task, 
PP- 7$» 79* 81. Pallas, Reise durch verschiedene Provinzen des Russischen 
Reichs, iii. 47 (Ostyak). Kidd, op. cit. p. 238 (Basuto). Majerus, 
' Brautwerbung und Hochzeit bei den Wabende (Deutsch-Ostafrika),' 
in Anthropos, vi. 899. The Yakut say that in ancient times " a 
bride concealed herself for seven years from her father-in-law, 
and from the brothers and other masculine relatives of her husband " 
(Sieroshevski, ' Yakuts ' in Jour. Anthr. Inst. xxxi. 93). 448 
THE HISTORY OF HUMAN MARRIAGE chap. 
This seems hard to explain if the taboo was intended to 
prevent incest but is easy to understand if it is merely an 
expression of sexual modesty, which, as already said, is 
readily influenced by habit and necessity. The birth of a 
child is a signal that the union is to be more or less perma- 
nent.1 Grandparental feelings may help to ease the rela- 
tions between those who had to avoid each other. And 
besides, as the brothers Goncourt said, " les petites pudeurs 
n'existent pas pour les meres."2 
Like the hypothesis that the rules of avoidance were 
intended to prevent incestuous intercourse, so Mr. Crawley's 
explanation could apply only to avoidances between persons 
of different sexes. He looks upon the avoidances between 
son-in-law and mother-in-law and between daughter-in-law 
and father-in-law as merely cases of that taboo between the 
two sexes which, according to him, is such a prominent 
feature of primitive society. The female sex is taboo to 
the male, " because of those qualities which spoil a man," 
and the male sex is likewise considered dangerous, and 
is consequently taboo, to the female. Hence the separation 
between brothers and sisters, boys and girls, as soon as 
puberty draws near ; hence the avoidance between engaged 
couples and between the married man and his wife ; hence 
also the avoidance between relatives by marriage of different 
sexes. The avoidance between son-in-law and mother-in- 
law " seems to be causally connected with a man's avoidance 
of his own wife."3 But is there really among primitive 
peoples such a widespread and radical separation between 
the sexes as Mr. Crawley assumes ? If there were, why 
should the avoidance between a man and his mother-in-law 
have attracted the attention of travellers in all parts of the 
world, whereas so very little is heard of the avoidance of a 
man and his wife ? There must after all be something 
peculiar about the relations between son- and mother-in- 
law, they could hardly be similar to those between men and 
women in general. Mr. Crawley says " it will be found 
that the mother-in-law taboo tends to disappear when the 
1 Cf. Crawley, op. oil. p. 407 sq. 2 Journal des Goncourt, iii. 5. 
3 Crawley, op. cit. p. 408 sqq. XII SEXUAL MODESTY 
449 
taboos between husband and wife are intensified, and 
vice versa." As sexual taboo must be kept up for safety, 
also in the case of married people, " a substitute to receive 
the onus of taboo is useful, and the best substitute is the 
mother-in-law; if the husband avoids her, his relations 
with his wife will be secure, and if the mother-in-law avoids 
him, her daughter's safety will be secured likewise."1 This 
is ingenious, but where is the evidence for the fundamental 
statement that the taboos between son- and mother-in-law 
and between husband and wife alternate with one another ? 
The avoidance between relatives by marriage of the same 
sex Mr. Crawley makes no attempt to explain. He has 
undoubtedly minimised its prevalence and importance.2 
Tn the Hiaina in Morocco, as we have seen, a man is still shyer 
of his father-in-law than of his mother-in-law. Among the 
Batoro. south of Lake Albert Nyanza, the avoidance between 
son- and father-in-law is even more rigid than that between 
son- and mother-in-law ; and among the Lendu, another 
tribe in Uganda, the father-in-law can never visit his son-in- 
law, except in the event of the serious illness of his daughter, 
whereas the mother-in-law may visit her son-in-law and his 
wife when two months have passed since the marriage.3 
I have no doubt that cases of this kind are exceptional, 
and believe that the common view is correct, that the rules 
of avoidance relating to persons of different sexes are more 
frequent or more rigid than those relating to persons of the 
same sex. But this is only what may be expected if they 
are due to sexual modesty, which is most strongly shown 
towards people of the opposite sex. The presence of a son- 
in-law is apt to call forth in his mother-in-law associations 
of ideas that make her blush both as a woman and as a 
mother ; and a daughter-in-law would naturally feel more 
1 Ibid. p. 412. 
2 He quotes, for instance (ibid. p. 406 sq.), Mr. Howitt's denial 
(• Notes on the Australian Class Systems,' in Jour. Anthr. lust. xii. 
503) of any taboos between a man and his father-in-law among 
the Australian natives. But there are such taboos among them, as 
appears from Curr, The Australian Race, iii. 461 (Mount Gambier 
blacks), and Strehlow, op. cit. vol. iv. pt. i. 90 (Arunta), 
* Cunningham, Uganda, pp. 54, 331. 450 
THE HISTORY OF HUMAN MARRIAGE chap. 
shy of her father- than of her mother-in-law. Among the 
Yakut the male relatives of the husband had formerly to 
avoid the young wife ; they said, " Ah, poor child, she is 
bashful." Nowadays they must avoid showing her the 
body uncovered above the elbow or the sole of the foot, 
and refrain from indecent expressions and vulgar vitupera- 
tives in her presence ; and it is a shame for her to show them 
her uncovered hair and foot.1 If sexual intercourse be- 
tween two persons related by marriage is regarded as in- 
cestuous, decency would for that reason also require 
particular reserve in their mutual relations. 
A very curious explanation of the avoidance between 
son- and mother-in-law has been given by M. Reinach. He 
says that if a man in his intercourse with his wife's family 
were allowed to see his mother-in-law, there would soon 
spring up between them such intimacy as to induce him to 
call her " mother." But if thus both the husband and wife 
called the same woman " mother " and behaved towards 
her as if they were both her children, it would appear to 
others that they were brother and sister and consequently 
had been guilty of the greatest of all crimes, incest. Now 
the savage knows only one means of avoiding such an 
appearance, and that is to have nothing at all to do with his 
mother-in-law.2 I should have imagined, however, that 
another alternative also would have occurred to him. If 
the use of the term " mother " for the mother-in-law gave 
rise to misunderstanding (I never heard that it did so among 
ourselves) he might-simply have refrained from using it. 
Some other attempts to explain the rules of avoidance 
may still be mentioned. According to Mr. Atkinson, they 
are due to jealousy. The avoidance between mother- and 
son-in-law is "a measure of protection for the marital 
rights of the husband of the former." The avoidance 
between the wife and her husband's relations applies to the 
relations in the ascending line, that is, his seniors, because 
1 Sieroshevski, in Jour. Anthr. Inst. xxxi. 93. Miss Czaplicka, Abori- 
ginal Siberia, p. in. 
2 Reinach, ' Le gendre et la belle-mere,' in L' Anthropologic, xxii. 
659 sq. XII SEXUAL MODESTY 
451 
" against his juniors he can defend himself, against his seniors 
he needs the protection of law." The avoidance between 
father- and son-in-law is a consequence of the unrestricted 
power and right which the patriarch in primitive times had 
exercised over his daughters. It originated " during that 
stage of the transition era, when, incest still lingering, the 
immigrant suitor was so far acknowledged that his entry 
into a group was not always delayed till the death of his 
proposed father-in-law. As they were thus possible rivals 
there was a chance of friction, only to be averted by the law 
in question."1 The last-mentioned explanation need not 
detain us here : it is based on Mr. Atkinson's view that the 
primitive social group consisted of a single patriarch and a 
group of wives and daughters, with all of whom he enter- 
tained sexual relations ; and this view, as will be seen in a 
subsequent chapter, is a mere construction for which there 
is no foundation in fact. Again, the suggestions that a man 
has to avoid his mother-in-law owing to the jealousy of his 
father-in-law and that a woman has to avoid her father-in- 
law owing to the jealousy of her husband might possess 
some plausibility only if it could be shown that the 
former avoidance were regularly connected with matrilocal 
marriage and the latter with patrilocal marriage ; for it is 
difficult to see why a man should be more jealous of his 
son-in-law or of his father-in-law than of other men, or 
require special precautions against them, unless they lived 
in particularly close contact with his wife. Now, for 
reasons already stated, the data relating to the connection 
between rules of avoidance and the place where the husband 
takes up his abode must necessarily be unreliable. But 
we have seen that the avoidance between son- and mother- 
in-law does occur among many tribes where the husband at 
marriage takes his wife to his own home. As for the 
avoidance between the wife and the husband's family, 
Tylor says that there are no cases of such avoidance where 
the husband lives with the wife's family.2 But, as Mr. 
Crawley remarks, there are such cases ;3 and in my own 
1 Lang and Atkinson, Social Origins and Primal Law, p. 269 sqq. 
8 Fylor, in Jour. Anthr. Inst, xviii. 247. 3 Crawley, op.cit. p. 406. 452 
THE HISTORY OF HUMAN MARRIAGE chap. 
notes I find several instances of the two kinds of avoidance 
being spoken of as co-existent among the same people.1 It 
may be added that Mr. Atkinson, in his explanations, 
ignores the avoidance between a wife and her mother-in-law. 
Professor Kohler, again, looks upon the avoidance 
between a man and his parents-in-law as a survival of a 
previous custom of group-marriage : it shows that individual 
marriage is something unusual and novel, and was formerly. 
illegitimate.2 This is quite in agreement with Kohler's ten- 
dency to interpret all sorts of customs as relics from a 
mythical group-marriage stage, for the existence of which we 
have no evidence whatever. But the most original ex- 
planation comes from Mr. Keane. He says that the strange 
reluctance of the mother-in-law to meet her son-in-law 
seems accounted for by a Patagonian practice which per- 
sisted till quite recent times. On the death of any young 
person the head of the family was required to despatch 
some aged woman, a mother-in-law by preference. Hence 
through fear of such a fate women acquired the habit of 
avoiding all contact with their sons-in-law, and the feeling 
continued after the motive had been forgotten.3 Mr. 
Keane does not tell us how it happened that the murder of 
the Patagonian mother-in-law caused a panic among 
mothers-in-law all over the world. 
In order to obtain a full and detailed explanation of the 
rules of avoidance between relatives by marriage, as they 
are found among different peoples, it would no doubt be 
necessary to examine those rules and the conditions under 
1 Mason, Ethnology of the Salinan Indians (California), p. 164. 
Strehlow, Die Aranda- und Loritja-Stamme in Zentral-Australien, 
vol. iv. pt. i. 102 (Western Loritja). Seligman, Melanesians of British 
New Guinea, p. 485 sq. (Southern Massim). Parkinson, Dreissig 
Jahre in der Sudsee, p. 67 (natives of Gazelle Peninsula, New Britain). 
Decle, Three Years in Savage Africa, p. 159 (Matabele). Torday and 
Joyce, Les Bnshongo, p. 116. Weeks, Congo Cannibals, p. 133 sq. 
(Boloki). 
2 Kohler, ' Rechtsphilosophie und Universalrechtsgeschichte,' 
in v. Holtzendorff, Enzyhlopadie der Rechtswissenschaft, i. 30. 
* Keane, Ethnology, p. 219. XII SEXUAL MODESTY 
453 
which they occur much more minutely than has been done 
hitherto. But I feel confident that they are in the main 
due to that feeling of sexual shame which a person is naturally 
apt to experience in the presence of a member of the family 
circle of his (or her) mate, as well as in the presence of a 
member of his (or her) own family circle. That there is such a 
tendency is beyond all dispute ; nor can there be any doubt 
about its origin. It must be an offshoot of that aversion 
to sexual intercourse between parents and children, brothers 
and sisters, and generally persons who have grown up in 
the same home, which is normally felt among all races of 
men. This feeling throws a veil of secrecy over the relations 
between husband and wife and imposes a taboo on sexual 
matters in all other relations inside the domestic circle. Nay, 
through the influence of early sentiments and habits, it has 
also become a motive force of sexual modesty outside that 
circle. Female coyness could never account for the peculiar 
sexual reserve between near relatives, nor do I think that it 
could account for the genuine shyness which characterises 
true modesty. As Dr. Havelock Ellis points out, it is the 
sign of sexual emotion, and it easily becomes an invitation.1 
Montaigne speaks of the "artifice" of virgin modesty.2 
" It is the most pudent girl," says Restif de la Bretonne, 
" the girl who blushes most, that is most disposed to the 
pleasures of love " ;3 and the same thought is more coarsely 
expressed in the Welsh saying, " The more prudish the 
more unchaste." 4 The man, again, is excited by the woman's 
coyness, and thus, as Stendhal puts it, modesty " is the 
mother of love." 5 At the same time he must lay restraint 
on his behaviour, he must not make the woman blush for 
shame ; but what is required of him is outward decency 
only, and it is doubtful how far female coyness could be the 
1 Ellis, op. cit. i. 45 57. 
2 Montaigne, Essays, book ii. ch. 15, vol. ii. 332. 
3 Restif de la Bretonne, quoted by Ellis, op. cit. i. 46. So also, 
according to Venette (La generation de I'homme, i. 375 sqq.}, " la 
femme timide aime plus que la hardie et 1'enjouee." 
4 Ellis, op. cit. i. 46. 5 Stendhal, De l'aniour, p. 56. 454 
THE HISTORY OF HUMAN MARRIAGE ch. xn 
cause of any deeper sexual bashfulness in a man than that 
resulting from a breach of etiquette. On the other hand, 
in the sexual shame felt in the presence of a near relative 
there is nothing pleasing or exciting when displayed by a 
woman and nothing conventional when displayed by a man. 
Why ? Because it is rooted not in love but in aversion. CHAPTER XIII 
COURTSHIP 
Wherever we are able to observe an external difference 
between the male and the female reproductive cells of plants, 
the male cell behaves actively in the union and the female 
cell passively.1 Tn this respect there is an analogy between 
plants and many of the lower animals. In the case of some 
lowly-organised animals, which are permanently affixed 
to the same spot, the male element is invariably brought to 
the female. There are other instances in which the females 
alone are fixed, and the males must be the seekers. Even 
when the males and females of a species are both free, it is 
almost always the males that first approach the females.2 
As Darwin points out, we can see the reason why, in the 
first instance, the male plays the active part :-" Even if 
the ova were detached before fertilisation, and did not 
require subsequent nourishment or protection, there would 
yet be greater difficulty in transporting them than the male 
element, because, being larger than the latter, they are 
produced in far smaller numbers."3 He adds, however, that, 
with respect to forms of which the progenitors were prim- 
ordially free, it is difficult to understand why the males 
should invariably have acquired the habit of approaching 
the females, instead of being approached by them. Perhaps 
the explanation may be that the seeker is more exposed to 
danger than the one sought after, and that the death of a male 
at the pairing time is less disadvantageous for the existence 
of the species than the death of a female. At any rate, we 
1 Sachs, Text-Book of Botany, p. 897. 
2 Darwin, Descent of Man, i. 343 sq. 3 Ibid. i. 343. 
455 456 
THE HISTORY OE HUMAX MARRIAGE chap. 
may say with Darwin that it is necessary that the males 
should be endowed with strong passions in order that they 
may be efficient seekers ; and the acquirement of such 
passions would naturally follow from the more eager males 
leaving a larger number of offspring than the less eager.1 
The sexual desire of the male is commonly considered to be 
stronger than that of the female.2 But, so far as mankind 
is concerned, Dr. Havelock Ellis thinks that there is now 
a tendency to minimise unduly the sexual impulse in women 
-as in earlier ages there was generally a tendency to credit 
them with an unduly large share of it-and that, roughly 
speaking, the distribution of the sexual impulse between the 
two sexes is fairly balanced.3 He also observes that before 
sexual union the male tends to be more ardent, whereas 
after sexual union it is the female that tends Jto be so.4 
This tendency has been noticed in some birds as well as 
among men. Mr. Selous remarks concerning gulls :-" In 
actual courting, before the sexes are paired, the male bird is 
usually the most eager, but after marriage the female often 
becomes the wooer. Of this, I have seen marked instances." 
He mentions especially the plover, kestrel hawk, and rook.5 
Sir R. Heron even states that with pea-fowl the first advances 
are always made by the female ; and something of the same 
kind takes place, according to Audubon, with the older 
females of the wild turkey.6 
In human courtship, also, the male generally plays a more 
active and the female a more passive part; the woman 
" requires to be courted." Yet, as among the lower 
animals, there are exceptions to the rule. Among not a few 
uncivilised peoples the women are said to be the courters, 
or to make proposals of marriage. 
Among the Indians of Paraguay, whose women according 
1 Ibid. i. 344. 
2 See, e.g., Burdach, Die Physiologic als Dvfahrungswissenschajt, 
i. 409 ; Tillier, L'instinct sexuel chez I'homme et chez les animaux, 
p. 71 sq. ; Idem, Le mariage, p. 173. 
3 Ellis, Studies in the Psychology of Sex (vol. iii.), Analysis of the 
Sexual Impulse, &c., pp. 179, 203. 4 Ibid. p. 193 sq. 
" Selous, Bird Watching, p. 112. 
G Darwin, op. cit. ii. 134. Xin COURTSHIP 
457 
to Rengger were endowed with stronger passions than the 
men,1 they were also allowed to make proposals.2 This is 
still the case among the Choroti of the Gran Chaco. During 
the algaroba season, when marriages are usually contracted 
among them, those young men and women who want to 
marry wander about from village to village. In each village 
dances are performed on four successive evenings. " Free 
love is connected with these dances and the girls play the 
most active part in them although they do not actively en- 
gage in the entertainment. After each dance, and par- 
ticularly after the last one, the avitsye, each girl may choose 
herself a temporary lover for the night. The avusye which 
is held on the fourth evening is supposed to be decisive : 
those girls who have found men whom they like and want to 
marry, then keep them, not only for that night but for ever. 
After this the young man is taken to the house and village 
of the girl, where he thenceforwards remains as her husband 
and becomes a member of his wife's family."3 Of the 
Sotegaraik or Tapieti, on the Pilcomayo, a recent 
German explorer writes :-" The women and girls dance 
behind the men, now separately, now in groups. On this 
occasion the girls also choose their husbands, quickly 
removing the spouse elect from the row of the dancers and 
then disappearing with him. There is no marriage cere- 
mony."4 Among the Bororo Indians of Matto Grosso the 
proposal of marriage likewise comes from the' woman. 
" The young man sits in the bahito (the men's house), the 
young girl sends him mingan of maize ; if he accepts, 
he is a bridegroom." But if he eats only a portion or does 
not eat any at all, he thereby signifies that he cither wishes to 
postpone his decision or rejects the proposal outright.5 
Among the Tarahumare of Mexico, says Dr. Lumholtz, 
1 Rengger, N aturgeschichte der Sdugethiere von Paraguay, p. n. 
2 Moore, Marriage Customs, p. 261. 
3 Karsten, Indian Dances in the Gran Chaco, p. 30. See also 
Nordenskiold, Indianliv i El Gran Chaco, pp. 79, 82, 84. 
4 Herrmann, ' Die ethnographischen Ergebiiisse der Deutschen 
Pilcomayo-Expedition,' in Zeitschr. f. Ethnol. xl. 129. 
5 Fric and Radin, ' Contributions to the Study of the Bororo 
Indians,' in Jour. Anthr. Inst, xxxvi. 390. 458 
THE HISTORY OF HUMAN MARRIAGE chap. 
" the custom of the country requires the girl to do 
all the courting. She is just as bashful as the young 
swain whom she wishes to fascinate, but she has to take the 
initiative in love affairs. The young people meet only at the 
feasts, and after she has got mildly under the influence of the 
native beer that is liberally consumed by all, she tries to 
attract his attention by dancing before him in a clumsy 
way, up and down on the same spot. But so bashful is she 
that she persistently keeps her back towards him. ... If 
occasion requires, the parents of the girl may say to the 
parents of the boy, ' Our daughter wants to marry your 
son.' Then they send the girl to the boy's home, that the 
young people may become acquainted. For two or three 
days, perhaps, they do not speak to each other, but finally 
she playfully begins to throw pebbles at him. If he does not 
return them, she understands that he does not care for her. 
If he throws them back at her, she knows that she has won 
him. She lets her blanket drop and runs off into the wood, 
and he is not long in following her." Sometimes a boy who 
very much likes a certain girl may make the first advances, 
but even then he has to wait until she throws the first pebbles 
and drops the blanket; for among these Indians " it is the 
woman who seeks the man, and the fair who deserve the 
brave."1 So also among the Pueblos " the usual order of 
courtship is reversed ; when a girl is disposed to marry she 
does not wait for a young man to propose to her, but selects 
one to her own liking and consults her father, who visits 
the parents of the youth and acquaints them with his 
daughter's wishes. It seldom happens that any objections 
to the match are made."2 Among the Indians of the 
interior of British Columbia " a dance was held at certain 
seasons to permit the young men and women to choose life 
1 Lumholtz, Unknown Mexico, i. 267 sq. Of another Mexican 
tribe, the Huichol, the same traveller says (ibid. ii. 93) that among 
the young people the man seeks the woman but among the persons 
of more mature age the woman seeks the man. * 
2 Bancroft, Native Races of the Pacific States, i. 547 sq. See also 
Schoolcraft, Archives of Aboriginal Knowledge, iv. 86 (Moqui) ; 
Solberg, ' Gebrauche der Mittelmesa-Hopi (Moqui) bei Namengebung, 
Heirat und Tod,' in Zeitschr. f. Ethvol. xxxvii. 629. XIII COURTSHIP 
459 
partners. A young man might choose a wife at a certain 
period of the dance by taking hold of her belt or by merely 
laying his hand upon her. If accepted, she allowed him to 
dance with her to the even, when the couple were led out 
before the people by the chief and proclaimed man and wife. 
A girl could choose a man in the same way."1 Among the 
Thompson Indians the girls generally touched the young 
men on either the head or the arm The man usually took 
to wife the girl who had touched him, but he was not com- 
pelled to do so ; and we are told that some girls who touched 
a man and were not accepted felt so ashamed that they 
committed suicide.2 
Mr. Batchelor says that it occurs among the Ainu that the 
young woman does the wooing and courting, in which case 
she gets the mother and father to call upon the parents of her 
choice; and in such a case, if the proposal is accepted, the 
bridegroom is removed from his own family to take up his 
abode close to the hut of his father-in-law.3 But Pilsudski 
states that though the custom of women trying to win men 
did formerly exist, it is no longer observed, and that Ainu 
women " are even very much displeased at any hint of such 
a thing." Their mythology ascribes this custom to Tungus 
women.4 In Borneo, also, there is a tribe who is accused 
by its neighbours of practising such a custom : it is a 
common statement made by the Kayans that " among the 
Kalabits the initiative in all love-making is taken by the 
women."5 
Among the Western Islanders of Torres Straits " it was 
the custom for the young women to propose marriage to the 
young men, but this has been interdicted by the mission- 
aries." Dr. Haddon believes that young men never made 
1 Teit, ' Indian Tribes of the Interior,' in Canada and its Provinces, 
vol. xxi. The Pacific Province, pt. i. 309. See also Idem, ' Lillooet 
Indians,' in Publications of the Jesup North Pacific Expedition, 
ii. 268. 
2 I clem, ' Thompson Indians of British Columbia,' in Publications 
of the Jesup North Pacific Expedition, i. 324. 
3 Batchelor, Ainu and their Folk-Lore, p. 223 sqq. 
4 Miss Czaplicka, Aboriginal Siberia, p. 103. 
6 Hose and McDougall, Pagan Tribes of Borneo, ii. 176, 246 sq. 460 
THE HISTORY OF HUMAN MARRIAGE chap. 
the first definite proposal to matrimony, although by 
scenting themselves they might indicate that they were 
ready to be sued. The custom is mentioned in the folk- 
tales.1 In various tribes of New Guinea, also, the marriage 
proposal or initiative generally comes from the girl.2 Among 
the Sulka, in the central part of New Britain, when a girl 
has made up her mind as to the young man she wants to 
marry, she mentions it to her father or some other near 
relative, who then goes to the man and proposes on her 
behalf.3 In many places in New Hanover and Northern 
New Ireland it is likewise the girl who takes the initia- 
tive and sends a go-between to the bridegroom elect to 
inform him about her choice.4 Among the Maori, when the 
young people gathered together at night in the play-houses, 
they made advances to each other, and these advances often 
came from the girls, " the recognised sign being a pinch, or 
the scratching of the finger-tip on the hand of the desired 
person." 5 Mr. Tregear even states that " in almost every 
case the first advances were made by the woman, either 
directly or through one of her friends " ; and in old Maoii 
legends girls are represented as quarrelling about men.6 
Among the Garos of Assam it is not only the privilege 
but even the duty of the girl to speak first. "If a male 
makes advances to a girl," says Dalton, " and the latter, 
rejecting them, chooses also to tell her friends that such 
tenders of affection have been made to her, it is looked on 
as an insult to the whole ' mahari ' (motherhood) to which 
1 Haddon, in Reports of the Cambridge Anthropological Expedition 
to Torres Straits, v. 222. 
2 Seligman, Melanesians of British New Guinea, pp. 76, 499, 709. 
Malinowski, ' Natives of Mailu,' in Trans. Roy. Soc. South 
A ustralia, xxxix. 560. Guise, ' Tribes inhabiting the Mouth of the 
Wanigela River, New Guinea,' in Jour. Anthr. Inst, xxviii. 209. 
3 Parkinson, Dreissig Jahre in der Sildsee, p. 177. On some islands 
at the south-west coast of New Britain girls occasionally propose 
to men (Vogel, Eine Forschungsreise im Bismarck-Archipel, p. 223). 
4 Parkinson, op. cit. p. 267. 
5 Best, ' Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 33. 
6 Tregear, The Maori Race, p. 285 sq. See also Colenso, Maori 
Races of New Zealand, p- 25. XIII COURTSHIP 
461 
the girl belongs, a stain only to be obliterated by the 
blood of pigs, and liberal libations of beer at the expense 
of the ' mahari ' to which the man belongs." 1 The only 
exception to this rule occurs when one daughter of a family is 
given in marriage to the son of her father's sister.2 
Among the Wangoni, a Zulu tribe, women have the right 
to propose. When a girl has had her puberty feast she goes 
to the house of the man she desires to marry, accompanied 
by her girl friends, all carrying green branches and singing, 
and declares in songs that he is her bridegroom elect. If the 
man shows himself unwilling to accept the proposal, they 
all go back, crying aloud ; whereas, if he accepts it, they 
shout for joy and take the girl, who is now looked upon as a 
bride, to her family. Next day the man pays a visit to the 
girl's father, and begins to negotiate with him for the bride 
price, which is no easy matter to settle. This custom, 
however, is nowadays on the verge of disappearing, as a 
father is generally inclined to sell his daughter to the man 
who sends him the largest number of oxen.3 Among the 
Konde people in " German East Africa " the daughters of 
chiefs are said to have the right to make proposals of 
marriage.4 Among the Bantu Kavirondo, if a girl is not 
asked in marriage, she may go off and offer herself to a 
man of another village, who, if he accepts her, has to 
negotiate for the payment of a marriage gift;5 and among 
the Nilotic Kavirondo, if a woman remains unmarried for a 
longer period than usual, she often goes off to a chief or 
wealthy man and states that she has come to stay and cook 
for him, in which case he generally takes her to wife, paying 
for her only a small present.6 Among some peoples the 
1 Dalton, Descriptive Ethnology of Bengal, p. 64. Cf. ibid. pp. 142, 
233 (Bhuiyas, Muasis); Playfair, Garos, p. 67. 
2 Playfair, op. cit. p. 68. 
3 Wiese, ' Bcitrage zur Geschichte der Zulu im Norden les 
Zambesi, namentlich der Angoni,' in Zeitschr.f. Ethnologic, xxxii. 191. 
Cf. Shooter, Kafirs of Natal, p. 52 ; Tyler, Forty Years among the 
Zulus, p. 200 sq. 
1 Fulleborn, Das Deutsche Niassa- iind Ruwuma-Gebiet, p. 346. 
Johnston, Fganda Protectorate, p. 748. Hobley, Eastern 
Uganda, p iS. 6 Johnston, op. cit. p. 791. Hobley, op. cit. p. 29. THE HISTORY OF HUMAN MARRIAGE chap. 
462 
father or parents of a girl take the initiative and choose 
a husband for the daughter.1 
The male very frequently fights with other males for the 
possession of the female. Thus, during the season of love, 
the males even of the most timid animal species engage 
in desperate combats. This fighting for a female occurs 
among insects,2 and is extremely prevalent in the order of the 
Vertebrata. Among the mammals lower than man there 
is certainly a normal co-existence of combat and courtship ; 
and there can be no doubt that our primeval human ancestors 
also fought for the possession of females. Fights of this 
kind seem even now to be of frequent occurrence among 
various savage peoples. 
Azara tells us that among the Guanas on the left bank of 
the Upper Paraguay the men often refrain from marrying 
until they are twenty years old or more, as before that age 
they cannot defeat their rivals.3 In his description of 
Brazilian savages, von Martius states that among the Muras 
on the Madeira the wives are mostly gained in a combat with 
fists between all the lovers of the same girl, and that the like 
is the case with the Passes 4 Among the Patwin of California 
it occasionally happened that men who had a quarrel about 
a woman fought a duel with bows and arrows at long 
distances.5 Speaking of the Northern Indians, Hearne 
states that " it has ever been the custom among those 
people for the men to wrestle for any woman to whom they 
are attached ; and, of course, the strongest party always 
carries off the prize. A weak man, unless he be a good hunter 
and well-beloved, is seldom permitted to keep a wife that 
a stronger man thinks worth his notice. . . . This custom 
1 Appun, ' Die Indianer von Britisch-Guayana, ' in Das Ausland, 
xliv. 124 (Arawaks). Coudreau, Chez nos Indiens, p. 127 (Rou- 
couyennes, of French Guiana). Hrdlicka, ' Notes on the Indians of 
Sonora, Mexico,' in American Anthropologist, N.S. vi. 69 (Yaqui). 
Teit, ' Thompson Indians,' in Publications of the Jesup North 
Pacific Expedition, i. 322. Kropf, Das Volk der Xosa-Kaffern, p. 132. 
2 Darwin, op. cit. i. 459, 501. 
3 Azara, Voyages dans V Amerique meridionale, ii. 94. 
4 v. Martius, Beitrage zur Ethnographic Anierika's, i. 412, 509. 
5 Powers, Tribes of California, p. 221 sq. XIII COURTSHIP 
463 
prevails throughout all their tribes, and causes a great spirit 
of emulation among their youth, who are upon all occasions, 
from their childhood, trying their strength and skill in 
wrestling."1 Richardson also saw, more than once, a 
stronger man assert his right to take the wife of a weaker 
countryman. " Any one," he says, " may challenge another 
to wrestle, and, if he overcomes, may carry off his wife as the 
prize. . . . The bereaved husband meets his loss with the 
resignation which custom prescribes in such a case, and seeks 
his revenge by taking the wife of another man weaker than 
himself."2 With reference to the Slave Indians, Mr. 
Hooper says :-" If a man desire to despoil his neighbour of 
his wife, a trial of strength of a curious nature ensues : they 
seize each other by the hair, which is worn long and flowing, 
and thus strive for the mastery, until one or another cries 
peccavi. Should the victor be the envious man, he has to 
pay a certain number of skins for the husband-changing 
woman."3 Among the Eskimo about Bering Strait Mr. 
Nelson was told by an old man that in ancient times, when 
a husband and a lover quarrelled about a woman, they were 
disarmed by the neighbours and then settled the trouble 
with their fists or by wrestling, the victor in the struggle 
taking the woman.4 
In the islands outside Kamchatka, if a husband found that 
a rival had been with his wife, he would admit that the rival 
had at least an equal claim to her. " Let us try, then," he 
would say, " which of us has the greater right, and shall have 
her." After that they would take off their clothes and begin 
to beat each other's backs with sticks, and he who first fell 
to the ground unable to bear any more blows lost his right 
to the woman.5 Among the Yukaghir, when a young man 
1 Hearne, Journey from Prince of Wales's Fori to the Northern 
Ocean, p. 104 sq. 
2 Richardson, Arctic Searching Expedition, ii. 24 sq. Cf. Mac- 
kenzie, Voyages from Montreal to the Frozen and Pacific Oceans, 
p. 145 ; Ross, ' Eastern Tinneh,' in Smithsonian Report, 1866, p. 310. 
3 Hooper, Ten Months among the Tents of the Tuski, p. 303. 
4 Nelson, ' Eskimo about Bering Strait,' in Ann. Rep. Bur. Ethnol. 
xviii. 292. Cf. Nansen, First Crossing of Greenland, ii. 319. 
