UNIVERSIT'9'd^' 
 CALIFORNIA 
 
 ! SAN DlGO
 
 THE PATRIAEOHAL THEOEY.
 
 THE 
 
 PATRIARCHAL THEORY 
 
 BASED ON THE PAPERS OF 
 
 JOHN FERGUSON McLENNAN 
 
 (Romphitb 
 
 BY 
 
 DONALD McLENNAN M.A. 
 
 OF THE INNER TEMPLE, BABBI8TBB-AT-LA\T. 
 
 LONDON 
 MACMILLAN AND CO. 
 
 1885
 
 CHABtBS DICKENS AND BVAN8, 
 CRYSTAL PALACE PBESS.
 
 PREFACE. 
 
 IN the last years of his life my brother was engaged, 
 amidst many difficulties from failing health and pro- 
 longed absences from England, in following up the 
 lines of inquiry first opened by his book on Primitive 
 Marriage and his essay in the Fortnightly Review on 
 the Worship of Animals and Plants. * 
 
 The views as to the early structure and subsequent 
 movements of human society embodied in these essays 
 were, on the whole, confirmed and enlarged by further 
 study ;f the range of available evidence was gradually ex- 
 tended by a systematic survey of all authentic accounts 
 of primitive peoples in the various quarters of the globe ; 
 at the same time new problems arose, and new points of 
 view suggested themselves as likely to furnish their satis- 
 
 * "The Worship of Animals and Plants," Fortnightly Review, 
 1869-70. 
 
 f Some of his views were modified more or less, and it will appear 
 from this work that one was abandoned the view expressed in 
 Primitive Marriage that Agnation was, at a certain stage, generally 
 prevalent, in stating which it may be believed that he yielded some- 
 what to the authority of Sir Henry Maine.
 
 vi PREFACE. 
 
 factory solution, and ultimately it seemed no hopeless 
 task, if only health and strength had revived, to under- 
 take a general work on the structure of the earliest 
 human societies. In particular, he felt that he was 
 able to give a much more consistent and intelligible 
 view of the condition of rude or undeveloped com- 
 munities than anything that had previously been 
 offered to the public. It was a part of his design to 
 set forth a theory of the Origin of Exogamy, and 
 to gather together the facts, very numerous and fall- 
 ing into several classes, by which that theory could be 
 supported. 
 
 But the inquiry into Exogamy and into the condi- 
 tions of rude society in which it must be supposed to 
 have originated led to another generalisation. As the 
 theory of the Origin of Exogamy took shape, and the 
 facts connected with it reduced themselves to form in 
 his mind, the conclusion was reached that the system 
 conveniently called " Totemism " from which his essay 
 on the Worship of Animals and Plants took its departure 
 must have been established in rude societies prior to 
 the origin of Exogamy. This carried back the origin 
 of Totemism to a state of man in which no idea of 
 incest existed. From that condition my brother hoped 
 to be able to trace the progress of Totemism neces- 
 sarily a progress upwards in connection with kinship 
 and with Exogamy. It may here be said that he had
 
 PREFACE. vii 
 
 for a time a hypothesis as to the origin of Totem ism, 
 but that he afterwards came to see that there were 
 conclusive reasons against it. At last, as far as I know, 
 he had none which should be easily intelligible to any- 
 one who knows the subject and knows what, on his view, 
 was involved in Totemism. To show its prevalence, 
 to establish some leading points in its history, to exhibit 
 it in connection with kinship and with Exogamy, and to 
 make out its connection with worship appeared to him 
 to be the matters primarily important. 
 
 It may be said that evidence of Totemism associated 
 with Exogamy was generally found in all rude societies 
 acknowledging kinship through women only ; that the 
 same association was found also, and almost as generally, 
 in rude societies which know kinship through males; 
 while his original essay had tended to show that the 
 worship of plants and animals in more advanced societies 
 acknowledging kinship through males was lineally des- 
 cended from Totemism. 
 
 The general conclusion from these and allied facts 
 taken as a whole, appeared to be that it was possible 
 to demonstrate that, Totemism preceding Exogamy, the 
 latter must have arisen in societies acknowledging no 
 kinship save through women; that all other facts bearing 
 on rude society may be interpreted as evidence of a 
 gradual progress from the condition of which Totemism 
 and female kinship are the mark ; and that thus it
 
 viii PREFACE. 
 
 was possible to exhibit the history of human society as 
 that of an evolution moving with very various rapidity 
 among different populations, but always beginning with 
 a condition in which the idea of incest did not exist, 
 and always tending upwards from that condition. 
 
 With regard to the part to be played in this research 
 by the theory of Exogamy, one other point maybe noticed. 
 It was found that Exogamy had an extension so great, 
 and of such a kind, as to imply that it arose from the 
 operation of general causes not limited to the circum- 
 stances of this or that particular population. Now 
 Exogamy is found alike in rude societies with kinship 
 through males, and in the still ruder societies which 
 have kinship only through females. It is perhaps 
 scarcely a legitimate hypothesis to suppose that it had 
 two separate origins for these two kinds of societies. 
 And the presumption seems to be that it began with 
 the ruder of the two ; in other words, that it began 
 before the recognition of male kinship. At any rate, 
 as has been said already, the explanation of its origin 
 which my brother designed to set forth was in accord- 
 ance with this presumption. And here the argument 
 from Totcmism came in to enforce and justify the con- 
 clusion by making it probable that every society in 
 which Exogamy has been found, must at one time have 
 known kinship through females only. Now the general 
 cause from which his theory derived the origin of
 
 PREFACE. ix 
 
 Exogamy was a scarcity of women. And the point 
 to be noticed is that in this connection my brother 
 had made large collections as to the prevalence of 
 infanticide and kindred practices, classified in rela- 
 tion to the systems of kinship with which they are 
 associated. 
 
 These indications will suffice to show that the 
 proposed research was of a very extensive and far- 
 reaching kind, and involved the use of a very large 
 apparatus of evidence. This being so, my brother 
 proposed to prepare the way for his larger work by 
 first issuing a critical essay by which he hoped to clear 
 out of the way a body of opinion, the prevalence of 
 which seemed to oppose an obstacle to the proper 
 appreciation of his constructive argument. 
 
 From the time of Plato downwards, theories of 
 human society have been current in which the family 
 living under the headship of a father is accepted as the 
 ultimate social unit. These theories have taken various 
 shapes, but in his opinion the most important, as well 
 as the most influential, shape to be taken account of is 
 that represented in the works of Sir Henry Maine. In 
 the hands of the more prominent amongst its older 
 advocates within the period of modern thought the 
 Patriarchal Theory, as it is called, was mainly a theory of 
 the source of sovereignty, and in this aspect it had gradu- 
 ally ceased to attract attention. With Sir Henry Maine,
 
 x PREFACE. 
 
 on the other hand, the theory becomes a theory of the 
 origin of society, or at least of the earliest stage of 
 society in which Comparative Jurisprudence is called 
 upon to take interest. And at the same time the theory 
 is expressly based on a comparative study of early 
 societies, so that it comes into direct conflict with every 
 theory of the origin of society which does not accept 
 the family as the primitive unit. 
 
 It was necessary, therefore, for my brother to take 
 notice of this theory, and to do so on the scale which 
 the intrinsic importance of the question demanded. To 
 make such a discussion a mere incident in a large and 
 complicated constructive work would not have tended 
 to convenience or clearness, and the only alternative 
 was to make it the subject of a separate essay. This 
 plan had the disadvantage that such an essay could not 
 but be polemical in form, a thing which has always a 
 somewhat ungracious look; but it was felt that the 
 enormous prestige which the Patriarchal Theory has 
 gained through Sir Henry Maine's advocacy made it 
 impossible to separate the argument from the form which 
 he has impressed on it. In English-speaking countries, 
 at least, this is the one form in which the theory is 
 current, and no discussion of it would have been useful 
 which did not closely follow the statements of the 
 author of Ancient Law, a book which, for more than 
 twenty years, has profoundly influenced the whole
 
 PREFACE. xi 
 
 teaching of Jurisprudence in our country. An inde- 
 pendent thinker who has arrived at conclusions funda- 
 mentally at variance with those set forth in so influential 
 a work cannot avoid bringing the points in dispute to a 
 direct issue. 
 
 The proposed criticism of the Patriarchal Theory 
 was first thought of in 1879, but the execution of the 
 project was deferred by increasing illness. Some notes, 
 however, were made from time to time, and in the 
 winter of 1880 I began, at my brother's request, actually 
 to write the book in co-operation with him, and we 
 finished together a draft of the first six chapters of the 
 present essay, and also of the ninth, the substance of 
 which had for the most part been published in his 
 Studies in Ancient History. His last illness cut short 
 the work at this point, and what remains I have had 
 to work out for myself, with the aid of some frag- 
 mentary notes, mostly relative to Agnation, expressly 
 prepared for this work, and for the argument pre- 
 sently to be spoken of of the hints that could be derived 
 from the collections also fragmentary formed for his 
 larger project. 
 
 Fortunately the notes referred to included a short 
 paper containing what appeared to me to be invaluable 
 suggestions for what I venture to think the most 
 important part of the whole discussion, namely, the 
 origin of Agnation. This paper (besides that it was
 
 Xll 
 
 PEEFAOE. 
 
 pointed out that, on Sir Henry Maine's statement, 
 Agnation might have been expected to fall with Patria 
 Potestas, unless something could be indicated capable 
 of making it survive ; and that at Eome Patria Potestas 
 was the longer-lived of the two) gave me the suggestion 
 that Agnation was originally the gentile relationship, 
 the suggestion that this could be supported by the 
 analogous operation of female kinship, the observation 
 that Agnation is none the less an exceptional pheno- 
 menon, and the indication of the retarding influences 
 which may have prevented its more general occurrence. 
 To bring out the full scope of these important hints, 
 I found it necessary to build up as best I could a 
 somewhat elaborate argument, in which the original 
 scheme of the work was necessarily transcended, so that 
 the latter part of the book is more constructive and less 
 purely critical than had been at first designed. Another 
 reason for this was the necessity for revising the whole 
 discussion in the light of Sir Henry Maine's latest 
 work, which made it needless to dwell at length on 
 several parts of the original plan, while on the other 
 hand it rendered it inevitable to go fully into the 
 whole question of the Levirate and of the family custom 
 of the Hindoos. For this part of the book of course 
 I alone am responsible. 
 
 I should add, perhaps, that I had hoped to bring 
 the book out much sooner. But the state of my health
 
 PREFACE. xiii 
 
 has compelled me to work slowly. And I had to 
 make some laborious researches. 
 
 It remains for me to express my gratitude to Pro- 
 fessor "W. R. Smith, of Cambridge, for his kindness in 
 reading a proof of this work, and for his readiness 
 on all occasions to give me help and save me trouble. 
 Some valuable notes which he has contributed to the 
 book are acknowledged at the proper places. 
 
 D. McLENNAN. 
 
 TEMPLE, November lith, 1884.
 
 CONTENTS. 
 
 
 CHAPTER I. 
 
 PA OB 
 
 THE PATRIARCHAL THEORY 1 
 
 CHAPTER II. 
 SIR HENRY MAINE'S LATER WRITINGS 15 
 
 CHAPTER III. 
 
 THE PATRIARCHAL FAMILY. THE AMOUNT OP PROOF OF IT THAT 
 
 MAY SUFFICE ......... 24 
 
 CHAPTER IV. 
 
 FLAN OF THIS WORK 31 
 
 CHAPTER V. 
 
 PATRIA POTESTAS AMONG THE HEBREWS ..... 35 
 
 CHAPTER VI. 
 
 PATRIA POTESTAS AMONG THE HINDOOS . . . . .51 
 
 CHAPTER VII. 
 
 PATRIA POTESTAS AMONG THE SLAVS . . . . .71 
 
 NOTE TO CHAPTER VIL 
 
 THB PATRIARCHAL THEORY AND ROYAL SUCCESSION ... 88 
 
 CHAPTER VIIL 
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES . 96
 
 xvi CONTENTS. 
 
 CHAPTEK IX. 
 
 PAGE 
 
 PATRIA POTBSTA8 IN IRELAND . . . . . . .120 
 
 CHAPTER X. 
 
 PATRIA POTESTAS CONCLUDED 132 
 
 CHAPTEK XI. 
 
 DERIVATIVE INSTITUTIONS. THE EVIDENCE AS TO THEIR ORIGIN 
 
 WHICH MAY SUFFICE 141 
 
 CHAPTER XII. 
 
 PATRIA POTESTAS AND AGNATION .181 
 
 CHAPTER XIII. 
 
 THE ORIGIN OF AGNATION 205 
 
 CHAPTER XIV. 
 
 EXAMPLES OF AGNATION 243 
 
 CHAPTER XV. 
 
 AGNATION CONCLUDED 259 
 
 CHAPTER XVI. 
 
 SONSHIP AMONG THE HINDOOS .... . 266 
 
 CHAPTER XVII. 
 
 SONSHIP AMONG THE HINDOOS .... 313 
 
 CHAPTER XVIII. 
 
 THE TUTELAGE OF WOMEN: THE HEIRSHIP OF SLAVES . . 340 
 
 CHAPTER XIX. 
 
 CONCLUSION .... 350
 
 THE PATRIARCHAL THEORY. 
 
 CHAPTER I. 
 
 THE PATRIARCHAL THEORY. 
 
 THE Patriarchal Theory, stated in its simplest form, 
 represents society as the enlargement of the family, 
 and the family as a group composed at first of a man 
 and his wife and children. When the children, and 
 afterwards their children and more distant descendants, 
 married, many such groups would be formed round the 
 original family, and all of these would acknowledge the 
 authority of the First Father, as chief or patriarch, as 
 long as he lived. However large the body of descendants 
 might become, they would constitute but one family, of 
 which the First Father would be the natural head. On 
 his death his descendants would naturally divide into as 
 many families as he had sons with offspring. Each of 
 these would resemble the original group absolutely 
 would be, that is, a collection of persons connected by 
 common descent, living under the authority of their 
 common progenitor. What had happened on the death 
 of the First Father would happen thereafter to every 
 family on the death of its head ; it would be resolved
 
 2 THE PATRIARCHAL THEORY. 
 
 into a series of families, each under the headship of that 
 son of the deceased from whom it was derived. 
 
 Should all the families descended from the same 
 First Father continue to hold together, they would in 
 time form a very large group in short, a tribe ; and 
 thus the Theory explains the genesis of tribes. It can 
 also be made to furnish a partial explanation of the 
 formation of gentes, clans, or houses, within the tribe. 
 These would begin to be formed as soon as some prin- 
 ciple of succession or election to the chieftainship 
 preserved families from dissolving on every death of a 
 chief. Or they might arise through the families des- 
 cended from some eminent progenitor distinguishing 
 themselves from the other families of the tribe by a 
 name indicative of such descent and of a special con- 
 nection between them. The Theory, moreover, explains 
 the genesis of nations. As there would be the 
 nucleus of a new tribe whenever a man descended 
 from the First Father separated, with his wife and 
 children, from the main body of his kindred, and settled 
 in a new district, in the course of generations after 
 many such separations had taken place the descendants 
 of the same First Father might constitute many tribes 
 and be the population of a large country. The tribes, 
 being united by ties of blood, would readily act together 
 for common purposes. By-and-by they would establish 
 some form of central government to facilitate such 
 action. Then they would have become a nation. 
 
 It has been usual to cite the history of Israel as an 
 illustration of this theory as stated above. Each of the
 
 THE PATRIARCHAL THEORY. 3 
 
 twelve tribes consisted of descendants of a son or 
 grandson of Jacob, the First Father ; and their union 
 constituted the Jewish nation. In Genesis, moreover, 
 the population of the world is represented as composed 
 of tribes and nations deduced from the sons of Noah 
 Shem, Ham, and Japhet. 
 
 The Patriarchal Theory, so stated, was most simple, 
 and agreeable to current prejudices. It used to be 
 generally accepted as palpably true, like the fact of the 
 sun moving daily round the earth. No one thought of 
 proving it, and but few of seriously doubting it. But 
 there were facts against it, and at length some of them 
 were noticed. Its explanation of the origin of gentes, 
 or clans, within the tribe was obviously insufficient. 
 Tribes like the Eoman Tribe, consisting of a number of 
 different stocks consisting of clans, each of which 
 differed in blood or origin from the others have been 
 exceedingly common ; and of their composition the 
 theory could give no account at all. What it could 
 prove was that the clans composing a tribe must 
 always and everywhere be all of the same stock 
 viz., that of the First Father of the tribe. The 
 insufficiency of this theory, however, such was its hold 
 upon the world, suggested, in the first instance, not 
 any doubt of its truth as a whole, but the need of making 
 some modification of it, or of finding some means of 
 supplementing it. And in a modified form, and supple- 
 mented to some extent, " Comparative Jurisprudence " 
 was promising to establish it for ever, beyond the 
 reach of doubt, at the time when research setting 
 
 B 2
 
 4 THE PATRIARCHAL THEORY. 
 
 ancient facts in new points of view it first came 
 seriously into question. 
 
 As thus modified and supplemented the Patriarchal 
 Theory holds its ground among us. And it is as thus 
 modified and supplemented that it is here to be the 
 subject of criticism. Its insufficiency in its old form 
 seems to be admitted. On the other hand, as now 
 presented, it lacks the simplicity which was formerly its 
 best recommendation. 
 
 The ingenious and learned author of Ancient Law, 
 indeed, in restating the Patriarchal Theory, ascribed to 
 it features which made it very different from the view 
 which had held possession of the world, and which 
 might reasonably have hindered its acceptance. The 
 Patriarchal Family, as he conceived of it complex, 
 artificial, strange does not look as if it could belong- 
 to the earliest history of man. 
 
 It is not merely a group of descendants with the 
 First Father at their head. It is a group of persons 
 living under a Patriarch who has over them despotic 
 power, and can sell any of them, or put him to death ; 
 and they are held to be related to him, and to one 
 another, not so much because of their being of his 
 blood as because of their common subjection to his 
 power. As to the composition of this group, the 
 Patriarch having children and other descendants, it 
 includes children and other descendants of his, and, 
 in theory, it is made up of his descendants. But the 
 stranger whom he adopts and he practises adoption 
 largely is in every respect as a son. On the other
 
 THE PATRIARCHAL THEORY. 5 
 
 hand, the tie of blood counts for so little that a son 
 who leaves him ceases to be a relation of him and 
 his family ; and a daughter, when she marries, becomes 
 as a stranger to them ; all descendants through 
 daughters being also excluded from relationship to. 
 the Patriarch and his house. The blood-connection 
 between the family and one-half of its kinsfolk is 
 always ignored, while a firm bond of relationship con- 
 nects people who have no blood in common. Primitive 
 as it is, this family includes slaves as well as children 
 and adopted persons. The slave, however, is not 
 within the relationship, because he is to remain a slave 
 for ever. As to the powers supposed to have been 
 wielded by the Patriarch over the members of the 
 Primeval Family, they are that assortment of powers 
 which in Kome was called Patria Potestas ; while the 
 system of relationship is what was conceived of as the 
 simplest form of the Eoman relationship, Agnation that 
 system which, in a more developed form, severed from 
 all rights of succession to family property, and even 
 from the family, every person connected with its head 
 through women only. 
 
 It is mainly on a proof of the universal prevalence 
 in the earliest times of the Roman institutions of Patria 
 Potestas and Agnation that this novel view of the 
 Patriarchal Family is founded. The Primeval Family of 
 the theory differs, however, from the ancient Roman 
 family by which it was suggested, in respect of the use 
 it made of adoption. It was freely enlarged by the 
 practice of adoption ; whereas in Rome adoption was
 
 6 THE PATRIARCHAL THEORY. 
 
 originally as the Will (to which it was prior in origin) 
 also was merely an expedient to enable a man who 
 had no child, and no prospect of having one, to choose 
 a young man, usually a near relative, to be to him as 
 a son to continue his family, to keep up his sacra, 
 and, on the other hand, to inherit his property. 
 
 The following extracts from Ancient Law wilJ be 
 
 o 
 
 enough to satisfy the reader of the accuracy of the 
 account just given of the view propounded in that work: 
 
 1. Effect of the Evidence. "The effect of the evi- 
 dence derived from Comparative Jurisprudence is to 
 establish that view of the primeval condition of the 
 human race which is known as the Patriarchal 
 Theory." Ancient Law, p. 122.* 
 
 2. The Primeval Family. " The eldest male parent 
 is absolutely supreme in his household. His dominion 
 extends to life and death, and is as unqualified over his 
 children and their houses as over his slaves ; indeed, 
 the relations of sonship and serfdom appear to differ in 
 little beyond the higher capacity which the child in 
 blood possesses of becoming one day the head of a 
 family himself." Ibid., p. 123. 
 
 " The Family is the type of an archaic society in all 
 the modifications which it was capable of assuming. . . . 
 We must look on the family as constantly enlarged by 
 the adoption of strangers within its circle, and we must 
 try to regard the fiction of adoption as so closely 
 simulating the reality of kinship, that neither law nor 
 
 * The paging is the same in all the editions, as far as this subject 
 is concerned. The writer has before him the first edition (1861), 
 and the seventh (1878). London : Murray.
 
 THE PATEIAECHAL THEORY. 7 
 
 opiiiion makes the slightest difference between a real 
 and an adoptive connexion. On the other hand, the 
 persons theoretically amalgamated into a family by their 
 common descent are practically held together by com- 
 mon obedience to their highest living ascendant, the 
 father, grandfather, or great-grandfather. The patriar- 
 chal authority of a chieftain is as necessary an ingredient 
 in the notion of the family group as the fact (or 
 assumed fact) of its having sprung from his loins ; and 
 hence we must understand that if there be any persons 
 who, however truly included in the brotherhood by 
 virtue of their blood relationship, have nevertheless de 
 facto withdrawn themselves from the empire of its 
 ruler, they are always, in the beginnings of law, con- 
 sidered as lost to the family. It is this patriarchal 
 aggregate the modern family thus cut down on one 
 side and extended on the other which meets us on the 
 threshold of primitive jurisprudence." Ibid., pp. 133-4. 
 
 3. Agnation. " Agnatic relationship is in truth the 
 connexion existing between the members of the Family 
 conceived as it was in the most ancient times." Ibid., 
 p. 147. 
 
 " The foundation of Agnation is not the marriage of 
 father and mother, but the authority of the Father. . . . 
 In truth, in the primitive view, Eelationship is exactly 
 limited by Patria Potestas. Where the Potestas begins, 
 Kinship begins, and therefore adoptive relatives are 
 among the kindred. Where the Potestas ends, Kinship 
 ends, so that a son emancipated by his father loses all 
 rights of Agnation. And here we have the reason why
 
 8 THE PATRIARCHAL THEORY. 
 
 the descendants of females are outside the limits of 
 archaic kinship." Ibid., p. 149. 
 
 " Cognates are all those persons who can trace 
 their blood to a single ancestor and ancestress." . . . 
 Agnates " are all the cognates who trace their connexion 
 exclusively through males," with or without the addi- 
 tion of persons brought into the family " by the arti- 
 ficial extension of its boundaries," as, e.g., by adoption. 
 Ibid., pp. 147-8. 
 
 "It is obvious that the organisation of primitive 
 societies would have been confounded if men had called 
 themselves relatives of their mothers' relatives." Ibid., 
 p. 149. 
 
 4. Universal Prevalence in Primeval Times of 
 Patria Potestas and Agnation. " The Patria Potestas, 
 in its normal state, has not been a durable institution. 
 The proof of its former universality is therefore incom- 
 plete as long as we consider it by itself." Ibid., p. 146. 
 
 " Hence comes the interest of Agnation for the 
 inquirer into the history of jurisprudence. The Powers 
 [Patria Potestas] themselves are discernible in compara- 
 tively few monuments of ancient law ; but Agnatic 
 relationship, which implies their former existence, is 
 discoverable almost everywhere." Ibid., p. 150. 
 
 The view disclosed in the passages quoted and it 
 is stated without reservations or ambiguities is, in 
 effect, that the family, much as it existed among the 
 Romans within the historical period, was primeval 
 and universal; and it is a proof of this that is put 
 forward as establishing the Patriarchal Theory. It
 
 THE PATRIARCHAL THEORY. 9 
 
 should be noticed that Agnation is declared to be the only 
 form of relationship consistent with the structure of 
 primeval society, and therefore the only form of rela- 
 tionship known in the most ancient times. Cognation 
 or natural relationship, had it been acknowledged, 
 would have led to the organisation of primitive societies 
 being confounded, and therefore in primitive times it 
 could not be recognised. 
 
 The Patriarchal Theory, in its new form, could not, 
 any more than in its old form, account for the union, 
 otherwise than as rulers and ruled, of different stocks 
 in the same society. The author of Ancient Law, 
 however, thought that an explanation of the actual 
 heterogeneity of societies which, according to the 
 theory, should have been homogeneous, was to be 
 found in the use in early times of a fiction analogous 
 to adoption and having a similar effect. His view is 
 propounded in vague terms, but the following passages 
 will show what it is, so far as it is defined : 
 
 The Genesis of Society. " In most of the Greek 
 states and in Rome, there long remained the vestiges 
 of an ascending series of groups out of which the State 
 was at first constituted. The Family, House, and Tribe 
 of the Romans may be taken as the type of them, and 
 they are so described to us that we can scarcely help 
 conceiving them as a system of concentric circles which 
 have gradually expanded from the same point. The 
 elementary group is the Family, connected by common 
 subjection to the highest male ascendant. The aggre- 
 gation of Families forms the Gens or House. The
 
 10 THE PATUIAHGHAL THEORY. 
 
 aggregation of Houses makes the Tribe. The aggrega- 
 tion of Tribes constitutes the Commonwealth." Ancient 
 Law, p. 128. 
 
 A Difficulty. The Heterogeneousness of Early 
 Communities. " It may be affirmed of early common- 
 wealths that their citizens considered all the groups 
 in which they claimed membership to be founded on 
 common lineage. . . . And yet we find that along 
 with this belief, or if we may use the word, this theory, 
 each community preserved records or traditions which 
 distinctly showed that the fundamental assertion was 
 false. . . . The composition of the state, uniformly 
 assumed to be natural, was nevertheless known to be 
 in a great measure artificial." Ibid., pp. 129, 130. 
 
 How the Difficulty is dealt with. " The earliest 
 and most extensively employed of legal fictions was 
 that which permitted family relations to be created 
 artificially, and there is none to which I conceive man- 
 kind to be more deeply indebted. If it had never 
 existed, I do not see how any one of the primitive 
 groups, whatever were their nature, could have absorbed 
 another, or on what terms any two of them could have 
 combined, except those of absolute superiority on one 
 side and absolute subjection on the other. No doubt, 
 when with our modern ideas we contemplate the union 
 of independent communities, we can suggest a hundred 
 modes of carrying it out, the simplest of all being that 
 the individuals comprised in the coalescing groups 
 shall vote or act together according to local pro- 
 pinquity ; but the idea that a number of persons
 
 THE PATRIARCHAL THEORY. 11 
 
 should exercise rights in common, simply because they 
 happened to live within the same topographical limits, 
 was utterly strange and monstrous to primitive 
 antiquity. The expedient which in those times com- 
 manded favour was that the incoming population should 
 feign themselves to be deduced from the same stock as 
 the people on whom they were engrafted ; and it is 
 precisely the good faith of this fiction, and the closeness 
 with which it seemed to imitate reality, that we cannot 
 now hope to understand." Ibid., pp. 130, 131. 
 
 That each of the larger groups was an " aggrega- 
 tion " of several groups of the order immediately lower ; 
 that the practice of creating family relations artificially 
 somehow helped on, or may have helped on, the aggre- 
 gation of groups that were not of the same origin ; 
 and that the expedient actually employed was " that 
 the incoming population should feign themselves to be 
 deduced from the same stock as the people on whom 
 they were engrafted " appears to be the view suggested. 
 It is not within the purpose of the present work to 
 consider it closely ; but the following observations may 
 be offered : 
 
 1st. That peoples not of the same origin who had 
 become united, in after times considered themselves of 
 the same stock, is unquestionable. 
 
 But did they use the pretence, known to them to be 
 false, that they were of the same stock to bring about 
 their combination ? or was this only a hypothesis by 
 which the fact of their living in combination was 
 ultimately accounted for ?
 
 12 THE PATRIARCHAL THEORY. 
 
 2nd. The populations which are said to have com- 
 bined on the pretence that they were of the same 
 descent, remained distinct from one another after their 
 union, each retaining its own name. The fiction 
 employed, therefore, had not the same effect as adop- 
 tion. It did not merge the one population in the other 
 so that no trace of heterogeneity was left. By adop- 
 tion, on the other hand, the adopted person became 
 merged in his new family, became a part of it, as if he 
 had been born in it, and nothing survived to show that 
 a stranger had been brought in. 
 
 3rd. Adoption itself can account for no appearance 
 of heterogeneity in Gens or Tribe, because, as has just 
 been said, it left no mark upon the family, nothing to 
 indicate that it had occurred. And there is another 
 reason why no such effect can be attributed to it. In 
 the best-known cases, the person who could be adopted 
 was a person of the adopter's blood. 
 
 " How," says Sir A. C. Lyell,* " does it come to pass 
 that in those primitive societies which assume as their 
 basis a common descent from one original stock, one so 
 constantly finds traces of alien descent ? How came a 
 variety of alien groups to coalesce into a local tribe ? 
 The fiction of male adoption is suggested as an answer, 
 but such adoption from alien stocks is quite unknown 
 throughout India, where the adoption of a son is always 
 
 * " Formation of Indian Clans and Castes," Fortnightly Review, 
 January, 1877. Sir A. Lyell's evidence as to the practice of adoption 
 among modern Hindoos is presumably good. On the value of his 
 speculation as to the origin of Clans and Castes it is unnecessary 
 here to offer any opinion.
 
 THE PATRIARCHAL THEORY. 13 
 
 made within the circle of affinity, ordinarily from the 
 nearest kindred." Adoption, in fact, does not in India 
 bring any alien blood into a family. 
 
 In Rome, too, the person brought into a family by 
 adoption usually belonged to the same Gens, that is 
 stock, as the adopter, and he took his place in the 
 adopter's family just as if he had been born in it. 
 
 Even if, at Rome, the adopted person belonged to 
 a different Gens from the adopter, he was, at any rate, 
 separated absolutely from the Gens of his birth and its 
 sacra, and introduced to the sacra of the adopter's Gens 
 and family. * He bore both the Gentile and the family 
 name of the adopter. Thus the adoption left no per- 
 manent trace there remained nothing to show that a 
 stranger had come into the family. 
 
 Neither in India nor in Rome, then, could adoption 
 account for the presence of Gentes of different stocks 
 within the same local tribe, or for the appearance of 
 families apparently of different stocks within the same 
 Gens. So far of adoption in advanced societies. 
 
 To pass to the primitive races (so called), we find 
 that among them, too, a person adopted takes his place 
 in the family of adoption as if he had been born in it. 
 When a captive taken in war is adopted to fill the place 
 of a person recently lost, he usually takes the very 
 name and place in the family of the deceased. In a 
 widely extended class of cases he takes the Gentile 
 name of the adoptive mother. In the cases more 
 
 * It is extremely doubtful whether this case ever occurred in 
 Rome, though some Roman lawyers say it did.
 
 14 THE PATRIARCHAL THEORY. 
 
 immediately comparable with those of Rome and India 
 cases in which kinship is through males and children 
 are of the stock of their father he takes the (totem) 
 Gentile name of the adoptive father, and is as much 
 under the bond of blood involved in that as if he 
 had by birth belonged to the father's stock. Among 
 primitive races, then, just as among more advanced 
 races, adoption leaves no trace of a stranger having 
 been brought into the stock. It produces, that is, no 
 permanent appearance of heterogeneity. 
 
 How far familiarity with a fiction which produced 
 the appearance of kinship could dispose ancient groups 
 everywhere, or nearly everywhere, to employ in uniting 
 with one another a fiction which did not produce the 
 appearance of kinship nor, so far as appears, any 
 effect whatever is a matter on which it would be idle 
 to speculate. 
 
 It cannot now be too much to say, however, that 
 the view propounded by way of supplement to the 
 Patriarchal Theory is itself so far wrapped in obscurity, 
 and so much open to doubt that the unfitness of the 
 theory by itself to explain the growth of society ought 
 to continue to tell against it. Here it will only be made 
 one of many reasons why it should be held indis- 
 pensable that the evidence for the Patriarchal Theory 
 should be full, clear, and strong.
 
 CHAPTER II. 
 
 SIR HENRY MAINE'S LATER WRITINGS. 
 
 SINCE the publication of Ancient Law much has been 
 written which has tended to raise doubts as to the 
 soundness of the Patriarchal Theory ; and, in the later 
 writings of the author of that work, it is evident enough 
 that he has at times been somewhat troubled about the 
 validity of his early impressions. The Patriarchal Family 
 of his theory has seemed to himself, considering it afresh, 
 a strange and inexplicable institution ; and he could not 
 but see that, if the reports of observers are to be trusted, 
 there have been many bodies of men among whom 
 it has been unknown. Excepting in a single passage 
 in his latest work, however,'"" reconsideration has never 
 carried him beyond the admission of a bare possibility 
 a possibility too faint, apparently, to be worth think- 
 ing seriously about, and which, at any rate, comparative 
 jurisprudence need not concern itself with that the 
 Patriarchal Family was not a primary social fact. It 
 has left him able to state as confidently as he did at 
 first the passage just referred to making no exception 
 * Early Law and Custom, pp. 286-288.
 
 16 THE PATRIARCHAL THEORY. 
 
 to this that, among all the more important of early 
 tribes, among all tribes that are worth any reasonable 
 person's notice, the Patriarchal Family, as he has 
 described it, with despotic power in the father, and 
 Agnation as the only bond of relationship, is to be met 
 with at the beginning of history, either actually existing, 
 or plainly traceable by its incidents and the marks it. 
 has left upon law and custom. The passages subjoined 
 are from his later works, and it will be found they bear 
 out what has just been said : 
 
 1. The Strangeness of the Patriarchal Family. 
 " The Patriarchal Family is not a simple but a highly 
 complex group, and there is nothing in the superficial 
 passions, habits, or tendencies of human nature which 
 at all sufficiently accounts for it. If it is really to be 
 accepted as a primary social fact, the explanation 
 assuredly lies among the secrets and mysteries of our 
 nature, not in any characteristics which are on the 
 surface. Again, under its best ascertained forms, the 
 Family Group is in a high degree artificially constituted, 
 since it is freely recruited by the adoption of strangers. 
 All this justifies the hesitation which leads to further 
 inquiry." Village Communities in the East and West, 
 pp. 15, 16. 
 
 The Patriarchal Theory. Description of the Patri- 
 arcJial Family. " The two societies, Eoman and 
 Hindoo, which I take up for examination . . . are 
 seen to be formed at what for practical purposes is the 
 earliest stage of their history, by the multiplication of 
 a particular unit or group, the Patriarchal Family.
 
 SIR HENRY MAINE'S LATER WRITINGS. 17 
 
 There has been much speculation of late among writers 
 belonging to the school of so-called prehistoric inquiry, 
 as to the place in the history of human society to which 
 this peculiar group, the Patriarchal Family, is entitled. 
 Whether, however, it has existed universally from all 
 time whether it has existed from all time only in 
 certain races or whether, in the races among whose 
 institutions it appears, it has been formed by slow and 
 gradual development it has everywhere, where we find 
 it, the same character and composition. The group 
 consists of animate and inanimate property, of wife, 
 children, slaves, land and goods, all held together by 
 subjection to the despotic authority of the eldest male 
 of the eldest ascending line, the father, grandfather, or 
 even more remote ancestor. The force which binds the 
 group together is Power. A child adopted into the 
 Patriarchal Family belongs to it as perfectly as the 
 child naturally born into it, and a child who severs his 
 connection with it is lost to it altogether. All the 
 larger groups which make up the primitive societies in 
 which the Patriarchal Family occurs, are seen to be 
 multiplications of it, and to be, in fact, themselves more 
 or less formed on its model." The Early History of 
 Institutions, pp. 310, 311. 
 
 Prevalence of the Patriarchal Family. "Among 
 the Aryan sub-races, the Hindoos may be as con- 
 fidently asserted as the Romans to have had their 
 society organised as a collection of patriarchally- 
 governed families." Ibid., p. 323. 
 