■* Steller, Beschreibung von dem Lande Kamtschatka, p. 348. 464 
THE HISTORY OF HUMAN MARRIAGE chap. 
finds a rival in the tent of the girl he visits at night, he com- 
pels him to come out and fight. The vanquished one goes 
off home, and the conqueror re-enters the tent. In olden 
times, according to the tales of the Yukaghir, rivals used to 
go away from the tents and fight until one or the other was 
killed.1 Romantic marriages in which a young man obtains 
his bride, not through serving for her, but through braving 
dangers and overcoming enemies and competitors, are very 
popular in the tales of the Chukchee, and occur even at the 
present day among them. In one story we are told of a girl 
who caused her suitors to run foot-races with her and took in 
marriage the one who succeeded in distancing her ; and 
sometimes a series of contests is said to take place before 
a man succeeds in obtaining his wife.2 
Fighting for women is of frequent occurrence among the 
Australian natives. Speaking of those near Herbert Vale 
in Northern Queensland, Dr. Lumholtz says that " if a 
woman is good-looking, all the men want her, and the one 
who is most influential, or who is the strongest, is accordingly 
generally the victor." Hence the majority of the young 
men must wait a long time before they get wives, as they 
have not the courage to fight the requisite duel for one with 
an older man.3 Narcisse Peltier, who during seventeen 
years was detained by a tribe of Queensland Australians, 
states that the men " not unfrequently fight with spears for 
the possession of a woman." 4 In the tribes of Western 
Victoria described by Mr. Dawson a young chief who cannot 
get a wife, and falls in love with one belonging to a chief who 
has more than two, can, with her consent, challenge the 
husband to single combat, and if the husband is defeated, 
the conqueror makes her his legal wife.5 " On the Hunter," 
says Mr. Ridley, " when a man seeks a wife he goes to a camp 
where men and women are sitting together round a fire, and 
throws in a boomerang. If one of the men throws back a 
1 Jochclson, Yukaghir, p. 63. 2 Bogoras, Chukchee, p. 579 sqq. 
3 Lumholtz, Among Cannibals, pp. 213, 1S4. 
4 Spencer, Principles; of Sociology, i. 601. 
5 Dawson, Australian Aborigines, p. 36. Cf. Ridley, Aborigines 
of A ustralia, p. 6, XIII COURTSHIP 
465 
boomerang at him he has to fight for the privilege sought ; 
but if no one challenges him, he quickly steps in and takes 
one of the young women for his wife."1 Among the abori- 
gines of New South Wales, says Dr. Fraser, it happens that 
two rivals fight for a girl, and that she is the prize of victory.2 
Among the Kiwai Papuans of British New Guinea, 
according to Dr. Landtman, most fighting inside a tribe is 
occasioned by the fair sex ; and in their tales about a 
mythical hero, by name Sido-the first man who was killed 
by another-the two men are, characteristically enough, 
represented as fighting for a woman.3 The principal causes 
of war among the Maori are well put in one of their aphorisms, 
" By women and land are men lost." In ancient days 
women were the causes of many sanguinary wars among 
them, although they were not an exogamous people.4 We 
are told that if a Maori lady had two or more lovers, a feud 
might take place between them, assisted by their relatives, 
and the conqueror seized the lady, who often fared badly 
in the scuffle.5 In Samoa and Fiji, also, women were one of 
the chief causes of fighting ; 6 and of the natives of Makin, 
of the Gilbert Group, it is said that " they have no wars, and 
very few arms, and seldom quarrel except about their 
women."7 
In some of the Bushman communities in South Africa 
" a youth who desired to take to himself a girl was obliged 
to prove himself a man by fighting with others for her and 
winning her by victory."8 Among the Mountain Damara, 
" if one person feels himself stronger than another he will 
1 Ridley, Kamilaroi, Dippil, and Turrubul, p. 157. 
2 Fraser, Aborigines of New South Wales, p. 26. 
3 Landtman, Nya Guinea farden, p. 81. 
4 Best, ' Notes on the Art of War, as conducted by the Maori of 
New Zealand,' in Jour. Polynesian Soc. xi. 13. 
Polack, Manners and Customs of the New Zealanders, i. 139 sqq. 
See also Colenso, op. cit. p. 25 sq. 
6 Pritchard, Polynesian Reminiscences, pp. 55, 269. 
7 Wilkes, Narrative of the United States Exploring Expedition, 
v. 72. 
8 Theal, Yellow and Dark-skinned People of Africa south of the 
Zambesi, p. 47. Cf. Lichtenstein, Travels in Southern Africa, ii. 48. 
VOL. 1 H II 466 
THE HISTORY OE HUMAN MARRIAGE chap. 
consider himself perfectly at liberty to take away by force 
any female to whom he takes a fancy."1 The chief cause of 
quarrels and fights on the Congo, says Mr. Weeks, is about 
women ; "if you push the matter to its real origin you will 
in nine cases out of ten find a woman at the bottom of it."2 
The people of Wadai are notorious for their desperate fights 
for women ; and among the young men of Baghirmi bloody 
feuds between rivals are far from rare.3 
In other cases the rivalry of the suitors takes less violent 
and more playful forms. Among the Maori, if a girl had 
two suitors with equal pretensions, a kind of " pulling- 
match," called he puna rua, was arranged in which the girl's 
arms were dragged by each of them in opposite directions, 
the stronger man being the victor.4 They also had the 
following custom, known as kai tamahine. " A party of 
young, active, and presentable men would form themselves 
into a party and go on a visit to some village where resided a 
young woman noted for her good looks and qualities. The 
visit was for the express purpose of showing themselves and 
their accomplishments to the girl, in the hope that she would 
accept one of them as a husband. The period of the visit 
would be quite a gay time, for the party of young men 
would give performances of various kinds, in order to exhibit 
their skill, grace, dexterity, and so forth, each endeavouring 
to excel his companions. They would perform haka, or 
posture dances, of various kinds, and play games of skill. 
Each would hope that the girl would select him as a hus- 
band."5 Among the Dongolowees of Kordofan, as we are 
informed by Dr. Felkin, if two men are suitors for a girl, 
and there is a difficulty in deciding between the rivals, the 
following method is adopted. The fair lady has a knife tied 
to each forearm, so fixed that the blade of the knife projects 
1 Stow, Native Races of South Africa, p. 260. 
2 Weeks, Among Congo Cannibals, p. 226. 
3 Barth, Reisen und Entdeckungen in Nord- und Central-Afrika, 
i"- 352. . 
1 Dieffenbach, Travels in New Zealand, ii. 36 sq. Taylor, Te Ika a 
Maui, p. 337. 
5 Best, Maori Marriage Customs,' in Trans, and Proceed. New 
Zealand Institute, xxxvi. 36. XIII COURTSHIP 
467 
below the elbow. She then takes up a position on a log of 
wood, the young men sitting on either side with their legs 
closely pressed against hers. Raising her arms, the girl 
leans forward, and slowly presses the knives into the thighs of 
her would-be husbands. The suitor who best undergoes this 
trial of endurance wins the bride, whose first duty after 
marriage is to dress the wounds she has herself inflicted.1 
The Kirghiz have a race called the " love chase," which may 
be considered a part of the form of marriage among them. 
" In this the bride, armed with a formidable whip, mounts 
a fleet horse, and is pursued by all the young men who make 
any pretensions to her hand. She will be given as a prize to 
the one who catches her, but she has the right, besides 
urging on her horse to the utmost, to use her whip, often 
with no mean force, to keep off those lovers who are unwel- 
come to her, and she will probably favour the one whom she 
has already chosen in her heart. As, however, by Kirghiz 
custom, a suitor to the hand of a maiden is obliged to give 
a certain kalym, or purchase-money, and an agreement must 
be made with the father for the amount of dowry which he 
gives his daughter, the ' Love Chase ' is a mere matter of 
form."2 Among some of the Pepos of Formosa " there 
was a custom of holding, on a certain day specially announced, 
a running race in which all young bachelors competed. The 
prize was the privilege of marrying the most beautiful girl 
of the tribe."3 
In ancient India it was a custom in royal circles when a 
princess became marriageable for a tournament to be held, 
and the victor was chosen by the princess as her husband. 
1 Wilson and Felkin, Uganda and the Egyptian Soudan, ii. 310. 
2 Schuyler, Turkistan, i. 42 sq. Valikhanof (Russians in Central 
Asia, p. 84) mentions a Kirghiz game in which the youth who 
finds most favour in the eyes of one of the girls has to perform some 
dexterous feat or sing a song ; if his performance is creditable the 
girl rewards him with a hearty kiss. 
3 Davidson, Island of Formosa, p. 581 sq. According to Dr. 
Muller (' Ueber die Wildenstamme der Insel Formosa,' in Zeitschr. f. 
Ethnol. xlii. 230), the form of marriage among some Formosa savages 
consists " in einem Wettbewerb zwischen den Brautigam und einen 
anderen (fingierten) Bewerber." 468 
THE HISTORY OF HUMAN MARRIAGE chap. 
This custom, known as the Swayamvara or " Maiden's 
Choice," is often mentioned in the legends.1 So also we meet 
with several instances of emulation for women in Greek 
legends and myths. Pausanias tells us that Danaus estab- 
lished a race for his daughters, and that " he that outran 
all the rest was to have the first choice, and take her whom he 
most approved ; he that was next in order was to have the 
second choice, and so on to the last; and those who had 
no suitors were ordered to wait till new ones came to the 
course." Icarus likewise proposed a race for the suitors of 
Penelope.2 And according to Pindar, Antaeus, father of a fair- 
haired and greatly-praised daughter, who had many suitors, 
stationed the whole company of them at the end of the race- 
course, saying that he should have her for his bride who 
should prove foremost in the race and first touch her 
garments.3 
On Palm Sunday, the day for presentiments of love, South 
•Slavonian youths wrestle with each other, believing that 
he who proves the stronger will get the prettier wife.4 
Arthur Young informs us of the following strange custom 
which prevailed in the interior of Ireland in his time :- 
" There is a very ancient custom here," he says, " for a 
number of country neighbours among the poor people to 
fix upon some young woman that ought, as they think, to 
be married ; they also agree upon a young fellow as a proper 
husband for her ; this determined, they send to the fair one's 
cabin to inform her that on the Sunday following ' she is to 
be horsed,' that is, carried on men's backs. She must then 
provide whisky and cider for a treat, as all will pay her a 
visit after mass for a hurling match. As soon as she is 
horsed the hurling begins, in which the young fellow ap- 
pointed for her husband has the eyes of all the company 
fixed on him : if he comes off conqueror, he is certainly 
married to the girl; but if another is victorious, he as 
certainly loses her, for she is the prize of the victor. . . . 
1 Samuelson, India, Past and Present, p. 48. 
2 Pausanias, Descriptio Graeciae, iii. 12. 1 sq. 
:i Pindar, Pythia, ix. 117. 
* Krauss, Sitte und Brauch der Sudslaven, p. 163 sq. XIII COURTSHIP 
469 
Sometimes one barony hurls against another, but a marriage- 
able girl is always the prize."1 
A suitor may also be anxious to display his manliness in 
the presence of his girl without any emulation with rivals. 
Pitre tells us that in Sicily " the youth who wishes to marry 
seeks to give some public proof of his valour and to show 
himself off. In Chiaramonte, not to speak of other districts, 
in evidence of his virile force, he bears in a procession the 
standard of some confraternity, a very high and richly 
adorned standard which makes its staff bend to a semicircle, 
of such enormous weight that the bearer must walk in a 
terribly bent position, his head and loins thrown back and his 
feet forward. On reaching the house of his betrothed he 
makes a great many proofs of his boldness and skill in 
wielding this accursed standard, which in that moment 
seems a plaything in his hands, but may yet prove fatal 
to somebody."2 
Among savages, as among ourselves, a young man tries 
by various means to gain the favour of the girl he desires 
to marry. Dr. Eylmann states that the courtship of a South 
Australian black does not fundamentally differ from that of 
a white man ; he, too, expresses his love by soft glances, a 
pleasant smile, and courtesies of various kinds.3 Among 
the Gila of Northern Mexico, " when a young man sees a girl 
whom he desires for a wife, he first endeavours to gain the 
good-will of the parents ; this accomplished, he proceeds 
to serenade his lady-love, and will often sit for hours, day 
after day, near her house, playing on his flute. Should the 
girl not appear it is a sign she rejects him; but if, on the 
other hand, she comes out to meet him, he knows that his 
suit is accepted, and he takes her to his house." 4 
In primitive human courtship, as Dr. Havelock Ellis 
observes, music frequently plays a very considerable, 
though not usually the sole, part, being generally found as the 
1 Young, ' Tour in Ireland,' in Pinkerton, Collection of Voyages 
and Travels, iii. 860. 
3. Pitre, Usi e costumi credenze e pregiudizi del popolo siciliano, ii. 24. 
3 Eylmann, Die Eingebovenen der Kolonie Sudaustr alien, p. 129. 
4 Bancroft, op. cit. i. 549. Cf. Fletcher and La Flesche, * Omaha 
Tribe,' in Ann. Rep. Bur. Ethnol. xxvii. 319. 470 
THE HISTORY OF HUMAN MARRIAGE chap. 
accompaniment of the song and the dance at erotic festivals.1 
Dancing is a frequent means of sexual excitement and 
method of courtship.2 Among the Hos in Chota Nagpur 
" it is at the village dances that most of the courting takes 
place."3 In the Mortlock Islands the young men who wish 
to marry saunter about in the villages neat and trim ; at 
night there is singing and dancing, and most marriages 
are concluded on such occasions.4 Among the Western 
Islanders of Torres Straits, " during the numerous and pro- 
longed dances of former days, the young women watched the 
active movements of the capering youths, admired their 
glossy skin, their frizzly hair, their numerous gay ornaments, 
and took delight in their wonderful activity. . . . Pre- 
eminence had its reward, for, as the former chief of Mabuiag 
put it-' In England, if a man has plenty of money, 
women want to marry him ; so here, if a man dances well, 
they want him too.' " " It is no wonder," says Dr. Haddon, 
" when the natives described to me their courtship customs 
that they generally began with a reference to dancing."5 
1 Ellis, op. cit. (vol. iv.) Sexual Selection in Man, p. 130. 
2 Ibid. (vol. iii.) Analysis of the Sexual Impulse, &c., p. 34 sqq. 
Klemm, Allgemeine Cultur-Geschichte der Menschheit, i. 257 sqq. 
Kulischer, ' Die geschlechtliche Zuchtwahl bei den Menschen in der 
Urzeit/ in Zeitschr. f. Ethnol. viii. 143 sqq. Wallaschek, Primitive 
Music, p. 199 sqq. Idem, Anfange der Tonkunst, p. 226 sqq. Hirn, 
Origins of Art, p. 233 sq. Groos, Play of Man, p. 273. Stoll, 
Das Geschlechtsleben in der Volkerpsychologie, p. 580 sqq. Whiffen, 
North-West Amazons, p. 198 n. 2. Ehrenreich, ' Ueber die Boto- 
cudos,' in Zeitschr. f. Ethnol. xix. 33. Powers, Tribes of California, 
p. 57. Teit, quoted supra, p. 458 sq. (tribes of the interior of British 
Columbia). Idem, ' Thompson Indians,' in Publications of the fesup 
North Pacific Expedition, i. 324. Geiseler, Die Oster-Insel, p. 41 
(Easter Islanders). Turner, Samoa, p. 95. Waitz-Gerland, Anthro- 
pologic der Naturvolker, vi. 754 (Australian aborigines). Strehlow, 
Die Aranda- und Loritja-Stamme in Zentr al-Austr alien, vol. iv. 
pt. i. 94. Bonwick, Daily Life and Origin of the Tasmanians, p. 38. 
v. Rosen, Trdskfolket, p. 272 sq. (Batwa in North-Eastern Rhodesia). 
See also infra, i. 501 sq. 3 Bradley-Birt, Chota Nagpore, p. 102. 
4 Kubary, ' Die Bewohner der Mortlock Inseln,' in Mittheil. 
Geograph. Gesellsch, Hamburg, 1878-79, p. 252. 
5 Haddon, Head Hunters, p. 158. Idem, in Reports of the Cam- 
bridge Anthropological Expedition to Torres Straits, v. 222. xin COURTSHIP 
471 
Similar statements are made with reference to the natives 
of Kaiser Wilhelm Land, among whom the young men 
carefully study the art of dancing and not infrequently for 
this purpose make long journeys to other tribes which are 
famous for their skill in dancing.1 
It is not in the human species alone that we find courtship 
consisting in a more or less prolonged peaceful activity on the 
part of the male. In many species of moths the males 
assemble round the freshly emerged female, which sits 
apparently motionless while the little crowd of suitors buzz 
around her for several minutes. Suddenly, and, as far as one 
can see, without any sign from the female, one of the males 
pairs with her and all the others immediately disappear. 
In these cases the males do not fight or struggle in any way.2 
In spiders of the family Attidce " the males vie with each 
other in making an elaborate display, not only of their 
grace and agility but also of their beauty, before the females, 
and . . . the females, after attentively watching the dances 
and tournaments which have been executed for their grati- 
fication, select for their mates the males that they find most 
pleasing."3 Among birds there is courtship by means of 
antics or dances, the display of unusual or beautiful forms or 
colours, and songs or love-notes. The male of the Argus 
pheasant, noted for the extreme beauty of his plumage, 
shows himself off before the female by erecting his tail and 
expanding his magnificent wing-feathers into a great, 
almost upright, circular fan or shield, which is carried in 
front of the body.4 Mr. Belt has seen the female of Florisuga 
mcllivora sitting quietly on a branch, and two males display- 
ing their charms in front of her :-" One would shoot up like 
a rocket, then suddenly expanding the snow-white tail like 
an inverted parachute, slowly descend in front of her, turning 
1 Neuhauss, Deutsch Neu-Guinea, i. 389 sq. 
3 Poulton, Colours of Animals, p. 291. 
3 Peckham, ' Observations on Sexual Selection in Spiders of 
the Family Attidae,' in Occasional Papers of the Natural History 
Society of Wisconsin, i. 60. 
4 Darwin, op. cit. ii. 99 sq. Cf. Forbes, A Naturalist's Wanderings 
in the Eastern Archipelago, p. 131. 472 
THE HISTORY OF HUMAN MARRIAGE chap. 
round gradually to show off both back and front. . . . The 
expanded white tail covered more space than all the rest of 
the bird, and was evidently the grand feature in the per- 
formance."1 The bullfinch " makes his advances in front 
of the female, and then puffs out his breast, so that many 
more of the crimson feathers are seen at once than otherwise 
would be the case "; whilst the male goldfinch, when courting 
the female, " sways his body from side to side, and quickly 
turns his slightly expanded wings first to one side, then to 
other, with a golden flashing effect."2 Among mammals 
outside the human species, however, courtship of this 
peaceful kind is hardly known Darwin says that proof 
is wanting that the males of mammals make any effort to 
display their charms to the female ;3 but as exceptions to 
this rule are mentioned the males of some baboon species, 
who are said to turn their coloured surfaces to the females.4 
So also there is very little that deserves to be called vocal 
art in the courtship of mammals, most of them confining 
their acoustic demonstrations to a passionate howl, roar, 
shriek, or growl, or to the simple call. But here again some 
monkeys form a notable exception : the howling apes collect 
in companies and frequently give concerts that last for 
hours.5 Thus we may say that the phenomena of court- 
ship in mankind more closely resemble those found among 
birds than those of other mammals. 
Where the male is the courter, as is generally the case, 
the female on her part does not remain completely passive. 
She may in fact display considerable activity in the pro- 
ceedings which form the prelude to the act of pairing. 
Speaking of the courting of some eider-ducks, Mr. Selous 
observes that " the females seemed themselves almost as 
active agents in the sport of being wooed as were their 
lovers in wooing them."6 And in the case of the stone- 
curlew, or great plover, the dancing, which is often 
1 Belt, The Naturalist in Nicaragua, p. 112. 
2 Darwin, op. cit. ii. 104 sq. 
3 Ibid. ii. 320. 4 Groos, Die Spiele der Tiere, p. 285. 
6 Ibid. p. 289 sqq. See also Darwin, op. cit. ii. 299. 
6 Selous, op. cit. p. 144. XIII COURTSHIP 
473 
immediately followed by intercourse, is common to both the 
male and the female.1 So also among savage peoples both 
sexes often take part in their erotic dances.2 In various 
animal species the female invites the male's approach, by 
call-notes or otherwise,3 although the invitation be followed 
by that resistance or fleeing which on the part of the female 
is a regular and most important feature of the performance 
of courtship. The doe, in her pairing-time, calls to the buck 
in clear tones that bring him to her side at once ; then 
she, " half in coyness, half in mischief, takes to flight at his 
eager approach, makes toward an open space, and runs in a 
circle. The buck naturally follows, and the chase grows hot 
and as exciting as a race of horses on a track. To the frequent 
high calls of the fleeing doe are added the deep, short cries 
of the panting buck ; but suddenly the roguish doe disap- 
pears like a nymph into the thicket near at hand, and the 
baffled buck stands with head erect and ears thrown forward ; 
then we see his head lowered as he catches the scent, and he 
too vanishes in the wood."4 Alternate seeking and fleeing 
are exceedingly common among birds. Thus " the female 
cuckoo answers the call of her mate with an alluring laugh 
that excites him to the utmost, but it is long before she gives 
herself up to him. A mad chase through tree tops ensues, 
during which she constantly incites him with that mocking 
call, till the poor fellow is fairly driven crazy."5 The female 
kingfisher often torments her devoted lover for half a day, 
coming and calling him, and then taking to flight. " But she 
never lets him out of her sight the while, looking back as she 
flies and measuring her speed, and wheeling back when he 
suddenly gives up the pursuit."6 
1 Ibid. p. 18 sqq. 
2 See, e.g., Wallaschek, Primitive Music, p. 213 ; Idem, Anfdnge 
der Tonkunst, p. 239 sq. 
3 Hacker, Der Gesang der Vogel, p. 49 sqq. 
4 Muller, quoted by Groos, Play of Animals, p. 284 sq. ; Idem, 
Die Spiele der Tiere, p. 306 sq. 
6 Idem, Play of Animals, p. 285. Idem, Die Spiele der Tiere, 
p. 307 sq. 
6 Idem, Play of Animals, p. 286. Idem, Die Spiele der Tiere, 
p. 308. 474 
THE HISTORY OF HUMAN MARRIAGE chap. 
Thus the female often must be long courted before she 
yields ; she has an instinctive impulse to prevent the male's 
approach, which can only be overcome by persistent pursuit 
and the exercise of all his arts.1 This is a feminine charac- 
teristic found in mankind as well as among the lower animals. 
Both in savage and civilised countries courtship may mean a 
prolonged making of love to the woman. Mariner's words 
with reference to the women of Tonga has a much more 
general application. " It must not be supposed," he says, 
" that these women are always easily won ; the greatest 
attentions and most fervent solicitations are sometimes 
requisite, even though there be no other lover in the way. 
This happens sometimes from a spirit of coquetry, at other 
times from a dislike to the party, &c."2 Among the 
Uppiliyans of the Trichinopoly District, Southern India, a 
girl " has to be asked in marriage a number of times, before 
consent is given, lest it be thought that she is yielding too 
easily."3 Among the Omaha " a man of twenty-five or 
thirty will court a girl for two or three years. Sometimes the 
girl pretends to be unwilling to marry him, just to try his 
love, but at last she usually consents."4 When the Apache 
lover has proposed to a girl by staking his horse in front of 
her dwelling, four days are allowed her for an answer. A 
ready acceptance is apt to be criticised with some severity, 
while a tardy one is regarded as the extreme of coquetry. 
Scarcely any girl will show that she favours the suitor, by 
leading his horse to water, before the second day, so as not 
to be suspected of having an unusual desire to be married.•' 
In Greenland it was considered the proper thing for a lady 
" to show no partiality whatever for her suitor however 
fond she might be of him." 6 Among the Bakongo " the 
1 Buchner, Liebe und Liebesleben in der Thierwelt, p. 39 sqq. 
Groos, Die Spiele der Tiere, p. 304 sq. Idem, Play of Animals, 
p. 283 sq. 
2 Mariner, Natives of the Tonga Islands, ii. 174. Cf. Fritsch, 
Die Eingeborenen Sud-Afrika's, p. 445 (Bushmen). 
3 Thurston, Castes and Tribes of Southern India, vii. 238. 
4 Dorsey, ' Omaha Sociology,' in Ann. Rep. Bur. Ethnol. iii. 260, 
5 Cremony, Life among the Apaches, p. 246 sq. 
6 Nansen, op. cit. ii. 317. XIII. COURTSHIP 
475 
young people go through a courtship. When the lover 
formally introduces himself, the young girl flies across the 
fields screaming as if she had seen a wolf, but she doe$ not 
run away far when the wolf pleases her/'1 We shall see 
subsequently that at the weddings of many peoples virgin 
coyness finds a ceremonial expression in the resistance made 
by the bride. 
At the end of the performance the female may also take 
an active part by accepting or refusing her lover or, if there 
are several lovers, by selecting one of them. We are told that 
she may exercise such a choice even when the rivals are 
fighting for her ; but in such cases force is probably the chief 
factor in determining the possession of the female by the 
male.2 In mankind the woman's liberty of choice is often 
hampered by social factors unknown among the lower 
animals ; but even among savages it is not the rule that 
women are married without having any voice of their own 
in the matter. This is a subject which will be treated in 
detail in a subsequent chapter. 
From this sketch of the principal features of courtship 
we shall proceed to a discussion of their origin. In a previous 
chapter I have dealt with the passion of masculine jealousy, 
which is the immediate cause of the combats of males for the 
possession of females ; and I ventured to suggest that its 
remarkable violence, even in otherwise timid animals, may 
possibly be due to its usefulness to the species as a means of 
preventing promiscuity.3 But it may be useful in other 
respects as well. Already Burdach observed that through 
the lights between rivals " the weak are excluded from 
propagation and a more vigorous procreation is secured."4 
And according to later writers the fighting acts as a sexual 
stimulant both on male and female. But how shall we 
explain the peaceful activities of the courting male ? 
In these activities certain secondary sexual characters 
play- a most important part, and the question of their 
genesis can therefore not be passed over in silence. The 
1 Johnston, George Grenfell and the Congo, ii. 678. 
2 Tillier, L'instinct sexuelle, p. 136. 
3 Supra, p. 334 sq. 4 Burdach, op. cit. i. 413. 476 
THE HISTORY OF HUMAN MARRIAGE cn.'XUi 
discussion of it may seem rather remote from the subject 
of human courtship ; but it will be found to have indirectly 
a bearing also on points which are of great importance 
for a right understanding of sex relations in mankind. It at 
once takes us to Darwin's theory of sexual selection. CHAPTER XIV 
SECONDARY SEXUAL CHARACTERS IN ANIMALS-FEMALE 
COYNESS 
Besides natural selection, which depends on the success of 
both sexes, at all ages, in relation to the general conditions of 
life, Darwin introduced another principle, sexual selection, 
which depends on the success of certain individuals over 
others of the same sex in relation to the propagation of 
the species. According to the former principle, those indi- 
viduals who are most successful in the struggle for existence 
survive the others, and characters useful to the species are 
thus inherited ; according to the latter, those individuals 
who have the greatest success in the struggle for mates have 
the most numerous offspring, and the characters which gave 
them the preference pass on to the new generation, and are 
afterwards intensified by the operation of like causes. The 
sexual struggle is of two kinds. In both it is carried on by 
individuals of the same sex ; but in one these individuals, 
generally the males, try to drive away or kill their rivals ; 
in the other, they seek to excite or charm those of the oppo- 
site sex, generally the females, who select the most attractive 
males for their partners. Therefore the characters acquired 
through sexual selection, and transmitted chiefly to offspring 
of the same sex, generally the males, are, on the one hand, 
weapons for battle, vigour and courage ; on the other hand, 
certain colours, forms, ornaments, sounds, or odours, which 
are felt to be pleasant. The secondary sexual characters of 
the latter sort are thus due to the preference given to them by 
the females. They have been acquired because they are 
477 478 
THE HISTORY OF HUMAN MARRIAGE chap. 
beautiful or otherwise agreeable, whereas the characters 
resulting from natural selection have been acquired because 
they are useful. " The senses of man and of the lower 
animals," says Darwin, " seem to be so constituted that 
brilliant colours and certain forms, as well as harmonious and 
rhythmical sounds, give pleasure and are called beautiful; 
but why this should be so we know not."1 According to 
Darwin, natural and sexual selection are thus two different 
sources from which animal characters have arisen. 
Far from co-operating with each other, these two kinds of 
selection seem even to work in opposite directions. Sexual 
selection, as described by Darwin, produces effects disadvan- 
tageous to the species. " It is evident," he says, " that the 
brilliant colours, top-knots, fine plumes, &c., of many male 
birds cannot have been acquired as a protection ; indeed, 
they sometimes lead to danger."2 When we consider what 
an important part is played by colours, as means of protection, 
in the whole animal kingdom, it is certainly surprising that 
many male animals display brilliant hues, which cannot fail 
to make them conspicuous to their enemies. The strong 
odours emitted by certain reptiles and mammals, during the 
pairing season, and the sounds produced by various species 
at the same period, have also the effect of attracting hostile 
animals that are searching for food. And the danger arising 
for the species from these secondary sexual characters is all 
the greater because they generally appear at the time when 
offspring are about to be produced. Thus, besides colours, 
structures, and functions, adapted in the most marvellous 
way to the requirements of each species, there are others 
highly dangerous, which, according to Darwin, depend upon 
an aesthetic sense, or taste, in the females, the origin of which 
we do not know. 
Darwin has shown how immense is the influence exercised 
by natural selection on the organic world. A disciple, there- 
fore, naturally feels perplexed when he is told of a series of 
facts, which, according to the explanation given by the master, 
seem actually opposed to the theory of natural selection. 
When realising the contradiction between the two theories of 
1 Darwin, Descent of Man, ii. 384. 2 Ibid. ii. 252. xiv SECONDARY SEXUAL CHARACTERS IN ANIMALS 
479 
selection, he is inclined to ask :-Can we, after all, be sure 
that the secondary sexual characters have no useful function 
to fulfil ? If they have, they, also, may be explained by the 
principle of the survival of the fittest. The larger size and 
greater strength of the males, and the weapons of offence or 
defence many of them possess, may easily be so accounted for, 
as, among the higher animals, the males generally fight with 
each other for the possession of the females. The point is 
whether the other secondary sexual characters can be due 
to the same cause. 
It is an established fact that the colours of flowers serve a 
definite end. Through them the flowers are recognised by 
insects in search of honey; and the insects, during their visits, 
involuntarily carry the pollen of one flower to the stigma of 
another, and thus effect cross-fertilisation, which is proved to 
be of great importance for the vigour and fertility of the next 
generation of plants. Now it is extremely interesting to note 
that brilliant colours are found only in species of flowers to 
which they are useful as means of attracting insects; they 
never occur in plants which are fertilised by the wind.1 
Wallace observes that plants rarely need to be concealed, 
because they obtain protection by their spines, or their hard- 
ness, or their hairy covering, or their poisonous secretions. 
Hence there are very few cases of what seems to be true pro- 
tective colouring among them.2 In animals, on the con- 
trary, colour is greatly influenced by their need of protection 
from, or warning to, their numerous enemies ; colours of 
other kinds must always, to a certain extent, be dangerous 
for the species. Is it probable, then, that, whilst gay colours 
occur only in the flowers of those plants to which they are of 
real use, conspicuous colours should occur in animals to 
which they are of real danger-merely because the females 
find them attractive ? 
Wallace, in his well-known criticism of Darwin's 
theory of sexual selection,3 suggests that the very frequent 
1 Muller, Fertilisation of Flowers, p. 14. 
2 Wallace, Tropical Nature, p. 223. 
3 Wallace, ' Colours of Plants and the Origin of the Colour- 
Sense. in Tropical Nature, p. 221 sqq. Idem, Darwinism, ch. x. 480 
THE HISTORY OF HUMAN MARRIAGE chap. 
superiority of the male bird or insect in brightness or 
intensity of colour is due to the greater vigour and activity 
and the higher vitality of the male. This intensity of 
coloration is therefore most manifest in the male during 
the breeding season, when the vitality is at a maximum. 
It would be further developed by the combats of the 
males for the possession of the females; and the most 
vigorous and energetic usually leaving the most numerous 
and most healthy offspring, natural selection would indirectly 
become a preserver and intensifier of colour.1 This view 
has been criticised by G. W. and Elizabeth G. Peckham 
in the record of their observations on sexual selection 
in spiders of the family Attidce. " First," they say, 
" we found no evidence that the male spiders possess 
greater vital activity ; on the contrary, it is the female that 
is the more active and pugnacious of the two. Second, we 
found no relation, in either sex, between development of 
colour and activity ; the Lycosidee, which are among the most 
active of all spiders, having the least colour development, 
while the sedentary orb-weavers show the most brilliant 
hues."2 But even if Wallace is right in his opinion that 
there is some connection between vigour and colour, it 
may be asked whether this connection, depending on some 
unknown physiological law, is so necessary that it takes place 
even when colour is positively disadvantageous to the species. 
Nothing of the kind is found in the vegetable kingdom. We 
know, as Wallace himself remarks, that colours which rarely 
or never appear in the species in a state of nature continu- 
ally occur among cultivated plants and domesticated animals 
-a fact which shows that the capacity to develop colour is 
ever present.3 Among wild plants such colour variations 
are never preserved except when they are useful. Is it not 
most reasonable to suppose that the like is the case with 
animals ? 
1 Idem, Tropical Nature, p. 193 sqq. 
3 G. W. and Elizabeth G. Peckham, ' Observations on Sexual 
Selection in Spiders of the Family Attidre,' in Occasional Papers of 
the Natural History Society of Wisconsin, i. 59. 
3 Wallace, Tropical Nature, p. 187. Xiv SECONDARY SEXUAL CHARACTERS IN ANIMALS 
481 
The truth seems to be that colour subserves the same pur 
pose in both of the great kingdoms of the organic world 
Just as flowers are coloured that insects may recognise 
where honey is to be found, and thus may be led to promote 
fertilisation, so the sexual colours of animals have been 
developed to make it easier for the sexes to find each other 
during the pairing time. Protective colours are useful so 
far as they conceal the animal from its enemies, but at the 
same time th^y conceal it from individuals of its own species. 
Sexual colours are therefore useful as well, because they make 
the animal more visible. It is quite in accordance with the 
theory of natural selection that, where such colours occur, 
the advantage from them should be greater than the dis- 
advantage. We can see the reason for the brilliant colours 
of humming-birds, as these birds, on account of their great 
activity, " are practically unmolested,"1 and for the bright 
hues of the rose chafers, who are saved from attack by a 
combination of protecting characters.2 But generally 
there is danger in sexual colours, so that nature has given 
them with the utmost cautiousness. They are not de- 
veloped till the age of reproduction, and they appear, in a 
great many species, only during the pairing season, or are 
concealed except during courtship.3 Usually they occur in 
males only, because of the females' greater need of pro- 
tection.4 The greatest advantage is won with the least 
possible peril. 
It is a fact of great importance that sexual colours occur 
exactly in those species whose habits make these colours 
most visible. Thus the nocturnal moths, taken as a body, 
are much less gaily decorated than butterflies, all of which 
are diurnal in their habits, although, according to Wallace, 
the general influence of solar light and heat is no adequate 
cause for the variety, intensity, and complexity of the 
colours. The females of the ghost moth are yellow with 
darker markings, whereas the males are white, that they may 
1 Ibid. p. 213. 
2 Wallace, Contributions to the Theory of Natural Selection, p. 73 sq. 
3 Poulton, Essays on Evolution, p. 380. 
4 Wallace, Contributions, &c., p. 259 sqq. 482 
THE HISTORY OF HUMAN MARRIAGE 
CHAP. 
be more easily seen by the females whilst flying about in the 
dusk ; and it is remarkable that in the Shetland Islands the 
male of this moth, instead of differing widely from the 
female, frequently resembles her closely in colour-as Mr. 
Fraser suggests^ because, at the season of the year when the 
ghost moth appears in these northern latitudes, the whiteness 
of the males is not needed to render them visible to the 
females in the twilight night.1 Both Darwin2 and Wallace 3 
think that, in this case, colour may be a means of recognition. 
Generally speaking, bright sexual colours are found almost 
exclusively in species which are diurnal or semi-diurnal in 
their habits, and are excluded from those parts of the body 
which move too rapidly to be seen.4 Moreover, they occur 
chiefly in species which, because of their manner of living, 
are to be seen at a distance; they seldom occur in sedentary 
or slowly moving terrestrial animals.5 The members of the 
lowly organised order Thysanura are wingless and dull- 
coloured. The Hemiptera, which usually lurk about plants 
and prey upon hapless insects, are not, as a rule, remarkable 
for conspicuous hues. The Orthoptera are all terrestrial 
in their habits, generally feeding upon plants, and although 
some exotic locusts are beautifully ornamented, their bright 
tints, according to Darwin, do not seem to fall under the 
head of sexual coloration. On the other hand, the dragon- 
flies, which live in the open air, possess splendid green, blue, 
yellow, and vermilion metallic tints, and the sexes often 
differ in their coloration. Every one has admired the 
extreme beauty of many butterflies, especially of the males. 
Amongst the Fishes, living in a medium through which bright 
colours may be observed at a distance, we often find, besides 
protective colours, conspicuous hues which are especially 
intense and visible during the pairing time. Among 
the Reptiles the little lizards of the genus Draco especially 
deserve attention: they glide through the air on their 
1 Fraser, 'Sexual Selection,' in Nature, iii. 489. 
2 Darwin, op. cit. i. 485. 3 Wallace, Darwinism, p. 270. 
4 Poulton, op. cit. p. 380. 
6 The Gallinaceae, however, form an exception ; though almost 
wholly terrestrial, they have the most pronounced sexual colours. 
But they are active and wander much. xiv SECONDARY SEXUAL CHARACTERS IN ANIMALS 
483 
rib-supported parachutes, and the beauty of their colours 
baffles description. Mammals, on the other hand, do not 
generally present the splendid tints so common among male 
birds ; and the brighter colours of certain arboreal mammals 
serve chiefly as means of concealment. 
These phenomena show that sexual colours have been 
evolved for the purpose of being seen. They could scarcely 
be due merely to the fact that coloration is connected with 
the degree of vitality, since the Mammals, for instance, are 
certainly not less vigorous than any of the other Vertebrate 
orders. It may perhaps be suggested that, as flying animals 
more easily escape their enemies than terrestrial, they may 
vdth less danger be decorated with conspicuous hues. But 
here we have to observe the most important fact that 
animals which do not possess sexual colours generally have 
some other means of making themselves discoverable. 
Flowers which need the help of insects for fertilisation 
attract them, in some cases, not by bright colours, but by 
peculiar odours. And as we do not find conspicuous colours 
in plants fertilised by the wind, so flowers have no perfume 
except where it is of real use. The most brilliant flowers, 
as a rule, are those which possess least odour, whilst many of 
them have no scent at all. White or very pale flowers are 
generally the most odoriferous. M. Mongredien gives a list 
of about 160 species of hardy trees and shrubs with showy 
flowers, and another list of sixty species with fragrant 
flowers; but only twenty of the latter are included among 
the showy species, and these are almost all white-flowered.1 
Most of the white flowers are scented only at night, or their 
perfumes are most powerfully emitted at that time; the 
reason being that white flowers are fertilised chiefly by night- 
flying insects. We arrive thus at two conclusions: first, 
that powerful odours and conspicuous colours as guides to 
insect fertilisers are, as a rule, complementary to each other ; 
secondly, that they occur alternately in the way most useful 
to the species. 
In the animal kingdom various odours and sounds are 
closely connected with the reproduction of the species. 
1 Wallace, Tropical Nature, p. 230 sq. 484 
THE HISTORY OF HUMAN MARRIAGE chap. 
During the season of love a musky odour is emitted by the 
submaxillary glands of the crocodile, and pervades its 
haunts. At the same period the anal scent-glands of snakes 
are in active function, and so are the corresponding glands 
of the lizards. Many mammals are odoriferous. In some 
cases the odour appears to serve as a defence or a protection, 
but in other species the glands are confined to the males, and 
almost always become more active during the rutting season. 
Again, a great many insects have the power of producing 
stridulous sounds. In two families of the Homoptera and in 
three of the Orthoptera the males alone possess organs of 
sound in an efficient state, and these are used incessantly 
during the pairing season. Some male fishes have sound- 
producing instruments, and the fishermen of Rochelle assert 
that the males alone make the noise during the spawning 
time. Of frogs and toads the males emit various sounds at 
the pairing time, as in the case of the croaking of our common 
frog. During the rutting season, and at no other time, the 
male of the huge tortoise of the Galapagos Islands utters a 
hoarse bellowing noise, which can be heard at a distance of 
more than a hundred yards. Professor Aughey states that 
on two occasions, being himself unseen, he watched from a 
little distance a rattle-snake coiled up with head erect, which 
continued to rattle at short intervals for half an hour; at 
last he saw another snake approach, and when they met they 
paired. Among Birds the power of song, or of giving forth 
strange cries, or even instrumental music, is exceedingly 
common, particularly in the males during the pairing season ; 
and almost all male mammals use their voices much more 
during that period than at any other time. Some, as the 
giraffe and porcupine, are stated to be completely mute 
except during the rutting season. 
The colours, odours, and sounds of animals, like the colours 
and odours of plants-so far as they may be assumed to be in 
some way connected with the reproductive functions-are, as 
a rule, complementary to each other. Stridulating insects are 
generally not conspicuously coloured. Among the Homo- 
ptera there do not seem to be any well-marked cases of 
ornamental differences between the sexes. Among crickets, xiv SECONDARY SEXUAL CHARACTERS IN ANIMALS 
485 
the Locustidae, and grasshoppers, some species are beautifully 
coloured, but Darwin says :-" It is not probable that they 
owe their bright tints to sexual selection. Conspicuous 
colours may be of use to these insects by giving notice that 
they are unpalatable/' Other species have directly protec- 
tive colours. The bright hues of stridulating beetles seem 
to be of use chiefly for protective and warning purposes ; 
whereas species belonging to the orders Neuroptera and 
Lepidoptera, often extremely conspicuously coloured, are not 
remarkable for any stridulous sounds. Frogs and toads, 
which have an interesting sexual character in the musical 
powers possessed by the males, are evidently coloured accord- 
ing to the principle of protection, or sometimes tinted with 
conspicuous hues in order to be more easily recognised by 
their enemies as a nauseous food. Of Reptiles, the Lacertilia 
excel mainly in bright tints; the Chelonia, Crocodilia, and 
Ophidia, in sounds and odours. Among Birds, in one 
instance at least, the male is remarkable for his scent. 
" During the pairing and breeding season," says Gould, 
with reference to the Australian musk-duck, "... this bird 
emits a strong musky odour " ; it is not ornamented with 
any conspicuous hues.1 Sexual colours and the power of 
song are generally complementary to each other among 
Birds. " As a general rule," Mr. Wood remarks, " it is 
found that the most brilliant songsters among the birds are 
attired in the plainest garb ; and it may safely be predicted 
of any peculiarly gorgeous bird, that power, quality, and 
sweetness of voice are in inverse ratio to its beauty of 
plumage."2 Thus, of the British birds, with the exception of 
the bullfinch and goldfinch, the best songsters are plain- 
coloured, and the brilliant birds of the tropics are hardly 
ever songsters. The wild camel in the desert of Kum-tagh 
has a reddish, sandy hue, and the males, " even during the 
rutting season, utter no sound, but find their consorts by 
scent."8 The musk-deer, well known for the intolerable 
1 Gou'd, Handbook to the Birds of Australia, it 383. 
2 Wood, Illustra'ed Natural History, ii. 257. 
3 Prejeva'sky, From Kulja to Lob-nor, pp. 94, 92. 486 
THE HISTORY OF HUMAN MARRIAGE chap. 
perfume which the males emit at the pairing time, is also 
entirely silent.1 
Moreover, as appears from what has just been said, the 
sexual colours, scents, and sounds in the animal kingdom 
are complementary to each other in the way that is best suited 
to make the animals easily discoverable. As bright colours 
would be of no advantage to flowers fertilised by night-flying 
insects, so they would be of comparatively little advantage 
to animals living among grass and plants, in woods and 
bushes ; whereas sounds and scents make the animal recog- 
nisable at a considerable distance. We have also seen that it 
is among flying and aquatic animals that sexual colours 
chiefly occur, whereas terrestrial animals excel in sounds and 
scents. Thus most of the stridulating insects are terrestrial. 
Whilst brightly-coloured lizards, living on trees or running 
from stone to stone, must attract attention by the brilliance 
of their covering, crocodiles inhabiting rivers and jungles, 
and frogs crawling among the grass, allure their mates, the 
former by emitting musky odours, the latter by producing 
loud sounds. The odour of the Australian musk-duck, 
which depends for its food and for its preservation from 
danger upon its powers of diving rather than upon those of 
flying, is, as Gould observes, often perceptible long before 
the animal can be seen.2 
Darwin remarks, as regards birds :-" Bright colours and 
the power of song seem to replace each other. We can per- 
ceive that, if the plumage did not vary in brightness, or if 
bright colours were dangerous to the species, other means 
would be employed to charm the females ; and melody of 
voice offers one such means."3 But if we accept Darwin's 
theory of sexual selection, we are compelled to suppose that 
that inexplicable aesthetic sense of the females has been 
developed in the way most dangerous to the species. Con- 
spicuous colours are admired by the females of those animals 
which, by means of such colours, are most easily discovered 
by their enemies, and sounds and odours are appreciated 
exactly in those species to which they are most perilous. If, 
1 Brehm, Thierleben, iii. 94, 2 Gould, op. cit. ii. 382 sq. 
3 Darwin, op. cit. i. 62. xiv SECONDARY SEXUAL CHARACTERS IN ANIMALS 
487 
on the contrary, we accept the explanation that, although 
sexual colours, odours, and sounds are in some ways hurtful 
to the species, they are upon the whole advantageous, inas- 
much as they make it easier for the sexes to find each other, 
we have a theory in accordance with all known facts, as 
well as with the great principle of natural selection. With 
reference to the singing of birds it is worth noticing that in 
various species the males, when calling or singing, are in the 
habit of selecting a free and open place from which their 
voices can be heard afar.1 It may be objected that it is not 
the females but the males that are the seekers, whilst the 
secondary sexual characters generally occur in the males 
only. But the females are not entirely passive during the 
pairing season ; and several of the statements collected by 
Darwin directly indicate that females are attracted by the 
sounds of their future partners. Moreover, among butter- 
flies and beetles there are species in which the females alone, 
or both females and males, during the pairing season attract 
the attention of the opposite sex through odours, sounds, or 
colours ; and among birds call-notes are also produced by the 
less developed voices of females.2 
In his work on ' Darwinism/ Wallace expresses the opinion 
that the various sounds and odours which are peculiar to the 
male serve as a call to the female, or as an indication of his 
presence ; and, as he says, " the production, intensification, 
and differentiation of these sounds and odours are clearly 
within the power of natural selection."3 Wallace has also 
shown the immense importance of colour as a means of 
recognition. The theory here set forth thus, in fact, very 
nearly approaches his views. The main difference is that the 
sexual colours have been classified under the head of " colour 
for recognition," whether the positive cause by which they 
1 Hacker, Der Gesang dev Vogel, p. 47. 2 Ibid. p. 49 sqq. 
3 Wallace, Darwinism, p. 284. Hacker {op. cit. p. 46 and passim) 
has more recently expressed a similar opinion with reference to the 
sinking of birds. He says, " Es wird Niemand, der selber das 
Friihlingsleben der Vosel Jahr fiir Jahr mit offenem Auge verfolgt, 
die Auffassung von sich weisen konnen, dass die urspriinglichste 
Bedeutung der einfacheren sexuellen Laute der mannlichen Vogel 
die Anlockung der Weibchen ist." 488 
THE HISTORY OF HUMAN MARRIAGE chap. 
have been produced be a surplus of vital energy or not. But 
the argument on which I have ba~ed my conclusions differs 
from his, although, as he points out, it is " in general harmony 
with it."1 It is the way in which the sexual colours, sounds, 
and odours are distributed among the different animal 
species that, in my opinion, is the strongest evidence against 
Darwin's theory of sexual selection, because this distribution 
is invariably such as to make the animal distinguishable at 
a distance. And now I am prepared to go a step further. 
It seems to me probable that these secondary sexual char- 
acters 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. They 
would thus influence the species not only quantitatively 
but qualitatively. Professor Hacker is of opinion that 
the singing of birds serves such a purpose,2 and I venture 
to believe that the same is equally true of sexual colours 
and odours. 
Indeed, so far as the mere propagation of the species is 
concerned, these characters would, in the case of gregarious 
animals and many others besides, be quite unnecessary and 
therefore harmful to the species, because there would be 
no difficulty in finding a mate from the immediate neigh- 
bourhood. Why then those sounds or colours or odours which 
are apt to attract enemies from a distance ? Among 
gregarious animals the pairing instinct is during the sexual 
season much stronger than the herding instinct; and if we 
knew more about their habits, it might be found that they 
commonly prefer mating with animals outside their own herd. 
As will be seen in a subsequent chapter, a preference for 
strangers is noticeable in our domestic animals; in certain 
wild species pairing is known to be avoided in the nest; 
and there are yet other arrangements in the animal world 
tending to prevent in-breeding.3 If the sexual sounds, 
colours, and odours serve the same purpose, the analogy 
between them and the colours and odours of the flowers 
1 See Mr. Wallace's 'Introductory Note.' 
5 HScker, op. cit. pp. 48, 90 sq. 3 See infra, on Exogamy. xiv SECONDARY SEXUAL CHARACTERS TN ANIMALS 
489 
of plants becomes as complete as possible. In many ^ases 
it may really be their main, if not their only, object. 
There are some further secondary sexual characters 
which according to Darwin must be regarded as ornaments 
-such as the great horns which rise from the head, thorax, 
and clypeus of many male beetles; the appendages with 
which some male fishes and reptiles are provided ; the combs, 
plumes, crests, and protuberances of many male birds; 
and various crests, tufts, and mantles of hair which are 
found in certain mammals. But some of these characters 
may also serve as means of recognition, or be of use to the 
males in their fights for females. Wallace suggests that 
crests and other erectile feathers may have been safeguards 
for the bird by making it more formidable in appearance and 
thus frightening away enemies ; that long tail- or wing- 
feathers might serve to distract the aim of a bird of prey; 
and that the ornamental appendages of birds and other 
animals are due to a surplus of vital energy, leading to 
abnormal growths in those parts of the integument where 
muscular and nervous action are greatest.1 
If the primary object of secondary sexual characters is to 
make it easier for individuals of different sex but of the same 
species to find and recognise each other, we can understand 
the immense variability of those characters in different 
species-it is precisely what might be expected. If, on the 
other hand, they were due to female choice, why should 
the characters which appeal to the females differ to such an 
extent that there are not two species in which they are the 
same ? This difficulty did not escape Darwin. " It is a 
curious fact," he says, " that in the same class of animals 
sounds so different as the drumming of the snipe's tail, the 
tapping of the woodpecker's beak, the harsh trumpet- 
like cry of certain water-fowl, the cooing of the turtle-dove, 
and the song of the nightingale, should all be pleasing to the 
females of the several species." And further, " What shall 
we say about the harsh screams of, for instance, some kinds 
of macaws; have these birds as bad taste for musical 
sounds as they apparently have for colour, judging by the 
1 Wallace, Darwinism, p. 293 sq. 490 
THE HISTORY OF HUMAN MARRIAGE chap. 
inharmonious contrast of their bright yellow and blue 
plumage ? "x 
It has been said that the hypothesis of sexual selection has 
often been placed in a false light by the introduction of the 
unnecessary supposition that the female animal is guided 
by aesthetic feelings in her choice of mate, and that Darwin 
himself-although many of his expressions undoubtedly 
give support to such an interpretation-also states that the 
colours or sounds or other characteristics of the male are an 
appeal to the female's sexual emotions, a stimulus to sexual 
excitement, an allurement to sexual contact.2 I think, 
however, that the difference between the two " theories " is 
more verbal than real. When Darwin speaks of the female's 
" aesthetic " appreciation of certain characters of the male, 
he evidently does not mean that she takes a disinterested 
pleasure in his beauty, but that she is attracted by it in a 
manner that leads to pairing. In no case does the difference 
affect either my criticism or the explanation I have given of 
the characters in question. This explanation implies, of 
course, that the female is allured by the colours she sees, the 
sounds she hears, or the odour she smells-otherwise they 
would be entirely useless. But instead of tracing their 
origin to an inexplicable tendency in the female to be 
attracted sometimes by certain colours, sometimes by certain 
sounds, sometimes by certain odours, it shows why she is so 
attracted in different species : the stimulus to which she 
responds is of such a nature as to make it possible for her to 
recognise the male at a distance. Thus her tendency to 
be affected by male characters of a certain kind has been 
acquired through natural selection owing to the fact that 
those characters are useful for propagation or-if I am right 
in my supposition that they are apt to prevent in-breeding- 
for the production of a vigorous offspring. They would, 
naturally, never have come into general existence unless the 
females had been allured by them-that is, unless the females 
as well as the males had possessed a sexual instinct- 
1 Darwin, op. cit. i. 74, 67. 
2 See Lloyd Morgan, Animal Behaviour, p. 263 ; Ellis, Psychology 
of Sex, (vol. iii.) Analysis of the Sexual Impulse, p. 19 sqq. xiv SECONDARY SEXUAL CHARACTERS IN ANIMALS 
491 
but, on the other hand, the females would not have had an 
occasion to be allured by them unless they had been useful 
to the species. Nor can I believe that the delight which the 
females undoubtedly take in them could have made them 
develop beyond the limit for their usefulness. For in 
animals in a state of nature there could be no luxurious 
evolution of characters so dangerous for the species as the 
sexual colours and sounds and odours for the most part are. 
They may, indeed, be useful not only as means of recognition 
and attraction, but as means of further stimulating the 
sexual instinct and of promoting important physiological 
processes; but among wild animals natural selection could 
never have allowed them to pass out of its control. 
Secondary sexual characters the primary object of which 
is to bring male and female together may also play a part in 
the prelude to pairing after that object has been achieved. 
A notorious instance of this is the singing of birds, and another 
instance is the display of sexual colours, although the bright 
surfaces displayed may also be such as have been previously 
concealed. There can be little doubt that these performances, 
as also the antics or dances of the courting males, are 
occasioned by the unwillingness of the female to pair at once, 
since it is hard to believe that the excited male would of his 
own accord for any length of time postpone the gratification 
of his desire. That they are due to a conscious endeavour 
on his part to overcome the coyness of the female by charm- 
ing her, we have no right to assume. They are most 
probably the spontaneous outcome of his excitement. Even 
the activity which is most suggestive of an express purpose, 
namely, the display of peculiarly coloured surfaces which are 
generally hidden, may be thus explained. Display in court- 
ship occurs in plainly coloured as well as in ornamental 
species ; and if a bright spot appears on a part of the body 
which ds exposed only in courtship, natural selection may 
leave it uninterfered with if harmless, and even preserve and 
develop it if useful as a sexual stimulant. The central 
problem in the theory of courtship is thus to explain the coy- 
ness of the female, which more than anything else makes it- 
to use a phrase of Darwin-" a prolonged affair." 492 
THE HISTORY OF HUMAN MARRIAGE chap. 
It should, first, be noticed that in various cases at least the 
coyness of the female may be regarded as a modified continua- 
tion of a previous absolute refusal of the sexual approaches of 
the male.1 As Mr. Heape observes, the sexual season of an 
individual male and an individual female of the same 
species is not necessarily coincident, and when the male has 
a sexual season all the year round this is not always the case 
with the female.2 In all animals which have been investi- 
gated coition is not allowed by the female until some time 
after the swelling and congestion of the vulva and surround- 
ing tissue are first demonstrated.3 And frequently also the 
male seems disposed to have intercourse when the heat of the 
female has already been succeeded by a marked repugnance 
to it.4 It is impossible, however, to believe that a quality 
so intimately connected with the all-important function of 
reproduction as feminine coyness could owe its origin merely 
to a whimsical tendency to continue playfully a rather 
embarrassing habit which no longer has a meaning. We 
must assume that the coyness of the female also when she is 
in heat serves some useful purpose, and it is therefore 
necessary to look for another explanation of it. 
Montaigne said long ago that nothing whets our taste so 
much as rarity and difficulty, and that the artifice of virgin 
modesty serves the object of increasing in us the desire to 
overcame all obstacles.5 Burdach made the same observa- 
tion the basis of a biological theory. He wrote that the 
male's endeavours to catch, overcome, and hold the resisting 
female make him more and more ardent and increase his 
fitness for procreation, while the physical and mental 
excitement of the female at the same time seems to promote 
her fecundity.6 On this point Burdach, although his book 
appeared some thirty years before the ' Origin of Species,' 
1 Cf. Burdach, Die Physiologic als Erfahrungswissenschaft,i. 411 ; 
Ellis, op. cit. vol. i. Evolution of Modesty, &c. p. 39 sq. 
* Heape, ' Sexual Season of Mammals,' in Quarterly Jour. Micro- 
scopical Science, xliv. 5, 12. 
3 Ibid. p. 55. 4 Tillier, L'instinct sexuel, p. 89. 
5 Montaigne, op. cit. book ii. ch. 15, vol. ii. 329, 332. 
6 Burdach, op. cit. i. 370, 41 r. XIV FEMALE COYNESS 
493 
was really more Darwinian than Darwin himself, who never 
seems to have realised the great importance of female 
coyness for the existence of the species.1 The first writer 
after Darwin, so far as I know, who emphasised this was 
Tillier (in 1889). Speaking of the long-winded procedure of 
courtship, he says :-" It is natural to think that its object 
and result are to make the male more ardent. The excite- 
ment which produces a delay in the pairing increases also 
perhaps in a certain measure the secretion of the glands and 
thus renders the chances of fecundation more numerous."2 
And in another place :-" Is it not possible to admit that this 
long-continued display leads to a kind of sexual excitement 
which is profitable to fecundation when the individuals 
arrive at the point of uniting definitely ? " 3 Some years 
later, although evidently without knowing Tillier's treatise, 
Groos likewise suggested that the primary object of courtship 
is to produce sexual excitement. He quotes Ziegler's 
statement that "in all animals a high degree of excitement 
of the nervous system is necessary to procreation," and 
proceeds :-" Since the sex impulse must necessarily have 
extraordinary strength, the interests of the preservation of 
species are best served by a long preliminary condition of 
excitement and by some checks to its discharge. The 
instinctive coyness of the female serves this purpose."4 
Similar views have been expressed by Havelock Ellis, Hirn, 
Lloyd Morgan, Hacker, and others.5 
In support of this theory it may be said that the activity 
of the male which is the consequence of the coyness of the 
female is not only, as already said, reflexes of excitement, 
but undoubtedly at the same time a sexual stimulant. 
Hacker points out that the call-cries and song of birds at 
1 Cf. Ellis, op. cit. (vol. iii.) Analysis of the Sexual Impulse 
p. 21. 
2 Tillier, L'instinct sexuel, p. 74. 3 Ibid. p. 127. 
4 Groos, Die Spiele dev Tiere, pp. 263, 265. Idem, Play of Man, 
p. 265. 
6 Ellis, op. cit. (vol. iii.) Analysis of the Sexual Impulse, p. 181. 
Hirn, Origins of Art, p. 192 sqq. Lloyd Morgan, op. cit. p. 265 sq. 
H3.cker, op. cit. pp. 48, 51, 57, 58, 89. Mortimer, Chapters on Human 
Love, p. 36. Marro, Lapubertd, quoted by Ellis, p. 181. 494 
THE HISTORY OF HUMAN MARRIAGE chap. 
the pairing season become methods of producing excitement, 
in the male by the muscular energy required and in the female 
through the ear.1 So also the males are excited by their 
own nuptial display, antics, and dances. Muscular move- 
ment is by itself calculated to produce sexual excitement, 
and dancing, as the highest and most complex form of muscu- 
lar movement, is therefore among the most various kinds of 
animals, from insects and birds to man, a prelude to coition.2 
It has been called " the spur of lust," or " a circle of which the 
devil himself is the centre." 3 And not only is movement 
itself a sexual stimulant, but even the spectacle of movement 
tends to be so. Hence, as Dr. Havelock Ellis points out, we 
so frequently find, both among the lower animals and savage 
peoples, that one sex alone, usually the male, takes an active 
part in dancing and similar movements calculated to stimu- 
late the sexual passion of the other sex as well.4 Sir Basil 
Thomson tells us of a case in which a young Fiji woman was 
so excited by a young man who surpassed all his fellows in 
the war-dance that she, unable to contain herself, rushed 
into the middle of the dancing ground, and clutching him, 
took his loin-cloth in her teeth-a most terrible breach of 
decorum.5 It is also highly probable that the females are 
excited by the males' display of their sexual colours.6 
Professor Hirn gives instances of the stimulating effect of 
brilliancy, and points out, for example, that the luminous 
spheres on the tails of pheasants and peacocks will, in virtue 
of their very smallness and brilliancy, provoke a stimulation 
in the mind of the hen before whom they are vibrated.7 
To lead to stimulating activity, however, can only be the 
immediate object of feminine coyness, not its ultimate 
object. The question now to be answered is, why such 
stimulating activity is of importance for the maintenance 
of the species. It is here the difficulties begin. Far from 
1 Hacker, op. cit. pp. 48, 51, 57 sq. 
2 Cf. Ellis, op. cit. (vol. iii.) Analysis of the Sexual Impulse, p. 47 ; 
Groos, Play of Man, p. 273. 
3 Burton, Anatomy of Melancholy, iii. 2. 2. 4, p. 540. 
4 Ellis, op. cit. (vol. iii.) Analysis of the Sexual Impulse, p. 48 sq. 
5 Thomson, Fijians, p. 240. 
6 See Wallace, Darwinism, p. 285. 7 Hirn, op. cit. p. 193 sqq. XIV FEMALE COYNESS 
495 
agreeing with one of the writers referred to above that the 
importance of the instinctive coyness of the female is " self- 
evident," I think, on the contrary, that in the present state 
of our physiological knowledge it is a problem which cannot 
be definitely solved. But hypothetical suggestions may 
prove to be useful. 
The first question which presents itself is whether the 
sexual instinct in animals is so feeble that it requires some 
other stimulus besides the mere presence of a suitable indivi- 
dual to lead to intercourse. Dr. Havelock Ellis says that 
we are deceived by what we see among highly fed domesti- 
cated animals, and among the lazy classes of human society, 
when we imagine that the sexual instinct is normally ever 
craving to be satisfied, and that throughout nature it can 
always be set off at a touch whenever the stimulus is applied. 
" So far from the instinct of tumescence naturally needing 
to be crushed, it needs, on the contrary, in either sex to be 
submitted to the most elaborate and prolonged processes in 
order to bring about those conditions which detumescence 
relieves. . . . The whole object of courtship, of the mutual 
approximation and caresses of two persons of the opposite 
sex, is to create the state of sexual tumescence."1 I cannot 
help thinking that this is a somewhat exaggerated statement 
of difficulties experienced by wild animals and savage men 
in obtaining tumescence, if by tumescence is merely meant 
the vascular congestion which is an essential preliminary to 
acute sexual desire.2 Dr. Malinowski is probably right 
in saying that in the majority of savage races sexual life 
is carried on very intensely; 3 the restrictions to pre- 
nuptial intercourse which we have found among many of 
them are certainly no evidence of a feeble sexual impulse. 
And among wild animals this impulse is probably intensified 
by being more or less limited to a special season of the year. 
It seems to me that the excitement resulting from the 
coyness of the female can hardly be needed in order to pro- 
duce acute sexual desire in the male, which is easily obtained 
1 Ellis, op. cit. (vol. iii.) Analysis of the Sexual Impulse, p. 44 sq. 
2 See ibid. p. 21. 
3 Malinowski, ' Baloma,' in Jour. Roy. Anthr. Inst. xlvi. 417. 496 
THE HISTORY OF HUMAN MARRIAGE ch. xiV 
without it; whilst in the case of the female it would be 
somewhat absurd to maintain that her coyness exists simply 
for the purpose of being overcome. But the prolonged 
excitement which follows from her coyness has another 
effect, which may be of the utmost importance for the 
species: it increases the secretions of the sexual glands. 
This was hinted at by Tillier, who pointed out that the 
excitement might thereby render the chances of fecunda- 
tion more numerous. So it does, because it is only when 
mixed with the prostatic fluid that the spermatozoa gain 
their full vitality. The profuse secretion of the prostate, 
resulting from prolonged excitement, would give motility to 
a larger number of spermatozoa than would otherwise gain 
the necessary vitality, and thus there would be an increased 
prospect of fecundation. Or, more generally and vaguely 
speaking, the prolonged excitement may be supposed to 
serve the same object as that enormous production of 
spermatozoa, estimated at 180 millions in a single ejacula- 
tion,1 which has never yet been satisfactorily explained. 
Again, as for the effect of courtship upon the female organism, 
we may assume, with Dr. Havelock Ellis, that the mucus 
which is so profusely poured out over the external sexual 
organs in woman during the excitement of sexual desire 
has for its end the lubrication of the parts and the facilitation 
of the passage of the intromittent organ.2 
I doubt whether at present anything more can be said 
about the biological importance of courtship and female 
coyness. 
1 Tigerstedt, Lehvbuch der Physiologic des Menschen, ii. 459. 
2 Ellis, op. cit. (vol. iii.) Analysis of the Sexual Impulse, p. 186 sq. CHAPTER XV 
PRIMITIVE MEANS OF ATTRACTION 
In the present and the next chapter I shall deal with means 
of various kinds by which primitive people endeavour to 
make themselves sexually attractive to persons of the 
opposite sex. They are used by both sexes-by men who 
court and by women who want to be courted. They all fall 
under the heading " self-decoration." 
The desire for self-decoration, although a specifically 
human quality, is exceedingly old. The ancient barbarians 
who inhabited the south of Europe at the same time as the 
reindeer and the mammoth brought to their caves brilliant 
and ornamental objects.1 Nowadays there are savages 
destitute of almost everything which we regard as necessaries 
of life, but there is no people so rude as not to take pleasure 
in ornaments. The women of the wretched Veddas in 
Ceylon decorate themselves with necklaces of brass beads, 
and bangles cut from the chank shell.2 The Fuegians " are 
content to be naked," but " ambitious to be fine."3 The 
Australian native, without taking the slightest pride in his 
appearance so far as neatness or cleanliness is concerned, 
is yet very vain of his own rude decorations.4 And of the 
Tasmanians Cook tells us that they had no wish to obtain 
1 Cf. Spencer, Principles of Sociology, i. 64. 
2 Emerson Tennent, Ceylon, ii. 443. 
3 Hawkesworth, Account of Voyages in the Southern Hemisphere, 
ii- 55- 
4 Eyre, Journals of Expeditions of Discovery into Central Australia, 
ii. 209 
497 498 
THE HISTORY OF HUMAN MARRIAGE chap. 
useful articles, but were eager to secure anything orna- 
mental. 
The predilection of savages for ornaments has been 
sufficiently shown by travellers in almost every part of the 
world. Feathers and-beads of different colours, flowers, 
rings, anklets, and bracelets are common embellishments. 
A fully-equipped Santal belle, for instance, carries two ank- 
lets, and perhaps twelve bracelets, and a necklace weighing 
a pound, the total weight of ornaments on her person 
amounting to thirty-four pounds of bell metal-" a greater 
weight," says Captain Sherwill, " than one of our drawing- 
room belles could well lift."1 We may without much 
exaggeration say with Grosse that " the primitive man 
attaches to his body all the ornaments he can get, and that 
he adorns all the parts of his body that can bear an orna- 
ment."2 
It is true that the things which people attach to their 
bodies are not all ornaments, even though they look as if 
they were.3 They may be trophies of war or of the chase, 
intended to show off strength, courage, or skill, or sub- 
stitutes for trophies having some resemblance to them.4 
They may be valuables worn chiefly for producing an im- 
pression of wealth.5 They may be badges, or indications 
of social status-like the necklaces of iron and ear-rings 
worn by Masai women in order that it shall be known that 
they are married.6 They may be signs of mourning-like 
the cassowary feathers worn by widowers among the Koita 
of Port Moresby.7 They may be charms for good luck-for 
example to secure favour with women or courage in war.8 
They may be used for curative purposes or serve as a pro- 
tection against illness, malevolent spirits, bewitching, or 
1 Sherwill, ' Tour through the RAjmahal Hills,' in Jour. Asiatic 
Soc. Bengal, xx. 584. 2 Grosse, Beginnings of Art, p. 84. 
8 See Finck, Primitive Love and Love Stories, p. 233 sqq. 
4 Spencer, op. cit. ii. 183 sqq. 
5 v. Barth, Ost-Afrika vom Limpopo bis zum Somalilande, p. 32. 
Schneider, Die Naturvolker, i. 107. Hirn, Origins of Art, p. 222. 
c Hollis, Masai, p. 282. 
7 Seligman, Melanesians of British New Guinea, p. 166. 
8 Parkinson, Dreissig Jahre in der Siidsee, p. 146 sqq. XV PRIMITIVE MEANS OF ATTRACTION 
499 
the evil eye.1 Among the East African Nandi, for instance, 
if a person dies his next younger brother or sister has to 
wear a certain ornament for ever after to escape being 
attacked by " the evil spirit or disease " ; 2 and in Morocco 
it is often impossible to distinguish between an ornament 
and an amulet. Certain gems, corals, and other ornamental 
objects are worn there as charms against the evil eye ; and 
the designs of Moorish decorative art largely serve the 
same purpose.3 But a recent American writer is certainly 
mistaken in his statement that when we subject the primi- 
tive custom of " ornamentation " to a critical examination 
we find in nearly every case that it is either not at all or 
only indirectly connected with the relations of the sexes. 