 " My suggestion is that the key to the Irish dis-
 
 18 THE PATRIARCHAL THEORY. 
 
 tribution of the Family, as to so many other things in 
 ancient law, must be sought in the Patria Potestas." 
 Ibid., p. 217. 
 
 " I think I can assign some at least plausible reasons 
 for believing that this perplexing fourfold division of 
 the Celtic family is ... a monument of that Power 
 of the Father which is the first and greatest landmark 
 in the course of legal history." Ibid., p. 216. 
 
 " Meantime, let me say something on the transmu- 
 tations which Patriarchal Power is observed, as a fact, 
 to undergo in the assemblages of men held together 
 by kinship which are still found making a part of Aryan 
 communities. The Joint Undivided Family, wherever 
 its beginning is seen in such communities, springs 
 universally out of the Patriarchal Family, a group of 
 natural or adoptive descendants held together by sub- 
 jection to the eldest living ascendant, father, grand- 
 father, or great-grandfather. Whatever be the formal 
 prescriptions of the law, the head of such a group is 
 always in practice despotic, and he is the object of a 
 respect, if not always of an affection, which is probably 
 seated deeper than any positive institution." Ibid., 
 pp. 115, 116. 
 
 " There can be no reasonable doubt that the House 
 Community of the South Slavonians is the Eoman 
 Gens, the Hellenic 761/09, the Celtic Sept, the Teutonic 
 Kin. It is also the Joint Family of the Hindoos, which 
 is itself a living though an extremely perishable institu- 
 tion."- " South Slavonians and .Rajpoots," Nineteenth 
 Century, Dec., 1877.
 
 SIR HENRY MAINE'S LATER WRITINGS. 19 
 
 In Early Law and Custom, published in 1883, this 
 passage reappears with a modification. " There can be 
 no reasonable doubt," it is there said, " that the House 
 Community of the South Slavonians corresponds to 
 one or other of the larger Roman groups" that is, 
 either to the Gens or the body of Agnates. 
 
 " I have, however, no doubt myself, from a variety 
 of indications, that these families [' natural families/ 
 consisting of the descendants of an ancestor still alive] 
 are, to employ a convenient term, patriarchal families 
 despotically governed by the eldest ascendant. . . . 
 The South Slavonians, like the Romans, maintain a 
 clear distinction between Agnatic and Cognatic relation- 
 ship, which they term respectively kinship through the 
 great blood and kinship through the little. Thus a 
 group of men connected with a common ancestor 
 through male descents (natural or adoptive) exclusively, 
 are kinsmen of the great blood ; they are kinsmen 
 of the little blood when they include also the descendants 
 of female relatives. Now the recognition of Agnatic 
 relationship is good evidence that patriarchal power 
 either exists or has once existed in a community ; 
 there may have been paternal power where there 
 is no Agnation, but where there is Agnation there 
 must almost certainly have been paternal power." 
 " South Slavonians and Rajpoots," Nineteenth Century, 
 1877. 
 
 This passage reappears, with a slight modifica- 
 tion in the first sentence quoted, in Early Law and 
 Custom, pp. 243, 244. It is there said that the families 
 
 c 2
 
 20 THE PATRIARCHAL THEORY, 
 
 are, "as a rule," despotically governed by the eldest 
 ascendant. 
 
 "The most recent researches into the primitive 
 history of society point to the conclusion that the 
 earliest tie which knitted men together in communities 
 
 was consanguinity or kinship It was regarded 
 
 as an actual bond of union, and in no respect as 
 
 a sentimental one If a man was not of kin 
 
 to another there was nothing between them. He was 
 
 an enemy to be slain, or spoiled, or hated The 
 
 tribes of men with which the student of jurisprudence 
 is concerned are exclusively those belonging to the races 
 now universally classed, on the ground of linguistic 
 affinities, as Aryan and Semitic. Besides these he has 
 at most to take into account that portion of the out- 
 lying mass of mankind which has lately been called 
 Uralian the Turks, Hungarians, and Finns. The 
 characteristic of all these races, when in the tribal 
 state, is that the tribes themselves, and all sub-divisions 
 of them, are conceived by the men who compose them 
 as descended from a single male ancestor. Such com- 
 munities see the family group with which they are 
 familiar to be made up of the descendants of a single 
 living man, and of his wife or wives ; and perhaps they 
 are accustomed to that larger group, formed of the 
 descendants of a single recently deceased ancestor, 
 which still survives in India as a compact assemblage 
 of blood relatives, though it is only known to us 
 through the traces it has left in our Tables of In- 
 heritance. The mode of constituting groups of kinsmen
 
 SIB HENRY MAINE'S LATER WRITINGS. 21 
 
 which they see proceeding before their eyes they believe 
 to be identical with the process by which the com- 
 munity itself was formed. Thus the theoretical assump- 
 tion is that all the tribesmen are descended from some 
 common ancestor, whose descendants have formed 
 sub-groups, which again have branched off into others, 
 till the smallest group of all, the existing Family, 
 is reached." The Early History of Institutions, pp. 
 64-66. 
 
 It will be seen that, in these passages, the Patriar- 
 chal Family of Ancient Law appears unchanged ; that 
 fresh examples of Patria Potestas are noted ; and that 
 the use of Agnation as furnishing a clue to the previous 
 existence of Patria Potestas, and therefore of the 
 Patriarchal Family, is still insisted upon. The Patriar- 
 chal Family or sure indications of it is found among 
 Romans, Greeks, Hindoos, Celts, Teutons, and Sla- 
 vonians ; while the Patria Potestas is seen on all hands 
 affecting ancient law, and is, in fact, " the first and 
 greatest landmark in the course of legal history." The 
 passage last quoted is an allegation that the Patriarchal 
 Theory has been held by all Aryan, Semitic, and Uralian 
 tribes, and an attempt to show how they came to 
 believe in it. 
 
 It should be added that, so far as the matters under 
 consideration go, Ancient Law itself has remained 
 absolutely unchanged since 1861 ; and, as regards the 
 " proof," from comparative jurisprudence, of the truth 
 of the Patriarchal Theory therein contained, that Sir 
 Henry Maine referred to it in his Village Com-
 
 22 THE PATRIARCHAL THEORY. 
 
 munities* as a proof to which he adhered in toto. 
 " I need not here repeat to you/' he says, " the proof 
 which I have attempted to give elsewhere." 
 
 The passage in Early Law and Custom referred to 
 in the first paragraph of this chapter, makes in some 
 respects a fresh departure from the author's other 
 writings. It occurs in a note appended to one of the 
 essays contained in that volume, and is as follows : 
 
 " The ' Agnatic ' Gentile groups, consisting of all the 
 descendants, through males, of a common male ancestor, 
 began to exist in every association of men and women 
 which held together for more than a single generation. 
 They existed because they existed in nature. Similarly 
 the group consisting of the descendants, through women, 
 of a single ancestress still survives, and its outline may 
 still be marked out, if it be worth anybody's while to 
 trace it. What was new at a certain stage of the 
 history of all or a portion of the human race, must 
 have occurred, not in connection with the Gens, 
 but in connection with the Family. There was always 
 one male parent of each child born, but prevalent 
 habits prevented his being individualised in the 
 mind. At some point of time, some change of sur- 
 rounding facts enabled paternity, which had always 
 existed, to be mentally contemplated ; and further, 
 as a consequence of its recognition, enabled the kinship 
 flowing from common paternity to be mentally con- 
 templated also. As to the new facts which led to this 
 recognition, all that, in my opinion, can be said of them 
 * Village Communities in East and West, p. 15. London, 1871.
 
 SIR HENRY MAINE '8 LATER WRITINGS. 23 
 
 is that they must have been, such as again to give free 
 play to an over-mastering emotional force. Believing, 
 as I do, that when Paternity reappeared, it reappeared 
 in association with Power and Protection, I require no 
 explanation of the fact that the kinship then recognised 
 was kinship through male descents only." Early Law 
 and Custom, pp. 287, 288. 
 
 Here we seem to have a period during which there 
 is an obscuration of paternity an inability to indi- 
 vidualise the father admitted for " all or a portion of 
 the human race." But the Patriarchal Family emerges 
 at the close of it. " Power and Protection " in a father 
 may no doubt mean less than Patria Potestas ; but 
 Early Law and Custom and, indeed, the essay in it 
 to which the note in which this passage occurs is 
 appended has so much to say of Patria Potestas, 
 and of the father's despotic power, that it is not to be 
 supposed that less is meant.
 
 CHAPTER III. 
 
 THE PATRIARCHAL FAMILY. THE AMOUNT OP PROOF OF 
 IT THAT MAY SUFFICE. 
 
 IN a passage already quoted, Sir Henry Maine has told 
 us that "the effect of the evidence derived from 
 Comparative Jurisprudence " is to establish the Patri- 
 archal Theory, as he has stated it. We are going to 
 give some reasons why it should be deemed necessary 
 that the evidence for the Patriarchal Family of his 
 theory, with its incidents of Patria Potestas and 
 Agnation, should be exceedingly clear and strong. 
 
 1. A sound theory of the origin of society should 
 explain at least the leading facts connected with the 
 growth of societies, easily and effectually ; and this, 
 as is frankly admitted by Sir Henry Maine, the 
 Patriarchal Theory cannot do. That fictions, or other 
 such expedients, should play some part in the forming 
 of societies would be in no way surprising. But Sir 
 Henry Maine has to throw the whole work of account- 
 ing for the commonest and most important facts of 
 social organisation upon a fiction, and that a fiction 
 of his own supposing his Patriarchal Theory being
 
 THE PATRIARCHAL FAMILY. 25 
 
 a mere stumbling-block to him. And even as to this 
 fiction he has to admit that its having been used 
 with good faith is " what we cannot hope to under- 
 stand" that its having been employed, that is, if 
 it was employed, is a thing utterly surprising and 
 unintelligible. Surely this failure of the theory raises 
 a presumption, or, at any rate, a feeling requiring to 
 be overborne by evidence, that the family of the 
 theory cannot have been the factor in social growth 
 it is said to have been. On this account alone it 
 would be proper to call for good evidence of its 
 prevalence, and for good reasons for thinking that, 
 where found, it was primordial. 
 
 2. A passage that has been quoted shows that 
 Sir Henry Maine himself thinks it only natural that 
 people should doubt whether the family of his theory 
 is to be accepted as a primary social fact. It is not 
 a simple but a highly complex group, he admits ; 
 "there is nothing in the superficial passions, habits, 
 or tendencies of human nature which at all sufficiently 
 accounts for it ; " if it be a primary social fact, " the 
 explanation assuredly lies among the secrets and 
 mysteries of our nature." 
 
 And, in truth, the Family held together by Power, 
 with blood-relationship recognised in it only to be 
 ignored no relationship at all through women acknow- 
 ledged, no relationship through males acknowledged 
 except in males subject to the father's Power, and 
 between those subject to that Power, a relationship 
 equally close whether they are related by blood or
 
 26 THE PATRIARCHAL THEORY. 
 
 not the Power, too, extending to life and death and 
 sale, and grown-up sons meekly submitting to it 
 propounded to us as the first form of the family, 
 might well be deemed apart from evidence a mere 
 fantastic imagination. Strangely complex as it is, the 
 evidence should be good indeed on which it is accepted 
 as having been primordial and universal. 
 
 As to the complexity of this family, no doubt too 
 much importance may be attached to the idea that 
 institutions in their beginnings are usually simple, and 
 become complex as they grow old. It is an idea borne 
 out, however, by many facts from many fields of nature. 
 No one could believe in the Ornithorynchus as a ger- 
 minal type of animal life. But the family of Sir Henry 
 Maine's theory is almost as curious a complex of types 
 as is the Ornithorynchus. Its head is head partly as 
 being the begetter, and partly as being the owner, 
 of its members ; so that the cementing principle is 
 neither kinship nor property, but a jumble of the two. 
 Kinship is not excluded, for in theory that is, partly 
 in fact, and partly by a fiction the family is made up 
 of the father's descendants, and he is the representative 
 of the family, not its owner ; and, on the other hand, 
 he has over the members of it, and over all that pertains 
 to it, an uncontrolled and unlimited power of disposal. 
 Then it may almost be said to be based upon fictions. 
 By a fiction, the wife, the mother of the family so far 
 as its members are begotten, is not the wife of her lord, 
 but his daughter, and the sister of her own children. 
 The children begotten are, in fact, property of the
 
 THE PATRIARCHAL FAMILY. 27 
 
 father, and, by a fiction, cease to be his children if he 
 sells them. By a further fiction, additional children, 
 who become in the full sense members of the family, 
 may be acquired by him by purchase, or a pretence of 
 purchase, and be to him even as sons and daughters 
 of his blood. Can anyone believe, excepting for 
 convincing reasons, that such a group as this was 
 elementary and primordial ? Is it not to be presumed 
 that, if found at all, it must have a history through 
 which its artificialities can be explained ? It would be 
 as easy to believe in Minerva springing full-armed from 
 the head of Jove as to believe, except under the 
 constraint of evidence, in such a family as rudimentary. 
 To return to our point, surely the evidence to compel 
 belief ought to be exceedingly full and good. Perhaps 
 this has now been sufficiently made out, but there is 
 something more which it would be improper altogether 
 to overlook. 
 
 3. There are many rude societies now existing in 
 which the family is radically different from the Patri- 
 archal Family ; and there is a great mass of evidence 
 which goes to show that an earlier family system, 
 founded solely on the recognition of blood-ties, though 
 of those only which men first learned to acknowledge, 
 everywhere preceded that family system, which, with 
 strange and peculiar incidents superadded, and blood- 
 relationship almost ignored, appears in Sir Henry 
 Maine's theory. It can be shown how this other family 
 system, founded on the recognition of kinship through 
 women, and through women only, would naturally give
 
 28 THE PATRIARCHAL THEORY. 
 
 place to a family system resembling (to speak loosely) 
 that of the theory to a family system in which the 
 father is head of the family, and the children are counted 
 of his blood or stock. And, indeed, the transition from 
 the one system to the other can be clearly illustrated by 
 actual cases, and many of the intermediate stages 
 mapped out. The actual heterogeneousness of societies, 
 too not that the agency of minor causes in producing 
 that need be called in question is simply and effec- 
 tually accounted for by means of the kinship acknow- 
 ledged in this form of the family. It has appeared that 
 the Patriarchal Theory fails to explain the hetero- 
 geneousness of societies ; and it seems impossible to 
 travel onwards from the Patriarchal Family depicted in 
 Ancient Law to the family founded on the recog- 
 nition of kinship through women only, and show how 
 the one can have changed into the other. 
 
 Comparing these two forms of the family, indeed 
 assuming both to have existed there cannot be a 
 question which is the earlier of the two, or which is, 
 so to speak, the more natural. In the one, blood- 
 relationship, struggling into recognition, is found 
 embodied in a system than which there can have been 
 no earlier system of blood ties. In the other, with 
 blood-relationship perceived, we find the recognition of 
 it stifled, and the Father's Power in the place of blood 
 as the actual measure of relationship. If the former 
 ignores one half of the actual blood ties, viz., those 
 arising through males, the latter as absolutely ignores 
 the other half, viz., those arising through females ; and
 
 TEE PATRIARCHAL FAMILY. 29 
 
 it acknowledges no blood tie, even through males, 
 merely because it is a blood tie. In the one, the system 
 of relationship is natural though imperfect ; and that of 
 the other seems almost unnatural that is, to take the 
 best view of it, it appears to be highly artificial. 
 
 That there is evidence of the prevalence of a 
 family system earlier than the Patriarchal Family 
 evidently and almost necessarily earlier if it generally 
 existed at all the incidents involved in which explain 
 the actual constitution of societies, which the Patriar- 
 chal Theory cannot do, is one reason more for carefully 
 scrutinising the evidence for every proposition which is 
 with that theory associated. 
 
 This is not the place to speak at length of the family 
 founded on kinship through women only, or to set forth 
 the evidence of its prevalence. It is not introduced 
 here with the purpose of overbearing the Patriarchal 
 Theory with counter-evidence. The evidence for it, 
 indeed, has never yet been fully exhibited. But much 
 of that evidence is before the world. The portions of 
 the earth " discovered," as we say, within the last three 
 hundred years, supply it in abundance. The study of 
 the ancient nations, too those very nations upon obser- 
 vation of which the Patriarchal Theory has been 
 founded has yielded an amount of it which is by no 
 means inconsiderable, and which inquiry is steadily 
 
 augmenting. 
 
 Sir Henry Maine has stated in one of his later 
 books that, even if there was an earlier kinship than 
 the kinship based upon Power, the fact could affect
 
 30 THE PATRIARCHAL THEORY. 
 
 Comparative Jurisprudence only remotely* so that, so 
 far as it is concerned, the inquiry whether there was 
 an earlier kinship need not be followed up. But if 
 Comparative Jurisprudence, in tracing the history of 
 society, neglects the earlier history of society, and 
 begins where it is convenient to begin, where a be- 
 ginning can be made without much trouble, or at some 
 arbitrarily chosen starting-point, does it not expose 
 itself to very serious chances of mistake ? Surely it 
 must be most unsafe to assume that nothing anterior 
 to the period of law-books, nothing that does not 
 bulk largely in early law-books, can have had any share 
 in the forming of societies. The conclusions of a 
 science that permits of any such assumptions, must be, 
 it would seem, very liable to error, and always liable 
 to be upset by the results of more thorough inquiry. 
 All that is now suggested, however, is that the 
 proof from early law-books and other such sources 
 for the propositions involved in the Patriarchal Theory 
 ought to be full, clear, and strong. 
 
 * Early History of Institutions, p. 57.
 
 CHAPTEE IV. 
 
 PLAN OF THIS WORK. 
 
 EXAMINING the evidence propounded in support of the 
 Patriarchal Theory, we find it to consist in some 
 measure of direct proof of the existence of the 
 Patriarchal Family as Sir Henry Maine has described 
 it, and in a much greater degree of evidence of certain 
 facts from which the former existence of that institution 
 is held to be a legitimate if not a necessary inference. 
 
 As to the direct evidence, Sir Henry Maine thinks 
 he has found the Patriarchal Family of the Eoman 
 type, with Patria Potestas for its leading feature, (1) 
 among the Hebrews, (2) among the Hindoos, (3) among 
 the Slavonians, (4) among the Irish. It is proposed 
 in separate chapters to examine the evidence adduced, 
 or statements made, with respect to each of these 
 peoples ; and on the other hand such evidence as 
 Comparative Jurisprudence allows appeal to being 
 alone insisted upon to see what is really indicated 
 by the evidence we possess about each. 
 
 Passing to what may be called the indirect evidence, 
 the first and by far the most important branch of it
 
 32 THE PATRIARCHAL THEORY. 
 
 consists of the prevalence of Agnation. Premising that 
 the Patria Potestas in its normal shape has not been, 
 and could not have been, a durable institution, Sir 
 Henry Maine argues that agnatic relationship implies 
 its former existence ; and states that, whereas cases 
 of Patria Potestas are rare, agnatic relationship is 
 discoverable almost everywhere, and that, in fact, it 
 would be difficult to say where it has not existed. 
 The inference from this, of course, is that Patria 
 Potestas has existed almost everywhere, and that, 
 virtually, it may be taken to have existed everywhere. 
 
 It is proposed to consider at some length the 
 connection alleged to exist between Agnation and 
 Patria Potestas ; and, as a preliminary thereto, to 
 discuss though necessarily in a brief and fragmentary 
 way the conditions upon which, in various circum- 
 stances, the connection between what we take to be a 
 derivative institution and that from which we think it 
 is derived can be held to be established. A full and 
 satisfactory discussion of this subject would be of use in 
 many inquiries. That which will here be ventured 
 upon will, it is hoped, be found not to be without 
 bearings upon the Patriarchal Theory. It may perhaps 
 suffice to show that certain propositions made in 
 Primitive Marriage with which the Patriarchal Theory 
 has to reckon that, for example, which connects 
 the Form of Capture and actual capture, and that 
 which connects the Levirate with Tibetan polyandry 
 stand, as regards the conditions of proof, upon a very 
 different footing from the connection alleged to exist
 
 PLAN OF THIS WORE:. 33 
 
 between Patria Potestas and Agnation. A different 
 account of the origin of Agnation will then be put 
 forward, and supported as far as can be done in a work 
 the purpose of which is chiefly critical. For the intro- 
 duction of this theory of Agnation apology can scarcely 
 be necessary. Its introduction will be fully justified, 
 it would seem, if it can be thought at all fit to compare 
 with Sir Henry Maine's account of Agnation. And, in 
 so far as it goes to show that Agnation was not a 
 primary form of kinship, it is directly subversive of the 
 Patriarchal Theory. 
 
 After this, the cases of Agnation which have from 
 time to time been specified by Sir Henry Maine will 
 come up for examination. It will appear that there is 
 not one of them, the Eoman case excepted, in which 
 relationship was clearly agnatic. And, indeed, there is no 
 proof that, before the date of the Twelve Tables, Agna- 
 tion was established even among the Komans outside 
 the Patrician class. Certain admissions incidentally 
 made in Sir Henry Maine's latest work, Early Laiv 
 and Custom, may, if we choose, save us trouble at this 
 point ; and, indeed, it will be found that the chapters 
 on Patria Potestas unavoidably treat of Agnation 
 also for some of the peoples with which they are 
 concerned. The work just mentioned, however, puts 
 at this point an additional labour upon us. In it in 
 the case of the Hindoos Sir Henry Maine has shifted 
 his ground, not founding upon the usual data of Com- 
 parative Jurisprudence, the law-books, or even upon the 
 earlier Hindoo writings, but taking his departure from
 
 3i THE PATRIARCHAL THEORY. 
 
 Ancestor-Worship. It becomes necessary to examine 
 minutely the account which he has thereupon given of 
 the Levirate and of certain curious varieties of sonship 
 curious, that is, as occurring among a people said to have 
 been " Patriarchal " acknowledged among the Hindoos. 
 And, that done, an attempt will be made to show what 
 the basis of family right among the Hindoos really was. 
 If it is successfully made out that Sir Henry Maine has 
 failed to account for the Levirate and Hindoo sonship 
 in general the Patriarchal Theory having to reconcile 
 itself with these the introduction of the former part of 
 this discussion (but that, indeed, could not be avoided) 
 will forward the purpose of this work. And, in so far 
 as, by the latter part of it, it is made probable that the 
 system of kinship through females only can be descried 
 among the prehistoric Hindoos, this part of it is sub- 
 versive of the Patriarchal Theory. 
 
 The other facts from which Sir Henry Maine has 
 made inference of Patria Potestas are the tutelage of 
 women and the heirship of slaves. But it will appear 
 that these need not long detain us.
 
 CHAPTER V. 
 
 PATRTA POTESTAS AMONG THE HEBREWS. 
 
 THE passage in Ancient Law* which affirms that 
 Patria Potestas existed among the Hebrews is as 
 follows : 
 
 " The effect of the evidence derived from Compara- 
 tive Jurisprudence is to establish that view of the 
 primeval condition of the human race which is known 
 as the Patriarchal Theory. There is no doubt, of course, 
 that this theory was originally based on the Scriptural 
 
 history of the Hebrew Patriarchs in Lower Asia 
 
 The chief lineaments of such a society [that is a society 
 organised on the patriarchal model], as collected from 
 the early chapters of Genesis, I need not attempt to 
 depict with any minuteness, because they are familiar 
 
 to us from our earliest childhood The points 
 
 which lie on the surface of the history are these : The 
 eldest male parent the eldest ascendant is absolutely 
 supreme in his household. His dominion extends to 
 life and death, and is as unqualified over his children 
 and their houses as over his slaves ; indeed, the relations 
 
 * Pp. 122-124.
 
 36 THE PATRIARCHAL THEORY. 
 
 of sonsliip and serfdom appear to differ in little beyond 
 the higher capacity which the child in blood possesses 
 of becoming one day the head of a family himself." 
 
 Further details are added of course on the authority 
 of Scripture, but without any particular Scripture being 
 referred to as to the father owning the flocks and 
 herds of all his children, and as to the law of in- 
 heritance. In the sentences omitted from the above 
 citation there is a reference to the controversy between 
 Locke and Filmer, and a suggestion that there is no 
 race of men that was not originally organised on the 
 patriarchal model. 
 
 Sir Kobert Filmer preceded Sir Henry Maine in 
 alleging, on the authority of Scripture, that Patria 
 Potestas existed among the Hebrews, and he set forth 
 the Scriptures on which he relied in support of his con- 
 tention, which the later writer thought it unnecessary 
 to do. To those who have studied the controversy 
 between Locke and Filmer* it may seem wonderful that 
 the truth of Filmer's main position could be thus lightly 
 
 * The reader who wishes to read this controversy for himself will 
 find Locke's part of it in Two Treatises on Government. In the 
 former the false principles and foundation of Sir Robert Filmer and 
 his followers are detected and overthrown. The latter is an essay 
 Concerning the true original extent and end of Civil Government. 
 No author or publisher is named. The book was printed in London, 
 in 1690. In the second essay will be recognised at once Locke's 
 famous Essay on Government. Filmer's Patriarcha, or the Natural 
 Power of Kings [by the learned Sir Robert Filmer, Baronet] was 
 printed in London, in 1680. It is bound, with separate title and 
 paging, in one volume [which is in 'the London Library] with 
 Filmer's Observations Concerning the Original and various Forms of 
 Government. London, 1696.
 
 PATRIA POTESTAS AMONG THE HEBREWS. 37 
 
 assumed by any one, and especially by any lawyer, who 
 had read Locke's masterly reply to the pleadings of his 
 opponent. 
 
 The question, however, being whether the Scriptures 
 do prove Patria Potestas or not, a short notice of the 
 matters discussed between Locke and Filmer will carry 
 us some way towards the settlement of it. 
 
 The main facts relied on by Filmer for Patria Potestas 
 in Israel were : (1), The sentence passed by Judah on 
 Thamar ; (2), Abraham's league with Abimelech ; (3), 
 Abraham's army of three hundred and eighteen soldiers 
 of his own family ; (4), The fifth commandment, which 
 Filmer takes to be the law enjoining obedience to kings, 
 and of which he says : It is delivered in the terms, 
 " Honour thy father," as if all power were originally in 
 the father. He inferred that the father was absolutely 
 supreme in his household, that he had the power of life 
 and death over his children, the power of making war 
 and peace, and generally all the powers of a King. 
 
 Locke, replying, pointed out, (1) as to the injunction 
 on which the duty of obedience to kings was founded, 
 that it was got by only partially quoting the command- 
 ment, which is " Honour thy father and thy mother ; " 
 and that, by the same method, it could be as easily and 
 as conclusively shown that all power was originally in 
 the mother. As illustrating the position of mothers 
 among the Hebrews, he cited Exodus xx. 12, "Honour 
 thy father and thy mother;" Exodus xxi. 15, " He that 
 smiteth his father or his mother shall surely be put .to 
 death;" Leviticus xx. 9, "Every one that curseth his
 
 38 THE PATRIARCHAL THEORY. 
 
 father or his mother shall surely be put to death ; " 
 Leviticus xix. 3, " Ye shall fear every one his mother 
 and his father " in which the woman is named first ; 
 Deuteronomy xxi. 18-21, setting forth how a son guilty 
 of habitual disobedience to the voice of his father or 
 the voice of his mother was to be dealt with ; and other 
 texts to the same effect. 
 
 (2) As to the case of Judah and Thamar, Locke 
 distinguished between an act and the right to do it. 
 ie Absalom," he said, " pronounced sentence against 
 Amnon and executed it too." He might have said 
 that what Judah did was no more than to declare the 
 well-known punishment of the offence. There was a 
 trial ; and the woman was acquitted. More to the point 
 is the observation that Judah was a younger son, whose 
 father and elder brethren were alive. He was not the 
 Paterfamilias. It was not he who, according to the 
 Patriarchal Theory, should have had the power of life 
 and death. If he had this power, how, consistently 
 with the theory, did he happen to have it? "Any 
 man," says Locke, "as well as Judah might have 
 right of dominion." 
 
 (3) In Genesis xlii. and xliii. is the story of Israel's 
 trouble about Benjamin, which shows the whole family 
 still clustered round the father. " Reuben," Locke re- 
 marked, " offered his two sons as pledges, and Judah 
 was at last surety, for Benjamin's safe return out of 
 Egypt which all had been vain, superfluous, and but a 
 sort of mockery if Jacob had had the same power over 
 every one of his family as he had over his ox or his ass,
 
 PATRIA POTESTAS AMONG TEE HEBREWS. 39 
 
 as an owner over his substance, and the offers that 
 Reuben and Judah made had been such a security for 
 the returning of Benjamin as if a man should take two 
 lambs out of his lord's flock and offer one as security 
 that he will safely return the other." 
 
 Putting aside the league with Abimelech, and the 
 magnitude of Abraham's following, as, by themselves at 
 least, illustrating nothing except the might and the 
 independent position of the patriarch, it will be seen 
 that Filmer's evidence for Patria Potestas among the 
 Hebrews consisted of the fifth commandment, mutilated 
 so as to distort its meaning ; and of what he (wrongly) 
 took to be an example of the exercise of the power of 
 life and death, not by a Patriarch, but by a person who, 
 had Patria Potestas really existed, would have been 
 subject to Patria Potestas himself. Filmer, that is, with 
 all the will in the world to find evidence for Patria 
 Potestas among the Hebrews, in fact found no evidence 
 of it whatever. It need scarcely be pointed out that 
 the fifth commandment throws no light upon the powers 
 possessed by the Hebrew father over his family (though 
 had he possessed the Roman Patria Potestas the com- 
 mandment might have been, as regarded him, unneces- 
 sary) ; nor that it does show, as many other facts do, 
 that a very high position was assigned to the mother 
 a position very different from that of sister to her 
 children, assigned to her by the Patriarchal Theory. 
 
 As to the story of Benjamin, upon which Locke 
 commented so acutely, it, at any rate, does not read as 
 if the father had the power of life and death over his
 
 - 
 
 40 THE PATRIARCHAL THEORY. 
 
 grown-up sons. The position of the sons seems very 
 distinguishable from serfdom. They show much defe- 
 rence for their father, no doubt, but they address him 
 like men who have a right to be listened to, and, for 
 the general good, press him, and almost coerce him, 
 into a course he was most averse to. And, if any 
 inference is to be made from the story, it can only be 
 that suggested by Locke that Reuben's sons were his 
 own and not his father's, and that Jacob had nothing 
 like the powers of an owner over Judah in other 
 words, that Jacob had not Patria Potestas. * 
 
 Further on in the history of the Israelites there is 
 no doubt whatever that the father had not Patria 
 Potestas. In Deuteronomy xxi. 18-21, provision is 
 made for the case of a man having a stubborn and 
 rebellious son, who will not hear the voice of his father 
 or the voice of his mother, and whom they have in vain 
 endeavoured by chastisement to correct. The father 
 and mother are to lay hold on the son, and bring him 
 before the elders of his city and unto the gate of his 
 place, and charge him before the elders with his offence. 
 The elders being satisfied, " all the men of his city " 
 were to " stone him with stones that he die." It will 
 be noted that father and mother were both required 
 to come forward, and that they appeared as accusers 
 merely. The idea of Patria Potestas is here excluded. 
 The mother was as necessary to the proceeding as the 
 
 * Giving a son as a hostage is, of course, no proof that the father 
 has the admitted power of life and death over his son. But a power 
 extending to life and death over young children only would not be 
 Patria Potestas.
 
 P ATRIA POTESTAS AMONG THE HEBREWS. 41 
 
 father, which by itself excludes Patria Potestas ; and the 
 power of life and death was with neither of them, nor 
 with both together, but the customary sentence was 
 executed by all the men of the city after a hearing 
 before the elders. 
 
 Going back beyond the time of the Patriarchs, we 
 find a piece of evidence which seems absolutely contra- 
 dictory of the whole Patriarchal Theory. The first 
 reference to marriage in the Scriptures (Genesis ii. 24) 
 mentions father and mother in a breath, and involves 
 that their son left them when he married. "There- 
 fore shall a man leave his father and his mother and 
 shall cleave unto his wife." The words, ascribed to 
 Adam, must be taken as embodying very early custom. 
 Now what do they mean ? Can leaving father and 
 mother (not the father only, observe) mean less than 
 leaving the household, leaving the family, of one's 
 birth ? 
 
 On the Patriarchal Theory, however, a man, when 
 he married, did not leave the household of his birth. He 
 was not separated from his father and mother ; he con- 
 tinued in their family, subject to his father's power, in a 
 condition scarcely distinguishable from serfdom. It was 
 the woman whom he married who left father and 
 mother giving up all relationship with them to 
 cleave unto her husband, to become a member of his 
 father's family, to become, with him, subject to his 
 father's Power. In a bride's case, on the Patriarchal 
 Theory, there was a real leaving. On her husband's 
 part there was none.
 
 42 THE PATRIARCHAL THEORY. 
 
 Even if the passage mean only what, no doubt, it 
 has commonly been taken to mean that a son, on 
 marrying, became emancipated, or, in familiar phrase, 
 set up house for himself, it is contradictory of the 
 Patriarchal Theory. For in that case it involves that 
 after getting a wife, at any rate, a man was free from 
 his father's authority ; while Patria Potestas extended 
 to all descendants and lasted as long as the oldest 
 ascendant lived. 
 
 But, indeed, unless by a man leaving his father and 
 his mother and cleaving unto his wife it can be meant 
 
 o 
 
 that he continued in his father's family (his mother 
 ranking in it as his sister) and brought his wife to 
 live in it, there is complete discordance between the 
 Patriarchal Theory and the earliest Scriptural record of 
 marriage custom. 
 
 This is all we are here concerned with. But it will 
 be well to go a step further and point to a marriage 
 system, altogether different from that of the Patriarchal 
 Theory, which may be that which is indicated in the 
 passage in Genesis. This marriage system seems to be, 
 in a subsequent portion of the Book of Genesis, clearly 
 disclosed as occurring among kindred of the Israelites ; 
 and, in the fullest sense of the words, it makes a man 
 leave his father and his mother to cleave unto his 
 wife. 
 
 It is what is known in Ceylon as beenah marriage 
 by which name, as having already been applied to it, 
 let us call it. In beenah marriage the young husband 
 leaves the family of his birth and passes into the family
 
 PATBIA POTESTA8 AMONG THE HEBREWS. 43 
 
 of his wife, and to that he belongs as long as the 
 marriage subsists. The children born to him belong, not 
 to him, but to the family of their mother. Living with, 
 he works for, the family of his wife ; and he commonly 
 gains his footing in it by service. His marriage involves 
 usually a change 'of village; nearly always (where 
 the tribal system is in force) a change of tribe so that, 
 as used to happen in New Zealand, he may be bound 
 even to take part in war against those of his father's 
 house ; but always a change of family. The man leaves 
 father and mother as completely as, with the Patriarchal 
 Family prevailing, a bride would do ; and he leaves 
 them to live with his wife and her family. That 
 this accords with the passage in Genesis will not be 
 disputed. 
 
 It is in Africa that beenah marriage is now most 
 prevalent ; and there are parts of Africa in which it 
 is quite commonly met with usually alongside of, and, 
 in some sense, contending with, a system of marriage 
 by purchase of the bride and her issue the two 
 systems, indeed, being generally in use even among 
 the same people, the one preferred in some cases, the 
 other in others. 
 
 Its occurrence must be familiar enough to students 
 of works of travel, and it would be superfluous here 
 to accumulate examples of it.* What is more to the 
 purpose is to point out that Jacob made a beenah 
 
 * Reference may be made, however, to Marsden's History of 
 Sumatra, because of the very interesting and instructive account it 
 gives of beenah marriage as practised in that island.
 
 44 THE PATMIABCHAL THEORY. 
 
 marriage into the family of Laban ; and that Genesis 
 xxiv. 1-8 shows that it was thought not improbable 
 that Isaac, as a condition of marrying into his father's 
 kindred, might have to do the same. 
 