He argues that, even in the few instances when the decora- 
tions of savages have subsequently come to be used as means 
of pleasing the women, " this was not as things of beauty 
but indirectly and unintentionally through their association 
with rank, wealth, distinction in war, prowess, and manly 
qualities in general."4 Very often such decorations have 
nothing whatever to do with either wealth or rank or 
manliness, and are worn by women as well as men. 
Dr. Karsten, again, suggests that the so-called ornament« 
of the South American Indians have been primarily 
used " for purely religious and magical ends, either to 
protect the wearer and to give him strength, or directly to 
conjure and exorcise the supernatural enemies by whom 
the Indian constantly believes himself to be surrounded " ; 
and that their character of being means of decoration is 
a secondary development.5 In many instances he has 
undoubtedly proved his case. But I think it is an indis- 
putable fact that savages, at present at least, practise 
ornamentation on a large scale as a sexual allurement 
1 See, e.g., Hollis, Nandi, p. 29 ; v. Rosen, Traskfclket, p. 302 
(Batwa of North-Eastern Rhodesia) ; Hirn, op. cit. p. 224 ; Finck, 
op. cit. p. 236 sqq. Karsten, Studies in South American Anthropology, 
i. passim. 
8 Hollis, Nandi, p. 29. 
3 See Westermarck, * Magic Origin of Moorish Designs,' in Jour. 
Anthr. Inst, xxxiv. 213 sqq. 
4 Finck, op. cit. pp. 232, 259. 5 Karsten, op. cit. i. 121. 500 
THE HISTORY OF HUMAN MARRIAGE chap. 
Burton, in his * Anatomy of Melancholy,' said long 
ago that " stronger provocations proceed from outward 
ornaments, than such as nature hath provided":1 and 
this is true of savage as well as civilised man. It is 
significant that in all parts of the world the desire 
for self-decoration is strongest in youth. Concerning 
the Dakota, Prescott states that both sexes adorn them- 
selves at their courtships to make themselves more attrac- 
tive, and that " the young only are addicted to dress."2 
The Oraon of Bengal is likewise particular about his personal 
appearance " only so long as he is unmarried."3 Among 
the Let-htas in Indo-China it is the unmarried youths who 
are profusely bedecked with red and white bead necklaces, 
wild boars' tusks, brass armlets, and a broad band of black 
braid below the knee.4 Among the Nagas of Assam, again, 
the women in some cases put aside on marriage the orna- 
ments which as girls it was their privilege to wear.5 Among 
the Bangerang tribe of Victoria the youth at about sixteen 
years of age " affected the use of red ochre, adorned his head 
with plumes, and made himself an opossum-rug, which he 
scored and coloured in the approved way. ... To please 
the ladies was evidently the chief object of his solicitude."6 
Speaking of the Encounter Bay tribe of South Australia, 
the Rev. A. Meyer says that " the plucking out of the beard 
and anointing with grease and ochre (which belong to the 
initiatory ceremony) the men may continue if they please 
till about forty years of age, for they consider it ornamental, 
and fancy that it makes them look younger, and gives them 
an importance in the eyes of the women."7 When Bulmer 
once asked an Australian native why he wore his ornaments, 
the latter answered " that he wore them in order to look well, 
1 Burton, Anatomy of Melancholy, iii. 2. 2. 3, p. 522. 
2 Schoolcraft, Indian Tribes of the United States, iii. 237 sq. 
8 Dalton, Descriptive Ethnology of Bengal, p. 249 sq. 
4 Colquhoun, Amongst the Shans, p. 76. 
5 Hodson, Naga Tribes of Manipur, p. 77. 
6 Curr, Recollections of Squatting in Victoria, p. 254. 
7 Meyer, ' Manners and Customs of the Aborigines of the En- 
counter Bay Tribe,' in Woods, Native Tribes of South Australia 
P 189. XV PRIMITIVE MEANS OF ATTRACTION 
501 
and to make himself agreeable to the women."1 In Fiji, 
says Mr. Anderson, the men " who like to attract the atten- 
tion of the opposite sex don their best plumage."2 Among 
the Wataveta " you.ig men ornament themselves pro- 
fusely " ; 3 and the young Thonga who goes, with two or 
three of his friends, to look for a wife in the villages " puts 
on his most brilliant ornaments and his most precious 
skins."4 Among the Fors, again, the girls " wear bells on 
their ankles and bangles round their necks and waists, and 
perfume themselves extensively in the hope of attracting 
lovers." 5 In British Nigeria, " when an unbetrothed girl 
arrives at the age of puberty, she advertises the fact by deck- 
ing herself out in her best clothes and ornaments, and 
parading the town or village, accompanied by a bevy of 
girls. This usually produces a suitor."6 Cruickshank says 
that on the Gold Coast " both men and women are at the 
greatest pains to render themselves mutually agreeable to 
each other. With this view, they decorate themselves with 
much care, and indulge in the greatest coquetry."7 At 
dances and festivals of many savages the young men and 
women endeavour to please each other-by painting them- 
selves with brilliant colours and decorating themselves with 
1 Brough Smyth, Aborigines of Victoria, i. 275. 
2 Anderson, Notes of Travel in Fiji and New Caledonia, p. 136. 
3 New, Life, Wanderings, &c. in Eastern Africa, p. 357. 
4 Junod, Life of a South African Tribe, i. 102 sq. 
5 Felkin, ' Notes on the For Tribe of Central Africa,' in Proceed. 
Roy. Soc. Edinburgh, xiii. 233. On the erotic influence of odour and 
the use of perfumes as sexual stimulants see Ellis, Studies in the 
Psychology of Sex, (vol. iv.) Sexual Selection in Man, p. 91 sqq. ; 
Stoll, Das Geschlechtsleben in der Volkerpsychologie, p. 802 sqq. Dr. 
Ellis observes that nearly all the perfumes of animal origin in use 
by civilised man are odours which have a specifically sexual object 
among the animals from which they are derived. This is particularly 
the case with musk-the most cherished perfume of the Islamic 
world and still the most popular of European perfumes-which is an 
odour chiefly emitted during the sexual season. Indeed, the sexual 
odours of most animals seem to be modifications of musk. 
6 Mockler-Ferryman, British Nigeria, p. 232. See also Adams, 
Sketches taken during Ten Voyages to Africa, p. 7 sq. (Fanti). 
7 Cruickshank, Eighteen Years on the Gold Coast, ii. 212. 502 
THE HISTORY OF HUMAN MARRIAGE chap. 
all sorts of ornaments.1 Thus among the Akikuyu in East 
Africa the young men come to the moonlight dances where 
the girls choose their own partners properly adorned and 
turned out, because if they did not, as they say, " none of 
the girls would dance with them."2 
In order to attach some object to their bodies many 
savages submit to more or less painful operations. The 
lips, the nose, and the lobes of the ears are especially ill- 
treated.3 The Shulis in the Egyptian Sudan bore a hole in 
the under-lip and insert in it a piece of crystal three or four 
inches long, which sways about as they speak.4 In Eastern 
Central Africa women wear a lip-ring ; they say it makes 
them look pretty, and " the bigger the ring, the more they 
value themselves." 5 Similar practices are found in other 
parts of Africa,® where they are likewise considered to 
improve the appearance,7 and among various tribes in 
America.8 Among the Pawumwa Indians of Brazil, as 
soon as the girls are old enough to be courted, they have their 
lips pierced and a plug is inserted in either lip, which is a 
sign that the boys and unmarried men may court the girl. 
1 Harmon, Journal of Voyages and Travels in the Interior of North 
America, p. 305 (Takulli). Hardenburg, Putumayo, p. 161 sq. 
(Huitoto Indians). Wallace, Travels on the Amazon, p. 281 (Uaup6s). 
Dalton, op. cit. p. 250 (Orions). Waitz-Gerland, Anthropologie der 
Naturvdlker, vi. 604 (Ysanel Islanders). Turner, Samoa, p. 121. 
Finsch, Neu-Guinea und seine Bewohner, p. 139 (Papuans of Hum- 
boldt Bay). 2 Routledge, With a Prehistoric People, p. 124. 
3 Stoll, op. cit. p. 98 sqq. Hovorka, ' Verzierungen der Nase,' 
in Mittheil. Anthrop. Gesellsch. Wien, xxv. 155 sqq Mathew, 
' Beitrage zur Ethnographic der Australier,' ibid, xxxvii. 18 sq. 
Harrison, ' On the Artificial Enlargement of the Earlobe,' in Jour. 
Anthr. Inst. ii. 190 sqq. 
4 Wilson and Felkin, Uganda and the Egyptian Soudan, ii. 62. 
6 Macdonald, Africana, i. 17. 
6 Barth, Reisen und Entdeckungen in Nord- und Central-Afrika, 
ii. 514. Livingstone, Missionary Travels and Researches in South 
Africa, p. 577. 
7 Petherick, Travels in Central Africa, i. 248 (Djour). Ankermann, 
Kulturkreise und Kulturschichten in Afrika,' in Zeitschr. f. Ethnol. 
xxxvii. 64. 
8 v. Martins, Beitrdge zur Ethnographic Amerika's zumal Brasiliens, 
i. 351. Wallace, Travels on the Amazon, p. 514. XV PRIMITIVE MEANS OF ATTRACTION 
503 
When they are married, the plug of the lower lip is taken out, 
a larger incision is made, and a piece of polished white 
quartz is inserted in the lip ; this serves as a kind of 
marriage certificate.1 In the north-western part of North 
America, when boys and girls approach puberty, they have 
their lower lip perforated for the labret;2 and Holmberg, 
who is probably our best authority on the Tlingit, says that 
the men undergo this operation to make themselves agree- 
able to the young women.3 It is about the same age that 
among various savages the septum of the nose is pierced for 
the reception of a piece of bone, wood, or shell, or some other 
small object ;4 and this, too, is said to be done for the sake 
of adornment.6 Of the Banaka people on the Gaboon 
Coast in West Africa Mr. Leighton Wilson writes:- 
" Their women disfigure their faces very much by making 
large holes in their ears, and through the cartilaginous parts 
of the nose. Weights are attached to make the hole large 
enough to pass the finger through. Pieces of fat meat are 
frequently worn in these holes, but whether for ornament 
or fragrance is not known. 1 inquired of one of them once 
why she did it, and received the laconic answer, * My 
husband likes it.' "6 To pierce, enlarge, or somehow 
mutilate the ear-lobes is a very widespread practice. Some 
American7 and African8 savages pull them down almost to 
1 Haseman, ' Some Notes on the Pawumwa Indians of South 
America,' in American Anthropologist, N. S. xiv. 341. 
2 Armstrong, Personal Narrative of the Discovery of the North-West 
Passage, p. 194. Lisiansky, Voyage round the World, p. 243. v. 
Langsdorf, Voyages and Travels, ii. 115. Dixon, Voyage round the 
World, p. 187. 
8 Holmberg, * Ethnographische Skizzen uber die Volker des 
russischen Amerika,' in Acta Soc. Scient. Fennicae, iv. 301. 
4 Franklin, Second Expedition to the Shores of the Polar Sea, p. 118 
(Eskimo). Holub, Seven Years in South Africa, i. 35 (Masarwas). 
Angas, Savage Life and Scenes in Australia, ii. 225. Seligman, 
Melanesians of British New Guinea, pp. 72, 73, 256, 489. 
6 Mathew, Two Representative Tribes of Queensland, p. 108. 
Eylmann, Die Eingeborenen der Kolonie Siidaustralien, p. 108. 
0 Wilson, Western Africa, p. 288. 
7 Carver, Travels through the Interior Parts of North America, 
p. 227/ v. Martius, op. cit. i. 319 (Botocudos), 620 (Arecunas). 
8 Johnston, Kilima-njaro Expedition, p. 429 sq. (Wataita). 504 
THE HISTORY OF HUMAN MARRIAGE chap. 
the shoulders. Among the Easter Islanders, says Beechey, 
" the lobe, deprived of its ear-ring, hangs dangling against 
the neck, and has a very disagreeable appearance, particu- 
larly when wet. It is sometimes so long as to be greatly in 
the way; to obviate which, they pass the lobe over the 
upper part of the ear, or more rarely, fasten one lobe to the 
other, at the back of the head."1 The Warega of the 
Belgian Congo look upon the enormous holes in the ear- 
lobes of their women as very ornamental.2 
The object of these practices is certainly obscure. Con- 
sidering that the parts of the body which are subject to them 
are its main entrance gates-the mouth, the nose, and the 
ears-it is tempting to surmise that their object may have 
been to provide the body with a guard, in the form of a per- 
manent charm, against the attacks of invisible enemies, 
whether spirits or impersonal evil influences, ever ready to 
enter and cause sickness or destruction or other injury.3 
I remember an occasion in Morocco when my Berber teacher, 
horrified by the foul odour emitted by a dog, closed with 
his fingers not only the nose but the ears, in order to shut 
out what is looked upon by the Moors as a sort of obnoxious 
miasma associated with the jnun, or evil spirits. Mr. Crawley 
suggests that the insertion of plugs or sticks or the like in 
the nose, lips, or ears was probably intended to secure the 
safety of those sense-organs by diverting evil influences 
from them " as lightning is diverted from an object by the 
lightning-rod " ; 4 but in the absence of any direct evidence 
for this suggestion 1 find it difficult to believe that those 
organs, which are not particularly sensitive or exposed to 
injury, would for the sake of their own safety be subjected 
to the painful process necessary for the reception of the 
amulet. Nor can I agree with Mr. Crawley that there is 
ample evidence that " savage mutilation " is never due to 
the desire for oan.ment, the savage being averse to the 
doctrine, " Il faut souffrir pour etre belle." 5 On the other 
hand, there is ample evidence that the practices in question 
1 Beechey, Voyage to the Pacific and Beering's Strait, i. 38. 
2 Delhaise, Les Warega, p. 170. 3 See Karsten, op. cit. i. 122 sqq. 
4 Crawley, Mystic Rose, pp. 135, 138. 5 Ibid. p. 135. XV PRIMITIVE MEANS OF ATTRACTION 
505 
are nowadays looked upon as ornamental, whatever may have 
been their original object. And as any theory as to their 
origin is merely conjectural, there is no justification for the 
dogmatic assertion that "it is erroneous to attribute* these 
practices to the desire for ornament." 
The teeth are often subject to ill-treatment of some kind 
or other and various reasons for the custom are in different 
cases given by the natives who practise it. The Dieri in 
South-Eastern Australia, on being asked why they knock 
out two front teeth of the upper jaw of their children, 
answer that when they were created the good spirit Mura- 
mura did this to the first child and, pleased at the sight, 
commanded that the like should be done to every male or 
female child for ever after.2 The savages of Northern 
Formosa knock out the eye-teeth of all the children when they 
reach the age of six or eight years, " in the belief that it 
strengthens their speed and wind in hunting."3 The Bantu 
Kavirondo, both men and women, usually pull out the two 
middle incisor teeth in the lower jaw, as it is thought that 
if a man retains all his lower incisor teeth he will be killed in 
warfare, and that if his wife has failed to pull out her teeth 
it might cause her husband to perish.4 The Nilotic tribes 
of Kavirondo, again, draw the six middle incisors in the 
lower jaw; and if a man does not submit to this operation 
it is said that his wife will die soon after marriage.5 But 
more often we are told that the extraction or mutilation of 
teeth is considered to improve the appearance. This holds 
true of various African peoples 6-for example, the Boloki 
1 See Stoll, op. cit. p. 247 sqq. ; Mathews, * Beitrage zur Ethno- 
graphic der Australier,' in Mitteil. Anthrop. Gesellsch. Wien, xxxvii. 
19 sq.; v. Jhering, 'Die kiinstliche Deformirung der Zahne,' in 
Zeitschr. f. Ethnol. xiv. 213 sqq.', Frazer, Totemism and Exogamy, iv. 
180 sqq. 
2 Gason, ' Manners and Customs of the Dieyerie Tribe,' in Woods, 
op. cit. p. 267. 
3 Taintor, Aborigines of Northern Formosa, p. 22. 
4 Johnston, Uganda Protectorate, p. 728. Hobley, Eastern 
Uganda, p. 20. 
6 Hobley, op. cit. p. 31. 
8 Ankermann, in Zeitschr. f. Ethnol. xxxvii. 64. 506 
THE HISTORY OF HUMAN MARRIAGE chap. 
on the Upper Congo River, who chisel the upper incisors 
to V-shaped points,1 and the Herero, who pull out the four 
middle teeth in the lower jaw and file the eye-teeth into the 
form of a swallow's tail.2 Of the former operation, as 
practised by the men among some of the Congo natives, 
Tuckey says that it is " principally done with the idea of 
rendering themselves agreeable to the women."3 Dr. Holub 
states that one of the Makalaka tribes, north of the Zambesi, 
and the Matongas, on its bank, " break out their top incisor- 
teeth from the sheerest vanity. Their women say that it is 
only horses that eat with all their teeth, and that men ought 
not to eat like horses."4 The Bagonjo in Uganda, who 
file the front teeth to a point, say that this helps them to 
laugh well; but " in days gone by the maiden who had 
her teeth unfiled was not desired in marriage."5 Speaking 
of the Malays of Naning; north of the Malacca lands, 
Newbold says that " the Malays consider the process 
of filing down the teeth, together with the subsequent 
operation of blackening them, indispensable for personal 
beauty " ;6 and Crawfurd makes a similar remark with 
reference to the same practice in the Malay Archipelago, 
where it is a necessary prelude to marriage, the common 
way of expressing the fact that a girl has arrived at puberty 
being that " she has had her teeth filed."7 In other coun- 
tries also the filing or mutilation of teeth takes place at that 
age.8 Among the Kadars, a wild jungle tribe in the Cochin 
State in South India, the operation of chipping " all or some 
1 Weeks, Among Congo Cannibals, p. 141. 
2 Dannert, * Uber die Sitte der Zahnverstiimmelung bei den Ova- 
herero,' in Zeitschr. f. Ethnol. xxxix. 948, 949, 952. ' Zahnver- 
stiimmelung der Hereros,' in Verhandl. Berliner Gesellsch. Anthrop. 
1908, p. 930 sqq. v. Francois, Nama und Damara Deutsch-Sud-West- 
Afrika, p. 198. 
3 Tuckey, Expedition to explore the River Zaire, p. 80 sq. 
4 Holub, Seven Years in South Africa, ii. 256. 
5 Cunningham, Uganda, p. 262. 
8 Newbold, Account of the British Settlements in the Straits of 
Malacca, i. 253. 
7 Crawfurd, History of the Indian Archipelago, i. 215 sqq. 
8 v. Francois, op. cit. p. 198 (Herero). Roscoe, Northern Bantu, 
p. 78 (Banyoro). See also v. Jhering, in Zeitschr. f. Ethnol. xiv. 218. XV PRIMITIVE MEANS OF ATTRACTION 
507 
of the incisor teeth, both upper and lower, into the form of 
sharp pointed, but not serrated cone," is performed on boys 
at the age of eighteen and girls at the age of ten or there- 
abouts.1 Among the Nicobarese the men blacken their 
teeth from the period of puberty, and this disfigurement is 
so favourably regarded by the fair sex that a woman "would 
scorn to accept the addresses of one possessing white teeth, 
like a dog or pig."2 It is at the same period of life that the 
Pelew Islanders,3 the natives of New Britain,4 and the 
Chaymas of New Andalusia,5 have their teeth blackened, 
which is considered an indispensable condition of beauty. 
Many savage men take particular pride in their hair- 
dress. Sometimes the hair is painted in a showy manner, 
sometimes decorated with beads and tinsel, sometimes 
combed and arranged with the most exquisite care. The 
Kandhs of Central India have their hair, which is worn very 
long, drawn forward and rolled up till it looks like a horn 
projecting from between the eyes. Around this it is their 
delight to wear a piece of red cloth, and they insert the 
feathers of favourite birds, as also a pipe, comb, &c.6 The 
men of Tanna, of the New Hebrides, wear their hair " twelve 
and eighteen inches long, and have it divided into some six 
or seven hundred little locks or tresses " ;7 and among the 
Latuka a man requires a period of from eight to ten years to 
perfect his coiffure.8 In North America Hearne saw several 
men, about six feet high, who had preserved " a single lock 
of their hair that, when let down, would trail on the ground 
as they walked." 9 Other Indians practise the custom of 
1 Anantha Krishna Iyer, Cochin Tribes and Castes, i. 24 sq. 
2 Man, ' Account of the Nicobar Islanders,' in Jour. Anthr. 
Inst. xv. 441. 
3 Dahlgren, ' Om Palau-oarna,' in Ymer, iv. 317. 
4 Powell, Wanderings in a Wild Country, p. 254. 
6 v. Humboldt, Personal Narrative of Travels to the Equinoctial 
Regions of the New Continent, iii. 224. 
0 Dalton, op. cit. p. 301. 
7 Turner, Samoa, p. 308. 
8 Baker, Albert N'yanxa, i. 198. 
a Hearne, Journey from Prince of Wales's Fort to the Northern 
Ocean, p. 306 note. 508 
THE HISTORY OF HUMAN MARRIAGE chap. 
shaving the head and ornamenting it with the crest 
of deer's hairs; and wigs are used by several savage 
peoples.1 
Fashions relating to the hair-dress may certainly owe 
their origin to different motives. Sometimes, as Frazer has 
pointed out, the savage may arrange his hair so as to assimi- 
late himself to his totem and put himself more fully under 
its protection. Thus the Buffalo clans of the Iowa and 
Omaha wear two locks of hair in imitation of horns ; the 
Small Bird clan of the Omaha " leave a little hair in front, 
over the forehead, for a bill, and some at the back of the head, 
for the bird's tail, with much over each ear for the wings " ; 
and the Turtle subclan cut off all the hair from a boy's 
head, except six locks which are arranged so as to imitate 
the legs, head, and tail of a turtle.2 Dr. Karsten believes 
that the hair customs of the South American Indians are 
based on the idea " that the hair is, as it were, the seat of 
the spirit or soul." 3 But there is abundant evidence that 
the hair-dress serves as a sexual stimulant. 
It is the young and unmarried who pay the greatest atten- 
tion to it. Among the Acholi in the Nile Province the young 
men, " the bucks of society," are much more elaborately 
ornamented than the old men and chiefs, the hair-dress 
being their distinguishing feature ; it consists of "a curiously 
worked cone of matted hair, with beads neatly stitched in 
a pattern round it, and an empty cartridge-case stuck in 
at the top."4 Young unmarried Zulus allow their hair to 
grow, and dress it in a variety of fantastic shapes ; " now 
it looks like a sugar loaf, now like two little hills with 
valleys between."5 Among the Bunjogees, a Chittagong 
Hill tribe, the young men " stuff a large ball of black cotton 
into their topknot to make it look bigger."6 In the Tenimber 
Group of the Malay Archipelago the lads decorate their 
long locks with leaves, flowers, and feathers, as Riedel says, 
1 Catlin, North American Indians, ii. 23. 
2 Frazer, Totemism, p. 26 sqq. 3 Karsten, op. cit. i. 59. 
4 Cunningham, Uganda, p. 352. 
5 Tyler, Forty Years among the Zulus, p. 60. 
6 Lewin, Wild Races of South-Eastern India, p. 240. XV PRIMITIVE MEANS OF ATTRACTION 
509 
" only in order to please the women."1 In the Admiralty 
Islands, according to Professor Moseley, " only the young 
men of apparently from eighteen to thirty, or so, wear the 
hair long and combed out into a mop or bush," whilst the 
boys and older men wear the hair short.2 Among the 
Takulli, an ethnic group of Athapascan Indians, " the 
elderly people neglect to ornament their heads, in the same 
manner as they do the rest of their persons, and generally 
wear their hair short. But the younger people of both 
sexes, who feel more solicitous to make themselves agreeable 
to each other, wash and paint their faces, and let their hair 
grow long."3 Sibree tells us that when king Radama 
attempted to introduce European customs among the 
Hovas of Madagascar, and ordered that all his officers and 
1 Riedel, De sluik- en kroesharige rassen tusschen Selebes en Papua, 
p. 292. 
2 Moseley, ' Inhabitants of the Admiralty Islands,' in Jour. 
Anthr. Inst. vi. 400. 
8 Harmon, Journal of Voyages and Travels in the Interior of 
North America, p. 288. Short hair is often regarded as a symbol of 
chastity. Every Buddhist " novice "-that is, a person admitted 
to the first degree of monkhood-has to cut off his hair, in order to 
prove that " he is ready to give up the most beautiful and highly- 
prized of all his ornaments for the sake of a religious life " (Monier 
Williams, Buddhism, p. 306) ; and in Mexico the religious virgins, 
as also men who decided upon a life of chastity, had their hair cut 
(Acosta, Natural and Moral History of the Indies, ii. 333 ; Bancroft. 
Native Races of the Pacific States of North America, ii. 251 sq.). A 
similar idea probably underlies the custom which requires that 
women, when they marry, shall be deprived of their hair, the husband 
trying in this way to preserve the fidelity of his wife (see Wilkes, 
Narrative of the United States Exploring Expedition, iii. 354 ; Waitz- 
Gerland, Anthropologic der Naturvblker, vi. 567 ; Palmer, * Notes on 
some Australian Tribes,' in Jour. Anthr. Inst. xiii. 286 ; de Rubru- 
quis, ' Travels into Tartary and China,' in Pinkerton, Collection of 
Voyages and Travels, vii. 32 ; Heriot, Travels through the Canadas, 
p. 335). This custom is not only found among savages but prevailed 
at Sparta and Athens (Rossbach, Untersuchungen uber die romische 
Ehe, p. 290) and among the Anglo-Saxons (Wright, W omankind in 
Western Europe, p. 68). Among the Tlascalans of ancient Mexico 
it was customary to shave the head of a newly-married couple, both 
man and woman, " to denote that all youthful sports ought in that 
state to be abandoned " (Heriot, op. cit. p. 333L 510 
THE HISTORY OF HUMAN MARRIAGE chap. 
soldiers should have their hair cut, this command produced 
so great a disturbance among the women of the capital that 
they assembled in great numbers to protest against the 
king's order, and could not be quieted till they were sur- 
rounded by troops and their leaders cruelly speared.1 
Scarcely anything has a greater attraction for the savage 
than showy colours. " No matter," says Dr. Holub, " how 
ill a traveller in the Marutse district may be, and how 
many bearers he may require, if he only has a good stock of 
blue beads he may always be sure of commanding the best 
attention and of securing the amplest services ; his beads will 
prove an attraction irresistible to sovereign and subject, to 
man, woman, and child, to freeman and bondman alike."2 
The practice of ornamenting one's self with gaudy baubles 
is indeed extremely prevalent. Of Santal men at a feast 
Sir W. Hunter says that, " if all the colours of the rainbow 
were not displayed by them, certainly the hedgehog, the 
peacock, and a variety of the feathered tribe had been laid 
under contribution in order to supply the young Santal 
beaux with plumes." 3 Especially does the savage delight 
in paint. We know that the practice of applying paint to 
the body existed in Europe already at the end of the 
Quaternary and during the Neolithic periods.4 Red ochre 
is generally looked upon as the chief embellishment. 
Goethe praises the incomparable power of yellowish-red 
over the emotions, and says that " the fondness of savage 
men for it has been noticed everywhere." 5 Yet, as Grosse 
observes, painting with red would hardly have been so 
generally diffused in the lowest stages of civilisation if the 
red colouring material had not been everywhere so easily 
and so abundantly procurable.6 Of the other colours 
black and white are probably most in use. The Dakota 
paint their faces red and black, " which they esteem as 
1 Sibree, The Great African Island, p. 211. 
2 Holub, op. cit. ii. 351. 8 Hunter, Rural Bengal, i. 185. 
* D^chelette, ' La peinture corporelle et le tatouage,' in Revue 
archdologique, ser. iv. vol. ix. 38 sqq. 
5 Goethe, Zur Farbenlehre, § 775, vol. i. 293. 
8 Grosse, Beginning of Art, p. 62 sq. xv PRIMITIVE MEANS OF ATTRACTION 
511 
greatly ornamental."1 Among the Guaycurus many men 
paint their bodies half red, half white.2 Throughout the 
Australian . continent the natives stain themselves with 
black, red, yellow, and white.3 In Fiji a small quantity of 
vermilion is esteemed " as the greatest possible acquisi- 
tion."4 
The practice of painting the body may also, no doubt, be 
traced to different motives. Its object may be to frighten 
enemies in battle,5 or to protect the body from the weather 
and from insects.6 Among the Hottentots, for instance, the 
coat of grease and clay was intended for the latter purpose 
as well as for ornament.7 The Cross River natives have their 
bodies painted partly for decorative purposes but partly 
also as " medicine." 8 In time of war the men of most 
Upper Congo tribes blacken their faces and necks with 
palm-oil and powdered charcoal in resemblance to a certain 
species of monkey, and explain that by so doing they derive 
"monkey cunning."9 The figures which the Kiwai Papuans 
in British New Guinea paint on their bodies when they are 
dancing may symbolise their totems.30 And many savages 
paint themselves for mourning, according to Frazer with a 
view to disguising themselves from the ghost of the dead 
man.11 In Morocco the well-known henna, a colouring 
1 Carver, op. cit. p. 227. 2 v. Martins, op. cit. i. 230. 
3 Waitz-Gerland, op. cit. vi. 738. 4 Wilkes, op. cit. iii. 356. 
■' v. Martius, op. cit. i. 321, 738. Joest, Tdtoivien, N arbenzeichnen 
und Kbvperbemalen, p. 19 sq. 
6 Bancroft, op. cit. i. 159. Heriot, op. cit. p. 305. von den Steinen, 
Unter den Naturvblkern Zentral-Brasiliens, p. 185 sq. Hahn, 'Die 
Ovaherero,' in Zeitschr. Gesellsch. Erdkunde Berlin, iv. 248. Joest, 
op. cit. p. 18 sq. 
7 Theal, Yellow and Dark-skinned People of Africa south of the 
Zambesi, p. 74. 
8 Partridge, Cross River Natives, p. 168. 
9 Ward, Voice from the Congo, p. 258. 
10 Landtman, Nya Guinea far den, p. 80. See also Frazer, Totemism, 
P 30. 
11 Frazer, * Certain Burial Customs as illustrative of the Primitive 
Theory of the Soul,' in Jour. Anthr. Inst. xv. 73. Idem, ' Folk-Lore 
in the Old Testament,' in Anthropological Essays presented to E. B. 
Tylor, p. no. 512 
THE HISTORY OF HUMAN MARRIAGE chap. 
matter produced from the leaves of the Egyptian privet 
(Lawsonia inermis), is not merely a favourite cosmetic 
among the women, but is also frequently used as a means of 
protection against evil influences on account of the bar aka, 
or holiness, ascribed to it. So also there is bar aka in walnut 
root or bark (swak), with which women paint their lips and 
teeth brownish, and in antimony (khol), with which they 
paint their eyes black; and though chiefly used by the 
women, all these paints are also on special occasions, for 
ritual purposes, used by men.1 Dr. Karsten maintains 
that among the South American Indians the painting of 
the body was originally intended to be a prophylactic 
against evil spirits.2 But with reference to the Toba girls' 
custom of painting their faces he writes himself ;-" When 
I inquired as to the reason for this, I sometimes simply 
received the answer that it was considered beautiful, some- 
times that it was done to attract the men. The girls paint 
themselves, I was told, when they are desirous of a man."3 
That the painting of the body is regarded as an em- 
bellishment or sexual attraction is suggested by various 
other statements, in addition to those already referred to. 
Among the Guarayos in South-Western Amazonia a suitor 
paints himself from head to foot, and armed with his battle 
club promenades for several days round the cabin of the 
mistress of his heart, until on a day of feasting and 
dancing the marriage is consummated.4 Among the 
Matacos of the Gran Chaco a man paints himself with red, 
blue, and black when he is courting a girl,5 and among 
the Onas of Tierra del Fuego he on a similar 
occasion paints small white spots in his face.6 Among 
the Witoto and Boro, of the North-West Amazons, 
men and women are painted, or paint themselves, before 
1 Westermarck, ' Popular Ritual of the Great Feast in Morocco,' 
in Folk-Lore, xxii. 132 sqq. Idem, Marriage Ceremonies in Morocco, 
passim (see Index, s.v. ' Antimony,' ' Henna,' * Walnut root or bark '). 
Idem, Moorish Conception of Holiness (Baraka), p. 66 sq. 
2 Karsten, op. cit. i. 8 sqq. 3 Ibid. i. 20. 
4 v. Martius, op. cit. i. 217. Church, Aborigines of South America, 
P- 115. 5 Baldrich, quoted by Karsten, op. cit. i. 26 n. 2. 
11 Gallardo Tierra del Fuego-Los Onas, p. 15r. XV PRIMITIVE MEANS OF ATTRACTION 
513 
a dance,1 and the same is done in other savage tribes.2 Thus 
the Easter Islanders paint themselves-chiefly with red 
and white-only at dances and festivals, which are occasions 
for great licentiousness.3 Among the Aht of British 
Columbia " some of the young men streak their faces with 
red, but grown-up men seldom now use paint, unless on 
particular occasions," and the women cease to use it about 
the age of twenty-five.4 Speaking of the North American 
Indians in general, Mr. Sinclair observes that they considered 
their body-painting to enhance the beauty of both men and 
women.5 Of the aborigines of the Upper and Middle Finke 
River in Central Australia, Schulze wrote that " to beautify 
themselves they rub their bodies with grease to make the 
skin shine, and try to ornament it by painting, and the 
application of tufts and cords ";6 and in his description of 
the South Australian natives in general Dr. Eylmann states 
that the desire to please is undoubtedly the chief motive for 
their practice of painting themselves.7 On Flinders Island, 
whither the remnant of the Tasmanian race were removed, 
a rebellion nearly burst out when orders were once issued 
forbidding the use of ochre and grease, for " the young men 
feared the loss of favour in the eyes of their countrywomen.''8 
Sparrman tells us that the two Hottentots whom he had in 
his service, when they expected to meet some girls of their 
own nation, painted their noses, cheeks, and the middle of 
the forehead with soot.9 Among the Bayaka of Congo 
both sexes paint the chest with cam-wood, confessedly to 
improve their appearance;10 whilst the Bambala beaux 
1 Whifien, North-West Amazons, p. 88 sq. 
2 Supra, p. 501 sq. 3 Geiseler, Die Oster-Insel, pp. 35, 41. 
4 Sproat, Scenes and Studies of Savage Life, p. 28. 
5 Sinclair, ' Tattooing of the North American Indians,' in American 
Anthropologist, N.S. xi. 399. 
6 Schulze, * Aborigines of the Upper and Middle Finke River,' 
in Trans, and Proceed. Roy. Soc. South Australia, xiv. 223. 
7 Eylmann, op. cit. p. 387. 
8 Bonwick, Daily Life of the Tasmanians, p. 25 sq. 
9 Sparrman, Voyage to the Cape of Good Hope, ii. 80. 
10 Torday, Camp and Tramp in African Wilds, p. 131. Torday 
and Joyce1, * Notes on the Ethnography of the Ba-Yaka,' in Jour. 
Anthr. Inst, xxxvi. 41. See also I idem, Les Bushongo, p. 279. 
VOL. I L L 514 
THE HISTORY OF HUMAN MARRIAGE chap. 
and belles paint their bodies red with clay, likewise for the 
express purpose of increasing their beauty.1 
Closely connected with the painting of the body is the 
practice of tattooing, which existed in Europe both in pre- 
historic2 and historic 3 times, and is extremely widespread 
among modern savages. There is hardly any visible part 
of the human body, except the eyeball, that has not been 
subject to it. The Sandwich Islanders tattooed the top 
of the head, the eyelids, the ears, the nose, and sometimes 
even the tip of the tongue ;4 the Easter Islanders the edges 
of the ears and the fleshy part of the bps ;5 the Abyssinian 
women the gums.6 Tattooing has the advantage over 
body-painting that the marks are indelible and permanent, 
whereas the latter has to be repeated from time to time. 
At the same time those marks are a certificate of courage.7 
Speaking of the tattooing of the Polynesians, Mr. Macmillan 
Brown observes that " the transition from temporary 
painting to the permanency of tattooing must have been 
greatly aided by the necessity of courage in the endurance 
of pain that is entailed, as well as the perennial desire of the 
human heart to be ' beautiful for ever.' "8 The Anda- 
manese, says Mr. Man, consider tattooing " primarily, as 
ornamental, and secondly, as proving the courage of the 
individual, and his (or her) power of enduring pain."8 
Sometimes it serves the object of inspiring fear in the enemy ; 
according to Crozet, the Maori chiefs studied to invent 
1 Torday, op. cit. p. 80. Joest says (op. cit. p. 20), " Eitelkeit ist 
die Mutter der Korperbemalung, ihr Zweck war und ist ein kosme 
tischer, sie ist ein Act der geschlechtlichen Zuchtwahl." I became 
acquainted with his work only after the present chapter of the 
first edition had been prepared for press. 
2 Dechelette, in Revue archeologique, ser. iv. vol. ix. 38 sqq. 
3 Herodotus, v. 6 (Thracians). Caesar, De bello Gallico, v. 14 
(Britons). See Lacassagne, Les tatouages, p. 9. 
1 Freycinet, Voyage autour du monde, ii. 580. 
5 Beechey, op. cit. i. 39. 0 Parkyns, Life in Abyssinia, ii. 29. 
7 Franklin, Journey to the Shores of the Polar Sea, p. 71. Bock, 
Temples and Elephants, p. 170. Dalton, op. cit. p. 251. 
8 Brown, Maori and Polynesian, p. 184. 
9 Man, ' Aboriginal Inhabitants of the Andaman Islands,' in 
Jour. Anthr. Inst. xii. 331. XV PRIMITIVE MEANS OF ATTRACTION 
515 
designs which made them look horrible and gave them a 
most fearful look.1 Sometimes also the tattooing records 
achievements in war ; Keyser speaks of a chief in New 
Guinea who had sixty-three blue tattoo lines on his chest 
which represented the number of enemies he had slain,2 
and Finsch says that among the Motu of Port Moresby only 
men who had killed somebody were allowed to tattoo their 
chests.3 In Uea, or Wallis Island, to the west of the Samoa 
Group, tattooing in the men " was always a mark of dignity, 
and sometimes the recompense of valour."4 Among the 
Tsalisens of Formosa it is permitted to no other men than 
chiefs and members of their families, whom it serves " both 
as an ornament and as a sign of high rank."5 D'Urville 
professed to see in the tattooing of the Maori an analogy to 
European heraldry, although it illustrated only the merits 
of the person decorated with it, not those of his ancestors; 
and we know that some distinctive part of it was a mark of 
identity, used in place of a signature.6 Among the North 
American Indians, whilst certain devices could be worn 
only for valiant deeds, other designs were tattooed on slaves.7 
In ancient Rome masters tattooed their slaves, especially 
such as were working on the fields, in order to find them in 
case they escaped ; and during the reign of Honorius certain 
classes of labourers engaged in public work were tattooed 
for a similar purpose.8 In Easter Island a young married 
man tattooed the vulva of his wife on his chest as a sign 
that he was married.9 The tattooed marks are in various 
1 Crozet, Voyage to Tasmania, &c., p. 39. See also Thomson, 
Story of New Zealand, i. 77 ; Brown, op. cit. p. 190. 
2 Keyser, Our Cruise to New Guinea, p. 44 sq. 
3 Finsch, Bekleidung, Schmuck und Tatowirung der Papuas der 
Sudostkiiste von Neu-Guinea, p. 22. Of. Seligman, Melanesians 
of British New Guinea, p. 130 (Koita). 
4 Mangeret, Mgr. Bataillon et les missions de I'Oceanie, i. 97. 
5 Davidson, Island of Formosa, p. 572. 
6 Robley, Moko', or Maori Tattooing, pp. 17, 18, 10 sqq. 
7 Sinclair, in American Anthropologist, N.S. xi. 399. 
8 Berchon, in Actes de VAcademic nationale des sciences, belles- 
lettres et arts de Bordeaux, 1885, p. 804. 
9 Geiseler, op. cit. p. 29. 516 
THE HISTORY OF HUMAN MARRIAGE chap. 
cases represented as tribal tokens or means of distinguishing 
clansmen from enemies j1 but it should be noticed that 
certain designs which are common to the whole tribe or 
clan easily come to be looked upon in that light, although 
they were not originally intended to be marks of recogni 
tion.2 Tattooing is further practised as a remedy for 
disease, or as a prophylactic against it.3 The Western 
Dene believed that one or two transversal lines tattooed 
on the arms or legs of a young man by a pubescent girl 
prevented premature weakness of these limbs.4 In Morocco 
a man whom I knew had a tattoo mark underneath either 
temple which was intended to prevent him from sleeping too 
much. In the same country many tattoo marks, represent- 
ing an eye or the five fingers, are obviously in their origin 
charms against the evil eye, even when they are nowadays 
looked upon as ornaments;5and elsewhere also certain tattoos 
are used as magical means of protection against similar 
influences6 or against dangers in war7 or as charms for good 
1 Agassiz, Journey in Brazil, p. 318. v. Martins, op. cit. i. 484, 501, 
&c. Ehrenreich, Beitrdge zur Volkerkunde Brasiliens, p. 11. 
Beechey, op. cit. ii. 401. Powers, Tribes of California, p. 109. 
Mackenzie, Voyages from Montreal to the Frozen and Pacific Oceans, 
p. cxx. Waitz-Gerland, op. cit. vi. 38. Klose, Togo unter deutscher 
Flagge, p. 480 sq. Ellis, Ewe-speaking Peoples of the Slave Coast, 
p. 146. Fiilleborn, Das Deutsche Njassa- und Ruwuma-Gebiet, 
PP- 75. 77- 
2 Cf. Fiilleborn, op. cit. pp. 75, 77 ; Livingstone, Last Journals, 
i. 49- 
3 Seligman, op. cit. p. 493 (Southern Massim). v. Rosen, Trask- 
folket, p. 301 (Batwa of North-Eastern Rhodesia). Fiilleborn, op. cit. 
p. 78 (natives of "German East Africa"). Baumann, Usambara, 
p. 222 (Wapara). Marquordt, ' Bericht iiber die Kavirondo,' in 
Zeitschr. f. Ethnol. xli. 755. Fouquet, article reviewed in L'Anthro- 
pologic, viii. 496 (Copts). Mallery, ' Picture-Writing of the 
American Indians,' in Ann. Rep. Bur. Ethnol. x. 418. 
* Morice, ' Notes on the Western D6n£s,' in Trans. Canadian 
Institute, iv. 182. 
5 See my article on the ' Magic Origin of Moorish Designs,' in 
Jour. Anthr. Inst, xxxiv. 221. 
0 Marquardt, loc. cit. p. 755. 
7 Ibid. p. 755. Mallery, loc. cit. p. 418. Cf. Colquhoun, Amongst 
the Shans, p. 213. XV PRIMITIVE MEANS OF ATTRACTION 
517 
luck in the chase1 or success in love or other advantages.2 
Some North American Indians have their totems tattooed 
on their persons; 3 but Gerland's theory that tattoos were 
originally figures of totem animals, though they are no 
longer so,4 is an arbitrary assumption. Even where they 
obviously represent animals, they may be quite unconnected 
with totemism ; the Punjab " practice of tattooing a scor- 
pion, a snake, a bee or a spider has its origin in sympathetic 
magic, which is supposed to protect people so marked/'5 
Marquardt observes that not one of the animals which the 
Samoans regarded as sacred figures in their tattoo patterns.6 
In some parts of West Africa every child is immediately 
after birth dedicated to some definite fetish by having a 
mark tattooed on its abdomen ;7 and the native priests of 
Dahomey have hierarchical tattoo marks showing to what 
god they are vowed and what rank they hold, and all these 
marks are considered so sacred that no layman is allowed to 
touch them.8 The eunuch priests of Attis, again, were 
tattooed with a pattern of ivy leaves.9 But it is only in 
rare cases that tattooing is known to have any religious 
significance. Cook's statement that in the South Sea 
Islands it was in no way connected with religion10 has been 
supported by various writers who have studied the practice 
on the spot.11 The Fijian belief that it was instituted by the 
1 Fiilleborn, op. cit. p. 78. 
2 Mallery, loo. cit. p. 418. Rose, ' Note on Female Tattooing in 
the Panjab,' in Indian Antiquary, xxxi. 297. 
3 Frazer, Totemism, p. 28 sq. * Waitz-Gerland, op. cit. vi. 36 sqq. 
5 Rose, in Indian Antiquary, xxxi. 298. 
6 Marquardt, Die Tatowirung beider Geschlechter in Samoa, p. 18. 
7 Bastian, Afrikanische Reisen, p. 71. 
8 Ellis, Ewe-speaking Peoples of the Slave Coast, p. 146. 
9 Frazer, Adonis Attis Osiris, i. 278. 
10 Waitz-Gerland, op. cit. vi. 38. 
11 Kubary, ' Die Bewohner der Mortlock Inseln,' in Mittheil. 
Geograph. Gesellsch. Hamburg, 1878-79, p. 240. Finsch, ' Ueber die 
Bewohner von Ponapd,' in Zeitschr. f. Ethnol. xii. 308 sq. Joest, 
op. cit. p. 78. Marquardt, op. cit. p. 16 sqq. Dr. Karsten (op. cit. 
i. 219) writes with reference to the South American Indians that 
" even in our own days most Indians are, no doubt, aware of tattoo- 
ing being in its essence a religious custom, believing that the indelible 
marks and patterns which they wear in the face or on other parts of 518 
THE HISTORY OF HUMAN MARRIAGE chap. 
god Ndengei and that its neglect will be punished after death,1 
and similar stories found among some other peoples,2 tell 
us nothing about the origin of the custom, since any usage 
practised from time immemorial may readily be ascribed to 
the command of a god. But the Tahitians have a legend 
which deserves being quoted in full because it seems to give 
us some real insight into the matter. 
Taaroa, their god, and Apouvaru had a daughter, who 
was called Hinaereeremonoi. " As she grew up, in order to 
preserve her chastity, she was made ' pahio,' or kept in a 
kind of enclosure, and constantly attended by her mother. 
Intent on her seduction, the brothers invented tattooing, 
and marked each other with the figure called Taomaro. 
Thus ornamented, they appeared before their sister, who 
admired the figures, and, in order to be tattooed herself, 
eluding the care of her mother, broke the enclosure that had 
been erected for her preservation, was tattooed, and became 
also the victim to the designs of her brothers. Tattooing 
thus originated among the gods, and was first practised by 
the children of Taaroa, their principal deity. In imitation 
of their example, and for the accomplishment of the same 
purposes, it was practised among men. . .The two sons 
of Taaroa and Apouvaru were the gods of tattooing. Their 
images were kept in the temples of those who practised the 
art professionally, and every application of their skill was 
preceded by a prayer addressed to them, that the operation 
might not occasion death, that the wounds might soon heal, 
that the figures might be handsome, attract admirers, and 
answer the ends of wickedness designed."3 
the body will protect them against all sorts of evil influences." But 
this seems to be a conjecture rather than a statement of observed 
facts. 
1 Williams and Calvert, Fiji, p. 138. Pritchard, Polynesian 
Reminiscences, p. 391. Seemann, Viti, p. 113. Wilkes, op. cit. 
iii. 355. Thomson, Fijians, p. 218. 
2 Wilkes, op. cit. v. 88 (Gilbert Islanders), v. Siebold, Ethno- 
logische Studien liber die Aino auf der Insel Yesso, p. 15. Egede, 
Description of Greenland, p. 132 sq. ; Nordenskiold, Den andra Dick- 
sonska expeditionen till Gronland, p. 468 (Greenlanders). 
3 Ellis, Polynesian Researches, i. 262 sq. XV PRIMITIVE MEANS OF ATTRACTION 
519 
Tattooing may, as we have seen, serve a great variety of 
purposes, but in most of the instances mentioned above it 
is restricted to one or a few marks with a specific meaning. 
When practised on a larger scale it is generally considered 
to improve the appearance of the person subject to it, and 
we have reason to believe that in such cases it is practised 
with this object in view. Kubary,1 Finsch,2 Joest,3 Mar- 
quardt,4 Kramer,5 and others agree that in the islands of the 
Pacific Ocean it is principally, if not exclusively, a means 
of embellishment; and from various other parts of the world 
we hear it spoken of as an ornament.6 The North American 
Indians, says Mr. Sinclair, " regarded it as enhancing the 
beauty of the fair sex, and the good looks of the braves."7 
Even to the European taste the incised lines and figures 
have in many cases a certain beauty. Speaking of the 
Gambier Islanders, Beechey assures us that their tattooing 
undoubtedly improves their appearance.8 Yate remarks 
that " nothing can exceed the beautiful regularity with 
which the faces and thighs of the New Zealanders are 
tattooed," the volutes being perfect specimens, and the 
regularity mechanically correct.9 Forster observed that 
among the natives of Waitahoo (Marquesas Islands) the 
punctures were disposed with the utmost care, so that the 
marks on each leg, arm, and cheek and on the corresponding 
1 Kubary, in Mittheil. Geograph. Gesellsch. Hamburg, 1878-79, 
p. 240. 
2 Finsch, in Zeitschr. f. Ethnol. xii. 308. Idem, Ueber Bekleidung, 
&c., p. 16. Idem, in Joest, op. cit. p. 39. 
3 Joest, op. cit. p. 53 sqq. 4 Marquardt, op. cit. p. 16. 
5 Kramer, Die Samoa-Inseln, ii. 65, 66 n. 1. See also Friedlaender, 
' Notizen uber Samoa,' in Zeitschr. f. Ethnol. xxxi. 25 sq. In Uea 
tattooing was simply an ornament in the women (Mangeret, op. cit. 
i. 97). 
6 Riedel, op. cit. p. 280 (natives of Tenimber and Timorlaut). 
Davidson, op. cit. p. 565 (Atayals of Formosa). Hobley, op. cit. 
p. 31 (Nilotic tribes of Kavirondo). Klose, op. cit. p, 480 sq. 
(Bassari in Togoland). Delhaise, Les -Warega, p. 170. Fiilleborn, 
op. cit. p. 75 (natives of "German East Africa "); &c. 
7 Sinclair, in American Anthropologist, N.S. xi. 399. 
8 Beechey, op. cit. i. 139. 
9 Yate, Account of New Zealand, p. 147 sq. 520 
THE HISTORY OF HUMAN MARRIAGE chap. 
muscles were exactly similar.1 Among the Tahitians, 
according to Darwin, the ornaments follow the curvature 
of the body so gracefully, that they have a very pleasing 
and elegant effect;2 and among the Easter Islanders " all 
the lines were drawn with much taste, and carried in the 
direction of the muscle."3 The fact that the tattooed lines 
follow closely the natural forms of the body in order to 
render them more conspicuous has been observed in the 
case of other peoples also.4 
The beauty which the tattooing is intended to produce, 
however, is by no means purely aesthetical: it is meant to 
serve the purpose indicated in the Tahitian legend-to be 
a sexual lure. Among the Maori, according to Taylor, it 
was the great ambition of the young men to have fine 
tattooed faces, " both to render themselves attractive to the 
ladies, and conspicuous in war." 5 A girl would seldom 
accept an untattooed man; to have a smooth unmarked 
face was to have enormous odds against one in love- 
making.6 And the profusely tattooed thighs and lower part 
of the back must also have proved very fascinating; for a 
Maori said to one of our informants, " The women do like 
to see a chap tattooed that way."7 The girls, again, had 
1 Forster, Voyage round the World, ii. 14 sq. 
2 Darwin, Journal of Researches, p. 481 sq. 
3 Beechey, op. cit. i. 39. 
4 Waitz-Gerland, op. cit. vi. 573. Jones, Grammar of Ornament, 
p. 13 note. Cf. the tattooed circle round the mouth of the Juris 
(Wallace, Travels on the Amazon, p. 510) and the female Arecunas 
(Brett, Indian Tribes of Guiana, p. 268) ; the rings round the eyes of 
the women in the Admiralty Islands (Moseley, in Jour. Anthr. Inst. 
vi. 401), of the Australians (Angas, South Australia Illustrated}, 
and the Patagonians (King and Fitzroy, Voyages of the Adventure and 
Beagle, ii. 135) ; the cicatrices like parallel ridges upon the chest, 
thighs, and shoulders of the Tasmanians (Bonwick, op. cit. p. 24) ; 
and the tattoos on the hands and feet of Egyptian women (Lane, 
Manners and Customs'of the Modern Egyptians, i. 54, 57). 
6 Taylor, Te Ika a Maui, p. 321. See also Robley, op. cit. p. 22. 
8 Tregear, Maori Race, p. 291. Cf. ibid. p. 258. 
r Rutland, ' Survivals of Ancient Customs in Oceania,' in Jour. 
Polynesian Soc. xiii. 102. xv PRIMITIVE MEANS OF ATTRACTION 
521 
usually horizontal blue lines tattooed on their lips before 
marriage, because " red lips were not considered sightly, 
but were disliked."1 Speaking of Polynesian tattooing 
generally, Mr. Macmillan Brown remarks that one of the 
motifs of it was sexual: "it was intended to increase the 
influence of the individual over the imagination of the other 
sex, an intention brought out more clearly by the body 
tattooings of the Samoans, Tongans, and other islanders."2 
In Samoa, according to Marquardt, " the men tattoo them- 
selves in order to please the women and the women in order 
to increase their attraction for the men ; they all do it simply 
to make themselves more beautiful so as to become more 
sought after."3 Until a young man was tattooed he could 
not think of marriage, but as soon as it was done he considered 
himself entitled to all the privileges of mature years.4 
" When it is all over," says Pritchard, " and the youths 
thoroughly healed, a grand dance is got up on the first 
available pretext to display the tattooing, when the ad- 
miration of the fair sex is unsparingly bestowed. And this 
is the great reward, long and anxiously looked forward to 
by the youths as they smart under the hands of the 
matai." 5 On the island of Santa Anna, in the Solomons, 
a youth has likewise to be tattooed before he is at liberty to 
marry,6 and in the same group of islands this adornment is 
also " absolutely indispensable to the girl's prospect of 
being sought in marriage."7 In Florida, for example, " no 
girl would be considered marriageable unless tattooed."8 In 
Fiji, if a girl were untattooed at the age of puberty, " her 
peculiarity would be whispered with derision among the 
gallants of the district, and she would have difficulty in 
finding a husband." When questioned upon the matter, 
1 Best, ' Maori Marriage Customs,* in Trans, and Proceed. New 
Zealand Institute, xxxvi. 64. Tregear, op. cit. p. 264. 
2 Brown, op. cit. p. 187 sq. See also ibid. p. 183 sq. 
3 Marquardt, op. cit. p. 16. See also ibid. p. 8. 
4 Turner, Samoa, p. 88. 6 Pritchard, op. cit. p. 144 sq. 
G Guppy, Solomon Islands, p. 136. 
7 Penny, Ten Years in Melanesia, p. 91. 
8 Codrington, Melanesians, p. 237. Cf. ibid. p. 234. 522 
THE HISTORY OF HUMAN MARRIAGE chap. 
a chief declared to Sir Basil Thomson that " physically 
there was the greatest difference in the world between 
mating with a tattooed and an untattooed woman, and that 
the idea of marriage with an untattooed woman filled him 
with disgust." Sir Basil, however, believes that the 
reason for the practice must be sought for in some unknown 
sexual superstition and not in ornamental purposes, because 
the marks were limited to a broad horizontal band covering 
those parts that were concealed by the liku, or apron ;1 
but this argument seems by no means conclusive, con- 
sidering that the tattooing may be older than the liku, and 
even apart from this, Sir Basil says himself that in 
addition to the tattooing of the buttocks barbed lines and 
dots were marked upon the fingers of young girls to display 
them to advantage when handing food to the chiefs, and 
that the raised cicatrices which may still be seen upon the 
arms and backs of the women " are intended only for 
ornament, and have no other significance."2 The Southern 
Massim in British New Guinea, whose women are always 
profusely tattooed, state as reason for this " that it makes 
the girl look nice and accentuates her good skin. ' '3 Speaking 
of some savages of the same colony, Chalmers remarks that 
the women tattoo themselves " to please the men " ;4 and 
in Samarai, a small island outside the south-eastern point 
of New Guinea, the tattooing of the young girls is said by 
Finsch to serve no other object but to make it easier for them 
to find a husband.5 In the Kei Islands, to the south-west 
of New Guinea, it is, on the other hand, according to Riedel, 
the men who " tattoo themselves in order to please the 
women." 6 When Mertens asked the natives of Lukunor, 
in the Carolines, what was the meaning of their tattooing, 
one of them answered, " It has the same object as your 
clothes, that is. to please the women."7 Bock remarks, " As 
1 Thomson, op. cit. p. 217 sqq. 2 Ibid. pp. 218, 220. 
3 Seligman, op. cit. p. 493. 
4 Chalmers, Pioneering in New Guinea, p. 166. 
5 Finsch, in Joest, op. cit. p. 41. 0 Riedel, op. cit. p. 228. 
7 Waitz-Gerland, op. cit. vol. v. pt. ii. 67. XV PRIMITIVE MEANS OF ATTRACTION 
523 
the Dyak women are tattooed to please their lowers, 
so the Laos men undergo the ordeal for the sake of the 
women."1 
In Morocco, although tattoo marks may be made, for 
example, to make a man a good shot, or to cure a swollen 
knee, or to act as charms against the evil eye, more elabo- 
rate patterns, at least, are uniformly regarded as orna- 
ments. A Berber from the tribe Aith Sadden, near Fez, 
who had a large tattoo on his right hand, told me that he 
had it made when he was a boy in order to be pleasing to 
the women ; and he added that many young men of his 
tribe have the right or left hand and the lower part of the 
forearm, as also one of their shoulders, tattooed for the 
same purpose, so as to find favour with the women with- 
out paying them anything. A beautifully tattooed girl 
is praised in their songs, and attracts many lovers who pay 
her well; and when given in marriage she fetches a high 
bride price. 
Among the Banabuddu, on the shores of the Victoria 
Nyanza, where the men used to tattoo a network of design 
on their breasts and the women during their youth had the 
same design made on their stomachs, a woman who refused 
having this done to her " would not be equally attractive 
to the opposite sex."2 Dr. Fiillebom observes that sexual 
considerations seem to be an important reason for tattooing 
among the Wahyao and Makua in " German East Africa," 
whose women are profusely tattooed on that part of the body 
which is concealed by their loin-cloth. When he asked 
for an explanation of this custom he was told " that it gives 
the man a more agreeable sensation to pass his hand over a 
surface covered with cicatrices than over a smooth one."3 
Young Kadiak wives are said to " secure the affectionate 
admiration of their husbands by tattooing the breast and 
adorning the face with black lines."4 Cranz states that 
among the Greenlanders the mother tattoos her daughter 
1 Bock, Temples and Elephants, p. 170. 
2 Cunningham, Uganda, p. 68 sq. 3 Fiillebom, op. cit. p. 78. 
4 Bancroft, op. cit. i. 72. 524 
THE HISTORY OF HUMAN MARRIAGE chap. 
in childhood because she is afraid that the girl will other- 
wise attract no husband.1 It is a common custom that the 
tattooing of either men or women is performed about the 
age of puberty.2 But among many peoples it is not accom- 
plished at once, so that the patient shall be able to bear the 
inflammation and pain at every stage of the process; and 
not infrequently it begins when the girls are quite 
young children, being constantly added to until they 
marry.3 Among the Nagas of Upper Assam it was the custom 
" to allow matrimony to those only who made themselves 
as hideous as possible by having their faces elaborately 
tattooed."4 The Makalaka girls in South Africa could 
not marry until about four thousand stitches had been 
made in the skin of the chest and stomach and a black fluid 
had been rubbed into the wounds.5 In Samoa great 
licentiousness was connected with the custom of tattooing ;6 
1 Cranz, History of Greenland, i. 129. 
2 Ellis, Polynesian Researches, i. 262 (Tahitians). Pritchard, 
op. cit. p. 393 (Tongans). Kramer, op. cit. ii. 63 sq. (Samoans). 
Thomson, op. cit. p. 217 (Fijians). Wilkes, op. cit. ii. 141 
(Samoans); iii. 355 (Fijians); v. 103 (Gilbert Islanders). Montgomery, 
Journal of Voyages and Travels by Tyerman and Bennet, i. 127 
(natives of Eimeo). v. Langsdorf, op. cit. i. 118 (Nukahivans). 
Beechey, op. cit. i. 139 (Gambier Islanders). Seligman, op. cit. 
p. 493 (Southern Massim). Bock, Head-Hunters of Borneo, p. 189 
(Dyaks). Schadenberg, ' Die Negritos der Philippinen,' in Zeitschr.f. 
EthnoL xii. 136. Davidson, op. cit. p. 565 (Atayals of Formosa). 
Fryer, Khyeng People of the Sandoway District, Arakan, p. 4. Forsyth, 
Highlands of Central India, p. 148 (Gonds). Wilson and Felkin, 
Uganda and the Egyptian Soudan, ii. 97 (Baris). Dobrizhoffer, 
Account of the Abipones, ii. 20. v. Martius, op. cit. i. 217 (Guarayos). 
Bancroft, op. cit. i. 47 (Eskimo), 98 (Tlingit). 
8 Ellis, Polynesian Researches, i. 262 (Tahitians). Angas, Poly- 
nesia, p. 328 (Marquesas Islanders). Idem, Savage Life and 
Scenes in Australia and New Zealand, i. 314 (Maori). Mont- 
gomery, op. cit. i. 127 (natives of Eimeo). Seligman, op. cit. p. 264 sq. 
(Roro-speaking tribes of British New Guinea). Man, in Jour. Anthr. 
Inst. xii. 331 (Andaman Islanders). Fytche, Burma, ii. 61. St. 
John, ' Ainos,' in Jour. Anthr. Inst. ii. 249. 
4 Dalton, op. cit. p. 39. 
6 Mauch, Reisen im Inneren von Sud-Afrika, p. 38 sq. 
6 Turner, op. cit. p. 90. XV PRIMITIVE MEANS OF ATTRACTION 
525 
and in Tahiti the chiefs prohibited it altogether on 
account of the obscene practices by which it was invariably 
accompanied in that island.1 
Under the general heading " tattooing " is often included 
the practice of scarification, and in the present connection 
it is unnecessary to draw a very rigid distinction between 
the two processes. Their ethnical distribution largely 
depends on the colour of the skin. " Scarification," says 
Grosse, " has found practice only among dark-skinned 
peoples, for the scars stand out plainly only on a dark skin. 
For a similar reason, tattooing has spread only among the 
fairer peoples."2 Warriors make scars on their bodies to 
record the number of enemies they have slain in battle.8 
1 Ellis, Polynesian Researches, i. 226.-In the work referred to 
above, Joest also expresses the opinion that the principal desire that 
induces men and women to tattoo themselves is to increase their 
attractions in the eyes of the opposite sex (see particularly p. 56)- 
Mr. Finck, on the other hand, in a violent attack on my view, writes 
(op. cit. p. 248) that the use of tattooing " as a stimulant of the passion 
of the opposite sex probably never occurred to a savage until it was 
suggested to him by a philosophising visitor." In support of this 
discourteous statement he gives a number of quotations relating 
to tattooing, in none of which he finds the slightest reference to the 
purpose of beautifying the body in order to attract the other sex ; 
and he says (ibid. p. 254) that this purpose is not even hinted at in 
the summary made by Mallery of the seventeen different motives 
to which the latter has been able to trace the practice of tattooing. Mr. 
Finck's own list is of course no evidence whatever, being anything but 
complete, and his representation of Mallery's views is grotesquely 
misleading. What Mallery really says (loc. cit. p. 418) is, that tattooing 
appears to be used, or to have been used very recently, for many 
purposes " besides the specific object of designating a tribe, clan, 
or family, and also apart from the general intent of personal orna- 
ment," and that the most notable of " those purposes " are the 
seventeen which he then enumerates. The importance Mallery 
attaches to " the general intent of personal ornament " is evident 
from the following statement:-" The mere idea of decoration as 
shown in what civilised people call deformations of nose, lip, ear, 
teeth, and in fact all parts of the body, is sufficient to account for the 
inception of any form of tattoo." 
2 Grosse, op. cit. p. 78. See also Parkinson, op. cit. p. 487. 
3 Stow, Native Races of South Africa, p. 535 (Bangw aketse). 526 
THE HISTORY OF HUMAN MARRIAGE chap. 
Scars or cicatrices are used as tribal marks,1 as a remedy for 
illness,2 as prophylactic charms,3 as mourning tokens4 or 
an expression of grief.5 Among the South American 
Indians, according to Dr. Karsten, scarification is practised 
as a means of ridding the body of evil spirits that have 
entered into it.6 
Most often, however, I believe, scarification is said to be 
partly or exclusively a kind of decoration.7 Some tribes 
of Madagascar, for instance, are in the habit of making marks, 
" which are intended to be ornamental," by slight incisions 
in the skin.8 The scarification which is practised by the 
natives throughout the Australian continent, in a large 
measure at least, serves the purpose of adornment.9 Among 
the South Australian Aldolinga the scars raised chiefly 
across the breast and the upper portion of the arms " are 
simply made to beautify their bodies."10 Barrington assures 
1 Ward, Voice from the Congo, p. 255. Mockler-Ferryman, 
British Nigeria, p. 229. Baumann, Usambara, p. 180. Howitt, 
Native Tribes of South-East Australia, p. 744. Brough Smyth, 
Aborigines of Victoria, i. 296. Mathew, Eaglehawk and Crow, p. 121 
sq. Mrs. Langloh Parker, Euahlayi Tribe, p. 125. 
2 Roscoe, Northern Bantu, p. 78 (Banyoro). Speiser, Two Years 
with the Natives in the Western Pacific, p. 112 sq. (natives of Port 
Olry in the New Hebrides). 
3 Howitt, op. cit. p. 746 (Yuin tribe). Hobley, op. cit. p. 20 
(Bantu Kavirondo). 4 Stair, Old Samoa, p. 117. 
5 Mathew, Two Representative Tribes of Queensland, p. 107. 
G Karsten, op. cit. i. 168 sqq. 
7 Theal, Yellow and Dark-skinned People of Africa south of the 
Zambesi, p. 237 sq. (Bantu). Stannus, 'Notes on some Tribes of 
British Central Africa,' in Jour. Roy. Anthr. Inst. xl. 317 sq. Barrett, 
' Notes on the Customs and Beliefs of the Wa-Giriama, etc., British 
East Africa,' ibid. xli. 23. Hobley, op. cit. p. 20 (Bantu Kavi- 
rondo). Speiser, op. cit. p. 112 sq. (natives of Port Olry in the New 
Hebrides). Turner, op. cit. p. 310 (natives of Tanna). Stair, op. cit. 
p. 116 (Samoans). Cf. Grosse, op. cit. p. 75. 
8 Sibree, op. cit. p. 210. 
8 Curr, Australian Race, i. 402 ; ii. 475. Eylmann, op. cit. p. 108 
sqq. Howitt, op. cit. p. 744. Brough Smyth, op. cit. i. 296. Mrs. 
Langloh Parker, op. cit. p. 125. Mathew, Eaglehawk and Crow, p. 121. 
Idem, Two Representative Tribes of Queensland, p. 106 sq. 
10 Krichauff, ' Customs of the " Aldolinga," ' in Proceed. Roy. 
Geograph. Soc. Australasia: South Australian Branch, ii. 79 sq. XV PRIMITIVE MEANS OF ATTRACTION 
527 
us that among the natives of Botany Bay " scars are, by 
both sexes, deemed highly ornamental."1 And the raised 
cuts of the tribes of New South Wales described by Palmer 
are said by him to be " merely ornamental and convey no 
idea of tribal connection," the women marking themselves 
in this manner " to add to their looks, and to make them- 
selves attractive."2 In some parts of the Solomons, 
Parkinson tells us, " a well healed scarification, which clearly 
shows the lines of the design, is the greatest ornament of 
men and women; the latter rise in price according to the 
beauty of the design."3 In Fiji " rows of wart-like spots 
are burned along the arms and backs of the women, which 
they and their admirers call ornamental."4 
There are thus various groups of very widespread practices 
which in a large measure must be regarded as means of 
stimulating the sexual instinct of the opposite sex. They 
may, as we have seen, serve other objects as well; and I 
have no doubt that superstitious beliefs, in particular, 
have played an important part in their origin. Even 
the fact that these practices are most prevalent about the 
age of puberty, as they would naturally be in their capacity 
of sexual stimulants, is probably more or less connected with 
superstitious ideas relating to the phenomena of pubescence, 
or possibly, as Professor Him suggests, with the change of 
social status accompanying sexual maturity.5 But it is 
both inconsistent and absurd to accept, like Mr. Finck, all 
other explanations of their origin and at the same time 
declare that the remarks of travellers regarding the addic- 
tion of savages to personal " ornamentation " are merely 
" the unwarranted assumptions of superficial observers, 
who, ignorant of the real reasons why the lower races paint, 
tattoo, and otherwise ' adorn ' themselves, recklessly in- 
ferred that they did it to ' make themselves beautiful.' " 6 
1 Barrington, History of New South Wales, p. n. 
2 Palmer,' Notes on some Australian Tribes,' in Jour. Anthr. Inst. 
xiii. 286. 