 Keeping to the former case, as being the clearer 
 and of itself sufficient, we find, first, that Jacob had 
 to buy his place in Laban's family as husband of 
 Laban's daughters, by service ; and, second, that the 
 children born to him belonged to Laban's family, and 
 not to him both notes of beenah marriage, and the 
 second denoting it beyond possibility of mistake. 
 Jacob, his wives concurring, stole away with them 
 and their children from his father-in-law. And Laban, 
 when he had overtaken him, claimed both the wives 
 and their chi]dren as his own. " These daughters," 
 we find him saying (Genesis xxxi. 43), "are my 
 daughters, and these children are my children." And 
 further on, after he had agreed tq let them go, it is 
 said that " Laban rose up and kissed his sons and 
 daughters, and blessed them." It is easy to under- 
 stand how they were Laban's and not Jacob's. What 
 could have made them Jacob's was purchase ; and 
 Jacob had not purchased. In Laban's days no marriage 
 arrangement was at all likely to be made that was not 
 well known and sanctioned by custom ; and, therefore, 
 it must be taken that beenah marriage was to what 
 extent cannot be known customary in the land of 
 Haran. The case of Eebekah proves that it was not 
 exclusively practised for she was purchased, and left 
 the family of her birth. But, as has been mentioned,
 
 PATBIA POTESTAS AMONG TEE HEBREWS. 45 
 
 beenah marriage is now seldom found without marriage 
 by purchase of the bride being found alongside of it 
 in some places the one, in other places the other, being 
 the more in vogue ; while, quite commonly, the one 
 or the other, according to circumstances, is preferred 
 in different cases by the same people. 
 
 If it be beenah marriage that is indicated in Genesis 
 ii. 24 there is at once an end of the Patriarchal Family, 
 so far as the Hebrews are concerned, and therefore 
 an end of it as a universal and primordial institution. 
 Auol if Jacob's was a beenah marriage which can 
 scarcely be seriously disputed why doubt that beenah 
 marriage is there indicated ? Marriage by purchase 
 of the bride and her issue can hardly be thought to 
 have been primeval practice. When we find beenah 
 marriage and marriage by purchase as alternatives, 
 therefore, it is not difficult to believe that the former 
 is the older of the two, and that it was once in sole 
 possession of the field.* 
 
 Putting aside this question of beenah marriage, how- 
 ever, the story of Laban and his family at any rate 
 enables us to decide whether the Patriarchal Family, 
 with its incidents of Patria Potestas and Agnation, 
 existed among the early Hebrews. 
 
 On the Patriarchal Theory, Laban's family should 
 have consisted only of his sons and their descendants 
 and his unmarried daughters. His daughters should 
 have been cut off from him by marriage ; and their 
 children, because included with them in another man's 
 * As to this, see p. 273, et seq.
 
 46 THE PATRIARCHAL THEORY. 
 
 family, should have been as strangers to him.* Laban's 
 daughters, their husband, and their children, all lived 
 with him, however, and he continued to regard his 
 daughters as daughters, and counted their children 
 among those belonging to him. " The organisation of 
 primitive societies would have been confounded," says 
 Sir Henry Maine, if men had acknowledged relationship 
 through women. Here is Laban, nevertheless, claiming 
 his daughter's children as his own. And, before that, 
 he had at once acknowledged relationship between him- 
 self and his sister Eebekah's son. " Surely," he had 
 said to Jacob, " thou art my bone and my flesh," Need 
 it be said that Laban's was clearly not a " Patriarchal " 
 family ? Laban had never dreamed of Agnation. What 
 authority he claimed over his household does not 
 appear ; but whatever it was, the limits of its applica- 
 tion were quite different from those of Patria Potestas. 
 
 * Jacob's wives, justifying to themselves the proposed desertion of 
 their father's household, say, " Are we not counted of him (their father) 
 as strangers ? For he hath sold us." But clearly this is only said to 
 justify to themselves what they were going to do. In fact, they had 
 all along been members of Laban's household ; and the sequel shows 
 that they were not counted by him as strangers. Plainly he had not 
 sold them. If he had done that, Jacob's right to go away with them 
 and their children would have been unquestionable, and it is clear 
 that he knew it to be open to question. Their initial outburst, " Is 
 there yet any portion or inheritance for us in our father's house 1 " 
 discloses a state of mind impossible to women brought up in a 
 " patriarchal " family. Such women would take it as of course that, 
 once married, they could have nothing to look for from the family of 
 their birth. Laban's daughters speak, be it observed, of the property 
 Jacob was about to carry off as theirs and their children's, seeming so 
 to justify his taking it away.
 
 PATRIA POTESTAS AMONG THE HEBREWS. 47 
 
 Jacob, on his part, had gone to his mother's brother, 
 feeling sure he would find protection as a relative. He 
 was most kindly received. In his marriages he only had 
 to submit to the custom of the country. In his circum- 
 stances, there could be no question of his getting a wife, 
 as his father Isaac had done, by purchase. And, as bear- 
 ing on Patria Potestas, it should be noticed that, by the 
 custom of the country, a family which took in a man to 
 be husband to one of its daughters, instead of compen- 
 sating the man's father for the loss of Potestas over 
 him, exacted a price in service for his admission. Patria 
 Potestas must, therefore, have been unknown in Haran. 
 One can the more easily believe it was unknown in 
 the household of the husband of Labau's daughters. 
 We have seen that, in Jacob's case, there is literally not 
 a trace of it. 
 
 Of minor facts which go to show that Patria Potestas 
 was unknown among the early Hebrews and their 
 kindred, only one or two need be added. 
 
 And, first, the concentration of all family property 
 in the hands of the Paterfamilias being among the 
 features of Patria Potestas, what is to be made of the 
 fact again from Haran that at the espousal of 
 Rebekah, the bridal gifts (the bride's price) were given, 
 not to the father of the bride, but to her brother and 
 her mother ? The father, if he had had Patria Potestas, 
 ought to have got them all, as " compensation for the 
 Patriarchal or Family authority which was transferred 
 to the husband ; " and, indeed, his wife and son should 
 have had no property independently of him. But he
 
 48 THE PATRIARCHAL THEORY. 
 
 got none. The wife and son were capable of having 
 property, and it was they who were compensated. Not 
 to dwell upon the other bearings of this which, indeed, 
 are obvious enough nothing could be more conclusive 
 against Patria Potestas. 
 
 Again, we find that each of Laban's daughters, on 
 her marriage, got a maid from her father ; and the maid 
 in each case seems to have become the daughter's 
 property, of which she could dispose at her own will. 
 Each daughter gave her maid to Jacob to bear children 
 for her, and the children were counted with those of the 
 mistress. * Similarly, in the house of Abraham, Sarah 
 gave to her husband her Egyptian maid to wife ; and 
 though Abraham, had he been a true Paterfamilias, 
 should have been king in his own house, it would 
 seem as if, without Sarah's consent, he could not have 
 begotten Ishmael. Afterwards, on his wife's order, he 
 
 o * 
 
 had, sorely against his will, to turn Hagar and Ishmael 
 out of doors. It does not look as if Sarah was in manu. 
 
 As bearing on the law of succession, and the repre- 
 sentative character sustained in primitive times, on the 
 Patriarchal Theory, by the head of the family, observe 
 that Abraham, though he had sons after Sarah's death, 
 left all that he had to Isaac, dismissing his other sons 
 with gifts in his lifetime. Abraham seems to have 
 acted as full proprietor of his estate. 
 
 As to Agnation, besides the evidence of its non- 
 existence already adduced, there is abundant Scriptural 
 evidence to show that, instead of its being established, the 
 * As to this, see note on page 273.
 
 PATRIA POTESTA8 AMONG THE HEBREWS. 49 
 
 relatives on the mother's side were anciently the closest 
 kindred. One case will prove this as well as twenty, 
 and, to take the first that comes to hand, see Judges 
 ix. 1-4. Of course Abimelech's pretension to rule arose 
 out of his being a son of Gideon. But it was his 
 mother's family that helped him to power, and they did 
 so because he was "their bone and their flesh." It is 
 Patria Potestas that is here being dealt with. But what 
 is the Patriarchal Family without Agnation ? 
 
 Sir Henry Maine says in one place* that the connec- 
 tion of the Patriarchal Theory with Scripture was rather 
 against its reception as a complete theory, because most 
 of the early inquirers into social phenomena were either 
 under the strongest prejudice against Hebrew antiqui- 
 ties, or were strongly desirous of constructing systems 
 without the assistance of religious records. The Scrip- 
 tures make it impossible, however, to accept as a com- 
 plete theory the Patriarchal Theory as it has been 
 enunciated by himself. Plainly, the Hebrews must be 
 excepted from it. The Scriptures not only do not 
 countenance it, but they contradict it, so far as the 
 Hebrews are concerned. It is not merely that they 
 contain nothing to suggest that the Family founded 
 upon Power ever existed among the Israelites. All the 
 evidence there is goes to show that the Eoman institu- 
 tion of Patria Potestas never prevailed among them ; 
 and, as regards their early history, which is what 
 concerns us, we have incidentally found complete 
 disproof of Agnation. It is needless to dwell upon 
 * Ancient Law, p. 122.
 
 50 THE PATRIARCHAL THEORY. 
 
 the indications that have appeared of their having had 
 in very early times a family system almost the converse 
 of the Patriarchal though these must count for some- 
 thing in estimating the claims of the Patriarchal Theory 
 to acceptance. So much may be said, notwithstanding 
 that it has to be borne in mind that Sir Henry Maine 
 has not adduced the evidence from Hebrew antiquities 
 and religions records upon which his opinion appears to 
 have been formed. As to impressions remaining with 
 us " from our earliest childhood," too much must not be 
 made of them. It is certain, however, that, in spite of 
 such impressions, the picture of the Hebrew family 
 sketched in Ancient Law comes upon most people 
 as a surprise.
 
 CHAPTER VI. 
 
 PATEIA POTESTAS AMONG THE HINDOOS. 
 
 IN Ancient Law Sir Henry Maine has stated that the 
 greater part of the legal testimony to the truth of the 
 Patriarchal Theory comes from the institutions of the 
 Romans, Hindoos, and Slavonians ; adding as if under 
 pressure of the evidence at his disposal that "the 
 difficulty at the present stage of the inquiry is to know 
 where to stop, to say of what races of men it is not 
 allowable to lay down that the society in which they 
 were united was originally organised on the patriarchal 
 model." His description of the Patriarchal Family 
 immediately follows the eldest male parent, the eldest 
 ascendant, absolutely supreme in his household, his 
 dominion extending to life and death, and being as 
 unqualified over his children and their houses as over 
 his slaves, while there is no property in the hands 
 of any of them that is not really his. It is quite clear, 
 then, that he considered Patria Potestas to have been 
 a feature of the family system of the Hindoos, though 
 he does not in Ancient Law affirm that there is 
 direct legal testimony to that effect. In his Early 
 
 E 2
 
 52 THE PATRIARCHAL THEORY. 
 
 History of Institutions (p. 323), however, he says 
 that " The Hindoos may be as confidently asserted as 
 the Eomans to have had their society organised as a 
 collection of patriarchally governed Families ; " and, in 
 another place (p. 310), that the two societies of Rome 
 and India " are seen to be formed at what, for practical 
 purposes, is the earliest stage of their history, by the 
 multiplication of a particular unit or group, the 
 Patriarchal Family." And a description of this unit 
 or group follows the latter passage : " The group consists 
 of animate or inanimate property, of wife, children, 
 slaves, land and goods, all held together by subjection 
 to the despotic authority of the eldest male of the eldest 
 ascending line, the father, grandfather, or even more 
 remote ancestor." In the latter passage he does seem 
 to speak as if there was direct evidence of Patria 
 Potestas among the Hindoos. At any rate, the 
 prevalence of that institution among them has always 
 been " confidently asserted " by him, and always 
 spoken of as if there was evidence for it of some 
 sort that must compel belief. 
 
 Let us see, then, what Hindoo law-books since 
 it is to law-books that Comparative Jurisprudence 
 makes appeal show as to the existence or non- 
 existence of Patria Potestas among the Hindoos. 
 
 This is not the place to discuss the comparative 
 antiquity and authority of Hindoo law-books. It must 
 suffice to say that, until recently, nobody ever doubted 
 that the most ancient and the fullest statement we 
 have of early Hindoo law is contained in the Manava
 
 PATEIA POTESTAS AMONG TEE HINDOOS. 53 
 
 Dharma Shastra, best known to us as the Code of 
 Manu ; and that, though this is now in question, the 
 reasons for allowing the work the authority that has 
 been assigned to it do not as yet appear to have been 
 shaken. From it will now be cited some passages which 
 bear upon Patria Potestas : 
 
 1. "Let no father who knows the law receive 
 a gratuity, however small, for giving his daughter 
 in marriage, since the man who . . . takes such a 
 gratuity is a seller of his offspring." Manu, chap, iii., 
 ver. 51. 
 
 From this it appears that a Hindoo father had no 
 power to sell his offspring. 
 
 2. Chap. iv. (On Economics and Private Morals), 
 ver. 180, declares it to be the duty of a housekeeper to 
 have no strife with his mother, father, son, wife, or 
 daughter. In vers. 184 and 185 of the same chapter, 
 it is declared that he must consider "his elder brother as 
 equal to his father, his wife and son as his own body, 
 his assemblage of servants as his own shadow, and his 
 daughter as the highest object of tenderness." When 
 offended with any of them, he is told to bear the offence 
 without indignation. 
 
 It was the duty of a housekeeper, that is, to be 
 self-restrained, forbearing, indulgent towards his family. 
 
 3. From chap. viii. (On Judicature) it appears that 
 all jurisdiction was vested in the king or his judges, and 
 in the father of a family none whatever. " Altercation 
 between man and wife, and their several duties " is 
 named (ver. 7) as one of the eighteen principal titles of
 
 54 THE PATRIARCHAL THEORY. 
 
 law to be daily dealt with by the courts. That is, 
 altercations between husband and wife, and questions 
 respecting their several duties were matters for settle- 
 ment in the courts of justice. 
 
 Here it may be seen that the Hindoo father was 
 not " absolutely supreme in his household," and that his 
 wife was not in manu. If he had a dispute with his 
 wife, the courts decided between them. 
 
 And the jurisdiction of the courts extended to all 
 the members of the family alike, and to all offences 
 which they could respectively commit against one 
 another. " Neither a father, nor a preceptor, nor a 
 friend, nor a mother, nor a wife, nor a son, nor a 
 domestic priest must be left unpunished by the king if 
 they adhere not with firmness to their duty." (Manu, 
 chap, viii., ver. 335.) 
 
 It appears from the same chapter that a father had 
 a certain power of correcting the members of his 
 family. "A wife, a son, a servant, a pupil, and a 
 younger whole brother may be corrected, when they 
 commit faults, with a rope or the small shoot of a 
 cane ; but on the back part only of their bodies, and 
 not on a noble part by any means. Who strikes them 
 otherwise than by this rule incurs the guilt or shall 
 pay the fine of a thief." (Chap, viii., vers. 299, 300.) 
 On the former of these verses Mr. Colebrooke observes :* 
 " May I quote a maxim of no less authority ? ' Strike 
 not, even with a blossom, a wife guilty of a hundred 
 faults/ " But, at any rate, these verses show that the 
 * Hindu Digest, Vol. II., p. 209.
 
 PATEIA POTESTAS AMONG THE HINDOOS. 55 
 
 Hindoo father's power over his family did not extend to 
 life and death. Even the power of correction allowed 
 him a power which, as regards children, fathers 
 probably have among every people was carefully 
 limited, and a penalty prescribed for any abuse of it. 
 From the same chapter (ver. 335, already quoted, and 
 ver. 389) it appears that, while his powers over his 
 family were thus limited, he was under strict legal 
 obligation to his family to stand by and support it, and 
 that he was liable to punishment if he did not, unless 
 he could plead, in regard to members of it whom he 
 forsook, that they had been guilty of deadly sin. 
 
 It may now be said that, as regards its essential 
 elements, the Code of Manu shows the Hindoo father 
 not to have possessed Patria Potestas, or any power 
 approaching to it, or capable of suggesting it. If the 
 prescriptions of " the law " were of any avail, he was 
 without the powers of life and death and sale. It 
 remains to see whether all the property of the family 
 belonged to the father, or whether its members could 
 have property independently of him. And 
 
 1. Manu ix. 194* says: "What was given before 
 the nuptial fire, what was given on the bridal pro- 
 cession, what was given in token of love, and what 
 was received from a brother, a mother, or a father, 
 are considered as the sixfold separate property of 
 a married woman." Property given to a woman 
 
 * Sir William Jones's translation. The words in italics in this 
 and other extracts are from the gloss of Kulluka, which Sir William 
 followed.
 
 56 THE PATRIARCHAL THEORY. 
 
 on her marriage was " inherited by her unmarried 
 daughter" (Manu ix. 131). As to her other property, 
 on her death, " the uterine brothers and the uterine 
 sisters, if unmarried,'' were to "equally divide the 
 maternal estate" (ix. 192). A further provision is 
 (ix. 195) that "What she received after marriage 
 from the family of her husband and what her affec- 
 tionate lord may have given her " that even that was 
 to be " inherited, even in his (her husband's) lifetime, 
 by her children." On the other hand, " Of a son dying 
 childless, and leaving no widow, the father and mother" 
 were "to take the estate "(ix. 217). When the wife 
 herself died childless, in certain cases the husband in- 
 herited her property, while in others so far was a wife 
 from being cut off by marriage from the family of her 
 birth it went to her father and mother (ix. 196, 197). 
 The Code enjoins on a woman not to make hoards from 
 the goods of her kindred, or even from the property of 
 her lord without his assent. It further reserves to her 
 what we call paraphernalia. 
 
 The sixfold enumeration of a married woman's sepa- 
 rate property appears not to have been restrictive, but 
 the precise limits of that property at the time when the 
 text of Manu was settled do not now concern us. It is 
 enough to know that it was fully recognised, and that, 
 from the notice it receives, it must often have formed 
 an estate of considerable importance. It, at any rate, 
 included the dos and the nuptial gifts of the husband 
 and his family, and all property given to the wife after 
 marriage by her own family, or by her husband and his
 
 PATBIA POTESTA8 AMONG THE HINDOOS. 57 
 
 family. In all such property of the wife the husband 
 had no right whatever. At the wife's death it went to 
 her children, even if the husband was then living ; and 
 even if she left no children it did not in every case 
 become his. When it went to her sons and daughters 
 it was their separate property just as it had been hers ; 
 which shows that sons and daughters, as well as wives, 
 were capable of holding property, and must often have 
 had property, independently of the head of the family. 
 For further proof of this, take the provision made for 
 the son of a Sudra woman by a man of one of the 
 three higher classes (ix. 155). Such a son was to 
 inherit no share of the family estate, but whatever the 
 father might give him was to be his. He might be 
 provided for, that is, by a gift inter vivos ; and, there- 
 fore, he was capable of holding property in his father's 
 lifetime. Still more must sons belonging to the higher 
 classes have had that capacity. 
 
 It now appears that the whole property of the 
 family did not belong to the father, and therefore that 
 there was no element of Patria Potestas to be found 
 among the Hindoos at the date of Manu. But further, 
 as to property : 
 
 2. Whether the sons of a family were co-owners 
 with their father in the undivided ancestral estate, or 
 had only a right of sharing in it at his death, or when 
 he divided it, has been much contested among Hindoo 
 jurists. There is evidence (but not in Manu) that 
 they could enforce a division of it against his wish. 
 Gautama alleged by Mr. Biihler to be an older writer
 
 58 THE PATRIARCHAL THEORY. 
 
 than Manu, and no doubt a great authority enumerates 
 sons who had done this among the classes of people 
 who were not to be allowed to partake of funeral 
 oblations ; * on which Mr. Buhler remarks : " From this 
 sutra it would appear that sons could enforce a division 
 of the ancestral estate against his [their father's] will, 
 as Yagnavalkya also allows (see Colebrooke, Mitakshara, 
 I. 6, 5-11), and that this practice, though legal, was 
 held to be contra bonos mores" A power of enforcing 
 partition would, it should seem, prove the joint owner- 
 ship of sons. It would conclusively prove the absence 
 of Patria Potestas. What Manu lays down as to the 
 proper period of partition is as follows : " After the 
 death of the father and the mother, the brothers may 
 divide among themselves the paternal and maternal 
 estate ; but they have no power over it while their 
 parents live, unless the father choose to distribute it " 
 (Manu ix. 104). Here the period of partition is post- 
 poned to the death of the mother in the event of her 
 surviving the father which also is a provision incon- 
 sistent with Patria Potestas. For if the verse proves 
 that sons had no power over the estate so long as the 
 father lived, it proves equally that they had no power 
 over it, after his death, while their mother lived.t But 
 
 * The Sacred Laws of the Aryan, Translated by Georg Buhler, 
 Oxford, 1879. Gautama's Institutes of Sacred Law, ch. xv., s. 19. 
 
 fr For authorities, and an argument founded on them, to show that 
 after the father's death the mother was anciently head of the family, 
 possessed, with full right of control, of the family estate, see Mr. 
 J. D. Mayne's Hindu Law and Usage, Madras and London, 1878, 
 pp. 124, 125.
 
 PATRIA POTESTAS AMONG THE HINDOOS. 59 
 
 it may merely lay down " a precept of perfection," and 
 be not inconsistent with the sons having the power to 
 enforce a partition if they were so wicked as to wish to 
 do so. It can scarcely be thought that property rights 
 in the estate were not possessed by them after their 
 father's death. But if they were then possessed by 
 them, there is nothing to show that they were not 
 possessed by them in his lifetime. It is to be gathered 
 from the older Hindoo text writers that partition was 
 usually made with the father's consent in his lifetime, 
 and that it was thought proper he should make it. He 
 divided not only property he had inherited, but also his 
 own acquisitions ; keeping a share to himself, and being 
 allowed some choice in the distribution of his acquisitions. 
 3. About the earnings of the members of an un- 
 divided family after the death of their father, there is a 
 good deal to be found in the Code of Manu ; and we 
 learn that each brother could keep to himself, if he 
 pleased, wealth acquired by him without using the 
 patrimony ; and that brothers who put their earnings 
 into the common stock (whose gains may from their so 
 doing be supposed to have been nearly equal) were to 
 have the property acquired divided ' equally between 
 them to the exclusion of the slight preference which 
 might be claimed by the eldest in a division of the 
 patrimony (ix. 204-9). The undivided brothers, in 
 short, were free to choose between holding each by 
 his own earnings, and making hotch-pot with one 
 another by throwing their earnings into the patrimony 
 the eldest, though having the control, getting no
 
 60 THE PATRIARCHAL THEORY. 
 
 advantage in a division over the others in the latter 
 case. As to the earnings of an undivided family while 
 the father lived, we find a rule laid down for one case 
 only (which may ha,ve been the commonest). It is that 
 " if among undivided brethren living with their father, 
 there be a common exertion for common gain, the father 
 shall never make an unequal division among them when 
 they divide their families" (ix. 215). This reads as if 
 the augmentation of the patrimony by gains arising 
 from the common employment not only involved that 
 such gains should be equally divided, but excluded 
 the father from showing any preference in the dis- 
 tribution of the patrimony itself. But taking it to 
 imply that there was to be an equal division only of the 
 family gains, it would be identical with the provision 
 made for the case of brothers who held together after 
 their father's death and made a common stock of their 
 gains ; and this shows that, before his father's death 
 just as after it a man had assured to him his fair 
 interest in the surplus profits of his labour which 
 accumulated in the hands of the head of the family. 
 What happened when, the father being alive and the 
 family undivided, there was not a common employment, 
 Manu does not tell us ; and it does not seem worth 
 while to go to more modern authors for light upon the 
 matter. It may be inferred that the father was not 
 bound to make an equal division of acquisitions in this 
 case ; but that, if the earnings of his sons came into his 
 hands, he would be expected to make a just division ; 
 and that is all that concerns us.
 
 PATEIA POTE8TAS AMONG THE HINDOOS. 61 
 
 As regards accumulations from joint earnings which 
 were in the father's hands, then, as well as in regard to 
 the ancestral estate, there was between father and sons 
 a sort of joint ownership ; while a son was capable of 
 holding, and must frequently have held, other property 
 as separate property. That is what the Code of Manu 
 shows us. Having shown this, we have shown the 
 absence of every element of Patria Potestas. 
 
 It remains, however, to consider a passage (Manu 
 viii. 416, 417) the only passage in Manu of which so 
 much can be said which, superficially considered, 
 may seem to carry some implication of one element of 
 Patria Potestas. It is regarded by Hindoo jurists 
 of every school, and beyond all doubt rightly regarded, 
 as bearing upon earnings only. It is as follows : 
 
 Ver. 416 : " Three persons, a wife, a son, and a slave, 
 are declared by law to have in general no wealth 
 exclusively their own ; the wealth which they may 
 earn is regularly acquired for the man to whom they 
 belong." Ver. 417: "A Brahmin may seize without 
 hesitation, if he be distressed for a subsistence, the 
 goods of his Sudra slave ; for, as that slave can have 
 no property, his master may take his goods." 
 
 Of course the words in italics, from the gloss of 
 Kulluka, convey what seemed to the commentator 
 necessary qualifications of propositions too wide in 
 their terms. The verses occur in the chapter on 
 Judicature and on Law private and criminal ; and, 
 as frequently happens in that chapter, the one is 
 merely a preamble to the other the first verse
 
 62 THE PATRIARCHAL THEORY. 
 
 prescribing nothing, and being, as is apt to happen 
 with preambles, and as we find to be the case 
 repeatedly in this chapter, a great deal larger than 
 was necessary to cover what the writer desired to lay 
 down. That was, that the owner of a slave need not 
 hesitate to appropriate the slave's peculium, since, 
 strictly speaking, all a slave had was his master's a 
 proposition harsh-sounding and liable to abuse ; so that 
 the commentator had to attach to it qualifications 
 which confine it to the Sudra slave of a Brahmin 
 master, who happened to be himself in distress. It 
 is, however, only with the preamble to this allowance 
 of the extreme right of a master that we have to do, 
 and of course with that only in so far as it refers to 
 the position in respect of property of a wife and of 
 a son. And, in the first branch of the verse, they 
 are said (not to have no wealth, but) to have no wealth 
 exclusively their own which Kulluka has qualified 
 by inserting the words, in general. In the second 
 branch of it, it is alleged that their earnings are 
 acquired for the head of the family ; to which Kulluka 
 has attached the qualifying word, regularly which 
 may be taken to be a variant of the words, in general 
 (regularly, as a rule, in general). 
 
 The verse, therefore, does not deny the capacity of 
 wife and son to possess property, but admits it. And, 
 except as to earnings, it does not suggest that the father 
 was even a sharer in the wealth of wife and son which 
 was not exclusively their own. 
 
 It must be borne in mind that much perhaps
 
 PATRIA POTESTA8 AMONG THE HINDOOS. 63 
 
 most of the property of Hindoos was in the position 
 here described.* We have an example of property 
 not being exclusively one's own, in the ancestral estate 
 as possessed by a father before he made partition with 
 his sons ; for, even if it be denied that his sons were 
 co-owners with him, it is certain that he could not 
 dispose of it at his pleasure. The interest in their 
 common property of individual brothers, when they 
 continued undivided after their father's death, putting 
 their gains into the common stock, is another example 
 of it. The case of brothers living undivided with their 
 father and pursuing a common industry the only case 
 involving earnings made in the father's lifetime and 
 before partition for which the Code of Manu makes 
 provision is a third ; and it is one of the cases which 
 the writer of the preamble must have had in view. 
 
 The separate property of a wife or a son is not, so 
 far as is disclosed in Manu, an example of wealth that 
 is not exclusively a person's own. Such property is 
 declared to be the property of the wife, or of the 
 son or daughter, without any qualification. If, how- 
 ever, in the interest of the heirs to whom it reverted, 
 it could not be wasted, the heirs were co-owners in it. 
 It was not the exclusive property of wife or son or 
 daughter. And yet the father, as such, had no right to it. 
 
 * "Among the Hindoos," says Mr. J. D. Mayne, "absolute, un- 
 restricted ownership, such as enables the owner to do anything he 
 likes with the property, is the exception. The father is restrained 
 by his sons, the brother by his brothers, the woman by her successors. 
 If property is free in the hands of its acquirer, it will resume its 
 fetters in the hands of his heirs." Hindu Law and Usage, p. 175.
 
 64 THE PATRIARCHAL THEORY. 
 
 But, as noticed already, it is not suggested that all 
 the property of wife and son which was not exclusively 
 their own belonged to the father so far as it was 
 not theirs. The claim made for him is confined to 
 earnings. And the writer of it had earnings, and 
 savings out of earnings, in mind when making his 
 initial flourish about the wife, the son, and the slave 
 having no property exclusively their own, whether he 
 had anything else in mind or not. It was savings out 
 of earnings he would naturally have been thinking of 
 when seeking to find authority for the pillaging of 
 slaves. And most probably he thought of nothing 
 else. 
 
 But had he been thinking dimly of something wider 
 and, no doubt, it was a case of exigency he was about 
 to provide for had he had in mind that the property 
 of wife or son was liable for debts incurred in time of 
 exigency for behoof of the family (Manu viii. 166), it is 
 to be observed that this liability carried with it no impli- 
 cation of Patria Potestas. For all the family property 
 was liable to be drawn upon in time of exigency. And 
 it was not the father only who could make it liable. 
 The family administrator, whoever he was, could do so. 
 The ground of liability was family necessity, and not 
 paternal right. 
 
 As to the father's right to receive the family earn- 
 ings, it has been shown that he was under regulation as 
 to the distribution of savings out of such earnings in one 
 case the case, no doubt, of most common occurrence 
 among the Hindoos. It is likely he was not unrestricted
 
 PATEIA POTESTAS AMONG THE HINDOOS. 65 
 
 as to this in any case. And his sons were, in fact, co- 
 owners with him of such savings. But, at any rate, the 
 father's right to receive earnings does not affect the fact 
 that wife and son had the capacity of holding other 
 property independently of him. 
 
 It is now plain that the passage under notice, though 
 words perhaps have not been weighed in it and it 
 belongs to a kind of writing in which words seldom are 
 weighed in which to make out somehow the appear- 
 ance of a reason for that which is to be done is what is 
 thought of does not carry the suggestion which might 
 in inconsiderateness be attached to it. When we ex- 
 amine it to see whether it can be regarded as claiming 
 the "Patriarchal" power over property for fathers, it 
 becomes clear that there is nothing to be said for that 
 view of it unless we overlook its words, and distort and 
 add to its meaning. For it makes no claim over all the 
 family property for the father. And it is not really at 
 variance with the prescriptions of the law. 
 
 As we find that the first branch of the preamble 
 (so to call it) refers to accumulations out of earnings 
 whether to anything more or not instead of this 
 passage making against what is elsewhere laid down 
 in Manu as to the property of wife or son, we get 
 from it additional proof that sons had an interest 
 assured to them in the property derived from their 
 labour when it remained in their father's hands. For 
 in saying that such property was not exclusively theirs 
 the writer admits that they had a property interest in 
 it. And he does not suggest or allow any encroachment
 
 66 THE PATRIARCHAL THEORY. 
 
 upon this or upon any property of son or wife but 
 implicitly declares its inviolability. 
 
 From the gloss of Kulluka it is obvious that there 
 were cases in which earnings were not in any sense 
 acquired for the head of the family, and therefore cases 
 in which the property arising from them was exclusively 
 the acquirer's own. Accepting the fact which nobody 
 will question it is unnecessary here to inquire what 
 the cases were, especially as the Code of Manu does not 
 help us in the matter. What concerned us here was 
 to show that the passage we have been considering 
 carries, as regards property, no implication of Patria 
 Potestas, and that appears to have been done 
 sufficiently.* 
 
 It should here be mentioned that after the separa- 
 tion which followed upon a division of the family 
 property, the members of the Hindoo family remained, 
 as they had been before, each other's nearest relatives 
 with possibility of heirship so that if, after that, the 
 father made acquisitions and had a son, the separated 
 brothers were free at his death to come in and make 
 hotch-pot with the latter. Separation, that is, had 
 not any of the effects of emancipation. 
 
 It has already appeared that the Hindoo wife was 
 by no means in manu. A few passages may be quoted 
 to show what her position in the family was : 
 
 " The law, abounding in the purest affection, for the 
 
 * The history of the family among the Hindoos is considered 
 at some length in chapters xvi. and xvii., on " Sonship among the 
 Hindoos."
 
 PATEIA POTESTAS AMONG THE HINDOOS. 67 
 
 conduct of man and wife " lays down, " Let mutual 
 fidelity continue to death," as the supreme rule for 
 the married pair. Mann ix. 101, 103. 
 
 " Married women must be honoured and adorned by 
 their fathers and brethren, by their husbands and by 
 the brethren of their husbands, if they seek abundant 
 prosperity. Where females are honoured there the 
 deities are pleased ; but where they are dishonoured all 
 religious acts become fruitless." Ibid., iii. 55, 56 
 from which also it appears that marriage by no means 
 cut off a woman from her family ; but this can be 
 proved by scores of passages. 
 
 " He who truly and faithfully fills both ears with 
 the Veda must be considered as equal to a mother ; he 
 must be revered as a father ; him the pupil must never 
 grieve. A mere dchdrya, or teacher of the gdyatri only, 
 surpasses ten upddhydyas ; a father a hundred such 
 dcharyas ; and a mother a thousand natural fathers." 
 Ibid,, ii. 144, 145. 
 
 " He shall be fined a hundred who defames his 
 mother, his father, his wife, his brother, his son, or his 
 preceptor." Ibid., viii. 275. 
 
 "If an elder brother act as an elder brother ought 
 he is to be revered as a mother, as a father." Ibid., ix. 
 110. 
 
 " The wives of his preceptor, if they be of the same 
 class, must receive equal honour with their venerable 
 husband." Ibid., ii. 210. 
 
 " Let every man constantly do what may please his 
 parents ; and on all occasions what may please his 
 
 F 2
 
 68 THE PATRIARCHAL THEORY. 
 
 preceptor. Due reverence to these three is considered 
 as the highest devotion ; and without their approbation 
 he must perform no other duty." Ibid., ii. 228, 229. 
 
 " Him by whom he was invested with the sacrificial 
 thread, him, who explained the Ve'da or even a part of 
 it, his mother, and his father, natural or spiritual, let 
 him never oppose." Ibid., iv. 162. 
 
 Such passages might easily be multiplied, but these 
 are enough to show that, in the Hindoo family, as 
 among the Hebrews, the mother was equal in honour 
 with the father, if not something more. When the two 
 are mentioned together, the mother is always mentioned 
 first. We have seen that the text in Manu relating to 
 partition ordains that sons should live together un- 
 divided till after the death of both father and mother ; 
 and there is not wanting authority to show that 
 anciently the mother had some control over the family 
 estate.* 
 
 It is now clear that the Code of Manu negatives 
 every element of Patria Potestas as regards the Hindoos 
 for the period at which it was drawn up. And it is not 
 being too venturesome to say that in no ancient collec- 
 tion of Indian laws is there a hint of it as actually 
 existing, or as having at a previous time existed in 
 India. 
 
 It will occur to most people, no doubt, that the Code 
 
 * As to the position of the mother in the Hindoo family, see also 
 the Vivada Chintamani, p. 225 ; Apastamba i. 4, 14, v. 6; ibid., ii. 
 2, 4, v. 13; ibid., ii. 6, 13, v. 1-5; and ii. 50, 51. It seems needless 
 to give further authority as to her right to possess separate estate.
 
 PATRIA POTESTAS AMONG THE HINDOOS. 69 
 
 of Manu and the same may be said of all other Hindoo 
 law-books is concerned with a social state by no 
 means primitive, in which it is idle to look for the 
 beginnings of things. But this is only what is to be 
 expected when we take comparative jurisprudence upon 
 its own ground. Going further back, however, we are 
 not aware that any trace of Patria Potestas, or of the 
 family system of the Patriarchal Theory, has ever been 
 pointed out in the ancient Sanscrit literature. It is 
 perhaps natural to conjecture that the rights of property 
 possessed by wives and children at the date of Manu 
 were once non-existent, and that they grew up by 
 degrees in previous ages. Even were this so, the fact 
 would no more prove that such rights were reared upon 
 Patria Potestas than the history of our own law as to 
 married women's property could prove that English 
 wives were formerly in manu. But if we push inquiry 
 back we meet with facts which tend to show rather that 
 a growing down of the rights of property of women and 
 children occurred among the Hindoos than that they 
 grew up. For there is abundance of evidence that 
 a provision like that of the Code of Gentoo Laws 
 (ch. ii. 8, 14), regulating succession in the case of 
 children born to a woman living in polyandry, was 
 much needed in India in the earliest times. This is not 
 the place to go into that evidence, and it must suffice 
 to say that striking proofs of the prevalence of poly- 
 andry among early Hindoos of saintly and of princely 
 stock are furnished by the Mahabharata. It will be 
 found a difficult task to reconcile them with the
 
 70 THE PATRIARCHAL THEORY. 
 
 primordial existence of the Patriarchal Family among 
 the Hindoos. 
 
 It is proper to add that were evidence of Patria 
 Potestas of later date than Manu forthcoming among 
 the Hindoos, it would not support Sir Henry Maine's 
 propositions. For that would show, not that Patria 
 Potestas is primordial, but that it is not primordial ; 
 not that it is early, but that it is late, appearing, if it 
 appear at all, after society has passed through a long 
 course, and institutions have become complex.
 
 CHAPTER VII. 
 
 PATEIA POTESTAS AMONG THE SLAVS. 
 
 IN Ancient Law the Slavonians, with the Romans 
 and Hindoos, are said to be the races which furnish 
 nearly all the legal testimony in support of the Patri- 
 archal Theory. What legal testimony has come from 
 the Slavonians is nowhere stated in that work ; but it 
 contains some notices of the Village Communities of 
 Russia ; and it quotes, as if adopting it, a statement* of 
 some of " the earliest modern writers on Jurisprudence," 
 that it was only the fiercer and ruder of the conquerors 
 of the empire, and notably the nations of Slavonic origin, 
 which exhibited a Patria Potestas at all resembling that 
 which is described in the Pandects and the Code. In 
 what the resemblance here spoken of consisted is not 
 disclosed, nor is it suggested that there was more than a 
 resemblance. 
 
 In a much later production than Ancient LawJ 
 
 * Ancient Law, p. 143. 
 
 t " South Slavonians and Eajpoots." The Nineteenth Century, 
 December, 1877. Reprinted, with some modifications, as chapter 
 viii. of Early Law and Custom, under the title, " East European 
 House-Communities."
 
 72 THE PATRIARCHAL THEORY. 
 
 Sir Henry Maine has identified the House-Community 
 of the South Slavonians, and also the Joint Undivided 
 Family of the Hindoos, with " the Roman Gens [in 
 Early Law and Custom, alternatively with either the 
 Gens or the body of Agnates], the Hellenic 7>o<?, the 
 Celtic Sept, the Teutonic Kin ; " and has declared him- 
 self satisfied not from knowledge, but as a matter of 
 inference that the natural families which now occur 
 among the South Slavonians are " Patriarchal " families, 
 despotically governed by the oldest ascendant. 
 
 Taking first the statements of the earlier work, their 
 vagueness, which must in any case have made them 
 difficult to deal with, is, as matters stand, peculiarly 
 perplexing. The author does not in any way suggest 
 to us what is the nature of the legal testimony contri- 
 buted by the Slavonians which he has had in mind ; 
 and he does not tell us what authority exists for a 
 statement about the paternal power among the early 
 Slavonians nor even who are the writers on juris- 
 prudence who professed to have knowledge of its nature 
 or limits. This is peculiarly perplexing, because, on 
 examining what is recorded of the early Slavonians, 
 what one finds is that there is a truly surprising absence 
 of information about their institutions, and, to all 
 appearance, nothing whatever to warrant the state- 
 ments of Sir Henry Maine and his authorities. 
 
 The Slavonic nations began to be known to the 
 Greeks in the latter half of the sixth century, Slavonic 
 tribes having by that time made their appearance in the 
 countries bordering on the Danube. Onwards from
 
 P ATRIA POTESTAS AMONG THE SLAVS. 73 
 
 that period pretty frequent notices of them occur in 
 the Byzantine writers. These will be found collected in 
 the voluminous work of Stritter. Frequent as these 
 notices are, however, a short passage in Procopius (A.D. 
 562), with a few sentences of later authors, contain all 
 the Byzantines have preserved to us concerning the 
 social habits or institutions of the Slavonians. Procopius 
 tells us that the Slavs and the Antes did not obey one 
 man, but lived in a state of democracy ; that they 
 followed their own ancient customs ; that they wor- 
 shipped the God of Thunder, regarding him as the only 
 Lord of the Universe, but worshipping besides rivers 
 and nymphs and other divinities ; that they lived in 
 poor and widely -scattered huts, and frequently changed 
 their place of habitation ; and that both peoples had 
 previously been called Spori as he thinks, because 
 their settlements were so far apart, and they occupied so 
 great a space. The Emperor Constantine Porphyro- 
 genitus, writing some four hundred years later than 
 Procopius, says that the Croats and Servians had no 
 princes except old men, who were their jupans, and that 
 the same might be said of all Slavish peoples ; but he 
 tells us nothing of their jupans* except that they 
 were old. From writers intermediate between these 
 two we hear of the love of liberty which distinguished 
 the Slavs, or rather of their repugnance to put up 
 with any master ; also of the devotion of Slavish 
 wives to their husbands, which often led them to kill 
 themselves at the husband's obsequies. And when to 
 * The name is said to be of Gothic origin.
 
 74 THE PATRIARCHAL THEORY. 
 
 this is added that they liked to establish themselves on 
 river banks, in woods, and in marshes, all has been told 
 that the Byzantine writers have handed down to us. 
 Taken altogether, it does not afford material for even a 
 conjecture about any Slavonic institution. 
 
 The Frankish and Northern annalists are no more 
 helpful to us in this matter than the Byzantine his- 
 torians ; and even the Chronicle of Nestor the oldest 
 Slav Chronicle extant, its author a monk of Kiev in 
 Russia, who lived in the second half of the eleventh and 
 early years of the twelfth centuries gives us scarcely a 
 hint as to the domestic economy or social system of the 
 early Slavs. Such as he is, however, Nestor is the only 
 early Slav authority worthy of being seriously looked 
 at. M. De Laveleye has relied upon a passage in the 
 " Libusen Sud," or "Judgment of Libusa" one of 
 the so-called national poems discovered at Koniginhof 
 in Bohemia in 1824 for proof of the antiquity of the 
 House-Community among the South Slavonians. But 
 whatever may be the antiquity of this institution, it is 
 not to be forgotten that those poems are of as doubtful 
 authenticity as the poems of Ossian ; that the period of 
 their composition is more uncertain still ; and that the 
 princess, whose judgment in the case of two brothers 
 who had quarrelled over the division of their inheritance 
 is taken to show that it was Slavonic custom not to 
 divide inheritances, is unquestionably a fabulous person, 
 a creation of the popular fancy the popular memory, 
 which has preserved the tradition of her and of many 
 other personages equally mythical, having completely
 
 PATBIA POTESTAS AMONG THE SLAVS. 75 
 
 lost hold of actual events and real personages of date 
 not much earlier than that to which she is assigned.* 
 
 Nestor fixes the coming of Eurik and his Varagian 
 followers into Russia at about two hundred years before 
 his own day (at A.D. 862), and he gives accounts of the 
 early inhabitants of Russia which apply to times more 
 remote. He begins, indeed, with the Flood, and traces, 
 though very briefly, the descent of the Slavs from 
 Japhet ; but this need not impair our faith in descrip- 
 tions which he cannot be suspected of having invented, 
 and which are consistent with what is otherwise known 
 of barbarous peoples. What can be gathered from him 
 that in any degree bears upon the present inquiry can 
 be quickly told. 
 
 The country was thinly peopled, and little better 
 than a wilderness. The small communities which formed 
 its population had at first little to do with one another. 
 In later times they sometimes combined against a 
 common enemy, for example to resist the predatory 
 onsets of the Varagians from beyond the Baltic. They 
 had given up wandering, and had settled down to 
 tillage ; and they lived in enclosed villages in happy 
 ignorance of the distinctions of rank. They were 
 known to each other by names derived from the physical 
 character of the districts in which they lived, or from 
 the name of some mountain or river. Thus, the people 
 of Kiev were Polians, inhabitants of the plains, and 
 
 * She is assigned to the latter part of the seventh century. The 
 Slavs made their appearance in the Danubian countries about a 
 century and a half before.
 
 76 THE PATRIARCHAL THEORY. 
 
 their near neighbours were Dreyvians or men of the 
 woods, while the Polovzans again were the people who 
 dwelt by the river Polovz. Their country abounding 
 in wild beasts, they were great hunters, and had a wealth 
 of skins in which, in somewhat later times, they paid 
 tribute to their conquerors, and which, indeed, for 
 centuries continued to form the basis of the Russian 
 currency. With a general resemblance between all, 
 each community had its own peculiarities of custom and 
 manners ; and Nestor himself a dweller at Kiev 
 gives a much more favourable account of the Polians 
 than he does of any of their neighbours. The Polians, 
 he says, were quiet and gentle. They showed much 
 respect to their parents and their relations, and to their 
 daughters-in-law, fathers-in-law, and brothers-in-law. 
 They had a form of marriage. The bridegroom did not 
 himself go to fetch his bride, but some one brought her 
 to him in the evening, and the price stipulated to be 
 given for her was sent the next day. Like other 
 heathen, they sacrificed to lakes, and springs, and 
 plants. The Dreyvians, on the other hand, lived 
 in a brutal way, like cattle. They killed one another. 
 They ate all unclean things. They really were 
 without marriage (that is, they did not marry after 
 an agreement, like the Polians) ; but they carried 
 away maidens by force, and took them to their beds. 
 The Radimitschans, the Viatitschans, and the Severians 
 had like customs living in the woods like wild 
 beasts, and eating all unclean things. They carried on 
 unchaste conversations before parents and daughters-in-
 
 P ATRIA POTE8TAS AMONG THE SLAVS. 77 
 
 law, paying no respect to them. They had no marriage ; 
 but they arranged merry games at which they played, 
 danced, and sang devilish songs, and, at the end, each 
 man carried away a woman who became his wife. Those 
 who could get them had two or three wives. They 
 burnt their dead, gathering the remains and putting 
 them into an urn, which they placed on a pillar at the 
 wayside. The Polovzans, again, smudged themselves 
 with blood ; ate carrion, moles, marmots, and all un- 
 clean things ; married step-mothers and daughters-in- 
 law ; and " committed all the other wicked practices of 
 their fathers." 
 
 Such is Nestor's brief account of the ancient manners 
 of his countrymen. It appears from it that, except 
 among the Polians, who had marriage by purchase, 
 probably with the form of capture, wives were usually 
 got by actual capture ; and that there were in use 
 among some of them friendly arrangements, such as 
 have been found among other peoples, to facilitate 
 wiving by this method arrangements such as would 
 prepare the way for capture to pass into a form. 
 
 Of the powers of fathers in the Slav household, 
 however, and of the constitution of the household, 
 whether among his predecessors or his contemporaries, 
 Nestor discloses nothing unless there is something to 
 be made out of a sentence which occurs in his account 
 of Kii, the fabulous founder of the town of Kiev. 
 The Polians, he there tells us, lived by themselves 
 apart, each one in his own place, with their families 
 over which they ruled. What is meant by families
 
 78 THE PATRIARCHAL THEORY. 
 
 here, and what is meant by ruling is not defined ; 
 and thus at first sight the passage looks vague enough 
 for every one to read his own meaning into it. From 
 the story it appears, however, that though "ignorant 
 people " declared that Kii was a ferryman about whose 
 ferry a little town grew up, Nestor believed him to 
 have been Cnaz or prince in his " family," and his 
 " family " to have been large enough to furnish forth, 
 or perhaps to form the nucleus of, an expedition to 
 Constantinople. This shows clearly that by " family," 
 Nestor, in the passage we are noticing, meant a 
 considerable body of kinsfolk with one of the kinsfolk 
 at their head. He narrates, too, that Kii and his two 
 brothers at first settled each on a separate mountain 
 not holding together as a family or in the same House- 
 Community ; and that they afterwards united, and, 
 coming into the plains, founded a little town which 
 was named after the eldest brother which is only 
 intelligible on the view of their being chiefs and rulers 
 over their kindred. Whether the Polian family which 
 lived apart was one man's family, however, or a House- 
 Community, or a clan, the passage leaves us altogether 
 uninformed as to the powers of its head. 
 
 Of the House-Community we find nowhere any 
 notice in Nestor. His mention of polygamy as an 
 occasional practice among certain Slavs, makes against 
 the existence of this family arrangement among them. 
 Apart from this, all his descriptions perhaps are 
 consistent with it. More than that can scarcely be said. 
 But it may be worth adding that it is no unlikely thing
 
 P ATRIA POTE8TA8 AMONG THE SLAVS. 79 
 
 that convenience at an early period established among 
 the Slavs an arrangement which, with varying 
 incidents, has been common among semi-barbarous and 
 barbarous peoples. 
 
 As to the relationships acknowledged among the 
 ancient Slavs, Nestor does not appear to. have known 
 of their varying from the system with which, as a 
 Christian of the Greek communion, he was familiar. 
 And it is but very rarely that his facts throw light 
 upon them. We find one, however, which seems 
 to exclude Agnation. The right-hand man of Vladimir 
 the Great, in his youth while he was fighting for a 
 kingdom, .and in the administration of his kingdom 
 afterwards, was his mother's brother, Dobrinia who, 
 plainly, was not only Vladimir's acknowledged relative 
 but, his own brothers being competitors and rivals, 
 the nearest relative he had to depend upon. Here 
 is testimony, though of course not " legal " testimony, 
 against the Patriarchal Theory. 
 
 Further testimony against it appears to be found in 
 the position assigned to Olga, widow of Igor, the son of 
 Kurik, after the death of her husband and during the 
 minority of her son ; and (since she may have been 
 Varagian) still more in the nature of the traditions 
 about her. Instead of living in perpetual tutelage, Olga 
 was her son's guardian, and regent of the kingdom, 
 with all the powers of a sovereign ; and Nestor gives 
 wonderful accounts of the revenges which taking up 
 the blood feud, woman though she was she took for 
 the murder of her husband.
 
 80 THE PATRIARCHAL THEORY. 
 
 We should scarcely find such a woman made a 
 popular heroine at a time when women were considered 
 unfit for all places of trust and condemned to life-long 
 subjection. 
 
 As to her being regent and guardian, Sir Henry 
 Maine, aware of the frequency of cases in which such 
 functions have devolved upon the sovereign's mother, 
 has accounted for them by saying that this has occurred 
 " doubtless out of respect to the overshadowing claims 
 of the mother." * But it is part of his theory that 
 women had no overshadowing claims ; that their lot in 
 life was perpetual tutelage ; and that the people with' 
 " overshadowing claims" were the husband's kinsmen on 
 the father's side. The fact he had to account for, and the 
 explanation he offers, alike show a weight allowed to 
 natural relations which the Patriarchal Theory denies to 
 them. 
 
 It may here be said that South Slavonian tradition 
 freely gives the sovereign's place to a woman ; and, for 
 example, Krok, the monarch with whom Bohemian 
 tradition begins, is succeeded by his daughter, the 
 Queen Libusa, whom we have already encountered as 
 the upholder of ancient custom. Sir Henry Maine, 
 aware of the frequency of such female successions, has 
 given an explanation of them also, and it also does not 
 seem consistent with his theory. 
 
 The explanation, he says, "no doubt is that the 
 circumstances of the time allowed unchecked play to 
 respect for the claims of blood ; the men being ex- 
 * Ancient Law, p. 240.
 
 PATBIA POTESTAS AMONG THE SLAVS. 81 
 
 hausted, a woman was taken rather than a new strain of 
 blood introduced." * 
 
 This explanation is simple and natural enough. 
 But it assumes (without warrant) that women have 
 never succeeded except when the men of the royal 
 stock have all been exhausted ; and it overlooks the fact 
 that the succession of a woman almost necessarily intro- 
 duces a new strain of blood. "We are not told of the 
 legendarv Krok that he left no male relatives behind 
 
 O it 
 
 him, but that, having a daughter, she was his successor ; 
 and her succession, according to the legend, did intro- 
 duce a new strain of blood, and passed a share of the 
 government at once to a person of new blood. This by 
 the way. What we are concerned to point out is that 
 this explanation ignores the Patriarchal Theory. That 
 theory does not allow " unchecked play to respect for 
 the claims of blood." The adoption of a son would be 
 entirely consonant with it ; and it is this which would, 
 in the case spoken of, be its natural resource not the 
 sovereignty of a woman. Than the sovereignty of a 
 woman, nothing could be more antipathetic to a theory 
 which considers a married woman as finis familise, and 
 puts women throughout life under supervision. 
 
 As to the particular case of Queen Libusa, it is to be 
 borne in mind that Libusa is undoubtedly a mythical 
 person. She was not a hard fact that a people brought up 
 under "Patriarchal" ideas had to make the best of. That 
 the popular fancy invented her, and made much of her, 
 proves that the early Slavs had no difficulty at all in 
 * Early Law and Custom, pp. 248, 249.
 
 82 THE PATRIAECRAL THEORY. 
 
 submitting to a female sovereign, or female chief, and 
 even that they took kindly to one. And, indeed, if 
 South Slavonian tradition be good for anything, the 
 Southern Slavs were well familiarised with female head- 
 ship. There could be no better proof of a people 
 being free from the ideas which are comprised in the 
 Patriarchal Theory. It should be said that women 
 appear to have been in the line of succession to the 
 throne in Eussia.* 
 
 So far, the little we have gathered from Nestor 
 which though little is not without weight is all against 
 the Patriarchal Theory. We learn further from the 
 Chronicle that the settlement of the warlike Varagians 
 in their country soon brought the Slav populations of 
 Kussia into contact, or closer contact, with the Greeks, 
 and that a Treaty (not noticed by the Byzantine writers, 
 but set forth at length by Nestor) was made with the 
 Emperor by Oleg, Rurik's successor, in the year 907. 
 Here at last we have a chance of getting some hints as 
 to the legal institutions of the Russians ; but, so far as 
 the present inquiry is concerned, there is scarcely any- 
 thing to be made of it. Levesque inferred from the 
 
 * We find Sir Jerome Horsey, in the time of the Czar Feodor, 
 speaking of the Czar's cousin, widow of a Duke of Holstein, as 
 " Queen Magnus, the next heir of the emperiall crown of Muscovvia," 
 and such she admittedly was. (Horsey was employed by Boris 
 Godounof, who was then aspiring to the throne, to persuade her into 
 coming with her daughter to Moscow ; which having been done, 
 both were promptly put into a nunnery.) This, notwithstanding that 
 during nearly six hundred years, the clergy, imbued with Byzantine 
 notions, had been steadily lowering the position of women, especially 
 of the women of the better classes.
 
 PATRIA POTE8TA8 AMONG THE SLAVS. 83 
 
 third article of this Treaty that, among the Russians, a 
 wife had a share in her husband's possessions ; but the 
 text of that portion of the article on which he founds is 
 too much in doubt for any inference to be made from it. 
 From the tenth article he inferred that they knew the 
 Will ; but perhaps it can hardly be taken to prove more 
 than that the Will was known to the Russians who 
 traded to Byzantium. The former of these articles gave 
 to the next relation of a murdered person a share of the 
 murderer's property when he had escaped ; and the latter 
 gave the property in Greece of a Russian who died there 
 without a Will or without having (probably wife or 
 children) to his relations in Russia. But it does not 
 appear who the " next relation " of the one article or the 
 "relations" of the other were. And it cannot even be 
 made out whether the power of willing existed when 
 there was a family (though apparently it did not). 
 
 A second Treaty with Greece, also preserved by 
 Nestor, was entered into by Igor, the son of Rurik, in 
 the year 945, and in connection with this there are two 
 things to be noted. The first is that the provision 
 relative to murder (differing, though perhaps only in 
 words, from that of the former Treaty) is that "the 
 relations " of the murdered person were to have the 
 right of killing the murderer, and of taking his property 
 if he escaped. The other point is more important. It 
 appears from the preamble of the Treaty that a married 
 woman Sphaindr, the wife of Oulieb was one of 
 Igor's deputies by whom the Treaty was negotiated. 
 This Sphaindr not being either a regent or a queen, we 
 
 G 2
 
 84 THE PATRIARCHAL THEORY. 
 
 cannot guess what explanation of her employment on 
 such a mission can be offered by an advocate of the 
 Patriarchal Theory. 
 
 Before the end of the tenth century, Russia, in 
 which there had long been Christians, all of a sudden 
 became a Christian country. Soon it swarmed with 
 clergy, and the clergy thenceforth had an immense 
 influence in its affairs and in the shaping of its insti- 
 tutions. An ordinance which assigns to them a most 
 extensive jurisdiction, both criminal and civil, is at- 
 tributed to Vladimir the Great, the first Christian 
 Monarch, but probably it is of much later date and 
 comprises powers of gradual growth. At any rate, the 
 first body of law published in Russia bears no trace of 
 Christian influence. Of course it is the more valuable 
 on that account. 
 
 This body of law the Laws of Jaroslav,* son of 
 Vladimir the Great was published in the year 1017. 
 That it was drawn up to some extent under Scandi- 
 navian influence need not be doubted. But we can feel 
 confident, from the nature of the matter dealt with, that 
 the article with which it opens, which alone concerns us, 
 embodied nothing that was not established Russian 
 custom, and the more confident because, in its second 
 branch, we find it conflicting with Scandinavian 
 usage. It is, moreover, in accord with what we have 
 gathered from Nestor. It deals with the blood feud, 
 and names those who should have the right of taking 
 
 * Euskaia Prawda. See Das alteste Recht der Russen, by 
 J. P. G. Ewer. Dorpat, 1826.
 
 PATEIA POTESTA8 AMONG THE SLAVS. 85 
 
 Christian influence must have aimed at 
 restricting the number of such persons. A monarch, 
 promulgating laws, would naturally seek to restrict 
 rather than to extend it. At any rate, this article, such 
 as it is, undoubtedly contains the best "legal testimony" 
 relative to the Patriarchal Theory which early Russia 
 has to offer. Its testimony is against the theory. 
 
 For it gives the right of revenging a murder to the 
 brother, the son, the father, the brother's son, and the 
 sister's son of the slain. Argument cannot be needed to 
 show that the duty of taking vengeance has everywhere 
 attached only to those who were in the fullest sense 
 relations of the slain. In Russia, therefore, a man's 
 sister's son must have been in the fullest sense his 
 relation. This article shows the absence of Agnation. 
 
 The second branch of this article provides for the case 
 of there being no person to take vengeance for a murder, 
 in that case imposing a fine payable to the prince. 
 The amount of the fine was the same whatever the rank 
 of the murdered person, and whatever his nationality 
 a complete departure from Scandinavian usage. 
 
 Isiaslav, son of Jaroslav,* was also a legislator. 
 He both modified and added to the Laws of his father, 
 but he left the first article untouched. 
 
 It can now be understood how perplexing is the 
 statement that important legal testimony to the 
 Patriarchal Theory is furnished by the Slavonic peoples. 
 Their oldest records so far as they are good for any- 
 thing negative Agnation and, to all appearance, the 
 * See Ewer, cited above.
 
 86 THE PATRIARCHAL THEORY. 
 
 tutelage of women. Women were competent to fill the 
 sovereign's place among the Slavs both of the North 
 and of the South. As to the paternal power among 
 the early Slavonians, there is simply an absence of 
 information about it. 
 
 In the centuries following the introduction of 
 Christianity, there were influences at work capable 
 of effecting, and which did effect, immense changes in 
 the social condition of Eussia the Church itself, which 
 introduced a foreign jurisprudence, and Eastern views 
 about the treatment of women ; the Tartar domination ; 
 the incessant wars of the princes between whom the 
 country was parcelled out ; the position of autocracy 
 finally arrived at by the Grand Prince, who thenceforth 
 treated even the highest of his subjects as his slaves. 
 From the boyar of the tenth century, his master's com- 
 panion and counsellor, with whom he lived on terms 
 almost of equality, to the noble of the sixteenth who, 
 in addressing the Czar, grovelled before him, saying, 
 " May I speak and not be whipped ? " is a tremendous 
 change, and it may be taken as a measure of what had 
 gone on throughout the whole social system. From 
 what is known of any Kussian institution in the six- 
 teenth century, generally speaking, no conclusion can be 
 reached as to what that institution was in pre-Christian 
 times. Nor will even those Kussian laws which are 
 assigned to the thirteenth century* be found helpful for 
 the purposes of the present inquiry. 
 
 * A collection of uncertain date (the Prawda of Novgorod), but 
 ascribed by Ewer to the period 1280-1300, contained in a volume of
 
 PATBIA POTESTAS AMONG THE SLAYS. 87 
 
 Here, then, so far as Ancient Law is concerned, 
 we may conclude. That work contains notices of the 
 Russian Village Community, but Sir Henry Maine has 
 changed his opinion as to the place of the Village 
 Community in the history of society. Once he was 
 disposed to identify it with the Roman Gens. He does 
 so no longer, and so no more need here be said of it 
 except that there appears to be little doubt that the 
 periodical redistribution of the land among all adult 
 males, which in Russia is its distinguishing and essen- 
 tial feature, was unknown in that country until after 
 the establishment of serfage. It is not known that 
 Village Communities (in the Russian sense) have at any 
 time existed among the South Slavonians. 
 
 uncertain history, is given in Ewer's "book already cited. These laws 
 form a supplement to the Laws of Jaroslav and Isiaslav. It is evident 
 that they consist of rules published at different periods, for they are not 
 altogether consistent with one another. It is always impossible to 
 make out satisfactorily to what extent those rules were innovations, or 
 what it was that they replaced.
 
 NOTE TO CHAPTER VII. 
 
 THE PATRIARCHAL THEORY AND ROYAL SUCCESSION. 
 
 IF, in the default of legal testimony, or of any direct 
 evidence, as to the social condition of the early Slavs, 
 one were casting about for means of determining what 
 that condition at one time was, the nomenclature of 
 relationships, or rather the terms of address between 
 relations, in use among the Slavonic peoples would 
 furnish evidence of the highest importance. Simply to 
 throw out this, and to indicate the kind of puzzle which 
 that nomenclature presents to an upholder of the Patri- 
 archal Theory is all that can be done here. To this day 
 a Russian addresses his first, second, third, and fourth 
 cousin, whether on the father's side or on the mother's, 
 as brother or sister, distinguishing a first cousin from a 
 brother or sister, however, as double-birth brother or 
 sister, a second cousin as triple-birth brother or sister, 
 and so on. Similar modes of speech, not confined to the 
 case of cousins, but running through the nomenclature 
 of relationships, occur among all the Slavonic peoples. 
 They can only be explained as exhibiting what Mr. 
 Morgan called the classificatory system of relationships 
 in stages of decay. How, consistently with the Patri- 
 archal Theory, to account for a man and his fourth 
 cousin on the mother's side having called each other 
 brothers is therefore the sort of problem which this 
 nomenclature raises. Those who know Mr. Morgan's
 
 ROYAL SUCCESSION. 89 
 
 voluminous work (Systems of Consanguinity and Affi- 
 nity of the Human Family] do not need to be told that 
 similar problems are raised by the nomenclature of 
 addresses in the case of numerous races not less im- 
 portant than the Slavs. It is not rash to say that a 
 solution consistent with the Patriarchal Theory is not to 
 be looked for. 
 
 The rule which, down to comparatively modern 
 times, regulated the succession to the throne in Russia 
 might also be made to throw light upon the ancient 
 social condition of the Slavs ; and it too makes a 
 serious difficulty for the Patriarchal Theory. This rule 
 familiar to us as being still in force among the Turks, 
 and which seems to have once been as well-known 
 among Aryans as it has been among Mongols gave the 
 throne to the oldest male of the royal family, so that 
 the brother of a deceased ruler succeeded in preference 
 to his son. The Patriarchal Theory involves the 
 succession of sons to their father and here we find 
 the son not succeeding his father. 
 
 Sir Henry Maine is of opinion that this rule was 
 arrived at upon considerations of policy. The origin 
 of it, he says, " is doubtless a simple calculation on 
 the part of rude men in a rude society that it is better 
 to be governed by a grown chieftain than by a child, 
 and that the younger son is more likely to have come 
 to maturity than any of the eldest son's descendants."* 
 He thinks that, in general, ordinary succession and 
 succession to the throne have had nothing to do with 
 one another t that the rule settled for the former, 
 and which was doubtless deemed equitable for it, has 
 commonly not been applied to the latter. And having 
 made regard to the royal blood account for the 
 
 * Ancient Law, p. 241. 
 
 f "The King and His Successor," Fortnightly Review, 1882. 
 Early Law and Custom, chap, v., on " Eoyal Succession and the 
 Salic Law."
 
 90 THE PATRIARCHAL THEORY. 
 
 sovereignty of females, and respect for the claims of 
 the mother for female regencies, he has attributed this 
 rule to a calculation which, in some of the cases for 
 which the rule was intended, would no doubt be a just one. 
 
 If it were admitted that rules for the succession to 
 chiefships and sovereignties have been derived from 
 rules for ordinary succession, the question would arise, 
 What system of ordinary succession gives heirship not 
 to the son but to the eldest male of the family ? It 
 would be found that this system of succession has 
 prevailed widely. This of itself would necessarily tell 
 against the Patriarchal Theory and that is what we 
 are here concerned with. But it is the system of 
 succession which arises with Tibetan polyandry. 
 
 As for Sir Henry Maine's explanation, views of 
 policy being apt to vary, an explanation of its kind 
 offered for a rule which has prevailed extensively, and 
 the policy of which is far from being absolutely good 
 it, though likely to work well in some cases, being 
 likely to work badly in as many others must always 
 be regarded as extremely doubtful. The best that can 
 be said for such an explanation of such a fact is that 
 often we cannot judge whether it is right or wrong, 
 probable or improbable. The utmost that can be 
 allowed is that there is something to be said for it. A 
 rule yields the results it is capable of yielding, whatever 
 be the way in which it originated. And nearly every 
 rule yields some good results. There would be an end 
 of inquiry if we were generally to assume that rules 
 were devised to accomplish that good which they effect. 
 And we know enough to be sure that this mode of 
 explaining is in most cases a bad one. An explanation 
 of this kind, therefore, can in general have no hold 
 upon belief, and cannot make any real difficulty for 
 a competing explanation of a different kind.* 
 
 * Of course, it is quite consistent with what is here said that it 
 should be beyond doubt that policy has had a secondary part in affect-
 
 ROYAL SUCCESSION. 91 
 
 Now, for the explanation of rules of chiefly and 
 royal succession which connects them with rules of 
 ordinary succession and which connects the Russian 
 rule with the system of ordinary succession which gives 
 heirship to the oldest male of the family a very strong 
 case can be made out. The nature of it may be 
 indicated here. 
 
 To begin, what men have done in a certain case 
 must outweigh any amount of speculation as to what 
 men in that case would do. Sir Henry Maine has him- 
 self remarked with surprise and as if it were a fault 
 that, in mediaeval Europe, in connection with the feudal 
 system, people tended to confound the law of succession 
 to the throne with the law of ordinary succession. And, 
 in fact, in connection with the feudal system, it seems 
 never to have been doubted that if a new case arose in 
 connection with royal succession, the law for ordinary 
 succession must govern it ; or, if an entirely new rule 
 was needed, that this rule would apply equally to 
 private succession and to succession to the crown. Sir 
 Henry Maine has shown how this was illustrated in the 
 controversy between Bruce and Baliol as to the succession 
 to the Scottish Crown. But it was perhaps even more 
 strikingly illustrated in another controversy which he 
 has discussed the controversy as to the succession to 
 the French Crown between the collateral male heir and 
 our Edward III. In this controversy it was never dis- 
 puted between the controversialists that the law which 
 regulated the succession to land should govern the 
 succession to the sovereignty also. What was disputed 
 was, whether the provision of the Salic Code as to Salic 
 
 ing chiefly succession. Whatever the rule for succession, policy might 
 exclude the lunatic, or the imbecile, or a person infirm in body. 
 Similarly, among the warlike Parthians, an injury to eyesight, which 
 disqualified a man for being a warrior, is said to have disqualified him 
 for being king though, among the Parthians, a child might be king 
 because he might grow up to be a warrior, and an old man of eighty 
 remained king.
 
 92 THE PATRIARCHAL THEORY. 
 
 land applied either to the soil or to the Crown of France 
 (see Shakespeare's " Henry V.," Acti., Scene 2), it being 
 admitted that, if it applied to the one, it also applied to 
 the other. That the provision of the Code in terms 
 applied to laud only was beyond doubt perfectly well 
 known. What such cases prove is, that mediaeval 
 Europe could not help believing that the law of royal 
 succession and the law of private succession should be 
 identical, and that whatever was settled for the one was 
 settled for the other also. Mediaeval Europe bulks 
 largely enough for us to be safe in inferring something 
 as to human tendency from it ; and its tendency was 
 not to devise special rules upon politic considerations 
 for the sovereignty when new rules for it were needful, 
 but to apply to it principles which were thought 
 good for succession in general, and, apart from cases 
 altogether unprovided for, to apply to it the ordinary suc- 
 cession law. No doubt the law for the succession of chiefs 
 and princes was well abreast of the general succession 
 law among the Northern peoples before what we call 
 feudalism began. Since it is not likely that the early 
 world was less under the influence of established systems 
 than mediaeval Europe, we get at once a strong pro- 
 bability that when in early times a rule of succession 
 for the chiefship came to be needed, the rule that would 
 be applied would be the rule which was already estab- 
 lished for the government of the family. And if the 
 family was governed by the eldest male, the eldest male 
 of the chiefly family would, as a matter of course, 
 succeed to the chiefship. 
 
 Chiefly successions would be rare, however, com- 
 pared with private successions whether to the property 
 of the family or to the government of the family. 
 They would also be greatly more important. There 
 would, besides, seldom be more than one or two persons 
 interested in making any change in them, whereas a 
 time would come when nearly all men might think
 
 ROYAL SUCCESSION. 93 
 
 themselves interested in procuring that men should be 
 succeeded by their sons rather than by their brothers or 
 by the sons of their sisters. And the interest at any 
 rate the immediate interest of the many as regards 
 chiefly succession would be that the settled rule should 
 remain. On all these accounts, there would not be in 
 the succession to chiefships the room there would be in 
 ordinary succession for the employment of those devices 
 by which systems may gradually be altered. What 
 might scarcely be noticed in the latter case would be 
 flagrant lawlessness in the former. The rules applicable 
 to the former might, therefore, remain unchanged when 
 the rules applicable to the latter had been radically 
 altered. The two might differ only for a little. But 
 until the change in private law was general and com- 
 plete, a change in the rule for the chiefship could 
 scarcely be thought of; and, even then, the persons 
 interested in making a change would be few, and the 
 interests opposed to it would often be powerful. The 
 two, therefore, might differ long ; and, if they did 
 differ long, they would almost necessarily come to be 
 regarded as independent of each other. And the longer 
 the divorce between them lasted, the more difficult 
 would it be to bring them together again. 
 
 That the rule for chiefly or royal succession should 
 often differ from the rule for ordinary succession is, there- 
 fore, what is to be expected if the former was got by the 
 application to the chiefship, when a rule for it became 
 needful, of the rule for ordinary succession which at the 
 time prevailed. And when we anywhere find the two dif- 
 fering, and that the rule for chiefly or royal succession can 
 be identified with an old rule of ordinary succession, it is 
 a reasonable supposition that this rule was previously 
 in that place the rule for succession in general. In 
 some cases it can be shown that this is the fact. And a 
 few examples of such verification are sufficient to con- 
 vert the supposition into a legitimate inference. In other
 
 94 THE PATRIARCHAL THEORY. 
 
 cases, though verification of the supposition may not be 
 possible, there is no alternative to it but an explanation 
 from policy which is manifestly ridiculous. Where we 
 find the chief succeeded by his sister's son, sons succeed- 
 ing to their father in all other cases, it is indefinitely 
 easier to believe that this happens because the sister's 
 son was formerly the heir in ordinary succession that 
 form of succession law having been very common than 
 to think that the chiefship was exceptionally provided 
 for to save it from a risk which all men were willing to 
 run for themselves with a view, that is, to ensuring that 
 the chief should never be of other than the chiefly blood. 
 