8 Parkinson, op. cit. p. 487. 
4 Williams and Calvert, op. cit. p. 137. 
5 Hirn, op. cit. p. 220. 8 Finck, op. cit. p. 232. 528 
THE HISTORY OF HUMAN MARRIAGE chap. 
In some cases these practices may stimulate the sexual 
instinct as expressions of strength, courage, or endurance,1 
but generally, I presume, they do so as means of attracting 
attention, of exciting through the charm of novelty,2 and, 
particularly, of expressing sexual desire or sexual intentions 
in those who adorn themselves. If, as Robert Burton put 
it, natural beauty is a much stronger loadstone when " those 
artificial enticements and provocations of gestures, clothes, 
jewels, pigments, exornations, shall be annexed unto it "3- 
the main reason for this is that they are invitations to love. 
Thus we can understand that even practices so repugnant 
to our taste as that of perforating the septum of the nose 
or removing teeth may be attractive, whilst their omission, 
if they have once been established as genuine customs, 
becomes positively repulsive. In Cochin China, where the 
women blacken their teeth, a man said contemptuously of 
the British Ambassador's wife that " she had white teeth 
like a dog " ;4 and the Abipones of Paraguay, who care- 
fully plucked out all the hairs with which our eyes are 
naturally protected, despised the Europeans for their thick 
eyebrows and called them brothers to the ostriches.5 But 
even where a certain kind of ornamentation has become 
general and is prescribed by custom for all the members of 
the community who are of the same sex, there is generally 
scope left for variety in the form of deviations or exaggera- 
tions,6 and thus the old custom may still remain a positive 
lure. As bright colours and other secondary sexual charac- 
ters in animals are marks of recognition and means of 
attraction, so self-decoration in mankind is a conscious 
indication of sexual feelings and therefore a sexual stimulant. 
And as display in animal courtship excites not only the 
1 Cf. Marquardt, op. cit. p. 16 sq , Tregear, op. cit. p. 258, and 
Brown, op. cit. p. 183 sq., with reference to tattooing. 
2 Walker observes {Beauty, p. 41) that " an essential condition of 
all excitement and action in animal bodies, is a greater or less 
degree of novelty in the objects impressing them." 
8 Burton, Anatomy of Melancholy, in. 2. 2. 3, p. 521. 
4 Waitz, Introduction to Anthropology, p. 305. 
6 Dobrizhoffer, op. cit. ii. 15. 
6 This is admitted even by Mr. Finck {op. cit. p. 245). XV PRIMITIVE MEANS OF ATTRACTION 
529 
individual courted but also the courter, so self-decoration 
for sexual purposes has in some measure a stimulating 
effect not only upon others but upon one's self. The native 
chief with whom Sir Basil Thomson discussed the tattooing 
of the Fijian women left him with the impression that, 
besides the other advantages the chief had mentioned, 
" tattooing was believed to stimulate the sexual passion 
in the woman herself."1 And I think it is a common 
experience among ourselves that there is self-excitement in 
finery calculated to make the person attractive to the 
opposite sex. Moreover, as the secondary sexual characters 
of animals chiefly occur in the males, so self-decoration in 
the savage world is very frequently more prevalent among 
men than women. The common notion that women are by 
nature vainer and more addicted to dressing and decorating 
themselves than men cannot be said to hold true of the 
lower races in general. 
Speaking of the Aleut on the Fur-Seal Islands of Alaska, 
Elliott says, " In these lower races there is much more vanity 
displayed by the masculine element than the feminine, 
according to my observation ; in other words, I have 
noticed a greater desire among the young men than among 
the young women of savage and semi-civilised people to be 
gaily dressed, and to look fine."2 Concerning the Cree, 
Mackenzie remarks that " the women, though by no means 
inattentive to the decoration of their own persons, appear 
to have a still greater degree of pride attending to the 
appearance of the men, whose faces are painted with more 
care than those of the women." 3 Among all the Indian 
tribes whom Richardson saw on his route through the 
northern parts of the fur countries, the women adorned their 
persons less than the men of the same tribes ;4 and the like 
1 Thomson, op. cit. p. 219. 
2 Elliott, ' Report on the Seal Islands of Alaska,' in Tenth Census 
of the United States, p. 21 sq. 
3 Mackenzie, op. cit. p. xciv. Cf. Harmon, Journal of Voyages 
and Travels in the Interior of North America, p. 319 sq. 
4 Franklin, Second Expedition to the Shores of the Polar Sea, 
p. 197. Cf. Mackenzie, op. cit. p. 126. 530 
THE HISTORY OF HUMAN MARRIAGE chap. 
is said of the Comanche.1 Among the Uaupes, Wallace 
observed that the men and boys appropriated all the 
ornaments.2 In Tierra del Fuego Lieutenant Bove found 
the men more desirous of ornaments than the women.3 
Proyart made a similar observation with regard to the 
Negroes of Loango ;4 and Schweinfurth5 and Barth,6 who 
had a vast experience of African races, agree that the women 
usually wear fewer ornaments than the men. Among the 
Gambier Islanders, according to Beechey, the women " have 
no ornaments of any kind, and appeared quite indifferent 
to the beads and trinkets which were offered them."7 The 
native women of Orangerie Bay of New Guinea, except 
that they are tattooed, adorn themselves less than the men, 
and none of them paint their faces and bodies, as the men 
frequently do.8 Among the Papuans of the Mimika dis- 
trict in the same island, according to Captain Rawling, the 
women are given to still less self-adornment than the men, 
and, with the exception of a few beads, wear nothing but the 
loin-cloth.8 In the Admiralty Islands young girls " some- 
times have a necklace or two on, but they never are decorated 
to the extent to which the men are," it being evidently not 
considered good taste for them to adorn their persons.10 
Among the aborigines of the New Hebrides,11 New 
1 Schoolcraft, Indian Tribes of the United States, i. 235. 
2 Wallace, Travels on the Amazon, p. 281. Cf. v. Martius, 
ofi. cit. i. 597. 
8 Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 129. 
4 Proyart, ' History of Loango,' in Pinkerton, Collection of 
Voyages and Travels, xvi. 575. 
6 Schweinfurth, Im Herzen von Afrika, ii. 7 sq. 
8 Barth, Reisen und Entdeckungen in Nord- und Central-Afrika, 
ii- 475- 7 Beechey, op. cit. i. 138. 
8 d'Albertis, New Guinea, i. 200. See also Miklucho-Maclay, 
' Ethnologische Bemerkungen uber die Papuas der Maclay-Kiiste 
in Neu-Guinea,' in Natuur kundig Tijdschrift voor Nederlandsch 
Indie, xxxvi. 296 sq. 
9 Rawling, Land of the New Guinea Pygmies, p. 57. 
10 Moseley, Notes by a Naturalist on the Challenger, p. 461. Idem, 
in Jour. Anthr. Inst. vi. 399. Romilly, Western Pacific and New 
Guinea, p. 115. 
11 Campbell, A Year in the New Hebrides, p. 145. XV PRIMITIVE MEANS OF ATTRACTION 
531 
Hanover,1 and the Australian continent 2 adornments are 
mostly monopolised by the male sex. Of the natives of 
Gippsland, for instance, it is said :-" The ornaments worn 
by the females were not much regarded by the men. The 
woman did little to improve her appearance ; ... if her 
physical aspect was such as to attract admirers she was 
content."3 
It has been suggested that the plainer appearance of the 
women depends upon their oppressed and despised position 
and the selfishness of the men.4 But it seems doubtful 
whether this is the true explanation, since, generally speak- 
ing, savage ornaments are not costly things. Of all the 
various kinds of self-ornamentation tattooing is the most 
laborious. Yet we find that among many savage peoples 
just this kind of self-decoration is more common or more 
elaborate among the women than among the men, or that 
it is entirely limited to the former.5 In Melanesia, for ex- 
ample, it is chiefly women who are tattooed, whereas in 
Polynesia it is chiefly men,6 although the Polynesian women, 
on the whole, occupy a higher social position than their 
Melanesian sisters.7 In Fiji, where women were fearfully 
oppressed, genuine tattooing was found on them only ;8 
but in Samoa, where they were held in much consideration,9 
1 Strauch, ' Bemerkungen uber Neu-Guinea, &c.,' in Zeitschr. f. 
Ethnol. ix. 43. 
2 Waitz-Gerland, op. cit. vi. 735. Breton, Excursions in New 
South Wales, &C., p. 210 sq. Bonwick, ' Australian Natives,' in Jour. 
Anthr. Inst. xvi. 204. Eylmann, op. cit. p. 404. Cf. Schulze, 
Aborigines of the Upper and Middle Finke River,' in Trans, and 
Proceed. Roy. Soc. South Australia, xiv. 229. 
3 Brough Smyth, op. cit. i. 275. 
4 Darwin, Descent of Man, ii. 372 sq. Avebury, Origin of Civili- 
sation, p. 43. Forster, op. cit. ii. 219. Mackenzie, op. cit. p. 126 sq. 
6 See, e.g., Frazer, Totemism and Exogamy, iv. 202 sqq.; Karsten, 
op. cit. i. 220 sqq. (South American Indians). 
6 Waitz-Gerland, op. cit. vi. 575. Rivers, History of Melanesian 
Society, ii. 436 sq. 
7 See Westermarck, Origin and Development of the Moral Ideas, 
i. 643 sq. ; Meinicke, Die Inseln des Stillen Oceans, i. 67. 
8 Mariner, Account of the Natives of the Tonga Islands, ii. 267. 
Williams and Calvert, op. cit. p. 145. Wilkes, op. cit. iii. 332. 
8 Wilkes, op. cit. ii. 148. Waitz-Gerland, op. cit. vi. 121. 532 
THE HISTORY OF HUMAN MARRIAGE chap. 
they were less extensively and less regularly tattooed than 
the men.1 
The comparative scarcity of female ornaments among 
many savages may to some extent be explained by Spencer's 
suggestion that ornaments partly derive their origin from 
trophy-badges, or Wundt's, that they indicate rank and 
fortune ; 2 but these explanations could only have a limited 
application. In so far as self-decoration is intended to 
stimulate the sexual instinct of the opposite sex, we must 
conclude that the greater desire of the men to make them- 
selves attractive by ornamentation is in the first place due 
to the likings of the women, and that the plainer appearance 
of the women is a consequence of the men's greater in- 
difference to their ornaments. Darwin has pointed out that 
among our domesticated quadrupeds individual antipathies 
and preferences are exhibited much more commonly by the 
female than by the male ; indeed, the male as a rule seems 
to be ready to pair with any female, provided she belongs to 
his own species.8 So also women seem generally to be more 
particular in their choice than men, when the union takes 
place without reference to interest. A Maori proverb says, 
" Let a man be ever so good-looking, he will not be much 
sought after ; but let a woman be ever so plain, men will still 
eagerly seek after her." 4 With regard to the Negroes of 
Sogno, Merolla da Sorrento wrote, " Women would have ex- 
perience of their husbands before they married them, in like 
manner as the men were to have of them; and in this particu- 
lar I can aver that they are commonly much more obstinate 
or fickle than men, for I have known many instances in which 
the men were willing to be married, while the women held 
back, and either fled away or made excuses."5 Among the 
Eastern Central Africans described by Mr. Duff Macdonald 
many cases are known of slave wives running away from 
1 Kramer, op. cit. ii. 63. 
8 Wundt, Ethik, i. 152 sqq. Cf. Hirn, op. cit. p. 220 sq.; Landt- 
man, Kulturens ursprungsformer, pp. 200, 202. 
3 Darwin, op. cit. ii. 290 sqq. * Taylor, op. cit. p. 293 sq. 
6 Merolla da Sorrento, ' Voyage to Congo,' in Pinkerton, op. cit. 
xvi. 236. XV PRIMITIVE MEANS OF ATTRACTION 
533 
free husbands, but none of slave husbands running away 
from free wives.1 In a subsequent chapter we shall see that 
endogamous restrictions, whether they have reference to 
race, nation, tribe, or rank, often apply to women only ; 
and there can be little doubt that this to some extent is 
due to the fact that the sexual instinct of women is more 
discriminating than that of men. If in countries with 
European civilisation women nevertheless are more particu- 
lar about their appearance and more addicted to self- 
decoration than the other sex, the reason for it may be 
sought for in the greater difficulty they have in getting 
married. But there is seldom any such difficulty in the 
savage world. Here it is, on the contrary, the man who 
runs the risk of being obliged to lead a single life. 
1 Macdonald, Africana, i. 141. CHAPTER XVI 
PRIMITIVE MEANS OF ATTRACTION 
(Concluded) 
We have still to consider a certain class of objects which 
men attach to their bodies, namely, such as fall under the 
heading of clothing. As the term implies, the purpose for 
which they are used is to clothe or cover the body, or some 
part of it; but the relation between self-decoration and 
clothing is very intimate. Not only are ornaments attached 
to clothes, but the latter are frequently used for purposes 
of ornamentation or allurement as well as covering the 
body. " The greatest provocations of lust are from our 
apparel," says Robert Burton j1 and Dr. Havelock Ellis 
justly observes that one of the chief sex allurements would 
be lost " and the extreme importance of clothes would 
disappear at once if the two sexes were to dress alike."2 
Indeed, it is in many cases impossible to decide whether a 
certain object attached to the body should be called clothing 
or ornament. And whilst clothes readily become decora- 
tions, there is, on the other hand, every reason to believe 
that mere decorations have also developed into clothes. 
The origin of clothing may no doubt be traced to several 
different sources. The most obvious one is the desire to 
protect the body from frost and damp.8 When man 
emigrated from his warm native home and settled down in 
less hospitable regions it became necessary for him to screen 
himself from the influences of a raw climate. The Eskimo 
1 Burton, Anatomy of Melancholy, iii. 2. 2. 3, p. 524. 
2 Ellis, Psychology of Sex, (vol. iv.) Sexual Selection in Man, p. 209. 
3 See Stoll, Das Geschlechtsleben in dev Volker psychologic, p. 484 sq. 
534 ch. xvi PRIMITIVE MEANS OF ATTRACTION 
535 
wrap themselves up in furs, and the wretched natives of 
Tierra del Fuego throw a piece of skin over their shoulders, 
or one of them, " on the side from which the wind blows."1 
But covering may also be used as a protection against 
influences of another kind. The custom of women veiling 
their faces, so common in Muhammadan countries, was 
probably adopted as a safeguard against the evil eye ;2 and 
in early Arabia very handsome men, too, protected them- 
selves in a similar manner, especially at feasts and fairs when 
they were particularly exposed to dangerous glances.8 The 
question which especially interests us in the present con- 
nection, however, is why man began to cover that part of 
the body which is most frequently concealed, although the 
rest of the body was left nude. 
In this case also one of the motives may have been the 
desire of protection. Some travellers among Brazilian 
Indians are of opinion that the string or strip of bast worn 
by certain tribes, partly at least, serves the object of pro- 
tecting the glans or the sexual mucous membrane against 
the attacks of insects or other injury ;4 and a similar 
suggestion has been made with reference to the tying up of 
the prepuce by the Nukahivans.5 But the dread of exposing 
the glans, which is found among some other South Sea 
1 Ross, Voyage of Discovery in the Southern and Antarctic Regions, 
ii. 285. Wilkes, Narrative of the United States Exploring Expedition, 
i. 121. 
2 See Westermarck, ' Magic Origin of Moorish Designs,' in J.our. 
A nt hr. Inst, xxxiv. 211. 
3 V7ellhausen, Reste des arabischen Heidentums, p. 196. 
4 von den Steinen, Unter den N aturvblkern Zentral-Brasiliens, 
p. 192 sqq. Krause (In den Wildnissen Brasiliens, p. 204) says that 
the explanation unanimously given by the KarayA is that the glans 
is tied up in order not to be seen by the women, but he thinks that 
the primary aim of the practice may be protection. For the practice 
itself see also Ehrenreich, Beitrdge zur Volkerkunde Brasiliens, pp. 11 
(Kar ay 4), 52 sq. (Yamamadi); Haseman, * Notes on the Pawumwa 
Indians of South America,' in American Anthropologist, N.S. xiv. 342. 
5 Fleurieu, quoted by v. Krusenstern, Voyage round the World, 
i. 156. See also Hovorka, ' Verstiimmelungen des mannlichen 
Gliedes bei einigen Vdlkern des Alterthums und der Jetztzeit,' in 
Mittheil. Anthrop. Gesellsch. Wien, xxiv. 140 sqq. 536 
THE HISTORY OF HUMAN MARRIAGE chap. 
Islanders as well,1 has also been attributed to fear of magic 
influences.2 Mr. Sommerville states that in Tanna, of the 
New Hebrides, " the closest secrecy is adopted with regard 
to the penis, not at all from a sense of decency, but to avoid 
the Narak (that is, bewitching force), the sight even of that 
of another man being considered most dangerous." The 
natives accordingly wrap it round with many yards of 
calico and other like materials, winding and folding them 
until a preposterous bundle eighteen inches or two feet long 
and two inches or more in diameter is formed.3 And, as 
Professor Him observes, the reproductive organs are not 
only, in virtue of their connection with the mysterious mira- 
cle of life, objects of superstitious awe, but, owing to the same 
consideration, they themselves may require particular 
protection against dangerous influences.4 The Bali and 
other Kamerun tribes believe that the evil eye is most dan- 
gerous to the male organs of generation, causing impotence ; 
hence the men take great care to conceal them when subject 
to medical examination.6 And besides the evil eye there are 
innumerable spirits that constitute so many dangers, for 
women in particular, and the possibility of being fecun- 
dated by whirlwinds or rain or other natural phenomena.6 
1 Cook, Journal of a Voyage round the World, p. 106 (Maori). 
Idem, Voyage to the Pacific Ocean, ii. 233 (Atooi of the Sandwich 
Islands). Rivers, History of Melanesian Society, ii. 433 (Gambier 
Islanders). Of the North Queensland aborigines Dr. Roth (North 
Queensland Ethnography : Bulletin 8. Notes on Government, Morals, 
and Crime, p. 7) tells us that, "except in those areas where introcision 
is practised, a man must never be seen with the preputium retracted." 
The Romans also considered that the glans should be carefully 
concealed by pugilists, athletes, and others who exposed their 
bodies (Stieda, ' Anatomisch-archaologische Studien. HI. Die In- 
fibulation bei Greichen und Romern,' in Anatomische Hefte, xix. 283). 
2 Rivers, op. cit. ii. 435. Karsten, Studies in South American 
Anthropology, i. 180 sq. 
8 Sommerville, 'Ethnological Notes on New Hebrides,' in Jour. 
Anthr. Inst, xxiii. 368. Cf. supra, p. 417. 
4 Hirn, Origins of Art, p. 218. See also Ratzel, Volkerkunde, i. 64. 
6 Plehn, * Beobachtungen in Kamerun,' in Zeitschr. f. Ethnol. 
xxxvi. 720. 
6 See Him, op. cit. p. 218 sq.; Landtman, Kulturens ursprungs- 
former, p. 200; Karsten, op. cit. i. 173 sqq. XVI PRIMITIVE MEANS OF ATTRACTION 
537 
Sinhalese women never expose their genitals because they 
fear that if they did, " a devil, imagined as a white and hairy 
being, might have intercourse with them."1 Among the 
Semangs of Malacca the women's girdles " are regarded 
both as a protection against the effects of ' hot rain ' and 
against 'pains in the waist.'"2 The Bantu Kavirondo, 
among whom young unmarried girls are usually nude, 
consider it very bad if a man takes hold of the dress of a 
woman who has borne children, even though it be her own 
husband; he who does so has to give a goat, which is 
killed and eaten, so as to remove the spell which will other- 
wise cause the woman's death. Yet if " a woman's dress 
is touched or torn off by an enemy, no curse is entailed."3 
Among several other peoples, also, only the married 
women cover themselves, either at once or, sometimes, 
after the birth of the first child, the unmarried ones remain- 
ing entirely naked.4 It has been said that the reason for 
this is the jealousy of husbands, which has consequently 
been regarded as the primary cause, or as one of the causes, 
of clothing.5 To the husband's mind, it is argued, the 
1 Ellis, op. cit. vol. i. Evolution of Modesty, &c., pp. 15, 56. 
2 Annandale and Robinson, Fasciculi Malayenses, i. 12. 
3 Hobley, Eastern Uganda, p. 21. 
4 Snow, Two Years' Cruise off Tierra del Fuego, ii. 46 (Fuegians). 
Church, Aborigines. of South America, p. 105 (Mojos of South-West 
Amazonia; some of the married women dress). Ling Roth, ' Abori- 
gines of Hispaniola,' in Jour. Anthr. Inst. xvi. 275. Guppy, Solomon 
Islands, p. 130 (natives of St. Christoval and the adjacent small 
islands) ; Elton, ' Notes on Natives of the Solomon Islands,' in 
Jour. Anthr. Inst. xvii. 96 (natives of the islands east of Guadal- 
canal). Turner, Samoa, p. 342 (New Caledonians). Finsch, Neu- 
Guinea, p. 96 (Papuans of Dorey). Petherick, Egypt, the Soudan and 
Central Africa, pp. 352 (Shilluk), 390 sq. (Dinka). Wilson and Felkin, 
Uganda and the Egyptian Soudan, ii. 49 ; Emin Pasha in Central 
Africa, p. 82 (Banyoro). Hutter, Wanderungen und Forschungen 
im Nord-Hinterland von Kamerun, p. 421 (Bali). Johnston, Uganda 
Protectorate, p. 770 (natives in parts of West Africa between the 
Niger and the Gaboon, especially on the Kamerun River, at Old 
Calabar, and in the Niger Delta). 
6 Letourneau, L'evolution de la morale, p. 149. Schurtz, Grundziige 
einer Philosophic der Tracht, p. 17. Thomas, Source Book for Social 
Origins, p. 549. Ellis, op. cit. vol. i. Evolution of Modesty, pp. 63, 538 
THE HISTORY OF HUMAN MARRIAGE chap. 
garment appears-illogically, though naturally-a moral 
and physical protection against any attack on his property ; 
and " as the conception of property also extended to the 
father's right over his daughters, and the appreciation of 
female chastity developed, this motive spread to unmarried 
as well as married women." It is difficult to believe, how- 
ever, that a feeling so grim and earnest as the jealousy of 
a married man would invent a merely symbolical means of 
protection utterly inadequate as a preventive of adultery. 
But the clothing of married women may have originated as 
a safeguard against supernatural dangers connected with 
pregnancy and child-birth, and would then, when generally 
accepted, be required by modesty and become an indication 
of their married state. Where garments of some sort are 
worn by all women, both married and unmarried, those of 
the former are often more seemly than those of the girls. 
Thus among the Shulis of the Egyptian Sudan, the 
married women wear a narrow fringe of string in front, the 
unmarried wearing nothing but bead ornaments.1 Among 
the natives of Tassai, New Guinea, the former use a larger 
and thicker kind of petticoat of pandanus leaf, divided into 
long grass-like shreds, reaching to the knee ; while that 
worn by the latter consists merely of single lengths made 
fast to a string which ties round the waist.2 In Fiji the 
liku-a kind of band made from hibiscus-bark-is before 
marriage worn very short, but after the birth of the first 
child is much lengthened ;8 and a similar practice occurs 
in other islands of the Pacific Ocean.4 
It has further been suggested that the fear of causing 
64, 80. Dr. Ellis thinks that when clothing had been once established 
the conception of women as property imported a new and powerful 
sanction to an emotion already based on more natural and primitive 
facts, that of sexual modesty. 
1 Wilson and Felkin, op. cit:ii. 62. Cf. ibid. ii. 97 (Baris) ; Shooter, 
Kafirs of Natal and the Zulu Country, p. 6. 
2 Macgillivray, Narrative of the Voyage of H.M.S. Rattlesnake, i. 263. 
3 Wilkes, op. cit. iii. 355. Seemann, Viti, p. 351. 
4 Forster, Voyage round the World, ii. 280. Waitz-Gerland, 
Anthropologie der Naturvolker, vi. 562. Cf. Dalton, Descriptive 
Ethnology of Bengal, p. 27 (Abors). XVI PRIMITIVE MEANS OF ATTRACTION 
539 
disgust has been the motive for covering the sexual centre. 
Dr. Havelock Ellis emphasises its close proximity to the 
points of exit of disgusting excretions.1 But this explana- 
tion could certainly not apply to the practice of concen- 
trating the covering on the sexual focus and leaving the 
anal region practically uncovered. There are indeed cases 
in which the clothing is behind and not in front; but they 
seem to be exceptional, and, as Dr. Ellis himself points out, 
there are probably other explanations for them. He also 
draws attention to the suggestion made by Lombroso and 
Ferrero, that the fear of the decomposition of the vaginal 
secretions causing disgust to men is the origin of modesty 
among savage women, as it remains the sole form of modesty 
among some prostitutes to-day.2 Such a fear may cer- 
tainly in some cases have led to clothing, as it in other cases 
leads to concealment by postures. Dr. Eylmann remarks 
that native women in South Australia in lying or sitting 
down avoid too much exposure, especially when they are 
in a state which hurts their vanity ;3 and of the Brazilian 
Mundrucus we are told by Tocantins that their women, in 
spite of being completely naked, carefully avoid any postures 
which might be considered indecorous, and that they do it 
so skilfully that nobody can notice when they have certain 
periods peculiar to their sex.4 As regards the origin of 
clothing in the case of males, a theory already set forth by 
St. Augustine deserves mention. He says that after the 
fall of our first parents " there began in the movement of 
their bodily members a shameless novelty which made 
nakedness indecent: it at once made them observant and 
made them ashamed; . . . and they sewed fig leaves 
together, and made themselves aprons."5 This explanation 
contains the psychological truth that nakedness becomes 
1 Ellis, op. cit. vol. i. Evolution of Modesty, pp. 51, 52, 80. 
2 Lombroso and Ferrero, La donna delinquents, p. 374. 
3 Eylmann, Die Eingeb^renen dev Kolonie Sudaustvalien, p. 128. 
4 Tocantins, ' Estudos sobre a tribu " Mundurucu "/ in Revista 
trimensal do Instituto Historico Geographico e Ethnographico do 
Brasil, vol. xl. pt. ii. 113. 
5 St. Augustine, De civitate Dei, xiv. 17 (Migne, Patrologies cursus , 
xli. 425 sq.). 540 
THE HISTORY OF HUMAN MARRIAGE chap. 
indecent when it suggests sexual intercourse and thereby 
hurts the feeling of sexual modesty. It is therefore con- 
ceivable that clothing has been adopted as a means of hiding 
the most obvious physical sign of sexual excitement. But this 
suggestion would be better supported by facts if the clothing 
used by savages were of a kind more suited to effect such 
concealment than it actually is, and if it were more needed 
for the purpose than it seems to be. In any case the com- 
mon theory that clothing owes its origin to the feeling of 
shame1 can have a meaning only if nakedness attracts 
attention to the sexual centre to such a degree as to suggest 
the idea of sexual intercourse, which, as we have seen, is 
the real cause of sexual shame. And this is evidently not 
the case among peoples who are used to nakedness. Whilst 
modesty relating to the sexual function may be said to be 
universal, there are a large number of peoples among whom 
both men and women, or one of the sexes, are in the habit of 
going completely naked and yet show no trace of shame, 
and other peoples who when they dress themselves are 
utterly neglectful of what we consider the first requirements 
of decency. 
Thus in the northern parts of the Californian Peninsula 
both men and women have been found in a state of nudity.2 
Among the Miwok, according to their own confession, 
persons of both sexes and of all ages were formerly abso- 
lutely naked.3 Lyman found the same to be the case with 
the Paiuches in Northern Colorado,4 Columbus with the 
aborigines of Hispaniola,5 Pizarro with the Indians of 
Coca,6 v. Humboldt with the Chaymas of New Andalusia,7 
and other travellers among various other South American 
1 Wundt, Ethik, i. 152. Schurtz, Grundzuge einer Philosophic dev 
Tracht, p. 9 sq. Stoll, op. cit. p. 486 sqq. ; &c. 
2 Baegert, ' Account of the Aboriginal Inhabitants of the Cali- 
fornian Peninsula,' in Smithsonian Report, 1863-64, p. 361. 
3 Powers, Tribes of California, p. 348. 
4 Waitz, op. cit. iv. 210. 
6 Ling Roth, in Jour. Anthr. Inst. xvi. 275. 
6 Waitz, op. cit. iv. 193 note. 
7 v. Humboldt, Personal Narrative of Travels to the Equinoctial 
Regions of the New Continent, iii. 230. XVI PRIMITIVE MEANS OF ATTRACTION 
541 
savages.1 In some Indian tribes the men alone go naked,2 
in others the women.3 In the northern parts of North 
America Mackenzie met a troop of natives of whom the 
men wore many ornaments and much clothing but showed 
not the slightest notion of sexual bashfulness in their dress. 
And of the Fuegians we are told that although they have 
the shoulder or the back protected by a skin, the rest of 
their body is perfectly naked.4 
It seems that in most Australian tribes both sexes, or 
at least the men, wear no clothing except in cold weather, 
when they throw something over the shoulders or the back.5 
1 de Lery, Histoire d'un voyage faict en la terre du Bresil, p. 121. 
Wied-Neuwied, Travels in Brazil, p. 59 (Puris at St. Fidelis). Wal- 
lace, Travels on the Amazon, p. 513 (Purupurus). v. Schiitz-Holz- 
hausen, Der Amazonas, p. 179 (Catamixis). Bleyer, ' Die wilden 
Waldindianer Santa Catharinas : die " Schokl6ng," ' in Zeitschr. f. 
Ethnol. xxxvi. 832. Church, op. cit. p. 105 (Mojos in South-Western 
Amazonia). Azara, Voyages dans l'Amerique mZridionale, ii. 83 
(some Indians in the neighbourhood of the river Paraguay). 
2 Herrmann, ' Die ethnographischen Ergebnisse der Deutschen 
Pilcomayo-Expedition, ' in Zeitschr. f. Ethnol. xl. 131 (Sotegaraik). 
Church, op. cit. p. 115 (Guarayos). Azara, op. cit. ii. 12, 42, 74, 126 
(Charruas, Pampas, Tupis, Payaguas). Charlevoix, History of 
Paraguay, i. 86 (Guaycurus). The same was often the case among 
the Nootka (Bancroft, Native Races of the Pacific States, i. 182) and 
the Patwin of California (Powers, op. cit. p. 220). 
3 History of the Life and Actions of Christopher Colon,' in 
Pinkerton, Collection of Voyages and Travels, xii. 101 (aborigines of 
Trinidad), v. Martius, Beitrdge zur Ethnographic Amerika's, i. 388, 
427, 504 (Mundrucus, Marauas, Juris). Wallace, op. cit. pp. 492, 
509 (Uaup6s, Curetus). Whiffen, North-West Amazons, pp. 72, 89 
(Witoto and Boro). 
Forster, op. cit. ii. 49g. King and Fitzroy, Narrative of the 
Voyages of the Adventure and Beagle, i. 23. Wilkes, op. cit. i. 121. 
Bove, Patagonia. Terra del Fuoco. Mari Australi, p. 129. 
Armstrong, Personal Narrative of the Discovery" of the North-West 
Passage, p. 33. Darwin, fournal of Researches, p. 228. See also 
Gallardo, Tierra del Fuego.-Los Onas, p. 131 sq. 
6 Curr, Australian Race, i. 93. Labillardiere, Account of a Voyage 
in search of La Perouse, ii. 27 sq. Bonwick, Daily Life and Origin 
of the Tasmanians, p. 104 sq. Oberlander, ' Die Eingeborenen der 
australischen Kolonie Victoria,' in Globus, iv. 278. Mathew, 
Australian Aborigines,' in four, and Proceed. Roy. Soc. N. S. 
Wales, xxiii. 391 Sq. Idem, Eaglehawk and Crow, p. 85. Idem, 542 
THE HISTORY OF HUMAN MARRIAGE chap. 
" They are as innocent of shame," says Palmer, " as the 
animals of the forests."1 Cook found the aborigines of 
Tasmania quite naked, and other travellers did the same, or, 
if not, attributed the little clothing they wore to the neces- 
sity of protecting themselves against the cold.2 Among 
various other peoples inhabiting islands in the Pacific 
Ocean,3 as also some tribes in Borneo4 and Sumatra,5 the 
people of Jarai bordering upon the empire of Siam,6 and the 
Nagas of Assam,7 both sexes were naked or, if they covered 
themselves a little, did so in a way which could have nothing 
to do with the feeling of decency ; whilst in many Pacific 
islands it was generally the men only who were completely 
devoid of covering.8 Of those on the south-west coast of 
Two Representative Tribes of Queensland, p. 95. Breton, Excursions 
in New South Wales, C^c., p. 211 sq. Henderson, Excursions and 
Adventures in New South Wales, ii. 104. Tench, Narrative of the 
Expedition to Botany Bay, p. 80. Schulze, ' Aborigines of the Upper 
and Middle Finke River,' in Trans, and Proceed. Roy. Soc. South 
Australia, xiv. 229. Stirling, ' Anthropology,' in Report on the 
Work of the Horn Scientific Expedition to Central Australia, iv. 37- 
Withnell, Customs and Traditions of the Aboriginal Natives of North 
Western Australia, p. 13. 
1 Palmer, ' Notes on some Australian Tribes,' in four. Anthr. Inst. 
xiii. 281 note. 