 Sir Henry Maine has justly remarked of the rule of 
 succession to the Sultanate that widespread rule for 
 which his explanation from policy has been offered that 
 it seems not to be derivable from the Mohammedan law 
 of inheritance. If, however, we suppose that it has 
 survived among the Turks from pre- Mohammedan 
 times, the supposition can easily be verified. And, if 
 we go on to suppose that it was once the rule for ordi- 
 nary succession among them, it can at least be shown 
 that it was once a rule of ordinary succession among the 
 races to which they belong. 
 
 But again, if the law of royal succession lags behind 
 the law of ordinary succession, it tends to follow it, and 
 to become identical with it as far as the case to which it 
 applies permits. The rule which preferred the oldest 
 male, once so common, is now found among no people 
 of any importance except the Turks ; and the natural 
 wish of fathers that their sons should succeed them in 
 the throne as in private successions has recently endan- 
 gered it as a rule even among the Turks. A mode of 
 nullifying it fratricide has long been in use among 
 them, as it is among many barbarous peoples who have 
 the same rule. In small tribal communities a form of 
 election has sometimes appeared with it in its decay ; 
 and this has doubtless been a device for averting the
 
 ROYAL SUCCESSION. 95 
 
 disputes about succession which arose when its justice 
 and policy had fallen into controversy. This expedient 
 gave the chief's son a chance of succeeding him, and it 
 enabled the tribe to settle in each case whether the old 
 rule should prevail or not. Of course every departure 
 from it which was thus sanctioned helped to break it 
 down as a rule, and so helped to prepare the way 
 for the succession of son to father. Another rule of 
 chiefly succession which has been mentioned, that which 
 gives the chiefship to a sister's son, appears to have 
 been nullified in some cases by means of an extraordi- 
 nary but effective expedient by the chief, that is, 
 marrying his own sister. 
 
 That the tendency of mankind to apply to a new 
 case the system they have already for similar matters 
 made the earliest rules for chiefly succession and the 
 contemporary rules for succession to the government or 
 to the property of the family identical a view corro- 
 borated by our finding that rules for chiefly or royal 
 succession are always, as nearly as circumstances 
 permit, and saving what are obviously expedients of 
 transition, identical with known rules of ordinary 
 succession identical frequently with the current rule 
 of ordinary succession, and always tending to approxi- 
 mate to it; identical in other cases with a more primi- 
 tive rule which can occasionally be shown to have 
 preceded the current rule in ordinary succession seems 
 to give a sufficient explanation of the rule which regu- 
 lated succession to the crown in Russia. In judging of 
 the sufficiency of this explanation it is difficult not to 
 think of the alternative one. No doubt rules for chiefly 
 succession formed in many different places, under every 
 variety of circumstances, upon considerations of policy, 
 might happen all to coincide, and might also happen all 
 to coincide with a discarded rule of ordinary succession. 
 But that this should have happened cannot but be 
 deemed very extraordinary.
 
 CHAPTEK VIII. 
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 
 
 IT now becomes necessary, by way of supplement to the 
 two preceding chapters, to consider how the existence of 
 the Joint Undivided Family among the Hindoos, and of 
 the House-Community among the Slavonians, bears upon 
 Patria Potestas. 
 
 And, to begin, let it be said that we know nothing 
 at all of the latter, and very little of the former, 
 except from comparatively recent descriptions. From 
 the early Hindoo writings it might be a justifiable 
 inference that undivided families were very much rarer 
 and smaller in dimensions among early Hindoos than 
 they are now. That, perhaps, is not a matter to dogma- 
 tise about ; but it is certain that those writings tell us 
 extremely little about the Undivided Family. And 
 their teaching was anything but favourable to it, for 
 they commonly inculcate upon fathers that they should 
 divide the family estate in their lifetime. Of Village 
 Communities, it may be said, those writings tell us 
 literally nothing. There is nothing in them from which it 
 can be gathered even that such communities were known.
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES, 97 
 
 Our knowledge of the Slavonian House-Community, 
 again, whatever may be the antiquity of that insti- 
 tution, is entirely derived from modern times. Its 
 existence in pre-Christian times, though anything but 
 improbable, is matter of conjecture ; and what it was 
 then nobody knows. It is from accounts of it written 
 after the Slavonians had been for many hundred years 
 Christians, and more or less in contact with other races, 
 that such inferences as can be made from it have to be 
 made. The Village Community is not known to have 
 ever existed among the South Slavonians. 
 
 It does not seem reasonable to think that con- 
 clusions as to the primitive social condition of man 
 are to be obtained from what we know of either of those 
 institutions. 
 
 It should be said that Undivided Families, or House- 
 Communities, or, to speak more generally, households 
 composed of the persons nearest to each other in ac- 
 knowledged kinship, are not found with one family 
 system only, with one system of kinship or inheritance 
 only. They are constantly found among peoples whose 
 material condition is not much advanced, whatever 
 their family system. But it would seem that if they 
 are more proper to one family system than to another 
 it is to the family system founded upon kinship through 
 females only. At any rate, as Mr. J. D. Mayne has 
 pointed out,* "the most perfect form of the joint family 
 now existing" is found among the Nairs of Malabar, 
 with kinship acknowledged through females only, and 
 * Hindu Law and Usage, pp. 179 and 192. 
 
 H
 
 98 THE PATRIARCHAL THEORY. 
 
 paternity unrecognised. A people which in the nine- 
 teenth century has House - Communities composed 
 chiefly or entirely of persons related to each other 
 through male descents may possibly have had, at an 
 earlier time, House-Communities composed of persons 
 related to each other through female descents. At any 
 rate, all that can be inferred from their now living in 
 House-Communities (that they acknowledge kinship 
 through males being otherwise known) is that their 
 material condition is such that this way of living still 
 suits them. And we must go to other sources when we 
 seek the means of concluding whether they have always 
 acknowledged relationship through male descents or not. 
 As to this, the present constitution of their House- 
 Communities cannot prove anything. 
 
 So much having been premised, we go on to state 
 Sir Henry Maine's view of the relation of the Joint 
 Undivided Family and of the House-Community to 
 Patria Potestas. It can be stated very briefly. 
 
 He tells us that the Joint Undivided Family,* 
 " wherever its beginning is seen " in Aryan Commu- 
 nities, springs universally out of the Patriarchal Family, 
 with despotic powers centred in the Paterfamilias ; and 
 that, as in such communities we find it springing from 
 a Patriarchal cell, so, when it dissolves, we see it 
 dissolving into a number of such cells. This he says 
 equally of the House-Community of the South Slavonians 
 and of the Joint Undivided Family of the Hindoos ; for 
 he holds the two to be identical, and both to be 
 * Early History of Institutions, pp. 116-118.
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 99 
 
 identical with the Gens (or alternatively with the body 
 of Agnates) of the Romans. The Hindoo and Slavonian 
 cases are the only cases of this kind he has considered. 
 
 Now, as to the former, it has already been shown 
 that evidence of Patria Potestas having at any time 
 prevailed among the Hindoos is altogether wanting ; 
 and especially that the "legal testimony," to which 
 alone Ancient Laiv made appeal, proves not the 
 prevalence of it, but the absence of it. Among the 
 Hindoos, as a fact, we do not see the Joint Undivided 
 Family either springing from a Patriarchal cell or dis- 
 solving into a number of such cells. We only know it 
 as springing from, and also as succeeded by, a family 
 system in which the paternal power was severely 
 restrained, and in which wives and children had rights 
 inconsistent with the Patriarchal Theory. 
 
 But is there not in the constitution of the Joint 
 Undivided Family among the Hindoos something that 
 makes so strongly for Patria Potestas as to impress 
 upon us irresistibly that there is a close connection 
 between the two ? No. Sir Henry Maine himself tells 
 us that the head of such a family is not a Paterfamilias; 
 that he is not the " owner of the family property, but 
 the manager of its affairs and the administrator of its 
 possessions ; " and as to the Power of the Father, the 
 utmost he can say is that each father or grandfather 
 " has more power than anybody else " over his wife and 
 his descendants. He allows, in fact, that there is 
 nothing at all approaching to, or suggestive of, Patria 
 Potestas to be found in the Hindoo Joint Undivided 
 
 H 2
 
 100 THE PATRIARCHAL THEORY. 
 
 Family. The head of the household is only an adminis- 
 trator ; the father only has (and it would be strange 
 indeed if less could be said of him) " more power than 
 anybody else " over his family. 
 
 One could understand it being inferred that, wherever 
 the Joint Undivided Family has been of common occur- 
 rence, Patria Potestas must have been unknown. As 
 has been said already, the most perfect specimens of 
 the Joint Family are found where fatherhood is not 
 recognised. 
 
 Passing now to the House - Community of the 
 Slavonians, it has to be recalled that, while a good deal 
 of evidence casting doubt on the Patriarchal Theory has 
 come from Slavonic sources, in the scantiness of our 
 knowledge of the early Slavs, we have no information 
 as to what the paternal power was among them. Sir 
 Henry Maine, being without means of showing what 
 was the constitution of the family out of which the 
 House- Community originally sprung among the Slavs, 
 has been content with inquiring into the constitution of 
 the family which issues from it now that is, of the 
 ordinary family, or, as he calls it, the " natural family," 
 now found, and quite commonly found, among the South 
 Slavonians. On his theory, the House- Community must 
 have sprung out of the Patriarchal Family, and that in 
 turn should issue out of the House-Community. Accord- 
 ingly he has given reasons on which he thinks it may be 
 concluded that the natural family now found among the 
 Southern Slavs is a Patriarchal family "despotically 
 governed by the oldest ascendant." He has arrived at
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 10 L 
 
 this conclusion, however, by inference. And he has put 
 aside testimony which makes that way of getting a con- 
 clusion unnecessary. 
 
 He gives two reasons for his inference. The first 
 is that there is in South Slavonian countries a great 
 respect for old age. The other is that in these countries 
 a distinction is maintained between agnatic and cognatic 
 relationship. 
 
 As to the former, respect for old age is common in 
 nearly every country, whatever the degree of parental 
 authority or the system of kinship ; and it has nowhere 
 been stronger than in Red Indian villages in, which, 
 relationship being counted only through women, a son 
 was not even a relative of his father. Add to this, that, 
 among the South Slavonians, it extends equally to old 
 women and old men, and there need be no hesitation 
 about declaring that the respect for age shown by the 
 South Slavonians is absolutely irrelevant to an inquiry 
 as to the powers of fathers among them. 
 
 As to the second, the fact founded upon is that, 
 among the South Slavonians, relations through male 
 descents are called relations of the great blood, and 
 relations through female descents relations of the little 
 blood. This shows, no doubt, some preference for 
 relations through male descents. But it cannot show 
 whether formerly that preference was greater than it is 
 now or less ; and still less can it show that at one time 
 relations through female descents were not acknowledged 
 as relations at all. A preference for relations through 
 male descents, that is, is not Agnation, and cannot prove
 
 102 THE PATRIARCHAL THEORY. 
 
 that relationship was formerly agnatic. And among the 
 South Slavonians, in point of fact, relationship through 
 women is most fully acknowledged and carries rights 
 of succession whereas Agnation denies all relationship 
 and, above all, all rights of succession to persons 
 claiming through women. Sir Henry Maine, however, 
 has concluded that he has here found a recognition of 
 agnatic relationship. Applying his formula, that this is 
 good evidence of Patriarchal Power subsisting or having 
 formerly subsisted, he then infers that the natural 
 families of the South Slavonians must be Patriarchal 
 families "despotically governed by the oldest ascen- 
 dant." Let us now inquire what they actually are."* 
 
 Though families of an ancient type, they are 
 extremely unlike the Patriarchal Family of Sir Henry 
 Maine's theory. The man, no doubt, is the head of the 
 house, and the wife is expected to "honour and obey" 
 him. But she is not without honour and influence 
 in her turn. She is absolute mistress in domestic 
 matters. She always has separate property, more or 
 less, and transmits it to her children. As regards the 
 children, the rights and duties of the spouses are the 
 
 * Such preference for relations through male descents as is found 
 among the South Slavonians, and the phrases which express it, 
 appear to be fully accounted for by the fact that such relations and 
 their children habitually live together with common interests in the 
 same House-Community ; while their relations through women are in 
 other House-Communities in which they have common interests with 
 strangers. If this be a true account of the preference, it has nothing 
 to do with Patria Potestas, and is among the results in working of 
 a family arrangement under which as will be seen by-and-by 
 parental power is curiously restricted.
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 103 
 
 same ; and if the father dies, the mother takes his place 
 till the eldest son is old enough to manage. The mother 
 naturally has most to do with the girls and little ones, 
 the father, though over all, taking for his special charge 
 the sons as they grow old enough to be useful. So far, 
 these households exhibit nothing unfamiliar ; nothing 
 very different from what, with the inevitable diver- 
 gences between theory and practice, varying from 
 country to country, from house to house, is to be 
 found in other parts of Europe. There are, neverthe- 
 less, between these and Western households remarkable 
 differences some of which it is important for us to 
 note. In some parts of South Slavonia, children are 
 allowed to keep certain of their acquisitions for them- 
 selves, but commonly among the South Slavonians 
 especially among the poorer folk so long as they 
 remain at home, they hand over all their earnings to 
 their father, who, on his part, provides for all the wants 
 of the family. On the other hand, a son on marrying 
 is free to leave his father who thenceforth has no 
 control over him or his affairs ; and, when he does 
 leave, he carries his proportional share of the family 
 possessions along with him. And, though it is at 
 marriage that this is usually done, the custom of the 
 country allows every grown-up son, whether married 
 or not, at any time to go forth from the paternal 
 household, taking with him his share of the family 
 possessions. Daughters, it should be said, are provided 
 for by dowry, and only become their father's heirs 
 (which they are in that case though married) when
 
 104 THE PATRIARCHAL THEORY. 
 
 there are no brothers. Sometimes a whole household 
 agrees to divide and break up ; and then the son with 
 whom the old people elect to live who is usually the 
 youngest gets on account of that a somewhat larger 
 share than that to which he would otherwise be entitled. 
 Enough has now been said to distinguish the Slavonian 
 from the Western family, and enough to show that 
 it differs essentially from the Patriarchal Family.* 
 
 Both wives and children have rights which the 
 Patriarchal Theory denies them. The wife has pro- 
 perty ; and, in succession to her husband, she may 
 become the head of the house. Failing brothers, a 
 married daughter may be her father's heir. The son, 
 instead of being a slave, has rights which, in families 
 of a more modern type, sons do not enjoy. The father, 
 no doubt, has that power over the young and helpless 
 which, subject to the limitations of law and opinion, 
 parents can everywhere exercise. But that is not 
 Patria Potestas. He keeps the common purse and 
 manages while his children are young there is nothing 
 very strange in that. What is strange is that his 
 sons, as they grow up, can leave him one by one, 
 each stripping him when he does so of a portion 
 of what in other countries would be accounted 
 his possessions. That his son, once grown-up, should 
 be free to assert independence of him would be con- 
 tradiction sufficient of the Patriarchal Theory. That 
 he should be then entitled both to independence and 
 
 * See Fedor Demelic, Le Droit Coutumier des Slaves Meridionaux. 
 D'apres le* Recherches de M. V. Bogichicli. Paris, 1877.
 
 V 
 
 / 
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 105 
 
 his share of the family possessions discloses a view 
 of family right which is separated by a gulf from that 
 view of it on which the " oldest ascendant " is lifelong 
 lord of his family and its property. But it is apparent 
 that in every important relation the Slavonian natural 
 family discountenances the Patriarchal Theory. It 
 shows, what a multitude of examples show, that a 
 family may be of ancient type, and yet be extremely 
 unlike that Eoman family which Sir Henry Maine has 
 called the Patriarchal. 
 
 When we come to inquire whether there is anything 
 in the constitution of the House - Community that 
 makes for Patria Potestas, the answer must be, as it was 
 in the case of the Joint Undivided Family of the 
 Hindoos, that there is nothing ; and that it is difficult 
 to believe that a high view of paternal power could 
 be entertained in a country in which such communities 
 have been very popular for ages. Slavonic writers 
 have insisted upon the dissimilarity between the House- 
 Communities and the Patriarchal Family, and upon the 
 inapplicability to the former of such words as patriarchal. 
 M. Fedor Demelic, from whose little tractate Sir Henry 
 Maine has chiefly derived his knowledge of those 
 institutions, has fully recognised the justice of this pro- 
 test ; and so, indeed, has Sir Henry Maine. Briefly, the 
 House-Communities are, among the South Slavonians, in- 
 dustrial partnerships formed, no doubt, among relatives 
 in the management of which, in general, all the grown 
 men and all the married women have a voice ; in which 
 a father and every grown-up son of his have precisely
 
 106 THE PATRIARCHAL THEORY. 
 
 the same position and rights ; having at their head an 
 elected manager who, though much controlled, has 
 considerable powers for management and discipline, to 
 which all, young and old, are subject, and which, 
 therefore, cross and cut down a father's control over 
 even his younger children. They are, therefore, societies 
 in which the Patriarchal Theory is set at open defiance. 
 It would seem to be highly improbable that, immediately 
 out of a family in which the " oldest ascendant " was 
 supreme, there ever issued a society in which he has 
 only equal rights with his grown-up son or grandson. 
 This would be despotism giving birth to democracy. 
 That the " oldest ascendant " should unless he happen 
 to be the elected head have only equal rights with 
 his descendants can in no possible way be made to 
 support the theory that he was at a former time 
 supreme over them all. 
 
 It will be proper, however, to show somewhat fully 
 what the South Slavonian House-Communities are, and 
 it will be well, by way of preface, to give some account 
 of the peasant household as it is (or till lately was) in 
 Russia. 
 
 Before the emancipation of the serfs, the peasant 
 household in Russia, Mr. Wallace tells us,* often con- 
 tained representatives of three generations ; all, young 
 and old, living together under the direction and authority 
 of the head of the house, called usually Khozain or 
 administrator, or in some districts Bolshak, "the Big 
 
 * Russia. By D. MacKenzie Wallace. Cassell, Fetter and 
 Galpin : London, 1877. Vol. I., pp. 134-7.
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 107 
 
 One." The Khozain's position was, generally speaking, 
 occupied by the grandfather or, if he was dead, by the 
 eldest brother ; but this rule was not strictly observed. 
 If the grandfather was infirm, or the eldest brother in- 
 capable, the place of authority was taken by some other 
 member it might be by a woman who was a good 
 manager and possessed the greatest moral influence. The 
 relations between the head of the house and the other 
 members depended upon custom and personal character, 
 and they consequently varied greatly in different families. 
 " The house with its appurtenances, the cattle, the agri- 
 cultural implements, the grain and other products in a 
 word, the house and nearly everything it contained 
 was the joint property of the family. Hence nothing 
 was bought or sold by any member not even by the 
 Big One himself, unless he possessed an unusual amount 
 of authority without the express or tacit consent of 
 the other grown-up males, and all the money that was 
 earned was put into the common purse." " The peasant 
 household of the old type/' Mr. Wallace proceeds, " is 
 thus a primitive labour association, of which the 
 members have all things in common, and it is not a 
 little remarkable that the peasant conceives it as such 
 rather than as a family. This is shown by the customary 
 terminology and by the law of inheritance. The head 
 of the household is not called by any name correspond- 
 ing to Paterfamilias, but is termed Khozain or adminis- 
 trator a word that is applied equally to a farmer, a 
 shop-keeper, or the head of an industrial undertaking, 
 and does not at all convey the idea of blood relationship.
 
 108 THE PATRIARCHAL THEORY. 
 
 The law of inheritance is likewise based on this conception. 
 When a household is broken up, the degree of blood 
 relationship is not taken into consideration in the dis- 
 tribution of the property. All the adult male members 
 share equally. Illegitimate and adopted sons, if they 
 have contributed their share of labour, have the same 
 rights as the sons born in lawful wedlock. The married 
 
 O 
 
 daughter, on the contrary being regarded as belonging 
 to her husband's family and the son who has pre- 
 viously separated himself from the household, are 
 excluded from the succession. Strictly speaking, there 
 is no succession or inheritance whatever, except as 
 regards the wearing apparel and any little personal 
 effects of a similar kind. . . . The members do not 
 inherit, but merely appropriate individually what they 
 had hitherto possessed collectively. Thus, there is 
 properly no inheritance or succession, but simply 
 liquidation and distribution of the property among the 
 members." 
 
 This statement of an excellent authority leaves little 
 need for comment. It is plain that the oldest, if fit, 
 was always the head of the family, but that capacity 
 was too important to be long dispensed with, even in 
 the case of the common progenitor. And a woman 
 might, as the most capable, come to have the manage- 
 ment. As to the powers of the manager, they were 
 what his personal character and the circumstances 
 of the family happened to procure him. But he no 
 more had the powers of a Paterfamilias than he had
 
 UNDIVIDED FAMILIES AND ROUSE-COMMUNITIES. 109 
 
 the name. As regards control over property, indeed, he 
 was so very far from having the position of a Pater- 
 familias that he could do nothing involving gain or loss 
 could neither sell nor buy without the consent of 
 the other members ; while he got, if from any cause the 
 household broke up, simply an equal share with the 
 others. When the family consisted, as it often did, 
 of one man and his descendants, it was identical in 
 composition with the Patriarchal Family. Observe 
 then the disparity between the powers ascribed to the 
 head in the one case and the powers actually possessed 
 by the head in the other. The Russian " Patriarch," 
 instead of being master in all things, governing all 
 despotically, stood in so different a relation to his 
 household that he was called administrator or manager ; 
 and while, in other respects, he had such power as he 
 could assert for himself, he needed for his actings, even 
 the most trifling, affecting property, the consent of his 
 grown-up descendants they being, as his co-partners, 
 equally interested with himself. Here, as in the South 
 Slavonian natural families, we find prevailing a view of 
 family right utterly different from that which underlies 
 the Patriarchal Theory. And the rule of inheritance, 
 or rather of partitioning, of these Russian households, 
 as explained by Mr. Wallace, clearly gives the sugges- 
 tion that married women and forisfamiliated sons were 
 cut off from succession, not on any profound theory of 
 the family, not for example through any fear of " society 
 being confounded," but on the simple ground that those
 
 110 THE PATRIARCHAL THEORY. 
 
 who had been conjoined in producing property, and 
 were the joint owners of it, were the persons who 
 should keep or divide it.* 
 
 The South Slavonian House- Community also is 
 regarded as an industrial association. And that is 
 precisely what it is. Sometimes it springs out of the 
 " natural family," through the members of such a 
 family, notwithstanding that they are free to separate, 
 electing to remain in association after the death of their 
 common progenitor; and, in that case, in its earliest 
 stages it must closely resemble the Russian peasant 
 household which has just been described. But it 
 appears very commonly to spring out of the ruins of 
 an older House-Community persons who had been 
 associated in a community which has been dissolved 
 recombining themselves, at their own choice, in a 
 number of new ones. The House-Community, which 
 is never large including on an average, according to 
 
 * The House-Community in Russia was kept alive, and the 
 authority of the chief as much as possible upheld, by the land- 
 owners, for reasons of their own, during the existence of serfage. 
 Since the emancipation of the serfs it has rapidly been disappearing. 
 It has already been stated that evidence of its antiquity seems not to 
 be forthcoming. Even the travellers of the sixteenth century have 
 not noticed it. Fletcher and Horsey appear not to have thought of 
 the Russian family as in any way different from the English ; and the 
 former tells us that " the commons pass over their lands by discent 
 of inheritance to whichever son they will, which they commonly do 
 after our gavel-kind, and dispose of their goods by gift or testa- 
 ment without any controlment." Fletcher's " Kusse Commonwealth " 
 in Russia at the Close of the Sixteenth Century. London, 1856. 
 Printed for the Hakluyt Society, p. 27. 
 
 But the more ancient the House-Community, the deeper does the 
 Patriarchal Family seem to be relegated into times of darkness.
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. Ill 
 
 M. Bogichich, from twenty to twenty-five persons 
 which never is large enough to form even a small 
 village, though all the inhabitants of a village are 
 occasionally of the same name comes after a time to be 
 of inconvenient size, and then it splits up. Thereupon, 
 while some of the members may go to live apart, each 
 with his own family, the majority reunite at their 
 pleasure in new communities. These are obviously 
 voluntary associations, founded upon express contract, 
 though subject to conditions which custom has estab- 
 lished. And though they consist of persons who have 
 previously been associated together, and are therefore 
 all or nearly all of the same stock while both con- 
 venience and family feeling tend to keep together those 
 who are most nearly related to each other they do not 
 necessarily consist of near relations. 
 
 Whether the House-Community originates in the 
 one way or in the other, the object of association is 
 the same. The Community strives to provide for the 
 families contained within it, not everything which their 
 standard of comfort requires, but necessaries lodging, 
 food, and clothing according to a customary not over- 
 lavish scale. While, therefore, there is combination in 
 labour, and a common purse among the associates, and 
 while much care is taken to secure that everyone does 
 his duty by the Community, there is need for separate 
 industry. And there is also room for private accumu- 
 lations. Each grown-up man works as he can for him- 
 self, to supplement what the Community gives him. 
 His wife and, when they are young, his children work
 
 112 THE PATRIARCHAL THEORY. 
 
 for him. And what is saved of the earnings of such 
 industry is his own. What comes to him by inheritance 
 also is his own. Moreover, the married woman has her 
 dowry reserved as her separate property ; and, as a rule, 
 even the younger folks, though most of their earnings 
 go to their father, are allowed to make their little 
 hoards. 
 
 The Community owns the land tilled in common 
 and the buildings upon it, the stock and the instru- 
 ments of labour, the products of the joint industry, and 
 the savings made out of them. A council, composed of 
 all the grown-up men and the married women (the 
 latter being in most places free both to speak and to 
 vote) considers and decides upon every matter of any 
 importance affecting the common interest unanimity of 
 decision being required in some districts ; in which case, 
 of course, a single person, if obstinate, can defeat the 
 wish of all his associates. And at the head of the 
 Community is an official called the Domacin, who dis- 
 tributes to the men their daily tasks and maintains 
 discipline ; who buys and sells not, however, except in 
 trifling matters, without the consent of the council 
 and who is, generally speaking, the representative of the 
 Community in its dealings with those outside it. There 
 is a Domacica too (usually the Domacin's wife), who has 
 similar powers and duties as regards the women. The 
 Domacin is elected, though, where one person is from 
 age and capacity marked out in the opinion of all for 
 the place, the form of election may be dispensed with ; 
 and he may be deposed, but the consent of all being
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 113 
 
 necessary for this, lie usually holds office till age makes 
 him unfit for it. He is generally one of the seniors, and 
 at any rate where all are near relations the oldest, if 
 capable, seems to be preferred ; but capacity is indis- 
 pensable, and therefore the choice of the household 
 frequently falls upon one of the younger men, and 
 sometimes even upon a married woman. His authority 
 varies with the trust reposed in him ; but he always has 
 the considerable powers which are deemed necessary to 
 secure the just performance of work and to maintain 
 order ; and these, while they extend over all, are 
 specially exercised over all the males, young and 
 old alike, who are fit for labour children of tender 
 years only being left to the control of their parents. 
 The Domacin can admonish any member ; he can chas- 
 tise none, unless it be the very young people. It is 
 in dealings with property that he is most under control, 
 but he is accountable to the council in all things ; and, 
 indeed, with such powers as are thought indispensable 
 for his office, he is only primus inter pares, holding his 
 place on condition of good behaviour. In a partition of 
 the common goods he gets only the same share as any 
 other, with, in some places, a present of some value 
 added. For such partitions there is no uniform rule 
 among the South Slavonians the distribution being 
 made in some districts per stirpes and in others per 
 capita. But excepting that in Herzegovina, and pro- 
 bably in other places, young women who have no 
 brothers share like men, and that everywhere some 
 provision is made for the helpless members of the house-
 
 114 TffE PATEIAECJTAL THEORY. 
 
 hold who are not entitled to share, it is the adult males 
 only who divide the common property between them. 
 
 This account of the South Slavonian House-Com- 
 munity, very imperfect as no doubt it is, suffices to 
 make clear the points that are material for the present 
 inquiry. Though composed of relatives, the persons who 
 form it need not be each other's near relations. It is 
 usually not a family, and it is a partnership the partners 
 being all the adult males. But the married women are 
 consulted in everything, and a married woman may 
 be the head of the association, with control over all 
 the men belonging to it. A young man, once come to 
 manhood, has in this association the same position and 
 rights in all respects as his father or grandfather ; and 
 a father's control over his son is, between a very early 
 age and manhood, in a great measure superseded by 
 that of the Domacin. No man has ampler rights than 
 the House-Community allows to grown-up sons. And 
 a father's power over even young children is greatly 
 restricted. This society undoubtedly has a lively sense 
 of the uses of authority. But it makes unusually 
 little of the Power of the Father. The powers of the 
 Domacin, again, considerable as they are, are all given 
 to him for the sa,ke of good management. He possesses 
 them in virtue of election ; and he holds them at the 
 pleasure of his associates to whom his relation is 
 simply that of an equal whom their own choice has 
 set over them and invested with managerial powers. 
 As to the women, a woman may be Domacin. And 
 the married women sit with the men in the council
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 115 
 
 which controls the Domacin. In this council, the 
 husband, the wife, and their adult son and no doubt 
 often grandfather and grandson meet each other and 
 their neighbours for consultation on the common 
 interest with equal rights. If the opinions of the old 
 have more weight than those of the young, that is only 
 what usually happens in all assemblies. But in some 
 districts, as has been said, any one even the youngest 
 of the associates has it in his power to thwart a 
 project though all the others, Domacin included, are 
 bent upon it. 
 
 The South Slavonian House-Community, therefore, 
 is not so much a family as an association of persons 
 who are relatives, and who need not be near relatives 
 an association quite commonly formed by persons who 
 have previously been associated together along with 
 others, and whose interests can conveniently be conjoined. 
 And, as has been said already, it is in all cases an 
 association which at every point sets " Patriarchal " 
 ideas at defiance. The family which issues out of it, 
 or is found alongside of it, too, is utterly different from 
 the Patriarchal Family. Paternal power is so far in 
 suspense in this House-Community that we should be 
 prepared to say of it, even more than of the Hindoo 
 Undivided Family, that the people among whom it has 
 been popular could not have had, or could have cared 
 nothing about such paternal powers as are denoted by 
 Patria Potestas. Economical considerations govern it, 
 and have taken from fathers not a little of the power 
 which modern societies allow to fathers, in order that 
 
 i 2
 
 116 THE PATRIARCHAL THEORY. 
 
 the powers necessary for control and management may 
 be possessed by an elected head whose powers, never- 
 theless, differ both in kind and in degree from those of 
 a Paterfamilias. And so greatly have the House-Com- 
 munities affected current notions that the relation of 
 the head of an ordinary family even to his children is 
 that of a Domacin rather than that of a father. 
 
 It would be unreasonable to make inferences from 
 the South Slavonian House-Communities against the 
 Patriarchal Theory. But it would be even less allowable, 
 were it possible to do so which clearly it is not to 
 make inferences from them in its favour. 
 
 It is obvious that, though Communities of a natural 
 type arise, on the whole the House-Communities have 
 drifted from the natural type of the Joint Family or 
 House-Community which we find in India, in per- 
 fection among the Nairs, with unrecognised fatherhood, 
 and more commonly, in a less stable form, based upon 
 kinship through male descents. It commonly happens 
 that those relations whom it suits to do so and they 
 are usually persons who have already been associated 
 together club their resources ; and the result is a 
 House-Community, in which partnership ideas pre- 
 dominate, in which all adult men are equal, and in 
 the management of which the women have their full 
 share. The South Slavonian House-Communities have 
 all grown out of a natural type of the Joint Family, 
 no doubt. They are not commonly identical in com- 
 position with the Hindoo Joint Family, but they some- 
 times are ; and it may be believed that at some former
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. 117 
 
 time they commonly resembled it more or less. They 
 would prove, if proof were needed, that joint families 
 can exist among people who acknowledge kinship 
 through male descents. And they would prove, if 
 proof were needed, that kinship through males has long 
 been acknowledged among the South Slavonians. But 
 they cannot inform us what paternal power, or what 
 limitation of kinship, accompanied kinship through 
 males among the early Slavonians ; or whether that 
 was the earliest kinship acknowledged among them ; 
 or even on what system of descents the earliest House- 
 Communities were based among them. For views about 
 such matters unless we choose to do without them 
 we must go to other sources. The same might be said 
 of the Joint Family of the Hindoos, but it is not 
 necessary to say it. We know a good deal about the 
 early condition of the family among the Hindoos ; and 
 it suffices to repeat that we do not find for it at any 
 time any proof of Patria Potestas. 
 
 The South Slavonian Communities, short-lived as 
 they usually are, cannot even give us suggestions (as 
 Sir Henry Maine thinks they can) as to the manner 
 in which chiefly authority arose. For such suggestions 
 we must look to communities which have produced 
 chiefs; not to communities which, so far as we know, 
 have never produced them. And, indeed, the history 
 of the growth of Power may be as hopefully studied in 
 the directorate of a joint-stock bank or of a railway 
 company as in a South Slavonian House-Community. 
 
 Sir Henry Maine at one time identified the Roman
 
 118 THE PATRIARCHAL THEORY. 
 
 Gens with the Village Community. He has since that 
 identified it with the Joint Family and with the House- 
 Community.* And in his latest workt he has given us 
 the alternative of identifying these with the Agnates of 
 Konian relationship. Between views so fluctuating a 
 selection must here be made ; and only a few words can 
 be said of his identification of the Joint Family and 
 House-Community with the Gens. What he says (or 
 rather what he formerly said) is that there can be no 
 reasonable doubt " that the House-Community is the 
 Gens ; " and that the Gens " actually survives " in the 
 Joint Family. 
 
 The Joint Family and House-Community grow up 
 within the limits of known relationship. They include 
 at most only a section of the Agnates. And they do not 
 last for which the reason is obvious. After a time 
 they become inconveniently large, and then they break 
 up. Convenience leads to their being established, and 
 favours them for a period, but it ultimately destroys 
 them. And, when they have been dissolved, they leave 
 no mark upon the future relations of those who have 
 been associated in them and their descendants though, 
 of course, their relationship is unaffected. The people 
 who compose them, among both Hindoos and South 
 Slavonians, acknowledge and their predecessors, so 
 far as we know, have always acknowledged relation- 
 ship both through male descents and through female 
 descents ; and are themselves but a section, often a very 
 
 * The Nineteenth Century, December, 1877, p. 799. 
 t Early Law and Custom, p. 239.
 
 UNDIVIDED FAMILIES AND HOUSE-COMMUNITIES. H9 
 
 small one, of those who are related to each other 
 through male descents being, in fact, relations who 
 have interests which are common, or interests which 
 can conveniently be conjoined. Both Hindoos and 
 South Slavonians, too, acknowledge, as the Romans did, 
 a wider relationship than can be traced, a relationship 
 which has left marks upon usage. The Hindoos have 
 the Gotra, which includes all persons who bear the 
 same family name, and marriage was forbidden between 
 persons of the same Gotra. Similarly, the South Slavo- 
 nians have the Clan (as some of them have it still, it 
 may be presumed they all had it formerly), consisting 
 of persons of the same name, and marriage is forbidden 
 within the Clan. 
 
 Now, the Roman Gens consisted of persons who 
 bore the same family name, and who had only a tradition 
 of relationship. And there is some reason to think that 
 marriage between its members was anciently forbidden. 
 It appears to correspond to the Gotra of the Hindo'os, 
 and to the Clan of the South Slavonians, and not to the 
 Joint Family or the House-Community. 
 
 It is clear, too, that Joint Families or House-Com- 
 munities cannot even throw light upon the early history 
 of Gotra, Clan, or Gens unless the Patriarchal Theory is 
 assumed.
 
 CHAPTER IX. 
 
 PATRIA POTESTAS IN IRELAND. 
 
 IN his Early History of Institutions (p. 216), in a 
 passage already quoted, in discussing what he calls a 
 perplexing problem relative to the Family among the 
 early Irish, Sir Henry Maine says he thinks he can 
 assign " some at least plausible reasons " for believing 
 that the organisation therein involved was " a monu- 
 ment of that Power of the Father which is the first and 
 greatest landmark in the course of legal history." He 
 has found corroboration of this view in the fact that the 
 father among the Irish had (along with various other 
 persons) "judgment and proof and witness" over his 
 son the power, if it was a power, thus enigmatically 
 referred to appearing to him to be Patria Potestas. Of 
 his solution of the problem he did not speak over -confi- 
 dently in the first instance. But he has since been 
 quite confident in asserting that he has found Patria 
 Potestas among the Irish. 
 
 The problem was raised in the first instance by a 
 passage in the Book of Aicill,* under the heading, 
 
 * Ancient Laws of Ireland, Vol. III.
 
 PATRIA POTESTAS IN IRELAND. 121 
 
 " What is the Reciprocal Right among Families ? " and 
 what it is may in a general way be gathered from the 
 following passage, taken from the preface to the third 
 volume of the Senchus Mor : " The most remarkable 
 custom described in the ' Book of Aicill ' is the fourfold 
 distribution of the family into the ' geilfine/ ' deirb- 
 fine/ ' iarfine/ and ' indfine ' divisions. . . . Within the 
 family seventeen members were organised in four divi- 
 sions, of which the junior class, known as the geilfine 
 division, consisted of five persons ; the ' deirbfine,' the 
 second in order, the ' iarfine/ the third in order, and the 
 ' indfine/ the senior of all, consisted respectively of four 
 persons. The whole organisation consisted, and could 
 only consist, of seventeen members. If any person was 
 born into the ' geilfine ' division, its eldest member was 
 promoted into the ' deirbfine ; ' the eldest member of the 
 ' deirbfine ' passed into the ' iarfine ; ' the eldest member 
 of the ' iarfine ' moved into the ' indfine ; ' and the 
 eldest member of the ' indfine ' passed out of the orga- 
 nisation altogether. It would appear that this transi- 
 tion from a lower to a higher grade took place upon the 
 introduction of a new member into the 'geilfine ' divi- 
 sion, and therefore depended upon the introduction of 
 new members, and not upon the death of the seniors." 
 
 Only men, or male persons, however, could be 
 included in the divisions, and the organisation might 
 consist of fewer, though it could not include more, than 
 seventeen persons. Divisions, too, are spoken of as 
 if they regularly existed in a group of four related to 
 one another. The passage from which we learn that,
 
 122 THE PATRIARCHAL THEORY. 
 
 on a new birth into the geilfine division, its numbers 
 being complete, the oldest member of that division 
 had to move into a higher division, assumes that the 
 four divisions would usually be in existence when such 
 a transference took place, and that the removal of the 
 oldest member from each of the three other divisions 
 would follow upon it. And when two of the four divisions 
 became extinct through the death of their members, 
 there was a provision made for new divisions being 
 formed in place of them as if to restore the organisation 
 to its normal state. It is laid down that, in this case, 
 the surviving divisions were not to inherit the property 
 of the divisions which were defunct, unless they were 
 able to form and did form, out of the family, two divi- 
 sions in place of those which had failed. The case of 
 the geilfine and indfine divisions having become extinct 
 is one actually put, and it is assumed that there would 
 be at hand members of the family as fit to be placed 
 in a new geilfine and a new indfine division as the 
 persons who had died. It should be added that it is 
 mainly through the laws of succession which connected 
 each of the divisions with all the others that we know 
 so far as we do know about the organisation. When 
 all the members of any of the divisions were dead, its 
 property was divided between the other divisions, and 
 the deirbfine division, the third in seniority, was the 
 most favoured as regards successions ; the iarfine 
 coming next, and the geilfine, the youngest division, 
 in the third place, while the oldest division, the indfine, 
 fared worst of all.
 
 P ATRIA POTESTA8 IN IRELAND. 123 
 
 The composition of this family organisation and its 
 purposes are discussed at length in Studies in Ancient 
 History (pp. 453 et seq.). But at least five different 
 theories of its composition have been propounded. And 
 it is anything but strange that there should be a conflict 
 of views on that subject. For the notices which we 
 have of this organisation are scanty and fragmentary, 
 while their meaning is often a puzzle in the English 
 version, and must have been still more of a puzzle in 
 the original. Enough can clearly be made out about it, 
 however, to enable us to judge whether Sir Henry 
 Maine is right in regarding it as " a monument of the 
 Power of the Father." It is only so far as they are 
 necessary for that, that the facts relating to it concern 
 us here. And now nothing need be added but a 
 passage from the Book of Aicill, which undoubtedly 
 throws light on the composition of the organisation, and 
 which has suggested to Sir Henry Maine a portion of 
 his theory. The passage runs : "If the father be alive, 
 and has two sons, and each of these sons has a family of 
 the full number, i.e., four, it is the opinion of the 
 lawyers that the father would claim a man's share in 
 every family of them, and that in this case they form 
 [literally, there are] two geilfine divisions; and if the 
 property has come from another place, from a family 
 outside, and there be in the family a son or brother of 
 the man from whom it has come, he shall not get it 
 .more than every other man of the family." This 
 passage shows that lawyers contemplated as possible 
 that two of the organisations might be formed among
 
 124 THE PATRIARCHAL THEORY. 
 
 one man's descendants, and it gives their view of what 
 the man's rights of property would be in that seemingly- 
 unusual case.* 
 
 Sir Henry Maine's theory is that any man of a sept 
 might become a root from which might spring as many 
 of the groups of seventeen persons as he had sons. 
 " As soon as any one of the sons had four children, a 
 full geilfine sub-group of five persons was formed ; but 
 any fresh birth of a male child to this son, or to any of 
 his male descendants, had the effect of sending up the 
 eldest member of the geilfine sub-group, provided 
 always he were not the person from whom it had sprung, 
 into the deirbfine. A succession of such births com- 
 pleted in time the deirbfine division, and went on to 
 form the iarfine and the indfine." On this view, the 
 fifth person in the geilfine sub-group was the man who, 
 with his descendants, made up the organisation. He 
 always remained in that sub-group, and was the " geil- 
 fine chief" or Paterfamilias. The geilfine sub-group 
 was his hand-family, or Family under Power ; while the 
 other groups were made up of the emancipated de- 
 scendants, " diminishing in dignity in proportion to 
 their distance from the group (the geilfine) which, 
 according to archaic notions, constitutes the true or 
 representative family." Sir Henry Maine has not 
 attempted to give any account of the purposes for 
 
 * The case put may have been a hypothetical one. "The 
 lawyers " are much given to pronouncing upon hypothetical cases ; 
 and the passage reads as if the law laid down was laid down 
 speculatively rather than known from actual practice.
 
 P ATRIA POTE8TAS IN IBELAND. 125 
 
 which the organisation existed, nor to show its relation 
 to Patria Potestas. 
 
 The objections to his theory are overwhelming, and 
 there are several, each of which must be fatal to it. 
 
 1. Sir Henry is obliged to assume that the eldest 
 member of the geilfine division (whom he takes to have 
 been the father) never left that division. But this is 
 contrary to what is expressly stated which is that, the 
 geilfine division being full, on a new member being 
 born into it, the eldest member was promoted into the 
 deirbfine. 
 
 2. On Sir Henry's theory, the extinction of the 
 geilfine division would mean that the Paterfamilias 
 and all his family under power were dead that the 
 family was extinct. But the extinction of the geilfine 
 division did not put an end to this Irish organisation. 
 The three other divisions divided its property, and 
 the organisation went on. Moreover, as has been said, 
 an extinct geilfine division could be replaced. And 
 when another division besides it was extinct, it and that 
 other held to be replaced as a condition of the organisa- 
 tion not collapsing. But a Paterfamilias and his family 
 under power would have been irreplaceable. A Pater- 
 familias and his family under power, therefore, cannot 
 have constituted the geilfine division, and the organisa- 
 tion cannot have consisted of a Paterfamilias and his 
 descendants. 
 
 3. But further, there is no authority for conceiving 
 that all the members of the other divisions had 
 origin ally been in the geilfine division. There is only
 
 126 THE PATRIARCHAL THEORY. 
 
 one case mentioned in which men passed out of one 
 division into another. That was when, with the 
 organisation complete, a new, that is a sixth person, was 
 born into the geilfine division. Then the senior member 
 
 o 
 
 of each of the three first divisions was, as the editors 
 of the Senchus MOT say, promoted, and the senior 
 member of the oldest division had to go out. 
 
 Again, while the geilfine was undoubtedly the 
 youngest division, there seems to be no authority to 
 show (what is essential to Sir Henry Maine's explana- 
 tion) that it was the first in dignity, and that the others 
 diminished in dignity with their remoteness from it. * 
 If the right of inheriting divisional property be a gauge 
 of dignity, it was inferior to two of the other divisions, 
 the deirbfine and the iarfine divisions both faring 
 better than it. This, at any rate, was the unlucrative 
 result of its being the youngest division. When any 
 division became extinct by the death of its members, its 
 property went to the other divisions in definite shares ; 
 and the rule of sharing, which preferred the junior to 
 the senior, and the nearer junior or senior to the more 
 remote, operated most favourably for the two middle 
 divisions. 
 
 So far as to the incompatibility between Sir Henry 
 Maine's theory of the organisation and the facts dis- 
 closed respecting it. It remains briefly to point out 
 
 * See Ancient Laws of Ireland, Vol. IV., p. 207 : " After this 
 the indfine land goes before the deirbfine land, for, in tribe lands, 
 the indfine land is the land in which it (the water) is detained in a 
 pond ; for the land in which it is detained is nobler with the Feini 
 than the land out of which it is drawn."
 
 PATEIA POTESTAS IN IRELAND. 127 
 
 some discrepancies between his theory and Patria 
 Potestas. 
 
 1. Taking the case put in the Book of Aicill upon 
 which he has founded, it will be seen that for the father 
 whom he regards as a root from which might spring as 
 many of the organisations as he had sons for the 
 father who, having two sons with four children each, 
 might, according to the lawyers, set on foot two geilfine 
 divisions he can find no place within the organisa- 
 tions sprung from. him. His explanation, that is, shuts 
 the true Paterfamilias out of the family. Of the 
 violence herein involved to the principle of Patria 
 Potestas he has himself remarked that it has no 
 analogy in Roman Law. 
 
 In the organisations as they really were, however, 
 the " oldest ascendant " would, in the case supposed, 
 have been the most important person at any rate as 
 regards property ; for, in the opinion of the lawyers, he 
 would have been entitled to share with every family 
 derived from him, as if he were included in it. And 
 the words suggest that there would have been more 
 families than two. 
 
 2. Emancipation from Patria Potestas disinherited 
 the emancipated. But our knowledge of the divisions 
 of the Irish family is mainly derived from the laws of 
 succession to property which connected each of them 
 with all the others. 
 
 3. Emancipation could only account for there being 
 two classes of descendants, the emancipated and the 
 unemancipated. Here we have, according to Sir Henry
 
 128 THE PATRIARCHAL THEORY. 
 
 Maine, four classes of descendants. Of this also it may 
 be said that it has no analogy in Roman Law. 
 
 4. What power was it that formed a man's descen- 
 dants into so many classes, that forced the men to pass 
 at the proper time from one class into another, that 
 established those rights of succession which existed 
 between them ? Were emancipated men still subject 
 to their father's power ? Or did men when emancipated 
 come under some other power, which was to control 
 their father as well as them ? Sir Henry Maine has left 
 these and other such questions in obscurity. All he 
 tells us is that it seems to have been a self-acting 
 principle that regulated the transference of men from 
 class to class. Need it be said that nothing like this 
 was ever found annexed to the Roman institution ? 
 Patria Potestas was exhausted with emancipation. 
 
 It would seem that Sir Henry Maine's theory is not 
 consonant either with the facts or with Patria Potestas. 
 And it leaves nearly every circumstance connected with 
 this family organisation of the Irish in obscurity. 
 
 His second piece of evidence of Patria Potestas 
 having existed among the Irish may be briefly dis- 
 posed of. 
 
 In the "Cain" Law of Social Connexions* the 
 question is discussed, How many kinds of Social Con- 
 nexions are there among the Feini ? The answer is 
 that there are eight the connexion of the chief with 
 his "aigillne" tenants; that of the Church with her 
 tenants of ecclesiastical lands ; that of the father with 
 * Ancient Laws of Ireland, Vol. II., p. 343, et seq.
 
 P ATRIA POTESTA8 IN IRELAND. 129 
 
 his daughter ; that of a sister with her brother ; that 
 of a son with his mother ; that of a foster-son with his 
 foster-mother ; that of a tutor with his pupils ; and that 
 of a man with a woman. The connexion of a father 
 with his son is not one of the eight. It is stated that 
 in each of these relationships one of the parties has 
 over the other "judgment and proof and witness" 
 which the Editors of the Senchus Mor have amplified 
 into the "power of pronouncing judgment and proof 
 and witness" under certain conditions as to credi- 
 bility as regards women witnesses over the other. In 
 another passage * it is said that " the literary foster- 
 father [or tutor] has power of pronouncing judgment 
 and proof and witness upon the foster-pupil, as has the 
 father upon his son, and the Church upon her tenant of 
 ecclesiastical lands ; " and but for this there would be 
 nothing to show that a man had over his son the power, 
 if power it was, which, as we have seen, his mother, his 
 foster-mother, and his tutor (the latter a stranger to the 
 family) alike possessed in relation to him. 
 
 Sir Henry Maine has seen in this passage a dis- 
 closure of Patria Potestas. But that his mother and 
 other persons might possess this power, or whatever it 
 was, over a man, as well as his father, is of itself enough 
 to exclude this suggestion. Had this been a power 
 possessed by fathers only, however, why must it have 
 been Patria Potestas ? Summarily to identify any and 
 every power possessed by a father over his children, the 
 nature and limits of which are unknown to us, with the 
 * Ancient Laws of Ireland, Vol. II., p. 349.
 
 130 THE PATRIARCHAL THEORY. 
 
 Eoman institution can lead only to error and confusion. 
 It is not always that the error can be detected so com- 
 pletely as it can be in this case. 
 
 For, if the passage first quoted leaves the meaning 
 of the phrase "judgment and proof and witness" in 
 doubt, we get further light as to its meaning from the 
 Crith Gablach,* where, after the statement that there 
 are seven divisions of social grades, the question is put, 
 What is the division of social grades derived from ? 
 The answer is, that it is derived from the similitude of 
 ecclesiastical orders, and that " good corroborative proof, 
 or denial, or evidence, or judgment is due from each of 
 them to the other." Drop out the word corroborative, 
 which the translators have inserted, and we have here, 
 about all the seven social grades, a statement identical 
 with that upon which the inference of Patria Potestas 
 has been founded. And it appears that the phrase 
 "judgment and proof and witness" refers to nothing 
 other than legal process. 
 
 It has been pointed out that the words "power of 
 pronouncing," in the passages we have been noticing, do 
 not belong to the text. They have been imported into 
 it in an attempt to give it something like meaning. 
 But the Editors of the Senchus Mor have not omitted 
 to state that, whatever "judgment and proof and 
 witness" might mean, they did not, in their opinion, 
 mean Patria Potestas.f "The provisions of the Irish 
 Family Law," they say, "do not appear to have any 
 
 * Ancient Laws of Ireland, Vol. IV., p. 299. 
 t Ibid., Vol. II., p. iv. of preface.
 
 PATEIA POTESTAS IN IBELAND. 131 
 
 connection with the ancient Roman Law. The Irish 
 Law demands for the mother a position equal with the 
 father, and there is no trace of the exercise of that 
 arbitrary power which was wielded by a Roman father 
 over the members of his family." This is a statement 
 that there is in the ancient laws of Ireland not only no 
 trace of Patria Potestas, but proof of its absence. 
 Having that upon such authority, it seems unnecessary 
 here to adduce the multitude of facts by which it can 
 be supported. The student can easily test the state- 
 ment for himself. He will find it very fully borne out. 
 
 K 2
 
 CHAPTER X. 
 
 PATRIA POTESTAS CONCLUDED. 
 
 BEFORE going further, it will be well to recapitulate the 
 results arrived at in the five preceding chapters. And 
 
 1. Though Sir Henry Maine professes that his de- 
 scription of the Patriarchal Family is taken from the 
 Scriptures, we have found that nothing like the family 
 which he has described is to be discovered in the 
 Scriptures, and that the history of the Hebrews is not 
 to be reconciled with the Patriarchal Theory. After the 
 Hebrews became a nation the Hebrew father certainly 
 had not Patria Potestas, and the laws assured a place of 
 high honour to the mother. Going further back, it 
 appears certain that Jacob had not Patria Potestas over 
 his family ; and it can scarcely be doubted that the 
 position to which he attained by service in his uncle 
 Laban's household, was much the same as that of a 
 beenah husband in Singalese or African villages. At no 
 time do we find among the Hebrews, either in law or in 
 practice, anything suggesting that a father had over all 
 descended from him those despotic powers which Sir 
 Henry Maine has so often recounted to us. To show the
 
 P ATRIA POTE8TAS CONCLUDED. 133 
 
 absence of Patria Potestas was the utmost that imme- 
 diately concerned us, and that has been done. But 
 from the older Scriptures we have learned that to some 
 effects at any rate in some families of the Hebrew 
 stock the mother rather than the father seems to have 
 been at one time the representative of the family ; and 
 that she certainly could hold property since the bride's 
 price was paid not to her husband but to her and her 
 son. And we have found an explanation of such facts 
 as this, and of such marriages as Jacob's, in the first 
 Scriptural reference to marriage custom which, if not 
 misinterpreted by us, is by itself enough to prove that, 
 so far as the Hebrews are concerned, the Patriarchal 
 Theory is utterly mistaken. At any rate the Scriptures, 
 while they appear to disclose to us more than one 
 ancient form of the Family, certainly do not exhibit the 
 Roman form. 
 
 2. We have found among the early Hindoos rela- 
 tions disclosed between the members of a family, which 
 are at no single point consistent with the Patriarchal 
 Theory. The father had certain powers over his house- 
 hold, no doubt ; but these were defined by law, and not 
 despotic but greatly limited, and he was answerable to 
 the courts of justice for any abuse of them. The 
 mother had in the family a position of the highest 
 honour. Both she and her children could hold property 
 independently of the father ; and it can scarcely be 
 doubted that, not only were adult sons co-owners with 
 their father of the ancestral estate, but that, at any rate 
 in some parts of India, they could force him to make
 
 134 THE PATRIARCHAL THEORY. 
 
 partition of it. In the Hindoo family as it appears to us 
 through Manu, the father, though head of the household 
 and custodian of its common property, and, though no 
 doubt honoured and submitted to, was not a Pater- 
 familias. The family disclosed to us is of an ancient 
 type, which is not the Roman type. 
 
 No doubt the oldest of Hindoo law-books deal with 
 a social condition which is very far from primitive. The 
 same may be said of law-books in general. But it is to 
 these that Comparative Jurisprudence, as represented 
 by Sir Henry Maine, commonly makes its appeal. Some 
 reasons have been given for believing that, as regards 
 India, an advocate of the Patriarchal Theory would only 
 lose by carrying inquiry farther back.* 
 
 3. As regards the Slavs, there is so remarkable a 
 dearth of information about their condition in early 
 times that we have found ourselves reduced to thinking 
 with bewilderment of the confident way in which Sir 
 Henry Maine has professed to find in their " legal insti- 
 tutions " evidence for the Patriarchal Theory. We have 
 found that such information as we have, while disclosing 
 nothing as to the power of fathers among the early 
 Slavs, gives not a little evidence against the Patriarchal 
 Theory. We have found among the Slavs women in 
 the place of Queen, of Regent, of Ambassador. We 
 have found relationship on the mother's side acknow- 
 ledged; and in the very first article of their earliest 
 laws which, however, are not very early, and perhaps 
 not perfect evidence about them we have seen, in the 
 * See chapters xvi. and xvii. on " Sonship among the Hindoos."
 
 P ATRIA POTESTAS CONCLUDED. 135 
 
 provision for the revenging of homicides, an acknow- 
 ledgment of a sister's son as among the nearest kindred. 
 There is, then, to say the least, no evidence that the 
 family among the early Slavs was of the Eoman type. 
 
 4. We have found that neither the Joint Family of 
 India nor the House-Community of the South Slavo- 
 nians has, within the times known to us, either sprung 
 out of, or dissolved into, a family of the Eoman type. 
 In both, too, the organisation deemed necessary has 
 been unfavourable to paternal authority, and has in- 
 volved its being to a great degree superseded. 
 
 As to the Kussian peasant household of recent times, 
 which has often been regarded (without evidence) as 
 giving something like the type of the ancient family 
 among the Slavs, we have found that what was chiefly 
 remarkable about it was the contrast between the 
 powers of its head and the powers of a Paterfamilias. 
 The head of such a household, even when he was the 
 common progenitor, instead of being despotic master, 
 was regarded as manager for all the adult members of 
 the household ; was actually called manager ; and, in 
 every transaction affecting property, was required to 
 consult with his co -partners. 
 
 5. As to the indications of Patria Potestas which 
 Sir Henry Maine believes he has discovered in Ireland, 
 it can scarcely be deemed too much to say that they 
 have been obtained through a most strange misappre- 
 hension of his authorities. His theory of the curious 
 fourfold organisation of the family described in the 
 Irish law-books neglects some, and is at variance with
 
 136 THE PATRIARCHAL THEORY. 
 
 others, of the few facts disclosed about that organisation. 
 And it is as difficult to reconcile it with Patria Potestas 
 as with the Irish texts. The phrase, constantly recurring 
 in the law-books, which he has relied on as furnishing 
 corroborative evidence, has been shown to be con- 
 nected with legal process. But, were it not so, the fact 
 that the mother and various other people had the power 
 denoted by it, if a power, equally with the father, shows 
 that it could not prove Patria Potestas, but would rather 
 tend to prove the absence of it. 
 
 All the cases therefore in which Sir Henry Maine 
 has averred, or seemed to aver, that direct evidence of 
 Patria Potestas was forthcoming have been examined, 
 and the result is that there is no evidence of Patria 
 Potestas to be found in any one of them ; and that, 
 excepting among the Slavs (as to whom nothing perhaps 
 can be said that is not liable to challenge), there is on the 
 contrary in every case evidence of the absence of 
 Patria Potestas. What can indirectly be established, 
 as by the prevalence of Agnation, and an argument 
 founded thereon, of course remains to be considered. 
 But it is now plain that it is by means of indirect evi- 
 dence, if at all, that the prevalence of Patria Potestas 
 must be made out. So far as direct evidence goes, Patria 
 Potestas appears to be of distinctively Eoman growth. 
 
 "The proof of its former universality is incom- 
 plete so long as we consider it by itself," is what Sir 
 Henry Maine has said of Patria Potestas. So far as 
 legal evidence or historical evidence goes, the fact 
 appears to be that this institution is absolutely un-
 
 PATEIA POTE8TA8 CONCLUDED. 137 
 
 paralleled, and that there were at the utmost only 
 three or four cases in which its existence could even be 
 alleged. Sir Henry's statement is the more curious, 
 seeing that he had not overlooked what Gaius has said 
 of Patria Potestas. 
 
 What Gaius has said seems well worth pondering 
 in an inquiry as to the diffusion of Patria Potestas. 
 Here it is briefly, in Sir Henry Maine's words : " Gaius," 
 he says, " describes the institution [Patria Potestas] as 
 distinctively Roman Among the races under- 
 stood to be comprised within the Eoman Empire, Gaius 
 could find none which exhibited an institution resem- 
 bling the Roman ' Power of the Father,' except only the 
 Asiatic Galatae." * 
 
 Sir Henry Maine, be it said, does not question the 
 evidence of Gaius. He has taken it as showing that 
 the institution which, among the Romans, proved so 
 durable was in its nature not durable, and so had 
 disappeared; and thus has almost seemed to make out 
 of its non-appearance anywhere a proof of its former 
 universal prevalence. 
 
 But the opinion of Gaius was that Patria Potestas 
 was distinctively Roman ; and his opinion is the 
 weightiest ancient opinion obtainable. He was one of 
 the few Roman lawyers who were deeply interested in 
 the antiquities of law, and given to tracing institutions 
 to their origins. If there was any one living in the time 
 of the Antonines likely to study carefully the laws and 
 institutions of the numerous peoples whom, by that 
 * Ancient Law, pp. 135-6.
 
 138 THE PATRIARCHAL THEORY. 
 
 time, Roman conquest had brought within the scope of 
 investigation, Gaius was the man. His statement, of 
 course, involves that he had made a study of them. 
 His opinion that Patria Potestas was distinctively 
 Roman was formed upon the examination he had made 
 of them. The validity of his evidence conceded, the 
 reasons for his opinion are obvious, and they might well 
 be deemed conclusive. With a vast area open to 
 observation, containing races in very various stages of 
 advancement, many of whom had been more or less 
 under observation for centuries,* Gaius had found Patria 
 Potestas nowhere, and had found only in one instance 
 an institution which even resembled it. And Patria 
 Potestas had in Rome proved a durable institution, 
 and, indeed, singularly tenacious of life. 
 
 A modern writer has, in some cases, facts at his 
 command not accessible to the ancients, which may 
 exclude what, apart from them, was a just conclusion. 
 But it has appeared that Sir Henry Maine is, equally 
 with Gaius, without examples of Patria Potestas. And 
 his views about it are only possible to one who deems 
 such new facts as are available to modern writers un- 
 worthy of notice. 
 
 * There can be no doubt that Gaius was aware of what Caesar 
 (lib. 6, c. 19) had said of the Patria Potestas of the Gauls, and had 
 rejected it as untrustworthy. Possibly he had more trustworthy 
 information. But is it too much to pronounce it incredible that, with 
 a system under which husbands and wives held their goods in 
 common, the survivor of them having the right to the whole of the 
 accumulated capital and interest, the husband should have the right 
 to put the wife to death whenever he pleased 1
 
 PATRIA POTESTAS CONCLUDED. 139 
 
 To show the full weight of the statement of Gains, 
 let us recall what was the area open to his observation 
 that is, the extent of the Koman Empire under the 
 Antonines and Marcus Aurelius. It included the whole 
 of Southern and Western Europe (including Britain) 
 as far as the Rhine and the Danube, with considerable 
 territories beyond the latter river ; Africa, to the Great 
 Desert and the Cataracts of the Nile ; Asia Minor and 
 Syria as far as the Euphrates. And even beyond 
 these limits there were countries, like Armenia and 
 Mesopotamia, which the Romans knew well, and 
 over which they exercised considerable authority and 
 patronage. 
 
 That a writer like Gaius, whom there is no reason 
 to suspect of carelessness in investigating facts, should 
 have found no Patria Potestas within this vast range, 
 is a fact most impressive when the question is raised 
 whether Patria Potestas was a primeval and a universally 
 prevalent institution. 
 
 Sir Henry Maine, while casting no doubt on the 
 statement of Gaius, and not dwelling on it further than 
 to point out to how great an extent Patria Potestas 
 had disappeared, remarks that the diffusion of Roman 
 citizenship among the races subject to Rome, which 
 began with the Empire, and was completed under 
 Antoninus Caracalla, " must have enormously enlarged 
 the sphere of the Patria Potestas." He makes no 
 attempt to show that such enlargement took place in 
 fact that the Roman institution did become naturalised 
 among the various races contained within the Empire.
 
 HO THE PATRIARCHAL THEORY. 
 
 And, of course, that, had he shown it, could not have 
 helped his main argument. His object was to exhibit 
 the singular fortune of this institution which, uni- 
 versal, ex hypothesi, at the beginning, but difficult to 
 find because, ex hypothesi, not durable in its nature, 
 was at last re-diffused from a people among whom it 
 had proved singularly durable, and re-established so 
 far as an enactment designed to confer a privilege 
 could establish it throughout the Koman world. He 
 does not seem to have noticed that he was raising up 
 a crop of difficulties for himself in dealing with the 
 races not at the time referred to within the Empire, 
 had instances of Patria Potestas occurred among them. 
 If so much diffusion were possible, could such instances 
 be assumed to have been of native origin ?
 
 CHAPTER XT. 
 
 DERIVATIVE INSTITUTIONS. THE EVIDENCE AS TO THEIR 
 ORIGIN WHICH MAY SUFFICE. 
 
 IN Ancient Law Sir Henry Maine relies for his proof 
 of the Patriarchal Theory chiefly on Agnation, which, 
 he declares, can be found (or traces of it) almost 
 everywhere, and which, he says, " implies the former 
 existence" of Patria Potestas wherever it is found. In a 
 more recent production* he has somewhat qualified his 
 original statement, according to which, wherever we 
 find Agnation the paternal powers must have formerly 
 existed. His later statement is that " where there is 
 Agnation there must almost certainly have been paternal 
 power." This change of terms shows a disposition to 
 avoid appearing to overpress the point, but what 
 change of view it involves is not altogether clear. It 
 may mean that, at the date of the later writing, Sir 
 Henry was not sure but only almost sure that there 
 was between the two the nexus which he had supposed 
 to exist between them ; or it may be an admission that 
 perhaps* Agnation could come into existence indepen- 
 
 * "South Slavonians and Rajpoots," Nineteenth Century, December, 
 1877. Early Law and Custom, p. 244.
 
 142 THE PATRIARCHAL THEORY. 
 
 dently of Patria Potestas. In case the latter be the 
 true meaning, it will be well to point out that, except in 
 the unqualified form, his proposition is of no value. 
 What shall determine our judgment, if we find Agnation 
 anywhere, as to whether it was preceded by Patria 
 Potestas or not, if it was not in the nature of things 
 necessarily preceded thereby ? 
 
 But why should Agnation either imply, or almost 
 certainly imply, the pre-existence of Patria Potestas ? 
 Certain reasons for this are assigned in Ancient Law, 
 and these will be examined by-and-by. Let us now, 
 however, consider, as briefly as may be beginning with 
 some of those cases which are simplest or easiest on 
 what grounds, on what kind and amount of evidence, 
 the belief that one institution or practice is derived from 
 another can reasonably be held with so much confidence 
 that we may venture to say the one implies the pre- 
 existence of the other. 
 
 The feu system of Scotland owes its existence to 
 accident, the statute Quia Emptores, designed to prevent 
 sub-infeudation, and which did prevent it in England, 
 having become inoperative in the northern part of our 
 island. It has flourished, and continues to flourish now 
 that feudalism has virtually disappeared, on account of 
 the permanency of the interests which it has established 
 the feus being perpetuities connected with the land, 
 often carrying the right to costly buildings and 
 through its convenience, and especially the superiority 
 of the feu as a building tenure over the lease. The 
 name and the terms employed, as well as the nature of
 
 DERIVATIVE INSTITUTIONS. 143 
 
 the tenure, still show the origin of the feu ; and though 
 it may already be regarded as an independent institu- 
 tion, it is, in fact, an example of the feudal system in 
 operation, which has been fixed and retained notwith- 
 standing that the system has passed away. 
 
 It is probable that the feu system will long cling to 
 the smaller perpetual holdings of land in Scotland. 
 Were it then to last, just as it is, till memory of the 
 feudal system having been established in Great Britain 
 has passed away, might it be confidently recognised 
 as an application of the feudal system, and the pre- 
 existence of the feudal system be inferred from it ? It 
 might, on condition that full accounts of the feudal 
 system remained ; that it was known to have been rather 
 widely diffused in Europe ; and, a fortiori, if it was 
 known to have been established among populations 
 likely to have influenced inhabitants of Britain ; and, 
 of course (though in such a case this need scarcely 
 be mentioned), that no preferable explanation was 
 forthcoming. For, in the first place, the feu, in the 
 case supposed, would not merely present some semblance 
 of being a feudal holding it would be feudal all over, 
 feudal in every incident. The chance of there having 
 been coincidence, the chance of so exact a resemblance 
 in all of so many particulars (including highly technical 
 terms) having been produced by other and unknown 
 influences the risk, that is, of mistake in accepting the 
 suggestion of resemblance, which in such cases is what 
 is to be feared, and what must be shown to be small 
 would be absolutely inappreciable.
 
 144 THE PATRIARCHAL THEORY. 
 
 Even if the known diffusion of feudalism left it- 
 improbable that feudalism had made its way into 
 Britain, there would be, in the case put, no room for 
 reasonable doubt as to the feudal origin of the feu. 
 The less improbable the state of knowledge made that, 
 however, the more confidently could this conclusion 
 be accepted. 
 
 And, a feudal origin once admitted, it would not 
 be difficult to judge between the two views between 
 which choice would then lie. Either feudalism had 
 been generally established, and a convenient application 
 of it had been preserved ; or a single application of it 
 had been somehow, at some time, introduced. Pro- 
 bability decides in such matters, and probability would 
 be immensely in favour of the former view. Again 
 there might be confident conclusion. It is only when 
 the intercourse of nations has become very free, and the 
 development of intelligence leads to the comparison of 
 institutions and seldom then that a people takes 
 what suits it of a foreign system and leaves the rest. 
 And what is likely to be taken in such a case is only 
 the idea, enough to give the use which is desired ; not 
 the inconvenient details and the unfamiliar foreign 
 terms. What we are always forced to reckon upon, 
 therefore, is that people have applied the systems they 
 have had and have grown familiar with. But, in the 
 case supposed, apart from this consideration, the appli- 
 cation would seem almost necessarily to imply the 
 establishment of the feudal system. It would bear 
 every appearance of being related to a system of which
 
 DERIVATIVE INSTITUTIONS. 145 
 
 it was an application. And it would seem to the last 
 degree unlikely that the correctness, to the minutest 
 detail, which it would exhibit could be found without 
 the system having been established and applications of 
 it made familiar by daily use. 
 
 With such perfect resemblance, then, as has, in the 
 case just put, been supposed between an existing insti- 
 tution or practice and the operation of an ancient 
 system which has been fully described, and is known 
 to have been considerably diffused in the same region in 
 which the former is found, we could not hesitate to 
 infer that the latter preceded the former in the par- 
 ticular place in which we find it. And we could not 
 infer less than that the ancient system had been long 
 enough and generally enough established in that place 
 for it to have the power of impressing general practice, 
 and so leaving an application of itself to future gene- 
 rations. It may be said, indeed, once for all that 
 whenever we come upon a practice in general use which 
 can be taken to be an application of an ancient system, 
 less than this cannot be inferred as to the former pre- 
 valence of that system. In practice, we have in general 
 but by no means always less perfect resemblance to 
 guide us than there would be in the case supposed. 
 That case shows, however, in a strong way and it is 
 chiefly for this reason it has been presented the great 
 value of resemblance, resemblance not in externals only, 
 but in idea or purpose, methods, and effects, as a means 
 of enabling us to connect a derivative with that from 
 which it is derived. And with less than perfect resem-
 
 146 THE PATRIARCHAL THEORY. 
 
 blance it being possible to exclude to a reasonable 
 extent the chance of mistake the indication of re- 
 semblance may often be distinct and strong enough to 
 carry us to a highly probable conclusion. 
 
 On the other hand, there are numerous cases in 
 which, the indication being not quite sufficient as re- 
 gards either distinctness or quantity, unless corroborative 
 evidence of some weight be obtainable, the notion that 
 an institution or practice has been derived from another 
 can only be at most a more or less amusing speculation. 
 
 Suppose, for example, as before, that at some future 
 time it should be unknown that the feudal system was 
 once established in Britain, and that there should then 
 survive, not the feu system as it is now, but some 
 remainder of it the system, that is, docked more or 
 less of the circumstances by which it could be identified, 
 and perhaps in some respects altered. It would then 
 depend upon the significance, and in some degree upon 
 the amount, of those original characteristics which 
 survived somewhat, too, upon the extent to which 
 methods or details had been modified or changed 
 whether or not there could be a hope of assigning the 
 remainder with any positiveness to its origin, and then 
 inferring the pre-existence of the feudal system in 
 Britain. It might easily happen that there would be 
 nothing left by which its feudal origin could be recog- 
 nised. To give any chance of that being recognised 
 there would have to be remaining something distinc- 
 tively feudal, something for which no other origin 
 could with equal probability be imagined. Even then
 
 DERIVATIVE INSTITUTIONS. 147 
 
 there might not be enough to yield more than a 
 surmise. 
 