2 Ling Roth, Aborigines of Tasmania, p. 128 sqq. Mrs. Prinsep, 
Journal of a Voyage from Calcutta to Van Diemen's Land, p. 79. 
Nixon, Cruise of the Beacon, pp. 25, 28. West, History of Tasmania, 
ii. 83. Bonwick, op. cit. pp. 24, 104. Breton, op. cit. p. 398. 
3 Wilkes, op. cit. iv. 277 (natives of Penrhyn Island); v. 46 (natives 
of Drummond's Island). Labillardiere, op. cit. ii. 274 (Solomon 
Islanders), 287, 289 (natives of the Louisiade Archipelago). Parkin- 
son. Dreissig Jahre in der Sudsee, p. 490 sq. (natives in the interior 
of Bougainville in the Solomons). Pfeil, Studien und Beobachtungen 
aus der Sudsee p. 48 (various tribes in the Bismarck Archipelago). 
4 Bock, Head-Hunters of Borneo, p. 183. 
5 Forbes, ' Kubus of Sumatra,' in Jour. Anthr. Inst. xiv. 122. 
• Crawfurd, History of the Indian Archipelago, iii. 5. 
7 Blochmann, ' Koch Bih&r, Koch Hcijo, and A'sdm, in the 16th 
and 17th Centuries,' in Jour. Asiatic Soc. Bengal, vol. xli. pt. i. 84 n. f. 
8 Powell, op. cit. pp. 74 sq. (natives of the Duke of York Group), 
250 (natives of New Britain). Parkinson, op. cit. pp. 342 (natives of 
Squally Island north-east of New Ireland), 431 (natives of 
Wuwulu and Aua west of the Admiralty Islands). Guppy, op. cit. XVI PRIMITIVE MEANS OF ATTRACTION 
543 
New Guinea, Gill wrote that they " glory in their nudeness, 
and consider clothing to be fit only for women."1 In one 
part of Timor, on the other hand,2 as also in a tribe of the 
Andamanese,8 it is the women who are devoid of any kind 
of covering. 
Passing to Africa, we meet with instances of the same 
kind. The Central African Pygmies, both men and women, 
often go naked in their forest homes, although in the pre- 
sence of strangers they don a small covering.4 The Bush- 
men of South Africa also went naked, except when they used 
a piece of skin barely sufficient to cover the back.5 Among 
the Bateso, a tribe of the Nilotic group living in the Central 
Province of the Uganda Protectorate, neither sex wear any 
clothing, although they are particularly fond of ornaments.6 
Sir Charles Eliot observes that, with the exception of those 
natives on the coast who have any pretensions to be 
Muhammadans and the people of Uganda, " the native 
races of East Africa show an entire absence of the feelings 
which lead the European and Asiatic to cover their bodies. . . 
Most natives appear to be simply in the state of Adam and 
Eve before the fall, . . . and to have no idea of indecency. 
p. 130 ; Elton, in Jour. Anthr. Inst. xvii. 96 (some Solomon Islanders). 
Rivers, History of Melanesian Society, ii. 442 (Banks Islanders). 
Earl, Papuans, p. 48; d'Albertis, New Guinea, ii. 380; Chalmers, 
Pioneer Life and Work in New Guinea, pp. no, 112 ; Landtman, 
Nya Guinea farden, p. 68 ; Gill, Life in the Southern Isles, p. 203 ; 
Waitz-Gerland, op. cit. vi. 568; Haddon, ' Ethnography of the 
Western Tribe of Torres Straits,' in Jour. Anthr. Inst. xix. 336 ; 
Idem, in Reports of the Cambridge Anthropological Expedition to 
Torres Straits, v. 272 (various tribes in New Guinea and neighbouring 
islands). Keate, Account of the Pelew Islands, p. 318. Kotzebue, 
Voyage of Discovery into the South Sea, iii. 191 (natives of Pelli of 
the Caroline Group). Among the Maori " men often went naked 
without any breach of modesty, or decorum " (Colenso, Maori Races 
of New Zealand, p. 47). , 1 Gill, op. cit. p. 230. 
2 Forbes, Tribes of Timor,' in Jour. Anthr. Inst. xiii. 406. 
3 Man, ' Aboriginal Inhabitants of the Andaman Islands,' in Jour. 
Anthr. Inst. xii. 330. 4 Johnston, Uganda Protectorate, pp. 538, 769. 
Kretzschmar, Siidafrikanische Skizzen, p. 225. Chapman, 
Travels in the Interior of South Africa, i. 78. Barrow, Account of 
Travels into the Interior of Southern Africa, i. 276. 
6 Roscoe, Northern Bantu, p. 272. 544 
THE HISTORY OF HUMAN MARRIAGE chap 
It is noticeable that even the races who are more or less 
clothed, such as the Wakamba, wear their garments so 
carelessly that they by no means fulfil our ideas of propriety. 
As a rule this nudity is confined to the men, but in Kavi- 
rondo it extends to both sexes. . . . Neither the Bantu 
nor the Nilotic Kavirondo are particularly low in the scale 
of civilisation."1 According to Sir Harry Johnston, com- 
plete nudity in the male is, or has until recently been, 
characteristic of " the Nilotic Negroes, the pagan Hamites 
(Gala and Bahima), the hybrid races between the Hamite 
and the Nile Negro, between the Hamite and the Bantu, 
and a few Bantu races who are either very much under the 
influence of neighbouring Masai or Gala tribes or have 
still retained in South Central Africa the impress of Bahima 
customs."2 The peoples among whom the men either go 
naked or dress without the least regard to decency as we 
understand it, or did so not long ago, include the Baris,3 
1 Eliot, East Africa Protectorate, p. 93 sq. Among the Nilotic 
tribes of Kavirondo unmarried men go naked, whilst married men 
who have a child wear a small piece of goat skin practically useless 
as a covering; nor wear unmarried girls any clothes, although 
married women wear a tail of strings behind but not in front (Hobley, 
op. cit. p. 31). Among the Bantu Kavirondo the men generally 
go naked, though the old men and chiefs wear a skin slung toga-wise 
from the shoulder ; young unmarried girls are usually nude, but 
the other women wear a fringe of black string in front and a tail 
behind (ibid. p. 21). 
2 Johnston, Uganda Protectorate, pp. 767, 769. He adds (ibid. 
p. 769 n. *) :-" The only Bantu tribes which formerly were, or at 
the present day are, without feelings of shame in regard to the 
exposure of the person in the male are the A-kamba, A-kikuyu, 
Wa-chaga, and other tribes in British East Africa living in close 
relations with the Masai or the Gala ; the Kavirondo, who were 
similarly influenced by the Nile Negroes ; the Bakonjo of Ruwenzori, 
who in this may have copied the Hima customs ; the Barundi of 
North Tanganyika likewise ; the Nkonde tribes of the north end of 
Lake Nyasa; the Mashukulumbwe (cf. Holub, op. cit. ii. 299) 
and Batonga of the Central Zambezi; and the Zulus of South and 
South Central Africa." Of the Bahima Mr. Roscoe writes (op. cit. 
p. 103), " Men go scantily dressed : they are often destitute of 
clothing except for a small skin-cape thrown over their shoulders." 
3 Wilson and Felkin, op. cit. ii. 96. XVI PRIMITIVE MEANS OF ATTRACTION 
545 
Shilluk,1 Dinka,2 Masai,3 Nandi,4 Madis,6 Wataveta,6 
Watuta,7 Washashi,8 Hottentots,9 and the Elik people of 
Old Calabar.10 Apud Masaios membram virile celare turpe 
existimatur, honestum expromere, atque etiam ostentare ; 
yet Sir Harry Johnston has never known them to be guilty 
of any gesture of deliberate indelicacy.11 So also among 
the ancient inhabitants of Lancerote, according to Bontier 
and Le Verrier, the men used no covering ; and in Teneriffe 
" the inhabitants went naked, except some few who wore 
goatskins."12 . Among the Bubis of Fernando Po,13 the 
natives of Loango14and Balonda,15 and some tribes on the 
Congo, for example the Bopoto,16 complete nudity prevails, 
or .prevailed, among the women. In parts of West Africa 
it is, or was, customary for young women to go entirely 
naked before marriage, and in Swaziland until quite recently 
not only unmarried women but very often matrons were 
stark naked.17 Even among the prudish Baganda, who 
made it a punishable offence at one time for a man to expose 
any part of his leg above the knee, the wives of the king 
would attend at his court perfectly naked.18 
1 Schweinfurth, Im Herzen von Afrika, i. 322. Petherick, op. cit. 
P- 352. 
2 Schweinfurth, op. cit. L 163. Petherick, op. cit. p. 390 sq. 
3 Last, ' Visit to the Masai People,' in Proceed. Roy. Geograph. Soc. 
N.S. v. 530. Eliot, op. cit. p, 137. 
4 Johnston, op. cit. p. 862 sq. Hollis, Nandi, p. 27. 
6 Felkin, ' Notes on theMadi or Moru Tribe of Central Africa,' 
in Proceed. Roy. Soc. Edinburgh, xii. 316. 
6 New, Lije, Wanderings, &c. in Eastern Africa, p. 357. 
7 Cameron, Across Africa, i. 285 sq. 
8 Kollmann, Victoria Nyanza, p. 187 sq. See also ibid. p. 188 sq. 
9 Theal, Yellow and Dark-skinned People of Africa south of the 
Zambesi, p. 73. 10 Johnston, op. cit. p. 770. 
11 Idem, Kilima-njaro Expedition, p. 413 note. Idem, Uganda 
Protectorate, p. 808. 
12 Bontier and Le Verrier, The Canarian, pp. 138, 139, xxxv. 
13 Moller, Pagels, and Gleerup, Tre dr i Kongo, i. 15. 
14 Wilson and Felkin, op. cit. ii. 53. 
16 Livingstone, Missionary Travels and Researches in South Africa, 
p. 3<>5- 16 Ward, Voice from the Congo, p. 256. 
17 Johnston, Uganda Protectorate, p. 770 sq. 
18 Ibid. p. 771. Eliot, op. cit. p. 130. 
VOL. I N N 546 
THE HISTORY OF HUMAN MARRIAGE chap. 
It must not be supposed that the peoples who attach no 
idea of shame to the entire exposure of the body are devoid 
of sexual modesty or are less modest than those who cover 
themselves.1 Mr. Kerr Cross remarks that " as a general 
rule, in Africa modesty is in reverse proportion to clothing."2 
M. Hyades assures us that the same feeling is very developed 
among the Yahgans of Tierra del Fuego, who are used to 
live naked : " they manifest it in their bearing and in the 
ease with which they show themselves without clothing, 
compared with the awkwardness, blushing, and shame 
which they exhibit, men and women, if one gazes at certain 
parts of their bodies."3 The Andamanese are said to' " pay 
no attention to their own nudity, although by nature they 
are modest."4 Parkinson observes that the modesty of 
a perfectly naked New Ireland girl is as great as, if not 
greater than, that which is found in most of our fashionable 
ladies in Europe. And he adds that " nakedness by itself 
causes no sexual excitement in a native."5 This is un- 
doubtedly true of all peoples who wear no clothing. " Where 
all men go naked, for instance in New Holland," says 
Forster, " custom familiarises them to each other's eyes, 
as much as if they went wholly muffled up in garments."6 
In an account of the Nagas in the southern mountains of 
Assam, written by a Persian historian of the seventeenth 
century, we read that " the women only cover their breasts as 
they say that it would be absurd to cover those limbs which 
every one might have seen from their birth; but this was 
not the case with the breasts, which since then had formed 
and should, therefore, be covered."7 That the nakedness 
of savages is not alluring, when one gets used to it, has been 
attested by many travellers. Lery, in his description of 
1 See, e.g., Bonwick, op. cit. p. 57 sq. ; Maupetit, ' La pudeur,' 
in Bull. Soc. d'Anthr. Paris, ser. vi. vol. v. 405 sqq. 
2 Cross, ' Notes on the Country lying between Lakes Nyassa and 
Tanganyika,' in Proceed. Roy. Geograph. Soc. N.S. xiii. 88. Cf. 
Johnston, Uganda Protectorate, p. 730 (Bantu Kavirondo). 
8 Hyades and Deniker, Mission scientifique du Cap Horn, vii. 239. 
4 Kloss, In the Andamans and Nicobars, p. 188. 
5 Parkinson, op. cit. p. 271. 8 Forster, op. cit. ii. 383. 
7 Blochmann, in Jour. Asiatic Soc. Bengal, vol. xli. pt. i. 84. XVI PRIMITIVE MEANS OF ATTRACTION^ 
547 
the voyage he made to Brazil in the sixteenth century, says 
that the stay among naked savages is not a provocation to 
lust, and that the nudity of their women proved to be much 
less exciting than our women's clothes.1 Speaking of the 
Fuegians, Captain Snow remarks :-" More harm, I think, 
is done by false modesty,-by covering and partly clothing, 
than by the truth in nature always appearing as it is. 
Intermingling with savages of wild lands who do not clothe, 
gives one, I believe, less impure and sensual feelings than 
the merely mixing with society of a higher kind."2 In his 
Africa Unveiled' Mr. Rowley observes that " when the 
sight becomes accustomed to the absence of raiment, your 
sense of propriety is far less offended than in England "; 3 
and Winwood Reade expresses the opinion that there is 
nothing voluptuous in the excessive deshabille of an equa- 
torial African girl, nothing being so moral and so unlikely to 
excite the passions as nakedness.4 Similar views are taken 
by various other travellers.5 They are confirmed by the 
experience made by our artists at home. Du Maurier 
observes that it is "a fact well known to all painters and 
sculptors who have used the nude model (except a few shady 
pretenders, whose purity, not being of the right sort, has 
gone rank from too much watching) namely, that nothing 
1 de Lery, op. cit. p. 121. With reference to the Uaupds, Wallace 
says (op. cit. p. 296) that " there is far more immodesty in the 
transparent and flesh-coloured garments of our stage-dancers, than 
in the perfect nudity of these daughters of the forest." 
2 Snow, op. cit. ii. 51. 3 Rowley, Africa Unveiled, p. 146. 
4 Reade, Savage Africa, p. 546. 
5 Speaking of the naked women of New Ireland, Zimmermann 
says (op. cit. ii. 103 sq.) :-" In derThat muss ich auch sagen, dass 
nach kurzer Zeit, nach ein er durchaus nicht lange dauernden Gewbh- 
nung an diese Sache, man gar nichts anstossiges mehr in diesem 
ganzlichen Mangel an Kleidung findet. . . . Ich habe sehr haufig 
bemerkt, dass ein Kleid irgend einer Dame, welches nicht nach der 
allgemeinen Mode geschnitten war, mir starker auffiel als mir der 
ganzliche Mangel an Bekleidung der Eingeborenen der tropischen 
Inseln aufgefallen ist; dazu kommt noch, dass die Leute dem 
Beobachter durchaus keine Veranlassung geben, an etwas unschick- 
liches zu denken." See also Grose, Voyage to the East-Indies, i. 244 ; 
Johnston, Kilima-njaro Expedition, p. 437. 548 
THE HISTORY OF HUMAN MARRIAGE chap. 
is so chaste as nudity. Venus herself, as she drops her gar- 
ments and steps on to the model-throne, leaves behind her 
on the floor every weapon in her armory by which she can 
pierce to the grosser passions of men."1 Flaxman said that 
the students in entering the academy seem to hang up their 
passions along with their hats. 
On the other hand, we hear of naked savages who are 
ashamed to cover themselves, looking upon clothing as some- 
thing indecent. The pious father Gumilla was greatly 
astonished to find that certain Indians on the Orinoco did 
not blush at their nakedness. He says that on the contrary, 
if the missionaries distributed to them, and especially to 
the women, cloth with which to cover themselves, they 
threw it into the river so as not to be obliged to make use 
of it; and when they were told to cover themselves, they 
answered, " We are not going to cover ourselves because 
it makes us ashamed."2 Speaking of the Chaymas of New 
Andalusia, who, like most savage peoples dwelling in very 
hot regions, have an insuperable aversion to clothing, 
v. Humboldt states that under the torrid zone the natives are 
ashamed, as they say, to be clothed, and flee to the woods 
when they are too soon compelled to give up their naked- 
ness.3 In an Indian hut at Mucura in Brazil, Wallace 
found the women entirely without covering and apparently 
quite unconscious of the fact. One of them, however, 
possessed a saia, or petticoat, which she sometimes put on, 
and seemed then " almost as much ashamed of herself as 
civilised people would be if they took theirs off."4 
Facts like these must certainly tend to discredit the 
popular view that modesty is the mother of clothing. And 
our confidence in it does not increase when we consider what 
sort of covering savages often use. 
Among the Wintun, a Californian tribe, a fashionable 
young woman wears a girdle of deerskin, the lower edge of 
which is slit into a long fringe with a polished pine-nut at 
the end of each strand, while the upper border and other 
1 Du Maurier, Trilby, i. 146. 
2 Gumilia, El Orinoco ilustrado, i. 137. 
3 v. Humboldt, op. cit. iii. 230. 4 Wallace, op. cit. p. 357. XVI PRIMITIVE MEANS OF ATTRACTION 
549 
portions are studded with brilliant bits of shell.1 When a 
Carib girl attained the age of ten or twelve years she 
assumed around the waist " a piece of cotton cloth worked 
and embroidered with, minute grains of shells of different 
colours, decorated in the lower part with fringe."2 Dr. 
von den Steinen describes the so-called uluri, which is worn 
by Carib, Nu-Arawak, and Tupi women, as a little ornamental 
triangular piece of bark-bast, which looks very coquettish 
when new ; and he observes that both this elegant garment, 
the red thread of the Trumaf, and the variegated flag of the 
Bororo, " attract attention, like ornaments, instead of 
drawing attention away."8 Among the Central Australian 
tribes described by Spencer and Gillen " the only covering 
takes the form of small aprons worn by women and pubic 
tassels worn by men. The latter articles vary to a consider- 
able extent in size. In the southern tribes-the Arunta, 
Loritja, and Kaitish-they are usually small and quite 
inadequate as a covering; in fact they are frequently 
coated with white pipe-clay, which serves the purpose, 
especially during the progress of corrobborees, when large 
numbers of men and women meet together, of drawing 
attention to the part which in other tribes they are designed 
to conceal. They are not often larger in size than a. five- 
shilling piece, and may be even smaller."4 Speaking of the 
perineal band made of bark cloth which is the one article of 
dress universally worn by both sexes among the Mafulu 
mountain people of British New Guinea, Mr. Williamson 
states :-" One is tempted to think, as regards both men 
and women, that, from the point of view of covering, the 
bands might be dispensed with altogether. This remark 
applies still more strongly to the case of young boys and 
unmarried girls, including among the latter big full-grown 
girls, who are in fact fully developed women, whose bands 
can hardly be regarded as being more than nominal." 5 In 
1 Powers, op. cit. p. 233. * Heriot, op. cit. p. 306 sq. 
3 von den Steinen, op. cit. p. 193 sqq. Cf. v. Martius, op. cit. i. 579, 
642, 702 sq. note. 
4 Spencer and Gillen, Northern Tribes of Central Australia, p. 683 sq. 
b Williamson, Mafulu Mountain People of British New Guinea, 
p. 26 sq. 550 
THE HISTORY OF HUMAN MARRIAGE chap. 
Tahiti a maro, composed of red and yellow feathers, was in 
Cook's time considered a present of very great value, and 
the women thought it " most ornamental " to enfold their 
loins with many windings of cloth.1 Seemann says that 
in Fiji the girls " wore nothing save a girdle of hibiscus- 
fibres, about six inches wide, dyed black; red, yellow, white, 
or brown, and put on in such a coquettish way, that one 
thought it must come off every moment."2 A similar 
practice is widespread in the islands of the Pacific, fringes 
made of cocoa-nut fibre or of leaves slit into narrow strips 
or filaments of bark, frequently dyed with gaudy colours, 
being in many of these islands the only garment of the 
natives. This costume, with its conspicuous tint and mobile 
fringe, has a most graceful appearance and a very pretty 
effect, but is far from being in harmony with our ideas of 
modesty. In the island of Yap, according to Cheyne, " the 
dress of the males, if such it may be called, is slovenly in the 
extreme. They wear the maro next them, and, by way of 
improvement, a bunch of bark fibres dyed red, over it."3 
In New Caledonia, in Forster's time, the natives only tied 
" a string round the middle and another round the neck ; "4 
whilst in some other groups the dress of the men consisted 
of nothing but a leaf5 or a mussel or shell.6 
In Kordofan the unmarried girls only tied round their 
loins a fringed leathern belt, which was frequently decorated 
with agates.7 Among the Nandi they " wear little more 
clothing than a very small apron of leather ornamented 
with beads and tassels."8 Among the Negroes of Benin, 
according to Bosman, they had no other garment than some 
1 Cook, Voyage to the Pacific Ocean, ii. 16 sq. Idem, Journal of a 
Voyage round the World, p. 44. 2 Seemann, Viti, p. 168. 
8 Cheyne, Description of Islands in the Western Pacific Ocean, 
p. 144. 4 Forster, op. cit. ii. 383. 
s Waitz-Gerland, op. cit. vi. 561, 565 (New Caledonia, New 
Hebrides, Ulaua). 
6 liid. vi. 567 (some parts of New Guinea). Labillardiere, 
op. cit. i. 2795^. ; Moseley, in Jour. Anthr. Inst. vi. 397 sq. ; Parkin- 
son, op. cit. p. 368 (Admiralty Islands). 
7 Pallme, Travels in Kordofan, p. 71. 
8 Johnston, Uganda Protectorate, p. 863 sq. XVI PRIMITIVE MEANS OF ATTRACTION 
551 
strings of coral twisted about the middle.1 In Lukungu 
the entire covering of most of the women consists of a 
narrow string with some white china beads threaded on it.2 
The Hottentot women, according to Barrow, bestowed 
their largest and most splendid ornaments upon the little 
apron, about seven or eight inches wide, that hung from the 
waist. " Great pains," he says, " seem to be taken by the 
women to attract notice towards this part of their persons. 
Large metal buttons, shells of the cypraea genus, with the 
apertures outwards, or anything that makes a great show, 
are fastened to the borders of this apron."3 The Bushman 
women of South Africa, met with by the same traveller, had 
as their only covering a belt of springbok's skin, the part 
which was intended to hang in front being cut into long 
threads. But the filaments, he says, " were so small and 
thin that they answered no sort of use as a covering ; nor, 
indeed, did the females, either old or young, seem to feel 
any sense of shame in appearing before us naked."4 
In these and similar instances-I have only chosen a few 
representative cases-it does not look as if the covering was 
adopted from a feeling of modesty, but in many of them, 
at least, it is quite obvious that it is used as a means of 
embellishment. This is also expressly pointed out by several 
of our informants, and some of them even go so far as to 
suggest that the primary object of the clothing was to 
serve as an ornament. De Laet says of the coast tribes of 
Guiana that men and women mostly remain naked, and 
only sometimes cover the pudenda, more from love of 
decoration than from a sense of shame.5 In his descrip- 
tion of the Upper Napo Indians in Ecuador Mr. Simson 
remarks, " Clothing with all savages is primarily looked 
upon as mere embellishment, though Indians who have 
frequent communication with more civilised men begin to 
show some shame when entirely nude."6 Dr. von den 
1 Bosman, * New Description of the Coast of Guinea,' in Pinkerton, 
op. cit. xvi. 524. 2 Moller, Pagels, and Gleerup, op. cit. i. 169. 
8 Barrow, op. cit. i. 155. 4 Ibid. i. 276 sq. 
5 de Laet, Novus orbis, book xvii. ch. 11, p. 641. 
6 Simson, Travels in the Wilds of Ecuador, p. 158. 552 
THE HISTORY OF HUMAN MARRIAGE chap. 
Steinen asserts that the Bakairl of Central Brazil " envy us 
our clothes as a valuable ornament."1 The pubic covering 
of a South Australian native is, according to Dr. Eylmann, 
decoration rather than clothing.3 Forster says of the natives 
of Mallicolo that "it is uncertain whether the scanty dress 
of their women owes its origin to a sense of shame, or to an 
artful endeavour to please " ; and of the men of Tanna, 
that " round their middle they tie a string, and below that 
they employ the leaves of a plant like ginger, for the same 
purpose and in the same manner as the natives of Mallicollo. 
Boys, as soon as they attain the age of six years, are pro- 
vided with these leaves; which seems to confirm what I 
have observed in regard to the Mallicollese, viz., that they 
do not employ this covering from motives of decency. 
Indeed, it had so much the contrary appearance, that in 
the person of every native of Tanna or Mallicollo, we thought 
we beheld- a living representation of that terrible divinity 
who protected the orchards and gardens of the ancients."3 
Parkinson regards the shell used by men in the 
Admiralty Islands as an ornament ;4 and a similar 
view is taken of the miniature pubic covering, consisting 
of a plaited capsule or a shell or cocoon, which is 
worn in some parts of Africa.5 Speaking of the very 
simple dress of the male Hottentot, Barrow says, " If 
the real intent of it was the promotion of decency, it 
should seem that he has widely missed his aim, as it is cer- 
tainly one of the most immodest objects, in such a situation 
as he places it, that could have been contrived."6 In the 
Bali countries of Kamerun any clothing worn by the natives 
is, according to Hutter, meant to serve either as decoration 
or as protection against the weather. When it is not called 
for from either point of view, for example when the people 
are working in the field or bathing, both men and women 
denude their bodies completely without finding it in the 
von den Steinen, op. cit. p. 190. 2 Eylmann, op. cit. p. 401. 
Forster, op. cit. ii. 230, 276 sq. 4 Parkinson, op. cit. p. 368. 
Ankernjann, in Zeitschr. f. Ethnol. xxxvii. 64. For this practice 
among the Wangoni see Fiilleborn, Das Deutsche Njassa- und 
Ruwuma-Gebiet, p. 152. 6 Barrow, op. cit. i. 154. XVI PRIMITIVE MEANS OF ATTRACTION 
553 
least indecent to do so ; and if anybody gives them a piece 
of cloth and asks them to cover their pudenda, they simply 
fail to understand the meaning of it. And yet, our inform- 
ant adds, these people are not low savages but a com- 
paratively civilised and intelligent race.1 Sir Harry John- 
ston states that among the Nandi peoples, who are quite 
indifferent to nudity in the men, " clothing is only worn 
for warmth or for adornment, and not for the purposes of 
decency." 2 And in his book on * The River Congo ' the 
same writer observes :-" Clothing was first adopted as a 
means of decoration rather than from motives of decency. 
The private parts were first adorned with the appendages 
that were afterwards used by a dawning sense of modesty 
to conceal them." 3 
A thing may of course be invented for one purpose and 
afterwards used for another purpose, and it is therefore 
conceivable that savages who now look upon their clothes as 
ornaments did not do so from the beginning. Yet there 
is every reason to believe that in many cases at least the 
pubic covering was originally adopted as a sexual lure. 
1 Hutter, op. cit. p. 375. 
2 Johnston, Uganda Protectorate, p. 862 sq. 
3 Idem, River Congo, p. 418. I became acquainted with this book 
only since the appearance of the first edition of the present work, 
where I already advocated the theory that covering, " at least in 
a great many cases," owes its origin to the desire of men and women 
to make themselves mutually attractive. This view has sub- 
sequently gained the support of Grosse (Beginnings of Art, p. 96 sqq.), 
Groos (Play of Man, p. 263), Bloch (Sexual Life of Our Time, p. 138), 
Parkinson (Dreissig fahre in der Siidsee, p. 271), and others. So, 
too, Dr. Havelock Ellis (op. cit. vol. i. Evolution of Modesty, p. 61) 
thinks there can be no doubt that the contention " that ornament 
and clothing were, in the first place, intended, not to conceal or 
even to protect the body, but, in large part, to render it sexually 
attractive, is fully proved." But he also writes (ibid. p. 47) that, 
although the statement that the primary object of clothes is to 
accentuate, rather than to conceal, has in it a large element of truth, 
it is by no means a complete account of the matter. In this I 
agree with Dr. Ellis, and in the present edition in particular I have 
taken care not to carry my theory to an extreme, admitting that, 
although the desire to be sexually attractive is probably one cause of 
the origin of clothing, there may be various other causes as well. 554 
THE HISTORY OF HUMAN MARRIAGE chap. 
If people attach objects to other parts of the body in order 
to stimulate the passions of the opposite sex, is it not all 
the more probable that the objects they attached to the 
sexual centre itself were intended to serve the same object ? 
While nakedness is not exciting if the eye is used to it, 
covering or half-covering readily becomes so. Montaigne 
asks :-" Why did Poppea invent the use of a mask to hide 
the beauties of her face, but to enhance it to her lovers ? . . . 
Why do people cover with so many hindrances, one over 
another, the parts where our desires and their own have 
their principal seat ? "x Virey gives the answer, " The 
less one sees, the more does imagination picture."2 So 
also, when Mr. Ward asked a native Congo chief the reason 
why the women of his district wore no costume, the import 
of his reply was, " Concealment makes the inquisitive 
hungry."3 The Khyoungtha of the Chittagong Hills have 
a tradition worth mentioning in this connection. " A 
certain queen noticed with regret that the men of the nation 
were losing their love for the society of the women, and were 
resorting to vile and abominable practices, from which the 
worst possible results might be expected. She therefore 
prevailed upon her husband to promulgate a rigorous order, 
prescribing the form of petticoat to be worn by all women 
in future, and directing that the males should be tattooed, 
in order that, by thus disfiguring the males, and adding 
■piquancy to the beauty of the women, the former might once 
more return to the feet of their wives."4 We can easily 
imagine that among a tribe of naked savages, when someone, 
whether man or woman, put on a bright-coloured fringe, 
a few gaudy feathers, a string with beads, a bundle of leaves, 
a dazzling shell, or a piece of cloth, this could not fail to 
1 Montaigne, Essays, book ii. ch. 15, vol. ii. 331. 
2 Virey, De la femme sous ses rapports physiologique, moral et 
littdraire, p. 300. " Nur das Verborgene reizt," says Zimmermann 
{op. cit. ii. 84), " und Diejenigen welche auf den Gesellschafts-Inseln 
die verhiillende Kleidung und den heimlichen Genuss und das 
Verbergen der natiirlichen Gefiihle einfiihrten, haben gewiss die 
Sitten nicht verbessert." 3 Ward, Voice from the Congo, p. 258. 
4 Lewin, Wild Races of South-Eastern India, p. 116 sq. XVI PRIMITIVE MEANS OF ATTRACTION 
555 
attract attention and served as an invitation. It is of great 
importance for our argument that among many peoples 
who generally go perfectly naked a covering is worn in 
circumstances which indicate that it is used simply as a 
means of sexual attraction. 
We have previously noticed the important part which 
dances and festivals play in the erotic life of savage peoples, 
and the endeavour of young men and women on these occa- 
sions to please each other by painting their bodies and 
decorating them with all sorts of ornaments. Then, also, 
it is the custom for them, in tribes who are otherwise naked, 
to put on a scanty covering. Among the Tasmanians, 
whose dances were performed " with the avowed intention 
of exciting the passions of the men," the women wore in the 
dance a covering of leaves or feathers, which, as among 
Australian natives on similar occasions, was removed 
directly afterwards.1 Among the Australian Pegulloburras, 
who generally go entirely naked, the women on festive 
occasions wear round the middle small fringes.2 Pre- 
paratory to their dances the Huitoto Indians of the Putumayo 
region used to paint themselves all over in various colours 
and adorn themselves with feather ornaments and neck- 
laces, and the few who possessed clothes generally wore them 
on these occasions.3 Speaking of the Witoto and Boro of 
the North-West Amazons, Mr. Whiffen remarks that " if 
for ordinary life the attire of the Indian is of the slightest, 
on the occasion of a festival or a dance the most elaborate 
sartorial preparations have to be made " ; and he also says 
of them that " their ornaments are more important than 
their dress, in fact their ornaments are their dress."4 
Wallace writes of the Uaupes that, " while dancing in their 
festivals, the women wear a small tanga, or apron, made of 
beads, prettily arranged. It is only about six inches square, 
but is never worn at any other time, and immediately the 
dance is over, it is taken off." Besides, their bodies are 
1 Bonwick, op. cit. pp. 27, 38. 
2 Curr, Australian Race, ii. 472. 
3 Hardenburg, Putumayo, p. 161 sq. 
•* Whiffen, op. cit. pp. 75, 79. 556 
THE HISTORY OF HUMAN MARRIAGE chap. 
painted.1 The same was the case with the Tahitian Areois, 
known for their licentious life and their lewd dances and 
pantomimes, who also sometimes, on public occasions, put 
on a girdle of the yellow ti leaves, which in appearance 
resembled the feather girdles of the Peruvian or other 
South American tribes.2 As to the South African Basuto, 
Casalis says that marriageable girls " frequently indulge in 
grotesque dances, and at those times wear, as a sort of petti- 
coat, long bands composed of a series of rushes artistically 
strung together/'3 
We have seen that among several peoples only the married 
women are clothed, whereas the unmarried ones go entirely 
naked. But among other peoples the case is just the 
reverse, the marriageable girls covering their nakedness, 
but later on, when married, laying off their garments-a 
practice which seems analogous to that of a married woman 
being deprived of her ornaments and her hair. Barth found 
this to be the case in various heathen tribes in the interior 
of Africa.4 In many parts of Australia young girls wear a 
sort of apron of fringe suspended from a belt round the 
waist, but give it up when they marry,5 or sometimes only 
after the birth of the first or second child.6 Thus among the 
natives of Botany Bay, New South Wales, according to 
Barrington, " the females at an early age wear a little apron, 
made from the skin of the opossum or kangaroo, cut into 
slips, and hanging a few inches from the waist; this they 
wear till they grow up and are taken by men."7 Mac- 
gillivray states that in many tribes of Torres Straits the 
1 Wallace, Travels on the Amazon, pp. 493, 281. v. Martins, 
op. cit. i. 597. 2 Ellis, Polynesian Researches, i. 235. 
3 Casalis, Basutos, p. 269. 
4 Barth, Reisen itnd Ent deckungen in Nord- und Central-A frika, 
ii. 467 sq. 
6 Freycinet, op. cit. ii. 748. Snow, cp. cit. ii. 46. Palmer, in 
Jour. Anthr. Inst. xiii. 281 note, 286. Mathew, in Jour, and Pro- 
ceed. Roy. Soc. N. S. Wales, xxiii. 392. 