 If, however, there remained enough of feudal 
 characteristic to give distinct suggestion of a feudal 
 origin, and there had been failure to account reasonably 
 for this otherwise, it would forthwith be at least a 
 plausible opinion more or less plausible according to 
 the circumstances that what we had come upon was a 
 remainder of feudalism. The more points there were at 
 which it gave feudal suggestion, the less would be the 
 likelihood of coincidence, the less the risk of mistake 
 in accepting that suggestion ; and the approach to a 
 positive conclusion might be correspondingly nearer. 
 (There are cases, however but the case supposed could 
 hardly be among them in which evidence got at a 
 single point may be absolutely convincing.) If now 
 there could be found other practices or institutions 
 which seemed also to give indication of feudal origin, 
 the accumulation of these indications with those 
 previously possessed might found a fairly probable 
 inference that each and all of the institutions or 
 practices dealt with had had a feudal origin. All this, 
 of course, upon condition that full accounts of the 
 feudal system remained, and that it was known to have 
 been considerably diffused. If the state of knowledge 
 made it probable, or left people free to think, that it 
 had been established in Britain if, for example, it had 
 been established in a neighbouring country, or among a 
 population likely to have influenced British institutions 
 an inference of fair probability might be arrived at 
 
 L 2
 
 148 THE PATEIAEGHAL THEORY. 
 
 the more readily. Where feudalism was known to have 
 been established, such an inference could be got more 
 easily still less would suffice for it. And the argu- 
 ment in another case might be greatly helped by 
 showing that, where feudalism had been established, 
 there appeared indications of its influence like those the 
 origin of which was being considered. 
 
 To pass now to a new case. The Common Law of 
 Scotland did not allow its ordinary rule applicable to 
 the property of spouses when one of them died and 
 there were no children which treated the personal 
 property of both as, in virtue of the marriage, their 
 joint property, and so divided it equally between the 
 surviving spouse and the relatives of the deceased to 
 take effect unless either there had been a child born 
 alive or the marriage had subsisted for a year and a 
 day. Failing the subsistence of the marriage for a 
 year and a day, and the birth of a child which lived 
 (though it did not survive) and, of course also, failing 
 the birth of a child of the marriage after the husband's 
 death the law severed the interests which on marriage 
 had become united ; so that simply what the surviving 
 spouse had brought into the common fund went to him 
 or her, while what had been brought by the deceased 
 went to his or her relatives. Now, in the old Celtic 
 marriage by handfasting it was agreed that a man 
 and woman should cohabit with a view to marriage 
 
 o 
 
 for a year and a day ; and they were husband and wife 
 without further ceremony within that time if a child 
 were born, or at the end of it, if the mother proved
 
 DERIVATIVE INSTITUTIONS. 149 
 
 then to be with child. Otherwise, they were then free 
 to leave one another and each to handfast with any 
 other person ; their interests, so far as they had become 
 common, being of course severed in the event of their 
 parting, and also in the event of one of them dying 
 within the period of contract while there was no child.* 
 It appears then that in marriage, when one of the 
 spouses died within what had been the handfasting 
 term, the Scots Law applied its ordinary rule only 
 when by custom handfasting would have turned into 
 marriage ; and that it did not apply it, but ordained a 
 severance of interests which would in the same cir- 
 cumstances have followed upon a handfasting where 
 handfasting would not have turned into marriage. 
 Handfastings which outlived the year and day (which, 
 except when one of the spouses died, must have 
 commonly happened) were marriages ; and within the 
 year and day, in a certain event, handfastings became 
 marriages. And in the cases which corresponded to 
 these last, and in all cases in which marriage outlived 
 the year and day the Scots Law applied its theory of 
 marriage, which gave each spouse an interest in the 
 property of the other; while it ordained that every- 
 thing should be as if there had been no marriage in 
 those cases in which there would have been no marriage 
 with handfasting. Marriage within the year and day 
 and Marriage not within the year and day were separate 
 
 * The accuracy of this statement of the contract may possibly be 
 questioned, but, even if it be, that is immaterial as regards the use 
 which is now to be made of it.
 
 150 THE PATRIARCHAL THEORY. 
 
 headings with the old text writers. And when within 
 the year and day handfasting would not have become 
 marriage, even when there had been a settlement, the 
 ordinary rule for marriage did not apply that is, the 
 settlement was of no effect except when there had 
 been a stipulation to the contrary. 
 
 Upon what conditions, now, in the case stated, 
 might we venture to conclude that the reason why 
 marriage did not produce the full effects of marriage 
 until it had lasted for a year and a day was that 
 anciently, during that period, marriage was commonly 
 inchoate ? Had handfasting, as above described, been 
 a prevalent practice, and been replaced by immediate 
 marriage, the ordinary consequences of handfasting 
 being nevertheless retained in the law, the law would 
 have been just what it was. That is plain enough. And, 
 on the other hand, no other explanation is known of 
 those curious provisions ; and it will be found extremely 
 difficult to suggest one that will bear examination. 
 Policy, indeed, might account for the year and day rule; 
 but can it also account for the exception to that rule ? 
 Clearly not in a satisfactory way. And if it were 
 known that the custom described commonly prevailed in 
 connection with marriage among the early Scots, and 
 the question were whether the provisions of the law 
 were derived from it, there could be no reasonable doubt 
 as to the answer. 
 
 Supposing that not known, however, how far would 
 the coincidence between the law and the operation of 
 the custom carry us ? A custom of which accounts
 
 DERIVATIVE INSTITUTIONS. 151 
 
 have come down to us prevailed of course to some 
 extent. Might it be inferred forthwith, from the exact 
 correspondence between the law and the operation of 
 the custom, that the latter must have at any rate pre- 
 vailed sufficiently to leave its mark upon the law ? 
 Most impressive as that correspondence is, so much 
 could not be maintained. The custom gives us, no 
 doubt, a perfect hypothesis to account for the law. 
 There is identity between the provisions of the law and 
 the operation of the custom. But there is in this case 
 (so to speak) resemblance only in effects ; and, perfect 
 as it is, that cannot carry us so far as a resemblance 
 which extends to idea and methods as well as to effects. 
 Besides, coincidence has always to be reckoned with and 
 guarded against. Before making such an inference, 
 therefore, we should have to assure ourselves that the 
 custom had had extension enough to make it not un- 
 likely that it should affect the law of marriage. Having 
 shown so much, we might trust to the evidence of 
 resemblance for the rest. The custom being peculiar to 
 the country where it occurred, more evidence would be 
 needed as to its extension there than would be needed if 
 it were more diffused. On the mere knowledge that 
 there had been a custom which would have operated as 
 has been described, however, it would be a very plausible 
 opinion an opinion clearly preferable to every other, 
 and extremely likely to be confirmed by inquiry if 
 evidence were obtainable that the provisions of the 
 law were derived from that custom, and therefore, that 
 it had been pretty generally prevalent. And unless the
 
 152 THE PATRIARCHAL THEORY. 
 
 evidence for the custom were wanting', not in volume 
 only but in weight, it would be more. It is the weight 
 of the evidence remaining as to the extension of the 
 custom, which must determine the degree of plausibility 
 or probability which would attach to such a conclusion. 
 It might be held with something like confidence 
 upon evidence a long way short of what would be 
 needed to show that the custom had been generally 
 established. 
 
 And, in considering evidence in such a matter, it 
 would be unreasonable to dwell upon the absence of 
 that which, from the nature of the case, could not be 
 forthcoming to shrink from a conclusion merely be- 
 cause only such evidence as happened to be recorded 
 had come down. The weight of what remains is what 
 would have to be looked at. Where the bulk of a 
 people do not read or write, contracts will not often be 
 made in writing; where there are no historians, customs 
 will die out without having been described ; and even 
 when a rude people has its annalist, the customs of the 
 people familiar to him, though matter of curious 
 inquiry to us are about the last things he intentionally 
 writes about. Nevertheless, we may have access to 
 evidence which may be weighty and even conclusive. 
 If a contract be in question, we may do pretty well if 
 there have come down to us some of the written contracts 
 made by the better sort of people especially if it be a 
 contract connected with marriage. We shall not do 
 rashly in inferring that the contracts made between 
 chiefs in connection with marriage had popular sanction
 
 DERIVATIVE INSTITUTIONS. 153 
 
 and it is the contracts made by people of condition 
 that are most likely to determine the subsequent law 
 or even in inferring that, though they may have varied 
 from, they coincided in the main with common usage. 
 
 It is through contracts made between chiefly families, 
 it may here be said, that we know of handfasting in 
 the Highlands where such contracts were made up to 
 rather recent times ; and if there be some risk of error 
 in thinking that the chiefs did not invent this singular 
 contract for their own exclusive use, and that they did 
 not run counter to usage in entering into it, there seems 
 to be no more than must be ventured whenever we 
 form conclusions on probable reasoning. That a contract 
 which, if not identical with, was barely distinguishable 
 from, that of handfasting, was the common portal to 
 marriage in some southern parts of Scotland a grievous 
 scandal in the eyes of the Eeformed Clergy till some 
 time after the Reformation, tends strongly to support 
 the view that handfasting, or something practically the 
 same, was anciently popular usage ; and would, of 
 course, be a weighty fact in any argument on the con- 
 nection between it and the Scots marriage law. 
 
 What we have now deduced is that where a custom 
 found in a single country has been so fully described to 
 us that we can understand how it must have operated, 
 and we find later in that country, in connection with 
 the same matter, exceptional arrangements, also peculiar 
 to the country, identical in operation, it is at once a 
 very plausible view likely upon inquiry to prove 
 preferable to any other that the latter have been
 
 154 THE PATRIARCHAL THEORY. 
 
 derived from the former ; that this may be taken as 
 made out upon evidence sufficient to show that the one 
 had a certain chance of affecting the other, and, there- 
 fore, upon evidence a good deal short of what would be 
 required to show that the custom was once generally 
 prevalent ; but that the degree of extension that can be 
 made out for the custom must determine whether a con- 
 clusion to this effect may be made positively or not 
 whether, that is, we can have a reasonable conclusion or 
 only a very plausible conjecture. In this case, never- 
 theless, as in the previous case, we see that, as a test 
 of connection, perfect resemblance between the things 
 compared is of very great value. In this case, sup- 
 posing the resemblance less than perfect, even if we 
 knew that the ancient custom had been generally 
 established though knowledge of that would carry us 
 over trifling differences there would always be room 
 for question as to whether the provisions of the law 
 were really derived from it. 
 
 Let us assume, now, the evidence for handfasting to 
 be thought strong enough for a confident conclusion 
 that the provisions of the Scots marriage law were a 
 remainder of it. Could we then, were we to find 
 marriage denied its effects in the same or similar circum- 
 stances in another country, venture to say with confi- 
 dence that handfasting, or something similar, must have 
 formerly prevailed there also ? Could we say : there 
 may have been handfasting where there is no such 
 denial to marriage of its effects, but where there is 
 such denial there must have been, or almost certainly
 
 DERIVATIVE INSTITUTIONS. 155 
 
 has been, handfasting ? Surely not. That view would 
 be pleadable ; it would be promising ; one could not be 
 sure but that it was right ; we should look hopefully for 
 means of verifying it ; and even without these, many 
 people might think it very plausible. But on belief in 
 the connection of the two things founded on a single 
 instance the custom being known to us only in that 
 instance to be confident about it would be absurd. 
 The custom would furnish a hypothesis for the new case 
 than which none could be better. But even a perfect 
 hypothesis cannot of itself carry us so far. And, on 
 the other hand, the second case might be such as to 
 shake our faith in the conclusion to which we had been 
 brought in the first. 
 
 It would be different if it could be made probable 
 that handfasting had left the remainder in a number 
 of instances. On finding just the same thing in a new 
 place and no other explanation for it then, with so 
 peculiar a provision, it might, even on a small number 
 of instances, be fairly probable that that, too, was a 
 remainder of handfastiug. If, in addition, it were 
 known historically that handfasting had been a common 
 custom, we might see our way to coming to a positive 
 conclusion to that effect. Given, for example, hand- 
 fasting known to have been so common that it would 
 be no surprise to find it in any new place or that, 
 from our knowing it to have prevailed somewhere near, 
 or among a kindred population, it would be no surprise 
 to find it in a particular new place in either case, on 
 finding what in Scotland and some other places had
 
 156 THE PATRIARCHAL THEORY. 
 
 been, for sufficient reason, taken to be a remainder of 
 it, we might reasonably infer a former prevalence of 
 handfasting. The more the cases in which it could be 
 believed to have left the remainder, the more positive 
 in the circumstances supposed would be our conclusion. 
 But that conclusion might, in such circumstances, be 
 reasonable, when the remainder had previously been 
 connected with the custom only in a single instance. 
 
 Let us next taking a case which has been matter 
 of controversy try to see on what conditions it may 
 be legitimate to regard the Levirate as a relic of 
 Tibetan Polyandry, from which it may be inferred 
 that Tibetan Polyandry formerly prevailed wherever 
 it is found.* 
 
 The Levirate (which we must be careful to distin- 
 guish from a very different and much less surprising 
 thing the custom whereby a brother succeeded to his 
 elder brother's estate and to his wife as forming part of 
 it) occurs only where monandry, separate property 
 rights of individuals, and the succession of sons to 
 fathers have been established the last so firmly that 
 even a fictitious son is preferred to a real brother and 
 where in general (in the great cases the Hindoo and 
 Hebrew certainly) marriage with a deceased brother's 
 wife is by rule forbidden. And the regular operation of 
 it was to impose upon a man whose brother (in the 
 
 * See " The Levirate and Polyandry," by J. F. M'Lennan, and 
 " A Short Eejoinder by Herbert Spencer," Fortnightly Revieio, 1877. 
 See also chapter iv. of Hindu Law and Usage, by J. D. Mayne. 
 Madras and London, 1878.
 
 DERIVATIVE INSTITUTIONS. 157 
 
 Jewish case, whose elder brother) had died, leaving a 
 widow and no children, the obligation to take the 
 widow and " raise up seed " to that brother. The son 
 begotten by the surviving brother upon the widow was 
 accounted his deceased brother's son, and as such 
 cutting out the Levir, his actual father he succeeded 
 to that brother's estate.* Adoption is an expedient a 
 " Patriarchal Family" might have resorted to, as modern 
 Hindoo families do, for the purpose in view, and it is an 
 expedient not inconsistent with the conditions and moral 
 sentiments proper to monandry. But what to ancient 
 Hindoos and ancient Hebrews seemed right was, that 
 the brother of the deceased should "perform the duty of 
 an husband's brother unto " the widow an expedient of 
 a kind repugnant to monandry and shocking to the 
 moral sentiments that grow round it. 
 
 Among the Hebrews the widow was at first her 
 brother-in-law's wife without any ceremony, and she 
 had a right to him as a husband ; afterwards no doubt 
 under the influence of ideas of propriety derived from 
 monandry a formal marriage became indispensable. 
 Among the Hindoos it was thought proper that the 
 intercourse should come to an end when one or at most 
 two sons had been born, and there never was a marriage. 
 
 * Among the Hindoos, and in early times (as the case of Judah 
 and his daughter-in-law shows) among the Hebrews also, failing a 
 brother, a near kinsman could act the part of a brother by the widow ; 
 but the Levirate has to be regarded as essentially an institution which 
 imposed a duty upon brothers. The brother was the proper person 
 to be Levir.
 
 158 THE PATRIARCHAL THEORY. 
 
 The Brahmin law-writers, though forced to tolerate the 
 Levirate, because they found it an ancient and firmly 
 rooted custom, loathed it as a practice " only fit for 
 cattle;" and, just as the Jews imposed a marriage 
 ceremony, they prescribe restrictions and regulations 
 (we do not know to what extent they were generally 
 submitted to) designed to mitigate the scandal of the 
 practice restrictions and regulations which, being 
 avowedly designed for that purpose, could not, even 
 had we not the Jewish case to enlighten us, be supposed 
 to have accompanied it at first, when it can have caused 
 no scandal. The feelings with which they regarded it 
 the feelings which a practice of monandry must produce 
 when they became general, put an end to it among 
 both Hebrews and Hindoos. 
 
 We have here, then, along with monandry, a practice 
 utterly antipathetic to it, which it ultimately destroyed, 
 and which therefore it may safely be said it could not of 
 itself have given birth to the essential things in it, 
 as ordinarily practised, being that, when a man died 
 leaving a childless widow, it was his brother's duty to 
 treat her as his wife, and that the brother's son by her 
 was accounted the dead man's son in such good faith 
 that he succeeded to the dead man's property. If it 
 must have originated independently of monandry, it is 
 in some custom which prevailed before monandry be- 
 came a general practice that we shall do well to look 
 for its origin. We are forced, or almost forced, to think 
 that among Hindoos and Hebrews there were other and 
 grosser marriage customs before monandry ; and, once
 
 DERIVATIVE INSTITUTIONS. 159 
 
 brought so far, there is no difficulty in thinking of the 
 custom which could have supplied a basis for it. With 
 Tibetan Polyandry, all the brothers of a family marry 
 one woman ; and the children of the brotherhood are 
 regarded as children by all the brothers, while they are 
 specially considered as the children of the eldest brother 
 perhaps in some degree because he may be the first 
 to marry (his younger brothers becoming partners with 
 him in marriage as they grow up as a matter of course), 
 but chiefly because he is the head or administrator of 
 the family. 
 
 Here we have a marriage system in which a man 
 as he grows up becomes by custom husband of his 
 brother's wife, and in which the children, by whom- 
 soever begotten, are regarded as children by all the 
 brothers, while they are specially the children of the 
 eldest and therefore a system which exhibits both the 
 conception of marriage and the conception of fatherhood 
 that appear with the Levirate. Repugnant to monandry, 
 the Levirate, in its ordinary working, was, as far as it 
 went, identical with Tibetan polyandry. Shocking to 
 the feelings of propriety engendered by monandry, it 
 exhibits without a shade of more or less the grossness of 
 Tibetan polyandry. If we are forced to think that 
 the Levirate had its origin in some prevalent marriage 
 custom of grosser mould that flourished anterior to 
 monandry, or along with it while it was yet not 
 generally established, Tibetan polyandry stands out 
 among marriage customs as the one which could 
 have left to monandrous generations, which deemed it
 
 160 THE PATRIARCHAL THEORY. 
 
 important that fathers should have sons to succeed 
 them, such a remainder as it is. 
 
 Tibetan polyandry is known, too, to have been 
 widely diffused throughout the world ; and that a relic 
 of it should long survive among a people who had 
 practised it before and along with monandry cannot be 
 thought surprising. Among a people who had practised 
 it and monandry along with it as occasion served 
 among whom monandry grew up alongside of it the 
 ideas and practices connected with it, even when 
 monandry had become general, could scarcely disappear 
 all at once. They would at any rate disappear in such 
 a case as polyandry itself would disappear gradually 
 and slowly. Would it be strange that, sons being desired 
 for heirs, polyandrous practice should survive for a con- 
 tingency in which the view of fatherhood associated 
 with it could be turned to account, and that it should, 
 monandry notwithstanding, receive the sanction of 
 custom ? If not, it is not surprising that Tibetan 
 polyandry should leave behind it a remainder in the 
 Levirate. Where it did so, from the nature of the case, 
 there might remain little besides to attest its former 
 prevalence. 
 
 In this case we have not, as in our initial sup- 
 position about the Scots feu system we had, an insti- 
 tution, otherwise extinct, surviving complete just as 
 it was originally in one application of it. But what 
 we have is not much less ; for the same view of mar- 
 riage, the same view of fatherhood and sonship and 
 these are all the elements in either appears in both the
 
 DERIVATIVE INSTITUTIONS. 161 
 
 Levirate and Tibetan polyandry. And Tibetan poly- 
 andry, as has been said already, is known to have been 
 widely diffused. In a polyandrous solution there is 
 nothing surprising. There cannot, in this case, be such 
 confident recognition that what we are trying to 
 account for is an example of that to which we assign its 
 origin as there might be in the former case. But there 
 can be confident recognition of the substantial identity 
 of the two. And, the Levirate being what it is, this is 
 enough to carry us far. It becomes at once very much 
 more than a plausible opinion that it was derived from 
 Tibetan polyandry. For the extreme difficulty (it 
 may almost be said the impossibility*) of suggesting for 
 it a monandrous solution that will bear thinking about 
 weighs heavily in favour of the polyandrous solution. 
 Were it admitted that polyandry either preceded or 
 went on side by side with monandry among either 
 Hindoos or Hebrews, there could be no reasonable doubt 
 that the Levirate was derived from it ; and not only in 
 the particular case. It being matter of common know- 
 ledge that polyandry has been considerably diffused, 
 and the Levirate being the peculiar institution it is, one 
 could not hesitate to believe that the one had been 
 derived from the other in every case. 
 
 Since, however, the question is, On what evidence 
 of the extension of the polyandry can we move on to 
 the conclusion that the Levirate was derived from it ? 
 the first thing to be considered is what it is that has to 
 
 * See chapters xvi. and xvii. on " Sonship among the Hindoos." 
 
 M
 
 162 THE PATRIARCHAL THEORY. 
 
 be done ? And what has to be done is to remove such 
 doubt as there can be and the difficulty of finding a 
 monandrous solution taken into account, it cannot be of 
 serious amount that the resemblance, or rather the 
 identity, perceived between the polyandry and the 
 Levirate is only some strange effect of coincidence 
 produced by influences unknown to us, and which we 
 cannot even imagine. That done, the Levirate itself 
 will do the rest. For its evidence as to the former 
 prevalence of polyandry once we feel sure it can be 
 relied upon is conclusive. It has, at most, therefore, 
 to be made probable that Tibetan polyandry has had 
 a certain chance of influencing marriage custom where 
 the Levirate has been found. To show, from the known 
 diffusion of polyandry, that it is not unlikely to have 
 preceded where the Levirate is found might almost be 
 enough : to show, for some leading example of the 
 Levirate, that there is some good evidence that poly- 
 andry preceded to show, for example, that there are 
 striking instances recorded which prove that it was 
 practised in early times, and, being then unusual, was 
 then justified as ancient custom ; and that there are, 
 besides, laws or usages (and one might suffice) which 
 seem to be, as much as the Levirate itself, of poly- 
 androus origin would, it would seem, be amply enough, 
 enough to justify a decided conclusion. 
 
 As has been pointed out in considering another 
 matter, it is idle in such a case to dwell upon the 
 absence of evidence which, from the nature of the case, 
 could not be forthcoming. And it is the weight of such
 
 DERIVATIVE INSTITUTIONS. 163 
 
 evidence as there is that is to be looked to, not the 
 quantity of it. If we find that even one law or usage 
 has come down which seems to be, as clearly as the 
 Levirate itself, of polyandrous origin, and as difficult to 
 account for otherwise, it is plain that the chance of 
 coincidence, which is what we have to be on our guard 
 against in this matter insignificant at best is so 
 materially diminished that it may be disregarded. 
 With a very few such it would become infinitesimal. 
 Add to even one such law or usage good evidence, even 
 if limited in quantity, of polyandry in ancient times 
 and then justified as ancient custom, and we may 
 without hesitation discard the fear of coincidence and 
 hold that the Levirate was derived from polyandry, and 
 that the former prevalence of polyandry may be inferred 
 from it. 
 
 Of course, the greater the quantity of evidence the 
 better. But it is a surprising thing that there is any 
 for what there is has come to us almost by miracle. 
 Brahmin poets and lawyers and their editors would have 
 suppressed it all had they been able. And what we 
 chiefly have to consider is how much room there is for 
 mistake as to the indication given by such evidence as 
 there is. 
 
 It need scarcely be said that such an argument as 
 has just been described would gain in conclusiveness 
 though possibly not so much as it might gain by being 
 materially strengthened in one case so far as it could 
 be repeated, for other peoples which have had the Levi- 
 rate. The evidence from all the cases could be accu- 
 
 M 2
 
 164 THE PATRIARCHAL THEORY. 
 
 mulated to dispel doubt as to the origin of the Levirate. 
 Every custom, too, which shows a " note " of the poly- 
 andry, or is akin to it, or which seems to have come out 
 of the layer of social condition to which the polyandry 
 can be assigned, which can be cited from peoples who 
 have had the Levirate, would add to the strength of 
 such an argument. And evidence pointing to the poly- 
 androus origin of the Levirate that would tell strongly, 
 that might even by itself be enough for belief, might 
 legitimately be brought from races that are backward 
 beyond what either Hindoos or Hebrews were when 
 history makes them known to us. 
 
 As the Patriarchal Theory is the subject of this 
 work, there can be no harm in adding that the 
 Patriarchal Theory has the Levirate to reckon with 
 and that, unless it can account for the Levirate con- 
 sistently with " Patriarchal " custom, it is at once out 
 of the field. 
 
 The cases which have, up to this, been examined 
 have been all of the class that is easiest to deal with 
 though neither of the simplest, nor of the most difficult, 
 order of cases of that class. They have been of the 
 class of cases in which the connection of the derivative 
 with that from which it is derived is suggested, and 
 more or less attested, by resemblances which can be 
 perceived between them resemblances in idea, purpose, 
 methods, results, and outward form. And, we have 
 found that, for confidence in these cases such con- 
 fidence as Sir Henry Maine has in the connection 
 between Patria Potestas and Agnation there is needed
 
 DERIVATIVE INSTITUTIONS. 165 
 
 a weight of corroborative evidence which is not usually 
 forthcoming. 
 
 When, indeed, the older institution is known to 
 have preceded in the place where the supposed deriva- 
 tive is found, and the resemblance is tolerably complete* 
 a conclusion as to the connection between them is not 
 very difficult to arrive at. But much must depend 
 upon the degree of resemblance. If, on the other hand, 
 that is not known, or not likely to be generally 
 admitted, even when there is no deficiency of resem- 
 blance, when even the later institution or usage is 
 substantially identical with the older, a good deal 
 varying according to the case and the chance it affords 
 to other explanations has to be done by way of 
 corroboration, and to exclude possibilities of mistake. 
 It must be known, first of all, that the older has been 
 so far diffused that it is not unreasonable to seek in it 
 an explanation of the later. It has then, at least, to be 
 shown that it has been so diffused that it is not im- 
 probable it should have had its day where the later is 
 found and upon that one could not in many cases 
 feel confidence. Unless, in some of the places where 
 the later institution or usage is found, we find either 
 direct evidence of the prevalence of the older, or other 
 unmistakable relics of it such as, by themselves, are 
 good evidence for proving its former prevalence in 
 general there cannot be room for anything approaching 
 to confidence. The methods by which an argument to 
 establish the connection between the two may be 
 strengthened, need not again be dwelt upon.
 
 166 THE PATRIARCHAL THEORY. 
 
 The cases of the class we have been dealing with 
 which are the least difficult are happily those which, for 
 throwing light upon the history of mankind, are of all 
 the class the most important. They are those in which 
 that which has to be accounted for is visibly a mimicry 
 or symbol. In other cases of this class it can be some- 
 times perceived, and more often surmised, that con- 
 venience has at least favoured the preservation of some 
 part, or of some use, of an old institution. But, owing 
 to our incapacity whatever may be its cause to drop 
 at once a mode of proceeding which has long been 
 customary, practices which have been long common or 
 general practices where they have prevailed, but have 
 been by degrees superseded as needless or ineffective, 
 often survive in a symbolism which itself is usually 
 inconvenient and ridiculous. That can in numerous 
 cases be shown on direct evidence, and can be illustrated 
 from nearly every field of human activity. For our 
 present purpose it may be taken as admitted and 
 beyond doubt. 
 
 When we come upon such symbolism in good 
 preservation not too far decayed to tell its tale plainly 
 when we find what is visibly and beyond mistake a 
 mimicry or counterfeiting, something feigned to be 
 done, not to be taken as done in earnest, and see that 
 it counterfeits something that mankind have been wont 
 to do in earnest, it is but seldom we need delay much 
 before accepting the indication given that there is now a 
 counterfeiting because there was once in the same place 
 a reality of the same sort that what we find people
 
 DERIVATIVE INSTITUTIONS. 167 
 
 feigning to do, predecessors of theirs did in earnest. 
 Where symbolism clear in its indication is found, 
 there is scarcely ever found any customary ex- 
 planation of it (though there is sometimes a half- 
 consciousness of the true one) : what is done people do 
 unthinkingly, in obedience to usage, because it was 
 done by their forefathers. And such explanation as is 
 offered by speculative persons in ignorance of the 
 tendency to symbolism is only rarely even stateable ; so 
 that in general the explanation from symbolism has 
 such advantage as can be derived from the absence of 
 any other fit to compete with it. It is not only sufficient, 
 and, so to speak, visibly sufficient, and based upon a 
 well-known tendency of mankind, but it has the field 
 to itself. Except as being a mimicry of something 
 that has become obsolete, what happens is unexplained, 
 if not inexplicable. It is evidently a mimicry. And it 
 being a way of mankind often to preserve in mimicry 
 their discarded usages, it may in general, as has been 
 said, be readily concluded that it is a mimicry of dis- 
 carded usage. A certain diffusion among mankind at 
 some time or among people who have the symbol of 
 that which is counterfeited must, of course, be shown ; 
 and also that where that occurred, it was common 
 practice so common that its leaving its impress upon 
 usage need not be deemed surprising. As to the amount 
 of evidence to these effects that may suffice for a 
 positive conclusion, that must vary with the case. 
 
 Where, instead of a complete symbol, we have only 
 what seems the remains of a symbol where the indica-
 
 168 THE PATRIARCHAL THEORY. 
 
 tion given is faint instead of clear as may happen in 
 the decay of a symbolism, of course there cannot be 
 anything like confidence felt in a conclusion reached 
 upon that alone. But, if there be clear cases of a 
 symbol as well as faint if it be known, not only that 
 the obsolete usage has been widely diffused, but that' 
 it has frequently been symbolised the previous pre- 
 valence of the usage among a particular people, may 
 be inferred with much probability even from a symbol 
 much decayed. When, on the other hand, we meet 
 with cases, as may occasionally be done, in which the 
 old usage, or a remainder of it, and its symbol are 
 found side by side, and the one, as it were, passing 
 into the other when, for example, in connection with 
 marriage, either a frequent practice of actual capture or 
 of abduction without leave (which is a modification of 
 actual capture) is found side by side with abduction by 
 arrangement/ 5 " that is, the form or symbol of capture 
 these may be, even by themselves, convincing, yield- 
 ing as good evidence as could be demanded for any 
 conclusion. 
 
 What are, on the whole, the most difficult matters 
 of the class we have been considering, are also, as it 
 happens, on the whole the least important. 
 
 We have dealt mainly with cases in which, of an 
 institution or usage which, after gradually shrinking, 
 has given place to something else, custom has retained 
 some part or some use what remains preserving more 
 or less recognisably the semblance of that from which 
 * See pp. 76 and 298-9.
 
 ' DERIVATIVE INSTITUTIONS. 169 
 
 it came. When we turn from these to cases in which 
 we are disposed to think that an old institution (instead 
 of shrinking) has been developed, or rather improved 
 and altered, so as to make it serve new uses of men 
 when the problem is, let us say, to make out the con- 
 nection between an existing institution and something 
 which looks as if it might have been its germ while 
 the conditions of proof remain precisely the same, on 
 the whole, much more care is needed to exclude possi- 
 bilities of mistake to exclude, that is, other explana- 
 tions. A real resemblance, indeed a resemblance, not 
 in outward look only, but in idea and purpose will 
 still be very helpful. The resemblances that are found, 
 however, are commonly vague and faint. And, with a 
 real resemblance, we have to be greatly on our guard 
 so strong, on the one hand, in the cases of this order 
 that we commonly have to do with (say, modern cases), 
 are the chances against so much of resemblance having 
 endured ; and so many, on the other hand, the possi- 
 bilities of the resemblance having come to the later 
 institution from some other source. Custom works 
 slowly and works gently changing only where there 
 must be change, sparing whatever can be spared, re- 
 gardless of congruousness of parts or symmetry of form. 
 But when men in conscious activity have, through 
 official agencies like courts of law or parliaments, been 
 transforming an institution for ages applying it to new 
 uses, discarding it for old uses, changing some of its 
 methods, replacing others by more efficient ones 
 though congruity and symmetry may not have been
 
 170 THE PATRIARCHAL THEORY. 
 
 much thought of, and there may sometimes be left a 
 point or two which can be with some confidence identi- 
 fied as belonging to the original, it is indeed surprising 
 if, on the whole, there remains a real resemblance to the 
 original at the end. And, on the other hand, before 
 institutions are much developed, societies have grown 
 rather complex. Races, each starting with its own 
 customs, have perhaps been mixed together. There has 
 been, at any rate, intercourse with foreign peoples, and 
 more or less imitation of foreign ways. Above all, 
 where the resources of older civilisations could be drawn 
 upon they will have been drawn upon, and only minute 
 investigation can disclose to what extent the borrowing 
 has gone. No doubt, it is possible to guard against all 
 those sources of mistake, and sometimes that is done. 
 And this order of cases has its favourable side. Where 
 they arise, much evidence as to the origin of an institu- 
 tion is often, with careful and skilled investigation, 
 obtainable. 
 
 A fair example of them, and an illustration of 
 what has been said of them, may be found in our 
 system of trial by jury and its origin. Trial by jury 
 was long customarily assigned, on the evidence of 
 some external resemblances, to an old English germ. 
 It seems now to be settled that it has come to us 
 through a form of process which came over with the 
 Normans, and which came to them somehow from the 
 Civil Law. 
 
 Where resemblance to another institution is alto- 
 gether wanting, however, the difficulty of finding an
 
 DERIVATIVE INSTITUTIONS. 171 
 
 origin for an institution is, in general, much greater 
 than it is in the worst cases in which some resemblance 
 is present. In these we have at least a clue to guide 
 our inquiries a hypothesis ready to our hands. There 
 may be risk of our accepting it too easily ; but, in any 
 inquiry, it is of great assistance to start with a hypo- 
 thesis. Where, on the other hand, the indication of 
 resemblance is wanting, unless, by taking trouble, we 
 can so far make out the history of the institution as to 
 find in it something to help us, we must be content to 
 be ignorant or to make surmises which, however 
 amusing, can lay no claim to probability. We may 
 find, however, something, in a prior form of that which 
 we are inquiring about, which gives us the clue of 
 resemblance, which found, the rest of its history may be 
 more easily and more hopefully followed up. Or we may 
 find it in its earlier history, where we have been tracing 
 it, so connected with another institution that we are put 
 on the alert to discover whether that other had to do 
 with its first appearance, and what it had to do with it. 
 The question arises : what influence coming from the 
 one institution, if any, could have originated the other ? 
 and, at the same time, this other question : does the 
 institution which we are disposed to think secondary 
 or derivative first appear along with the other in 
 all cases in which we find it ; or does it sometimes 
 appear independently of that other, or in connection 
 with something else ? It is natural to look at the 
 former question first, since it may give us a hypothesis ; 
 while the latter puts us upon fresh inquiry. And it
 
 172 THE PATRIARCHAL THEORY. 
 
 may turn out that no plausible theory as to the con- 
 nection of the institutions is to be had ; or that 
 there are half-a-dozen that are equally plausible, 
 which is much the same thing ; or, on the other 
 hand, that a theory can be framed which looks so 
 sufficient that it may be taken provisionally as a 
 hypothesis. 
 
 If this last should happen, the theory may connect 
 the two institutions either through something that is 
 essential to, or distinctive of, the older, or through some- 
 thing which, though found with it, belongs to it only 
 accidentally, and which might happen to occur apart 
 from it. In the latter case, it may be possible to find 
 the derivative apart from the institution in connection 
 with which its origin has been inquired into ; and 
 such verification as the hypothesis admits of, consists 
 in so finding it, and then finding along with it the 
 accidental element in that institution with which the 
 hypothesis connects it. It need scarcely be said that 
 we can never get, in such a case, anything like, or 
 equivalent to, "the power of predicting," except for 
 that which is truly the cause ; and that the occur- 
 rence of the derived institution in a new region, 
 without the institution to which we first traced it, 
 cannot raise any presumption that the latter also 
 prevailed in that region. It may make it worth our 
 while to look for traces of this institution. It may 
 raise some hope of finding them. But that is all. The 
 conditions affecting the former case, as the question 
 which is presently to be considered comes under them,
 
 DERIVATIVE INSTITUTIONS. 173 
 
 may, at the risk of being tedious, be set forth with more 
 minuteness. 
 