6 Taplin, ' Narrinyeri,' in Woods, Native Tribes of South Aus- 
tralia, p. 15. Brough Smyth, Aborigines of Victoria, i. 275. Curr, 
Australian Race, iii. 19. 
7 Barrington, History of New South Wales, p. 23 sq. XVI PRIMITIVE MEANS OF ATTRACTION 
557 
women have a petticoat of fine shreds of pandanus leaves, 
the ends worked into a waistband, upon the construction 
of which much labour is expended; but it is only " some- 
times put on, especially by the young girls, and when about 
to engage in dancing." Under this, however, another 
covering is usually worn.1 Among the Tupi tribes of Brazil, 
as soon as a girl became marriageable " cotton cords were 
tied round her waist and round the fleshy part of both arms ; 
they denoted a state of maidenhood, and, if any one but 
a maiden wore them, they were persuaded that the Anhanga 
would fetch her away. ... It cannot have been invented 
for the purpose of keepiifg the women chaste till marriage, 
for these bands were broken without fear, and incontinence 
was not regarded as an offence."2 Very generally in the 
savage world, where climate does not put obstacles in the 
way, boys or girls or both sexes go naked till they reach the 
age of puberty, covering being thus resorted to at the same 
period as most ornaments ;3 but for this there may be 
other reasons besides the desire to be attractive.4 A very 
1 Macgillivray, op. cit. ii. 19 sq. 
2 Southey, History of Brazil, i. 2405^. Cf. v. Martins, op. cit. i. m. 
3 Krause, In den Wildnissen Brasiliens, p. 204 (Karaya). Hase- 
man, ' Notes on the Pawumwa Indians of South America,' in 
American Anthropologist, N.S. xiv. 342. Thomson, Fijians, p. 217. 
Codrington, Melanesians, pp. 234, 240. Speiser, Two Years with the 
Natives in the Western Pacific, p. 235 (natives of Pentecost in the 
New Hebrides). Rawling, op. cit. p. 57 (Papuans of the Mimika 
district in Dutch New Guinea). Waitz-Gerland, op. cit. vi. 42 
(Polynesians). Bonney, ' Aborigines of the River Darling,' in Jour. 
Anthr. Inst. xiii. 127; Cameron, ' Notes on some Tribes of 
N. S. Wales,' ibid. xiv. 358 ; Bonwick, ' Australian Natives,' ibid. 
xvi. 209 ; Angas, Savage Life and Scenes in Australia, i. 98 sq. 
Ried61, op. cit. p. 463 (Damanese). Caillie, Travels through Central 
Africa to Timbuctoo, i. 351 ; Nachtigal, Sahara und Sudan, i. 221 ; 
Reade, Savage Africa, pp. 45, 245 sq. ; Roscoe, Northern Bantu, 
p. 164; Moller, Pagels, and Gleerup, Tre dr i Kongo, i. 128 ; 
Chapman, Travels in the Interior of South Africa, i. 36 (various 
African peoples). 
4 Dr. Karsten suggests (op. cit. i. 164) that the cotton cords of the 
Tupis " were magical things which were put on the girl at the critical 
epoch in her life, when she became a woman, in order to protect her 
against the Anhanga." 558 
THE HISTORY OF HUMAN MARRIAGE chap. 
interesting statement, made by Lohmann, is that among 
the Salivas only harlots clothe themselves; and they are 
said to do so in order to attract through the unknown.1 
In conformity with other ornaments, what we consider 
decent covering is said to be more common among savage 
men than women. " If dress were the result of a feeling of 
shame," Professor Waitz observes, ' we should expect it 
to be more indispensable to woman than to man, which is 
not the case."2 In America, according to v. Humboldt- 
among the Caribs, for instance-the men are often more 
decently clothed than the women.3 The same is stated of 
the Nagas of Upper Assam; 4 and, with particular refer- 
ence to African savages. Barth remarks, " I have observed 
that many heathen tribes consider a covering, however 
poor and scanty it may be, more necessary for man than 
w'oman."5 Whether this is the rule among savage peoples 
is doubtful. At any rate, the egoism of the men cannot be 
blamed for the nakedness of the women. For a savage Eve 
may pluck her clothes from the trees. 
It is finally worth noticing that the use of a covering is 
not the only way in which savage men and women pay 
attention to the pubic region, or draw attention to it. In 
many of the Pacific islands this part of the body is tattooed, 
even so profusely as to give the person, at a distance, the 
appearance of being dressed.6 In such cases tattooing has 
been said to answer the purposes of decency in the absence 
1 Lohmann, quoted by Bastian, Rechtsverhdltnisse bei ver- 
schiedenen Volkern der Er de, p. 174. 
2 Waitz, Introduction to Anthropology, p. 300. 
3 v. Humboldt, op. cit. vi. 10. 4 Dalton, op. cit. p. 41. 
5 Barth, Reisen, ii. 473. Cf. Moller, Pagels, and Gleerup, op. cit. 
i. 269. 
6 Finsch, ' Die Bewohner von Ponap6,' in Zeitschr. f. Ethnol. 
xii. 311, 314. Waitz-Gerland, op. cit. vol. v. pt. ii. 188 (Vaitupu); 
vi. 34 (Samoa). Kramer, op. cit. ii. 65 (Samoa). Wilkes, op. cit. iii. 355 
(Fiji). Mariner, op. cit. ii. 266 (Tonga). Cook, Voyage to the Pacific 
Ocean, ii. 192, 232 (Atooi of the Sandwich Islands). Tautain, 
' Sur le tatouage aux iles Marquises,' in L' Anthropologic, vii. 247. 
Hale, U.S. Exploring Expedition, vol. vi. Ethnography and Philology, 
p. 39 (Polynesians). XVI PRIMITIVE MEANS OF ATTRACTION 
559 
of clothing j1 but its primary object may have been a differ- 
ent one. De indigenis Tenimberis et Timorlaonis dum 
loquitur Riedel, adulescentes et puellas dicit saepe consulto 
abradere pilos pubis nulla alia mente, nisi ut illae partes 
alteri sexui magis conspicuae fiant.2 The sexual organs 
of both men and women are frequently subject to mutila- 
tions or other practices, the purpose of which may have been 
to make the person more attractive to the opposite sex. 
This is distinctly said to be the case with the artificial 
perforatio penis, followed by the insertion of some suitable 
object, which is in vogue among the Dyaks of Borneo and 
some other peoples. It is intended to increase the libido 
sexualis of the women and is therefore highly appreciated 
by them ; hence a man who has submitted to several such 
operations and consequently can carry several inserted 
objects is particularly sought after.3 A very obscure form 
of mutilation, on the other hand, is the Australian sub- 
incision, generally practised at the age of puberty and 
consisting in cutting the lower wall of the urethra so that it is 
slit completely open from below, the cleft sometimes ex- 
tending only halfway back, sometimes the whole way 
back to the scrotum. Various explanations of it have been 
suggested, the object most frequently assigned to it being 
that of limiting population by decreasing the chances of 
conception; but nothing can be definitely stated as to 
its origin.4 Following a suggestion made by myself, Mr. 
1 See Kramer, op. cit. ii. 65. 
2 Riedel, op. cit. p. 293. Cf. Zimmermann, op. cit. ii. 189 sq. 
(Papuans). 
8 Miklucho-Maclay, ' Ueber die kimstliche Perforatio Penis bei 
den Dajaks auf Borneo,' in Verhandl. Berliner Gesellsch. Anthrop. 
1876, p. 23. Idem, ' Perforatio glandis penis bei den Dajaks auf 
Borneo und analoge Sitten auf Celebes und auf Java,' ibid. p. 24 
sqq. Hovorka, in Mittheil. Anthrop. Gesellsch. Wien, xxiv. 133 sqq. 
4 Schuermann, * Aboriginal Tribes of Port Lincoln,' in Woods, 
Native Tribes of South Australia, p. 231. Miklucho-Maclay, ' Ueber 
die Mika-Operation in Central-Australien,' in Verhandl. Berliner 
Gesellsch. Anthrop. 1880, p. 85 sqq. Curr, The Australian Race, i. 75. 
Mathew, in Jour, and Proceed. N. S. Wales, xxiii. 411. Stuart, 
' The " Mika " or " Kulpi " Operation of the Australian Aborigines,' 
ibid. xxx. 115 sqq. Bancroft, ' Note on Mutilations practised by 560 
THE HISTORY OF HUMAN MARRIAGE chap. 
Mathew expresses the view that its aim is ornamentation 
and increased virility of appearance, and in support of this 
opinion he quotes a statement to the effect that in 
Central Australia, on the occasion of the rite of subincision 
being undergone, young men who have been operated upon 
once and even twice previously will voluntarily come 
forward and call upon the operators to enlarge the incision 
to the utmost; which shows that a pride is taken in the 
enlarged appearance.1 It is only when a young man has 
been subject to this operation and thereby become a kutyi 
that he is considered to be a " thorough man " ; and among 
the Dieri a kulji alone has the privilege of appearing before 
the women in a state of perfect nudity.2 Mr. Mathew 
observes that where subincision is practised, female intro- 
cision-that is, the forcible laceration and enlargement of 
the vaginal orifice-becomes inevitable, and that the latter 
therefore is a consequence of the operation performed upon 
the men.3 In fact, as Dr. Roth points out, female intro- 
cision only takes place in those districts where male intro- 
cision (subincision) is practised, both being equally indis- 
pensable preliminaries to marriage; but according to him 
the former practice may have led to the latter. "It is 
possible," he says, " that the cutting of the perineum and 
general laceration, &c., of the female was originally a matter 
Australian Aborigines,' ibid. vol. xxxi, p. xxvi. sq. Eylmann, op. cit. 
p. 118sqq. Hovorka, inMittheil. Anthrop. Gesellsch. Wien, xxiv. 132 sq. 
Stirling, loc. cit. pp. 26, 27, 33 sq. Stoll, op. cit. p. 527 sqq. Messrs. 
Spencer and Gillen, who give a detailed description of the practice 
{Native Tribes of Central Australia, p. 251 sqq.), state that at the pre- 
sent day, and as far back as their traditions go, the Arunta natives at 
least have no idea of its having been instituted with a view to prevent- 
ing or even checking population ; and they add [ibid. p. 264) that, 
as a matter of fact, it does not do it. 
1 Mathew, Eaglehawk and Crow, p. 121. 
2 Howitt, ' Diery and other kindred Tribes of Central Australia,' 
in Jour. Anthr. Inst. xx. 86 sq. Gason, Dieyerie Tribe of Australian 
Aborigines, p. 21 sq. 
3 Mathew, Eaglehawk and Crow, p. 121. Cf. Spencer and Gillen, 
op. cit. p. 263. On female introcision in Australia see ibid. p. 92 sqq., 
and Spencer and Gillen, Northern Tribes of Central Australia, 
P- *33 sqq. XVI PRIMITIVE MEANS OF ATTRACTION 
561 
of convenience for the male, the mutilation in her case subse- 
quently coming to signify her fitness, capability, or experience 
in the art of the full enjoyment of copulation, and that, 
on the principle of a form of mimicry, the analogous sign 
was inflicted on the male to denote corresponding fitness 
on his part."1 Thus even according to this theory sub- 
incision and female introcision ultimately owe their origin 
to the preference given to one of these practices by the 
opposite sex and to the other practice following as a conse- 
quence of it. In Ponape, one of the Caroline Islands, 
again, a lad, on reaching marriageable age, is subject to 
semi-castration, according to Dr. Finsch both to escape 
the possibility of orchitis and because the girls consider men 
thus disfigured handsomer and more attractive than others ; 
and the same custom prevails in Niutabutabu, of the Tonga 
Islands.2 
The most widely spread of all the mutilations of the 
sexual organs is circumcision. It is in use among all 
Muhammadan peoples, among most of the tribes inhabiting 
the West Coast of Africa, among the Kafirs, among nearly 
all the peoples of Eastern Africa, among the Christian 
Abyssinians, Bogos, and Copts, throughout the various 
tribes inhabiting Madagascar,3 and in the heart of the Black 
Continent, among the Mangbettu and Akka. It is practised 
very commonly in Australia, in many islands of Melanesia4 
and the Indian Archipelago,5 and in Polynesia universally. 
It has also been met with in some parts of America, e.g., in 
Yucatan,6 on the Orinoco,7 and among certain tribes of the 
1 Roth, Ethnological Studies among the North-West-Central Queens- 
land Aborigines, p. 179 sq. See also ibid. p. 175. 
2 Finsch, in Zeitschr. f. Ethnol. xii. 316. Christian, Caroline 
Islands, p. 74. 
8 Sibree, op. cit. p. 217. 
4 Waitz-Gerland, op. cit. vi. 560 sq. Parkinson, op. cit. p. 182 
(New Britain). Rivers, History of Melanesian Society, i. 292. 
A usual mode of " circumcision " in Melanesia and Polynesia is to 
split the prepuce longitudinally without the removal of any part. 
5 Schmidt, Liebe und Ehe im alten und modernen Indien, p. 186 sq. 
8 Lafitau, Masurs des sauvages ameriquains, i. 412. 
7 v. Martius, op. cit. i. 582 note. 562 
THE HISTORY OF HUMAN MARRIAGE chap. 
Rio Branco in Brazil.1 The Jews, Muhammadans, Abys- 
sinians, and a few other peoples being excepted, it is per- 
formed when the boy attains manhood; and where it is in 
use it is generally considered a necessary preliminary to 
marriage, " uncircumcised " being a bad word and the 
women often refusing all intercourse with such a man.2 
Many different explanations of this custom have been 
suggested.3 A frequent belief is that it was adopted from 
hygienic motives. But, as Andree points out, circumcised 
and uncircumcised peoples live in the same neighbourhood 
side by side, without any difference in their physical con- 
dition ;4 Sturt observes that in Australia " you would meet 
with a tribe with which that custom did not prevail, between 
1 Wallace, Travels on the Amazon, p. 517. For the distribution 
of circumcision see Andree, ' Die Beschneidung,' in Archiv f. An- 
thropologie, xiii. 53 sqq. ; Idem, Ethnographische Parallelen und 
Vergleiche, Neue Folge, p. 166 sqq.; PloSs-Renz, Das Kind, ii. 157 sqq.; 
Stoll, op. cit. p. 499 sqq.; Zaborowski, ' La circoncision, ses origines 
et sa repartition en Afrique et a Madagascar,' in L'Anthropologie, 
vii. 653 sqq.-The prevalence of circumcision in ancient Japan 
has been inferred from the present notion that it is a shame 
" pour les Japonais d'etre kawakamuri, c'est-&-dire d'avoir le gland 
reconvert par le prepuce. . . . Tout Japonais qui se respecte se 
pr6sente soit aux bains publics soit a une visite medicale, avec un 
gland entierement d^couvert " (Laloy, ' Deformations des organes 
g^nitaux chez les Japonais,' in L'Anthropologie, xiv. 375). 
2 Andree, in Archiv f. Anthropologie, xiii. p. 75. Bastian, Rechtsver- 
haltnisse, p. xx. Kramer, Die Samoa-Inseln, ii. 61. Ellis, Yoruba- 
speaking Peoples of the Slave Coast, p. 67. Weeks, ' Anthropological 
Notes on the.Bangala of the Upper Congo River,' in Jour. Roy. Anthr. 
Inst. xl. 402. 
3 See, e.g., Hovorka, in Mittheil. Anthrop. Gesellsch. Wien, xxivi. 
135 ; Waitz-Gerland, op. cit. vi. 41, 784 ; Spencer, Principles of Soci- 
ology, ii. 67 ; Muller, Allgemeine Ethnographic, p. 337 sq. ; Burton'1 
' Notes on the Dahoman,' in Memoirs Anthr. Soc. London, i. 318 ; 
Reade, Savage Africa, p. 539 sq. ; Modigliani, Un viaggio a Nias, 
p. 702 ; Thomson, Fijians, p. 217 ; Meiners, Allgemeine kritische 
Geschichte der Religionen, ii. 467; Trusen, Die Sitten und Gebrauche 
der alten Hebraer, p. 117 sqq. ; Frazer, Magic Art, i. 92 sqq. ; Idem, 
Folk-Lore in the Old Testament, ii. 330 ; Gray, ' Circumcision (Intro- 
ductory),' in Hastings, Encyclopaedia of Religion and Ethics, iii. 
664 sqq. 
* Andree, in Archiv f. Anthropologie, xiii. 78. XVI PRIMITIVE MEANS OF ATTRACTION 
563 
two with which it did."1 Moreover, as Spencer remarks, 
while circumcision does not exist among the most cleanly 
races in the world, it is common among the most uncleanly ;2 
among the Herero and Bechuanas, for example, the boys 
are circumcised, though these peoples are described as 
exceedingly filthy in their habits.3 It seems more probable 
that circumcision has something to do with the superstitious 
fear of sexual " uncleanness."4 Mr. Purcell states that in 
Australian tribes practising it "no man, woman, or youth, 
that has been circumcised, will take food from a boy of 
his own tribe who has not undergone the rite. Should a 
man from an uncircumcising tribe visit these people, they 
won't eat in his presence, as they don't consider him clean." 5 
So also in Fiji uncircumcised youths were regarded as un- 
clean, and were not permitted to carry food for the chiefs.0 
The Yoruba-speaking peoples of the Slave Coast call circum- 
cision " the cutting that saves."7 The Muhammadans 
regard it as "cleansing"; by circumcision "the boy 
becomes clean, and capable of performing religious exercises, 
of praying and entering the mosque." 8 
The most satisfactory explanation which has been sug- 
gested for this practice is, in my opinion, that it at once 
makes the boy a man and gives him the appearance of sexual 
maturity,9 or that it by giving him such an appearance is 
supposed to make him a man capable of procreation.10 
Grown-up girls would then no longer have any objection 
to him, and circumcision, also, might thus be regarded as 
a means of sexual attraction, whether intended to be so or 
1 Sturt, Narrative of an Expedition into Central Australia, ii. 140. 
2 Spencer, Principles of Sociology, ii. 67. 
3 Galton, Narrative of an Explorer in Tropical South Africa, 
p. 192 sq. Andersson, Lake Ngami, p. 465. 
4 Cf. Crawley, Mystic Rose, p. 137 sq. 
6 Purcell, ' Rites and Customs of Australian Aborigines,' in 
Verhandl. Berliner Gesellsch. Anthrop. 1893, P- 287. 
9 Williams and Calvert, op. cit. p. 144. 
' Ellis, Yoruba-speaking Peoples of the Slave Coast, p. 67. 
8 Ploss-Rcnz, Das Kind, ii. 167. Klunzinger, Upper Egypt, p. 195. 
9 Floss-Renz, op. cit. ii. 147. Andree, Ethnographische Paralleled, 
Neue Folge, p. 211 sq. von den Steinen, op. cit. p. 198 sq. 
10 Parkinson, op. cit. p. 182 (New Britain). 564 
THE HISTORY OF HUMAN MARRIAGE chap. 
not. In Tanna, one of the New Hebrides, where the prepuce 
is so cut as to leave a wing on each side forming a large 
lump underneath, it is considered that " the larger the 
operation, the more of a man does it make the boy."1 Dr. 
von den Steinen observes that the Brazilian practice of 
stretching and tying up the prepuce has the same effect 
as circumcision and may therefore be performed from the 
same motive, although he also regards it as a protection 
against insects.2 Among the Pawumwa Indians it is 
especially unmarried men who are addicted to it.3 
Among many peoples of Africa and in certain tribes 
of the Malay Archipelago and South America the girls 
also undergo a sort of circumcision, which is likewise looked 
upon as a necessary preliminary to marriage.4 Sunt 
autem gentes, quarum contrarius mos est, ut clitoris et 
labia minora non exsecentur, verum extendantur, et saepe 
longissime extendantur. Atque ista etiam deformatio 
insigne pulchritudinis existimatur.5 De indigenis Ponap&s 
haec adnotat Finsch: labia interna longius extenta et 
pendentia puellis et uxoribus singulare sunt incitamentum, 
quae res eodem modo se habet apud alias gentes, ut apud 
Hottentottas.6 Speaking of this practice among the 
Bawenda in Northern Transvaal, a German missionary says 
that it is considered a matter of great importance for the 
marriage of a girl. Before the young man pays the bride 
price he tries to satisfy himself as to the accuracy of the 
information he has received on the matter; for " the more 
the labia pudendi protuberate the more pleased they are. '7 
1 Gray, ' Some Notes on the Tannese, ' in Report of the fourth 
Meeting of the Australasian Association for the Advancement of 
Science, held at Hobart, in 1902, p. 659. 
2 von den Steinen, op. cit. p. 199. 
3 Haseman, in American Anthropologist, N. S. xiv. 342. 
4 See Ploss-Bartels, Das Weib, i. 246 sqq.; Floss-Renz, Das 
Kind, ii. 220 sqq.; Gray, in Hastings, op. cit. iii. 667 sqq. 
5 Ploss-Bartels, Das Weib, i. 243 sq. Ploss-Renz, Das Kind, 
ii. 215 sqq. 6 Finsch, in Zeitschr. f. Ethnol. xii. 316. 
7 Wessmann, ' Reife-Unsitten bei den Bawenda in Nord-Irans- 
vaal,' in Verhandl. Berliner Gesellsch. Anthrop. 1896, p. 3^3- See 
also Stannus, * Notes on some Tribes of British Central Africa, in 
four. Roy. Anthr. Inst. xl. 317. XVI PRIMITIVE MEANS OE ATTRACTION 
565 
Among means of attraction directly connected with the 
sexual centre may also perhaps be counted certain attitudes 
exhibited by women among naked tribes. These attitudes, 
for example their postures when they sit on the ground, 
have been regarded as signs of modesty; and their object 
may undoubtedly be to conceal something which otherwise 
might attract unpleasant attention. But at the same 
time they imply consciousness of certain facts, and the 
exhibition of this consciousness is hardly a mark of undiluted 
bashfulness. There is a close connection between conceal- 
ment and allurement, and coquetry borders on modesty. 
Stirling says of some Australian natives, " As to the women 
it was nearly always to be observed that, when seen in 
camp without clothing, they, especially the younger ones, 
exhibited by their attitudes a keen sense of modesty if indeed 
a consciousness of their nakedness can be thus considered."1 
Dr. Eylmann states that among those natives both the 
younger men and women are more conceited of their pudenda 
than of any other part of their bodies, the face alone ex- 
cepted.2 • • 
M. Reinach finds it difficult to believe that evolution can 
transform a sentiment into its contrary and make modesty 
grow out of a sort of " exhibitionism."3 And many a reader 
will probably share his scepticism and ask how it is that 
clothing has come to be required by decency if it was origin- 
ally adopted as a lure. No doubt, as everybody knows, 
modesty is very closely associated with clothing. Even 
peoples living in warm climates, whose covering is as scanty 
as possible, may be utterly ashamed to appear without it. 
The women of the tribes of South Andaman are so modest 
that they will not remove their small apron of leaves, or 
put anything in its place, in the presence of any person, 
even of their own sex.4 The Fijians, " though almost 
naked, . . . have a great idea of modesty, and consider it 
extremely indelicate to expose the whole person. If either 
1 Stirling, loc. cit. p. 36 sq. 2 Eylmann, op. cit. p. 127. 
3 Reinach, ' La prohibition de 1'inceste et ses origines,' in 
L'Anthropologic, x. 69. 
4 Man, in Jour. Anthr. Inst. xii. 330 sq. 566 
THE HISTORY OF HUMAN MARRIAGE chap. 
a man or woman should be discovered without the maro, 
or liku, they would probably be killed."1 Among the 
Pelew Islanders the women have an unlimited privilege 
of striking, fining, or, if it be done on the spot, killing any 
man who makes his way into their bathing-places.2 In 
those Australian tribes where a covering is worn by the 
women they will retire out of sight to bathe.3 The women 
of Nukahiva have only one small covering, but are so 
tenacious of it that the most licentious will not consent 
to take it off; and the men, who wear nothing but a string 
round the prepuce, are utterly ashamed if it happens to 
fall off.4 The Admiralty Islanders, according to Moseley, 
always cover themselves hastily on removing their pubic 
shell for barter, and evidently consider that they are ex- 
posing themselves " either indecently or irreligiously " if 
they show themselves perfectly nude.5 Facts of this sort, 
however, can be easily explained without the supposition 
that the covering which is now required by decency also 
was at first adopted from a similar motive. 
" The ideas of modesty," Forster truly remarks, " are 
different in every country, and change in different periods 
of time." 6 Speaking of the mouth ornament worn by the 
women of Port des Frangais in Alaska, which makes the 
lower part of the mouth jut out two or three inches, La 
Perouse says, " We sometimes prevailed on them to pull 
off this ornament, to which they with difficulty agreed ; 
they then testified the same embarrassment, and made 
the same gestures, as a woman in Europe who discovers 
her bosom." 7 Von Humboldt wrote :-" A woman in some 
parts of Asia is not,permitted to show the ends of her fingers ; 
while an Indian of the Caribbean race is far from considering 
herself naked, when she wears a guajuco two inches broad. 
Even this band is regarded as a less essential part of dress 
1 Wilkes, op. cit. iii. 356. 2 Semper, Die Palau-Inseln, p. 68. 
3 Curr, The Australian Race, i. 99. 4 Lisiansky, op. cit. p. 85 sq. 
5 Moseley, in Jour. Anthr. Inst. vi. 398. Cf. Labillardiere, 
op. cit. i. 279 sq. ; v. Krusenstern, op. cit. i. 156. 
8 Forster, op. cit. ii. 383. 
7 La Perouse, Voyage round the World, ii. 142. XVI PRIMITIVE MEANS OF ATTRAC'l TON 
567 
than the pigment which covers the skin. To go out of the 
hut without being painted with arnotta, is to transgress 
all the rules of Caribbean decency."1 In Tonga a man 
would consider it very indecent not to be tattooed.2 A 
Muhammadan woman who is used to veil her face would 
blush exceedingly for showing it bare. The Tubori women 
in Central Africa, who wear only a narrow strap to which 
is attached a twig hanging down behind, feel greatly 
ashamed if the twig happens to fall off.3 Savages in 
Sumatra and Celebes who carefully cover the knee consider 
it highly improper to expose it.4 In China a woman must 
not show her foot to any other man than her husband ; it 
is even improper to speak of a woman's foot, and in decent 
pictures this part of the body is always concealed under the 
dress.5 
These instances are sufficient to show how relative and 
conventional the ideas of modesty often are. Where it is 
the custom to paint or tattoo the body, people are ashamed 
to appear unpainted or untattooed. Where women are 
in the habit of covering their faces, such a covering is con- 
sidered indispensable for every modest woman. Where 
people for some reason or other have come to conceal the 
foot, the knee, or the ends of the fingers, their modesty 
prevents the exposure of the part which is hidden. Nobody 
would doubt that in these cases it is the covering that has 
caused the feeling of modesty and not modesty that has 
caused the covering. Or are we to believe that the shame 
we should feel if we walked about barefooted is the original 
cause of our wearing boots ? The feeling of shame, as 
Bain puts it, "is resolved by a reference to the dread of 
being condemned, or ill-thought of, by others."6 Custom 
is a tyrant who threatens with condemnation, disgrace, or 
1 v. Humboldt, op. cit. vi. 12 sq. 2 Mariner, op. cit. ii. 267. 
3 'Dr. E. Vogel's Reise nach Central-Afrika,' in Petermann's 
Mittheilnngen, 1857, p. 138. 
4 Crawfurd, History of the Indian Archipelago, i. 209. 
6 Stricker, ' Der Fuss der Chinesinnen,' in Archiv f. Anthropolo- 
gic, iv. 243. 
6 Bain, Emotions and the Will, p. 211. 568 
THE HISTORY OF HUMAN MARRIAGE chap. 
scorn anybody who acts in defiance of its rules. Whatever 
be the foundation for a certain practice, and however trivial 
it may be, people have a tendency to disapprove or deride 
any marked deviation from it for the simple reason that such 
a deviation is unusual. The young ladies of Balonda, 
wholly unconscious of their own deficiency, could not main- 
tain their gravity at the sight of the naked backs of Living- 
stone's men. " Much to the annoyance of my companions," 
he says, " the young girls laughed outright whenever their 
backs were turned to them, for the Balonda men wear a 
dress consisting of skins of small animals, hanging before 
and behind from a girdle round the loins."1 By degrees 
the custom may be associated with religion and then becomes 
even more powerful than it was before. We are told of a 
Fijian priest who, like all his countrymen, was satisfied 
with a masi, or scanty hip-cloth, but on hearing a descrip- 
tion of the naked inhabitants of New Caledonia and of their 
idols, exclaimed contemptuously, " Not have a masi and 
yet pretend to have gods ! "2 And, as Peschel remarks, 
" were a pious Mussulman of Ferghana to be present at our 
balls, and see the bare shoulders of our wives and daughters, 
and the semi-embraces of our round dances, he would 
silently wonder at the long-suffering of Allah, who had not 
long ago poured fire and brimstone on this sinful and shame- 
less generation."3 
Like the uncovering of any other part of the body, so 
also the exposure of the pubic region, if usually concealed, 
would be apt to cause a feeling of shame as a transgression 
of the rule of custom. But this is by no means the complete 
explanation of the shame felt at such exposure. Sexual 
modesty, which leads to the concealment of the sexual 
function, is apt to be hurt by the sight of a part of the body 
which too openly directs the thought to that function. 
Such is not generally the case with nakedness when the eye 
is used to it, but it is different with a nakedness to which 
1 Livingstone, Missionary Travels and Researches in South Africa, 
P- 3°5- 
2 Williams, quoted by Peschel, Races of Man, p. 171. 
3 Ibid. p. 171. xvi PRIMITIVE MEANS OF ATTRACTION 
569 
the eye is not used.1 This explanation may seem somewhat 
contradictory to the theory that clothing originated as a 
sexual lure. If modesty forbids the exposure of a part 
which is usually covered, because of its association with 
the sexual function, why did it not for the same reason 
forbid the alluring covering of that part when previously 
exposed ? The answer to this question lies in the fact that 
modesty draws a distinction between discreet invitation 
and coarse and open provocation. The former is pleasing 
and therefore not to be ashamed of, the latter is repulsive 
and therefore regarded as shameful. But the limit between 
that which modesty allows and that which it forbids is not 
fixed and well defined. As we have seen, there are naked 
savages who are ashamed to cover themselves, and on the 
other hand there are clothed savages who are not ashamed 
to show themselves naked. All depends on what is felt as 
shocking and what is not. Thus it is not correct to say, as 
does M. Reinach, that the modesty which avoids nakedness 
is the opposite of the feeling which led to covering as a means 
of attraction. The desire to avoid something repugnant is 
not the opposite of the desire to be attractive. 
The inferences made in this and the preceding chapter 
may perhaps seem to be opposed to the marked tendency 
1 If some peoples, like the Takulli (Harmon, op. cit. p. 286), 
Aleut (Georgi, Beschreibung alter Natitfnen des russischen Reichs, 
p. 364 sq. ; Dall, Alaska, pp. 139, 397), and the Ita Eskimo (Kane, 
Arctic Explorations, ii. 114), who cover almost the entire body, 
nevertheless feel as little shame in regard to nakedness " as the very 
brute creation," the reason for it is that the warmth of the summer 
sun or the heat of their dwellings in the winter induces them to 
throw off all their clothes, so that they are used to complexe nakedness 
as well as to clothing. On the cast coast of Greenland the Eskimo 
men and women, when indoors, are perfectly naked with the excep- 
tion of the natit, a narrow band about the loins, of dimensions " so 
extremely small as to make it practically invisible to the stranger's 
inexperienced eye " (Nansen, First Crossing of Greenland, i. 338 ; 
ii. 277). The fact that peoples who are used to extensive clothing 
are particularly ashamed to expose the pubic region shows that 
modesty with regard to that part of the body cannot be due to the 
influence of habit alone. 570 
THE HISTORY OF HUMAN MARRIAGE chap. 
in modern anthropology to look for superstitious origins 
of savage customs. With this tendency I am in full sym- 
pathy. My research work in Morocco has convinced me 
that in very many cases the belief in magic forces is at the 
bottom of customs which have never before been traced to 
such a cause ; and I have little doubt that in the genesis 
of practices which we have now discussed superstition has 
played a larger part than is known at present.1 But at the 
same time it should be remembered that the sexual impulse 
is even more primitive than the belief in mysterious or 
supernatural forces and agents. We have therefore no 
right to assume, without direct evidence, that what is now 
looked upon as a sexual stimulant originally was some- 
thing else ; and even if it is known to have been so, it may 
from the beginning have been a sexual stimulant as well. 
An object may be attached to the body both as a charm 
and as an ornament; and the little apron of a savage girl 
may have been adopted for the double purpose of conceal- 
ing something unpleasant or disgusting and of positively 
increasing her attractions.2 
It has been argued that savages would not endeavour to 
ornament themselves in order to become attractive to the 
opposite sex unless their girls enjoyed great liberty in the 
choice of a mate; that such liberty is generally wanting 
in the savage world; and that consequently the practices 
which have been interpreted as means of attraction must 
have another meaning. This argument, however, suffers 
from several defects. Men are not indifferent to the feelings 
of women and the impression they make upon them even 
though they can force to submission the women they desire. 
Where a girl is not allowed to choose her husband she may 
nevertheless be able to choose her lover. Though not con- 
sulted when asked in marriage, it may lie in her power to 
1 This has recently been much emphasised by. Dr. Karsten 
{op. cit. passim), who, in fact, assigns superstitious motives to all 
sorts of " self-decoration " as practised among the South American 
Indians. See also Frazer, Totemism and Exogamy, iv. 194, 200-202, 
207 ; Landtman, Kulturens ursprungsformer, p. 203. 
2 Cf. Ellis, op. cit. vol. i. Evolution of Modesty, p. 59. XVI PRIMITIVE MEANS OF ATTRACTION 
571 
influence the proceedings in some way or other, and if given 
away against her will she may manage to dissolve the 
union. But very frequently savage women have consider- 
able liberty to choose their husbands ; and we have reason 
to suppose that under more primitive conditions they were 
freer in that respect than they are at somewhat higher stages 
of culture. This subject will be discussed in a following 
chapter. 
END OF VOL I.