 When we deduce the secondary institution, then, out 
 of something essential or peculiar to the institution to 
 which it has been traced, the question arises whether 
 it has always first appeared along with this institution. 
 It is by showing this that the hypothesis may be 
 verified. And, however plausible it may be, unless 
 verification can be extended to a good many cases, one 
 cannot reasonably have any feeling like confidence in 
 the hypothesis. 'The chances of having to give up our 
 confidence are too great for us to be justified in feeling 
 any confidence. For it may turn out that a totally 
 different account can be given of the institution which, 
 even as a hypothesis, may be preferable. And a single 
 clearly hostile fact upsets a hypothesis. We know that 
 hypotheses which have been thought beautifully com- 
 plete have, in countless numbers, been set aside for 
 others which, as soon as they had been stated, were 
 perceived to be better ; or have been discarded as clearly 
 bad because of some fact which had not been known or 
 had been overlooked. The position of a hypothesis 
 which is unverifiable the means of testing which are 
 not to be had if there be such is, of course, very bad. 
 For while there may appear reason for giving it up 
 altogether, it can never possibly get beyond being a 
 more or less plausible speculation. And if circumstances 
 only permit of a hypothesis being tested in a very few 
 cases, we must be content with claiming for it a corre- 
 
 ' O 
 
 spending degree of probability.
 
 174 THE PATRIARCHAL THEORY. 
 
 Of facts which are to be considered hostile only one 
 or two need be mentioned. If we find the institution 
 we are inquiring about existing independently of that 
 from which our hypothesis derives it, and in connection 
 with something else, that is primd facie a hostile fact. 
 It may be possible to explain this ; but, unless that be 
 done, it is a fact which stops the way against the hypo- 
 thesis. For the explanation a subsidiary hypothesis is 
 required ; and that, too, has to be tested and verified. 
 It is indispensable, that is, to show that, in a number of 
 cases in which the two institutions have been found 
 together, the derived has outlived the parent institution 
 and outlived it apparently for the reason suggested. 
 This sufficiently done, what was primd facie a hostile 
 fact may be turned into a friendly one. But to 
 find the supposed derivative in independence of that 
 from which a hypothesis deduces it must always, in 
 the first instance, raise doubt as to the goodness of 
 the hypothesis. If it is often so found, there is 
 much probability that we are altogether on the wrong 
 track that we have been seeking the origin of our 
 institution in the wrong place that the appearance 
 of special connection between the two institutions 
 to which we have given ourselves up has been 
 delusive. 
 
 Again, if instances of the supposed derivative are 
 found to be much more numerous than instances of that 
 which we take to be the parent institution, that also is 
 primd facie a hostile fact. The latter institution, when 
 fully developed, should perhaps always, unless interfered
 
 DERIVATIVE INSTITUTIONS. 175 
 
 with, have produced the former ; but, as it may some- 
 times have failed of that, it should have occurred, if 
 anything, the more frequently of the two. Where, 
 then, it is found more rarely, explanation is necessary. 
 It may be possible to give it. But here again the 
 probability arises that we have got on a wrong track. 
 
 Only one more example of hostile facts need be 
 given. If that which we take to be the originating 
 institution is ever found without the supposed deri- 
 vative, obviously that has to be reckoned with. It 
 will supply a ready test of the goodness of our 
 hypothesis. In all probability it will dispose of it 
 altogether. 
 
 Seeing that so much has to be done, it must be 
 evident that, in inquiries in which evidence is not often 
 to be had in abundance, there can be confident con- 
 clusion in the end only in rare cases. And, instead of 
 having in the supposed derivative, when found apart 
 from the institution from which we are deducing it, an 
 easy means of proving the former prevalence of that 
 institution, it will be fortunate for us if we are able to 
 show that its being so found apart does not force us to 
 give up our hypothesis. 
 
 All this is elementary and has been intentionally 
 pitched low. It will be seen by-and-by that it has not 
 been set down without necessity. 
 
 Of an institution associated with another through an 
 accidental nexus, a sufficiently good illustration may be 
 found in primogeniture. The origin of primogeniture 
 in Europe has been, upon a certain amount of evidence
 
 176 THE PATRIARCHAL THEORY. 
 
 (the sufficiency of which is unimportant to us here), 
 traced back to the feudal system. But if the feudal 
 system did establish primogeniture in Europe, this 
 happened through those who moulded that system 
 having resolved that fiefs should be impartible, and 
 held by one person, and through it appearing reason- 
 able or convenient that what one only was to have 
 should go to the eldest. The impartibility of fiefs 
 was not of the essence of feudalism and still less 
 that there should be a preference of the eldest ; and 
 therefore, on this view, it was by accident that primo- 
 geniture arose in connection with feudalism. If this 
 account of the origin of primogeniture in Europe be a 
 true one, primogeniture might arise wherever a feeling 
 prevailed that some subject of inheritance ought not to 
 be divided (or where dividing it had never been thought 
 of) ; and the truth of the account is to be verified (so 
 far as anything can be done for it by verification) by 
 showing that succession to things regarded as not 
 divisible has been given to one son in other cases, apart 
 from feudalism. 
 
 Verification of the hypothesis, however, though 
 necessary, cannot carry us very far in such a case. To 
 be sure that the hypothesis is applicable to the case is 
 what is most important, and verification of it cannot 
 help in this. For primogeniture in Europe may have 
 been a result of impartibility, and may have been found 
 with the feudal system, and yet may not have first 
 arisen in Europe in connection with the feudal system. 
 It may have arisen earlier. In such a case, a high
 
 DERIVATIVE INSTITUTIONS. 177 
 
 degree of probability is always likely to be unattain- 
 able. 
 
 In cases of this class, it will usually be found that 
 the secondary institution is of a different nature from 
 that in connection with which it originated ; is without 
 resemblance to it ; and has its own separate objects and 
 effects. Special connection between the two is not 
 readily suspected. It has always to be made out pretty 
 closely upon evidence. It need scarcely be added that, 
 from finding primogeniture in Africa, we should get no 
 indication of feudalism having prevailed there. It is 
 impartibility that would be indicated. 
 
 Of an institution connected with another, not by an 
 accidental nexus, but essentially, we have, according 
 to Sir Henry Maine, an exceedingly strong example in 
 Agnation. We need not think, then, of other examples. 
 Agnation, on Sir Henry's view, proceeded from Patria 
 Potestas ; arising with it necessarily, that is, always ; 
 determined by its very nature, and being, therefore, it 
 might be thought, dependent thereupon. Patria Potestas, 
 on that view, always produced Agnation, and nothing else 
 ever produced it the one necessarily or invariably issued 
 from and accompanied the other. Now, to establish a 
 connection of this kind, what may suffice ? Speaking 
 generally, it may be said, as strong reasons as are 
 needed for any proposition whatever. We have already 
 seen what are the particular questions which arise. 
 Does the hypothesis seem a sufficient one ? Is it the 
 best hypothesis that is to be had ? And it should be, 
 in this case, not only the best, but beyond all doubt the
 
 178 THE PATRIARCHAL THEORY. 
 
 best. Then, has it been verified in many cases ? And, 
 in this case, an unusual amount of verification is 
 clearly indispensable. Lastly, is there an absence of 
 hostile facts ; or have facts which were primd Jade 
 hostile been sufficiently accounted for ? If all these 
 questions could be answered satisfactorily, Sir Henry 
 Maine would undoubtedly have reason for confidence. 
 But confidence as to what ? If we knew nothing of 
 his case we should expect him to be confident that 
 Patria Potestas and Agnation were always to be found 
 together that wherever he found the one he would 
 also find the other. 
 
 But the question we had to put at starting 
 was, why should Agnation imply the pre-existence of 
 Patria Potestas ? Sir Henry Maine thinks that Agna- 
 tion can often be found in independence of Patria 
 Potestas, where Patria Potestas is not to be found or 
 even to be traced, and in connection with other forms 
 of the paternal power. What he is confident about is 
 that, where he so finds it, he can infer the pre-existence 
 of Patria Potestas. And it is only as a means of 
 recovering lost cases of Patria Potestas that its connec- 
 tion with Patria Potestas is of importance to him. 
 That Agnation should be found without Patria Potestas, 
 and in connection with other forms of the paternal 
 power, however, and especially that it should be so 
 found "almost everywhere," is primd facie a fact hostile 
 to his hypothesis and, the nature of the hypothesis 
 considered, a fact extremely hostile. Primd facie, this 
 raises a strong presumption that the explanation of
 
 DERIVATIVE INSTITUTIONS. 179 
 
 Agnation is to be looked for in some other direction. 
 No doubt such seemingly hostile facts can sometimes be 
 accounted for and made friendly. The difficulty of 
 doing this with Agnation seems unusually great, But 
 we have seen how it is to be attempted. There is 
 needed a new theory or hypothesis to explain how 
 Agnation survived that which gave it being, and on 
 which the main hypothesis made it seem absolutely 
 dependent. And, at the least, it is indispensable to 
 show that in some cases in which Patria Potestas and 
 Agnation have been found together, Agnation survived 
 Patria Potestas, kept up by the influence indicated in 
 this new hypothesis. It should appear, too, that this 
 influence was not likely to have acted in exceptional 
 cases only. With a second theory so verified, there 
 might, after all, be more or less confident inference of a 
 pre-existence of Patria Potestas from Agnation. 
 
 It may be well to point out here, with reference to 
 Sir Henry Maine's hypothesis, that Patria Potestas and 
 Agnation dealt with entirely different matters, and did 
 not act within the same limits. The one embodied the 
 powers of the father. The other defined relationship, 
 and through that determined rights of inheritance. 
 Patria Potestas acted only within the family was con- 
 fined to the descendants of a single living Paterfamilias. 
 Agnation had a far wider sphere ; it connected the 
 members of a family with one another, but it also 
 connected the family with many other families. It 
 might be not unreasonable to suppose that the father's 
 power was placed so high as it was at Rome simply in 
 
 K 2
 
 180 THE PATRIARCHAL THEORY. 
 
 the interest of fathers ; and, in fact, Patria Potestas 
 was extended to plebeian fathers at Rome as a privilege. 
 Whether it could pass its bounds and produce effects 
 far beyond them effects, too, with which the degree of 
 a father's powers has no obvious connection is one of 
 the questions we shall immediately have to consider.
 
 CHAPTER XII. 
 
 PATRIA POTESTAS AND AGNATION. 
 
 IN considering the hypothesis which derives Patria 
 Potestas from Agnation it will be convenient to invert 
 the natural order, and notice first what can be gone 
 through most quickly and easily the facts which bear 
 upon the hypothesis. The hypothesis, as has been said 
 already, involves that Agnation necessarily or invariably 
 proceeded from and accompanied Patria Potestas. And, 
 nevertheless, its importance to Sir Henry Maine entirely 
 arises out of his thinking that Agnation, or traces of it, 
 can be found almost everywhere, in independence of 
 Patria Potestas, and in connection with lower forms of 
 the paternal power. He says that its being so found 
 " implies the former existence " of Patria Potestas. 
 
 To begin, then, Patria Potestas and Agnation are 
 found together only in a single case. They occurred 
 together at Rome, and have never been shown to have 
 occurred together anywhere else. And to prove that 
 the one always arose out of, and was thereafter to be 
 found along with, the other, there is the fact that we 
 find them together in one case.
 
 182 THE PATRIARCHAL THEORY. 
 
 To prove what it is equally necessary to prove, 
 however that Agnation, so originated, could outlast 
 Patria Potestas there is no fact at all. And the only 
 fact that bears upon the matter tends, and tends de- 
 cidedly, the other way. Sir Henry Maine supposes that 
 Agnation, though originated as he thinks it was, could 
 outlive Patria Potestas ; and that it constantly did 
 survive Patria Potestas for so long a period that every 
 trace of Patria Potestas having occurred along with it 
 had time to disappear. And yet at Rome, where alone he 
 can show us the two together, it was Patria Potestas that 
 showed the greater power of lasting. The last vestiges 
 of the two disappeared from the law together. But, in 
 fact, Agnation went first. The paternal powers were 
 susceptible of abridgment and restriction in various 
 ways short of extinction. The wife might become free 
 from them ; the children also ; and yet they might 
 remain for the slaves. And it was thus gradually that 
 they perished. But Agnation is perfect, or it ceases to be 
 Agnation. And the moment the ties of blood through 
 women received civil effects Agnation was no more. 
 This took place under the Praetors, long before the 
 disappearance of Patria Potestas. As was to be 
 expected, too, considering the difference in kind of the 
 matters with which each had to do, the enactments 
 which destroyed the one were mostly unrelated to those 
 which broke down the other. Patria Potestas at Rome 
 (where alone we know it) proved, in fact, an extremely 
 durable institution. And if a judgment may be formed 
 upon the one case which affords us the means of judging,
 
 PATEIA POTESTA8 AND AGNATION. 183 
 
 it was everywhere more likely to outlast Agnation than 
 Agnation to outlast it. At any rate, as the matter 
 stands, there seems a lack of reason for believing with 
 confidence that Agnation, if produced as it is said to 
 have been, would generally be the survivor, and that 
 it would survive Patria Potestas for ages. 
 
 We have now got the length of seeing that the 
 hypothesis which derives Agnation from Patria Potestas 
 is a naked hypothesis ; one, too, from its nature, need- 
 ing a great deal of verification and support, while it has 
 not in the slightest degree been verified or supported. 
 And, if cases of Agnation found with lower forms of the 
 paternal power than Patria Potestas were really plentiful 
 their being so found being unaccounted for it might 
 be dismissed forthwith as manifestly a bad hypothesis. 
 The facts would be that Agnation occurred, indifferently, 
 so far as could be seen, along with various degrees of the 
 paternal power; that it occurred with the exceptional 
 degree of paternal power called Patria Potestas in the 
 one case in which it has been found ; and that, in this 
 case, it did not survive Patria Potestas. For a hypo- 
 thesis which derived it, on the strength of this one case, 
 from Patria Potestas, and accounted for the numerous 
 cases where it occurred, in which the paternal power was 
 lower, by attributing to it a longevity indefinitely 
 greater than that of Patria Potestas, nothing could be 
 said. For a reason that will appear hereafter, however, 
 this will not be insisted upon here. 
 
 A hypothesis that cannot be verified or supported 
 cau never be more than a mere speculation, upon which
 
 184 THE PATRIARCHAL THEORY. 
 
 nothing can reasonably be founded ; but it may be 
 plausible or the reverse. Let us now see which of the 
 two this hypothesis is. 
 
 Sir Henry Maine's argument on the connection 
 between Patria Potestas and Agnation, as given in 
 Ancient Law, is contained in the following passages : 
 
 (a) "The foundation of Agnation is not the mar- 
 riage of father and mother, but the authority of the 
 father. All persons are Agnatically connected who are 
 under the same paternal power, or who have been under 
 it, or who might have been under it if their lineal 
 ancestor had lived long enough to exercise his empire. 
 In truth, in the primitive view, Relationship is exactly 
 limited by Patria Potestas. Where the Potestas begins, 
 kinship begins, and, therefore, adoptive relatives are 
 among the kindred. Where the Potestas ends, kinship 
 ends ; so that a son emancipated by his father loses all 
 rights of Agnation. And here we have the reason why 
 the descendants of females are outside the limits of 
 archaic kinship. If a woman died unmarried, she could 
 have no legitimate descendants. If she married, her 
 children fell under the Patria Potestas, not of her 
 father, but of her husband, and thus were lost to her 
 own family." Then follows the well-known passage to 
 the effect that early societies would obviously have been 
 confounded if men had called themselves relatives of 
 their mothers' relatives. Ancient Law, p. 149. 
 
 (b) "The Parental Powers proper are extinguished 
 by the death of the parent, but Agnation is, as it were, 
 a mould which retains their imprint after they have
 
 I 
 
 PATRIA POTESTAS AND AGNATION. 185 
 
 ceased to exist The Powers themselves are dis- 
 cernible in comparatively few monuments of ancient 
 law, but Agnatic Relationship, which implies their 
 former existence, is discoverable almost everywhere." 
 Ancient Law, p. 150. 
 
 Those passages, like many other explanatory passages 
 in Sir Henry Maine's writings, do not yield a meaning 
 that can be stated in plain terms. 
 
 That we are to understand from them that Patria 
 Potestas, out of its very nature, determined relation- 
 ship, and that the relationship it determined was always 
 Agnation is, however, clear enough. And it is also 
 clear that Sir Henry says it was subjection to the same 
 paternal power, that made the first relationship. 
 
 How, then, was this primitive relationship expanded 
 into Agnation as we know it ? Patria Potestas, at its 
 widest stretch, could cover but a small part of the limits 
 of Agnation for Agnation extended, roughly speaking, 
 to all persons descended through males from the same 
 male ancestor. The relationship limited by Patria 
 Potestas must, at the utmost, have been very far 
 short of the extent of Agnation. It is in vain we 
 try to make out from Ancient Law how Sir Henry 
 Maine connected those two. He tells us, indeed, that 
 "Agnation is, as it were, a mould" which retains the 
 imprint of the paternal powers after they have ceased 
 to exist. But figures of that kind are nearly always 
 delusive ; and it is easy to see that this one is. It is 
 inapt even for suggesting what Agnation was rela- 
 tionship subsisting between descendants through males
 
 186 THE PATRIARCHAL THEORY. 
 
 of the same male ancestor. But, at best, it only points 
 to the fact that the relationship which, at Rome, con- 
 nected families, and the persons included in them, with 
 one another, was transmitted through the male members 
 of the families. "Whatever the origin of Agnation may 
 have been, how otherwise could it be transmitted ? That 
 relationship was transmitted from father to children is, 
 however, literally all there is to say for the view that 
 the Power of the Father, acting as it were through a 
 mould, was the determinant of Roman relationship. Sir 
 Henry Maine should have at least shown us here how 
 Patria Potestas acquired the capacity of leaving an 
 imprint. For he assumes that it had not this capacity at 
 first. He tells us that at first relationship began and 
 ended with Patria Potestas. 
 
 Here occurs the question whether his primitive 
 relationship can, even by a strain of language, be 
 termed relationship. It was constituted by subjec- 
 tion to the same paternal power. It began and ended 
 with Patria Potestas. A son therefore lost it when he 
 was emancipated, or when he emancipated himself. A 
 daughter lost it when she married, because she then 
 passed out of her father's Potestas. Her children, we 
 are told, never had relationship to her family because 
 they were never under that Potestas. (But, when a 
 Paterfamilias had " children, grandchildren, and great- 
 grandchildren," might not his female descendants, 
 or many of them, have married, and their children 
 been born within the Potestas ? Did not the Patriarch 
 Abraham marry his sister-german ?) And further,
 
 PATRIA POTE8TA8 AND AGNATION. 187 
 
 when the Paterfamilias died, his Potestas being at an 
 end, the relationship which it had made ended too. 
 Those who had been members of his family ceased to be 
 related to each other. In its time of primitive vigour, 
 that is, Patria Potestas left no imprint behind it. All 
 the relationship then known was the connection made 
 between people by their living under the same family 
 chief. And that lasted only as long as they lived under 
 him. The connection between a ship's crew when at 
 sea under the same captain, or that of a regiment under 
 the same colonel, has as good a claim to be called 
 relationship as the connection between the members of 
 Sir Henry Maine's primitive family.* 
 
 It was, therefore, something that seems a good deal 
 short of true relationship that Sir Henry Maine had 
 to connect with Agnation. He leaves us in Ancient 
 Law to puzzle out for ourselves how the one could 
 grow into the other. And to guide us, there is nothing 
 except that (on his own assumption) the primitive 
 relationship was not " as it were, a mould " which 
 retained the imprint of paternal powers which had 
 become extinct, but a passing connection, which might 
 
 * What Sir Henry Maine says of the emancipated son is that he 
 lost " all his rights of Agnation." But this cannot mean less than 
 that he lost relationship altogether. For otherwise relationship and 
 Patria Potestas were not commensurate relationship did not end 
 where Potestas ended in the primitive family. And (though it is not 
 stated) the relationship of the family must have been at an end when 
 the Paterfamilias died. For otherwise relationship did not begin and 
 end with Patria Potestas in the primitive family even the primitive 
 relationship extended beyond the primitive family.
 
 188 THE PATRIARCHAL THEORY. 
 
 be broken in various ways, and which, on the death of 
 the Paterfamilias, came completely to an end, leaving no 
 trace of itself behind. Did Patria Potestas acquire in 
 time a capacity which it had not at first ? Or did some- 
 thing help it and was it in virtue of that help that 
 Agnation was produced ? Or did relationship grow up 
 in time after its own ways, independently of Patria 
 Potestas, but somehow determined by it ? Of these and 
 other such views that may occur to us Ancient Law 
 leaves us to take our choice.* 
 
 Of course, it is the exclusions of Agnation that 
 
 * A passage in Sir Henry Maine's latest work, Early Law and 
 Custom, which has already been quoted (supra, page 22), appears to 
 show that Sir Henry is now of the opinion the only rational opinion, 
 though not embodied in the account of relationship given in Ancient 
 Law that reflection upon the facts of blood-connection introduced 
 systems of relationship. Casually, while gently blaming Mr. Lewis H. 
 Morgan for over-facility of assertion and for a lax (though convenient) 
 use of the term gens, he reminds him that " what was new at a certain 
 stage of the history of all or a portion of the human race [apparently, 
 the recognition of relationship through the father] must have occurred, 
 not in connection with the gens, but in connection with the family ; " 
 and adds that, " at some point of time, some change of surrounding 
 circumstances enabled paternity, which had always existed, to be 
 mentally contemplated, and further, as a consequence of its recogni- 
 tion, enabled the kinship flowing from common paternity to be 
 mentally contemplated also." That "when paternity reappeared," 
 the kinship recognised was kinship through male descents only, is 
 still, in this passage, ascribed to Patria Potestas (the reappearance of 
 paternity "in association with Power and Protection"); but its 
 modus operandi is not explained. 
 
 The passage gives, or rather suggests to us, a perfectly good reason 
 for the acknowledgment of relationship between persons descended 
 from the same ancestor. Apart from this, the view it gives is that 
 relationship began in the family, and was, through the influence of
 
 PATRIA POTESTAS AND AGNATION. 189 
 
 chiefly need to be accounted for. And it is when we 
 come to consider what is said of these that the difficul- 
 ties in the way of the solution suggested in Ancient 
 Law, and the need in connection with them of explana- 
 tions that are not given, can be most plainly seen. 
 That persons actually of the same family, under the 
 same father, should be each other's relatives is not 
 surprising, whatever our theory of the origin of relation- 
 ship ; and that the children of such persons should be 
 each other's relatives would not be surprising. That 
 which is remarkable in connection with Agnation, that 
 which any theory of Agnation ought clearly to account 
 
 Patria Potestas, traced through male descents only the view of 
 Ancient Law, though without the explanation which accompanies 
 it, and without alternative explanations. It leaves us, therefore, to go 
 back to Ancient Law for explanation of what is the really important 
 matter the limitation on relationship ascribed to Patria Potestas 
 the exclusion of a man's descendants through females from relation- 
 ship with his family. Everywhere else, it may be said, throughout 
 Sir Henry Maine's writings of date subsequent to Ancient Law, 
 the connection which he alleges between Patria Potestas and Agnation 
 is taken as established, without there being anything to establish it 
 except the demonstration contained in that work. 
 
 It being admitted that relationship was introduced by mental con- 
 templation of the facts of blood-connection, is there room for the 
 suggestion that relationship through the mother escaped notice 1 
 Surely not. The blood-connection with the mother is more apparent 
 than the blood-connection with the father, and, if there was a time when 
 circumstances prevented the male parent from being ".individualised 
 in the mind," it would during that time have to itself all the mental 
 contemplation of which men were capable. If we trust to inference 
 we must conclude that the kinship flowing from common maternity 
 would have the first chance of being acknowledged. And we know 
 as a fact that numerous peoples have got this kinship without ever 
 getting beyond it.
 
 190 THE PATRIARCHAL THEORY. 
 
 for, is its denial of relationship with a family to the 
 descendants of its women. Now, all we are told as to 
 this in Ancient Law is, that relationship was, in the 
 primitive view, exactly limited by Patria Potestas ; that 
 while, if a woman died unmarried, she could have no 
 legitimate descendants the children of a married woman 
 fell under the Patria Potestas, not of her father, but of 
 her husband ; and that such children were thus lost to 
 the family of their mother, and " outside the limits of 
 archaic kinship." 
 
 But it need not be said that within times known to 
 us, mere difference of Patria Potestas did not exclude 
 persons from relationship with one another. If it had 
 done that, the children of orphan brothers would not have 
 been each other's relatives whereas the children of all 
 males descended from the same ancestor were each other's 
 Agnates, notwithstanding that each family formed by 
 them was subject to a separate Potestas. How then, in 
 the case of descendants through males, came mere dif- 
 ference of Patria Potestas not to have the effect of 
 restricting relationship, if it is enough (it is all we are 
 told of) to account for the exclusion from, relationship 
 of alt descendants of married women ? The primitive 
 view, by assumption, treated men and women alike 
 the reason which excluded women and their descendants 
 from relationship sufficing to exclude men and their 
 descendants also. Did Patria Potestas, in the advance 
 from the primitive condition, learn to discriminate be- 
 tween descendants through males and descendants 
 through females, and to treat them differently so that
 
 PATRIA POTESTAS AND AGNATION. 191 
 
 it made no bar to relationship between the former, while 
 excluding the latter from relationship ? It would, seem 
 that this must have happened. And, if so, how did it 
 happen ? At any rate, we have attributed to the Potestas 
 exercised by the husband of a female an effect which 
 Patria Potestas did not have (or did not continue to have) 
 when exercised by male descendants of her family. And 
 it was indispensable that it should have been explained 
 to us how Patria Potestas could have an effect in the one 
 case which it had not in the other. Ancient Law gives 
 no explanation of this ; and that means that it does not 
 enable us to see how it may have been that Patria 
 Potestas was the foundation of Agnation. 
 
 It appears, therefore, that, instead of a theory of 
 Agnation with some claim to be considered plausible or 
 satisfactory as a hypothesis, Ancient Law comes very 
 far short of giving us a hypothesis at all ; since it leaves 
 both the inclusions of Agnation and its exclusions 
 unaccounted for. What it gives us as an account of 
 Agnation is simply the suggestion that Agnation was 
 " as it were, a mould " which retained the imprint of the 
 paternal powers after they had become extinct ; and the 
 suggestion that, in primitive times, when Patria Potestas 
 limited relationship, the children of a married woman 
 would be under the Potestas of her husband, and not 
 under the Potestas of her father. It has already been 
 shown that no help in inferring the origin of Agnation can 
 be had by means of the former suggestion. It remains 
 for us to see what is to be said as to the validity, on Sir 
 Henry Maine's own view, of the latter suggestion.
 
 192 THE PATRIARCHAL THEORY. 
 
 Admitting, then, that Patria Potestas moulded rela- 
 tionship, and that, on the primitive view, relationship 
 was exactly limited by Patria Potestas, it cannot be valid 
 in any degree or to any effect, except on the assumption 
 that marriage always, or as a rule, involved a change of 
 Potestas ; that marriage carried the bride away from her 
 Paterfamilias ; that marriages did not occur between 
 persons who were under the same Potestas did not 
 occur, that is, between persons who were of the same 
 relationship. This, however, is a large assumption. It 
 requires to be justified ; and, if it is not justified, 
 Ancient Law fails to account even to the small degree 
 to which it has tried to do so for the exclusions of 
 Agnation. It need not be said that this assumption 
 has not been justified in Ancient Law. 
 
 As to the need for justification, surely it should have 
 been shown to us what there was to prevent the marriage 
 of persons living under the same Potestas or, to use 
 the more suggestive equivalent, persons of the same 
 relationship. We have the primitive Paterfamilias 
 represented to us as governing his " children, grand- 
 children, and great-grandchildren," besides persons 
 adopted by him and their descendants ; and, therefore, 
 as having under his Potestas a number of marriageable 
 men and women. Marriages might therefore occur 
 within the Potestas or relationship, unless there was 
 something to prevent them. They might even be 
 frequent. And, if they were anything like common, 
 then, in the primitive relationship to which we are 
 referred for the basis of the relationship ultimately
 
 P ATRIA POTESTAS AND AGNATION. 193 
 
 acknowledged a man's descendants through males 
 would be his descendants through females also to such 
 an extent that that relationship could not, it would 
 seem, develop into Agnation. It was necessary then 
 to show us that persons of the same Potestas or rela- 
 tionship did not marry one another which can only 
 be done by showing us some reason why they should 
 not marry one another. And it was the more necessary 
 that this should have been attempted because there is 
 no lack of instances of marriage in early times between 
 persons of the same family or who were each other's 
 near relations. To take one which has been mentioned 
 already, the story of Abraham shows us brother and 
 sister-german marrying. And, in Abraham's family, 
 we find also a marriage between uncle and niece 
 between a man and his brother's daughter. Compared 
 with such cases, the marriage of more distant relatives, 
 such as cousins-german, seems natural and easy. And 
 yet what Ancient Law assumes, in assuming that 
 marriage always carried the bride into a strange Potestas 
 or relationship, is that no such marriages took place 
 within the Potestas. 
 
 It need scarcely be added, as to the children of 
 women who had not been given in marriage of 
 whom Sir Henry Maine remarks that they would be 
 illegitimate that apparently there could have been 
 nothing to take them, at any rate, out of the Potestas 
 of the mother's father.* 
 
 * Among the Hindoos the child of an unmarried woman belonged 
 to her father, and was among his heirs. 
 
 o
 
 194 THE PATRIARCHAL THEORY. 
 
 As to the possibility of justifying the assumption 
 that there was no marriage within the Potestas, we 
 know of prohibitions of marriage only in connection 
 with the acknowledgment of relationship founded on 
 blood ; and then as applying, not to persons living under 
 the same chief or of the same group, but to all persons 
 counted to be of the same kindred whether living in 
 the same group or not. And if ever there was no rela- 
 tionship except that which could be made by subjection 
 to the same family chief, the question why there should 
 have been either prohibition of marriage between the 
 persons connected by it, or abstinence from inter- 
 marriage among them, will be found extremely difficult 
 to answer. The answer, which, of course, could only be 
 conjectural, would have to be consistent with, or re- 
 concilable with, the class of rather formidable adverse 
 facts one or two of which have been mentioned. 
 
 It remains to point out that it is entirely by means 
 of this assumption that Sir Henry Maine, in Ancient 
 Law, makes any approach he has made towards 
 accounting for the exclusions of Agnation. For Patria 
 Potestas supposing it somehow capable of moulding 
 or " founding " relationship would not have founded 
 Agnation unless women, when they married, had 
 passed into a strange relationship. It is not Patria 
 Potestas, therefore, but a system of not marrying 
 within the acknowledged relationship, which appears, 
 on the face of his own exposition, to have been the 
 determining cause of Agnation. It need scarcely be
 
 PATEIA POTE8TA8 AND AGNATION. 195 
 
 said that a custom which forbids marriage between 
 persons of the same relationship is well-known under 
 the name of exogamy. 
 
 Of the further question, how Agnation could survive 
 Patria Potestas, and even survive it for ages, Sir Henry 
 Maine has nowhere taken any notice. But it was in- 
 dispensable for him to show us how this could happen. 
 A married woman's children were lost to her family in 
 respect of her husband's Potestas, and of that only 
 this is what he tells us. Why then should the 
 descendants of married women continue to be lost to 
 their families when husbands ceased to have Patria 
 Potestas? The fact which had determined their exclusion 
 from relationship gone, the exclusion should have 
 ceased unless, indeed, there was by that time at 
 work for it some other influence which by itself was 
 of force enough to keep it up. If there was any such 
 influence it should have been indicated, and shown, 
 by sufficient evidence, to have possessed the capacity 
 attributed to it. The effect of this having been neg- 
 lected is that the theory is left unfit to account for 
 what is declared to be by far the most numerous and 
 most important class of cases of Agnation. 
 
 Here we must repeat that Sir Henry Maine can 
 show us Patria Potestas and Agnation together in 
 one case only, and that, in that case, Patria Potestas 
 survived Agnation. In that case, too, Agnation was 
 broken down Patria Potestas still surviving by the 
 circumstance which could most plausibly be thought 
 
 o 2
 
 196 THE PATRIARCHAL THEORY. 
 
 of as likely to keep it up after Patria Potestas was 
 gone. At Rome, Agnation was destroyed by its relation 
 to the law of inheritance. * 
 
 To the preceding criticism which, if valid, if only 
 in the last branch of it, shows that Sir Henry Maine 
 has entirely failed to make out, even by way of 
 hypothesis, any connection between Patria Potestas 
 and Agnation a word or two should be added as 
 to the view of primitive relationship from which, in 
 Ancient Laiu, he sets out. That the relationship 
 of primitive times (if it is to be so called) was con- 
 fined to the family, to those who were actually living 
 under the same family chief, and that the primitive 
 father was an unsocial being who acknowledged no ties 
 except the tie of power which bound his dependents 
 to him is, of course, sheer assumption. And all we 
 know of man's nature and ways makes it an exceedingly 
 improbable assumption. It is from an assumption 
 derived from a supposed pre-social state of man, 
 however, that Ancient Law leaves us to grope our 
 way towards the relationship acknowledged in the 
 rather artificial Roman society, t 
 
 * It may occur to some people that, the kinship once established, 
 the force of custom might keep it up. But custom tends to support 
 everything that is established. It would tend to support Patria 
 Potestas as well as to support Agnation. How can we know that 
 it would support Agnation after it had let Patria Potestas go down ? 
 There is only one way of knowing we must be able to show, by 
 means of cases in which Patria Potestas and Agnation occurred 
 together, that it actually did so. 
 
 t In such a state of man as is supposed in Ancient Law there
 
 P ATRIA POTESTAS AND AGNATION. 197 
 
 Sir Henry Maine's essay, nevertheless, yields us a 
 most valuable hint, if we go on to search for a better 
 account of the origin of Agnation. 
 
 For, as has been shown, he tacitly assumes that 
 marriage would carry a woman out of the " relation- 
 ship " of her birth, out of the " relationship " to 
 which the male members of her family continued to 
 belong, and that her children would be born in a 
 strange " relationship." And if Patria Potestas ever 
 excluded the descendants of females from relation- 
 ship, it is quite clear that it was through marriage 
 carrying women out of their original relationship that 
 it got the chance of doing so. 
 
 Then it is beyond question that Patria Potestas 
 whatever it may have done at first did not, as time 
 went on, have the effect upon relationship in the case 
 of the descendants of males which is attributed to it 
 in the case of the descendants of females. Difference 
 of Patria Potestas did not disconnect the former from 
 one another, though it is said to have had the effect 
 of disconnecting the latter from all of the former. 
 But, both at first and at all subsequent times, its 
 effects in the case of either must have been the same. 
 It follows that difference of Patria Potestas cannot 
 have had the disconnecting or excluding effect attri- 
 
 could be no such thing as right. A father's power would be what he 
 happened to be able to make it. However great, could it be identi- 
 fied with a recognised and legal right like Patria Potestas ? There 
 could no more be paternal right than there was true relationship in 
 Sir Henry Maine's primitive family.
 
 198 THE PATRIARCHAL THEORY. 
 
 buted to it in the case of the descendants of females, 
 unless it be in consequence of some effect of a lasting 
 kind produced by Patria Potestas in its earliest days, 
 when, by supposition, the relationship founded by it 
 did not last. 
 
 Now the important part, as regards Agnation, of 
 the effect attributed in Ancient Law to Patria Potestas 
 in its earliest period is that it restricted relationship 
 to the Potestas. For any approach to accounting for 
 Agnation, then, it was found necessary to assume that 
 a person could belong only to one "relationship." We 
 have seen that it was also found necessary to assume 
 that there was no marriage within the Potestas or 
 relationship ; that there was no marriage between 
 acknowledged relatives. 
 
 Without both these assumptions Ancient Law could 
 not even have made a beginning which is all it has 
 done of accounting for Agnation. But, between 
 them, they account for the children of a married 
 woman being, in the early days when relationship 
 was a transitory thing, cut off from the relationship 
 of their mother's birth. 
 
 How, originating when they are said to have been 
 originated, these conditions could manage to last needed 
 to be explained, and has not been explained. But it is 
 easy to see that they can give no real help in accounting 
 for Agnation unless, adding to the theory of Ancient 
 Law, we assume that they did last, their sphere of 
 operation expanding as relationship expanded or 
 rather, unless it is assumed that, with true relationship
 
 PATRIA POTE8TA8 AND AGNATION. 199 
 
 in existence, people could belong only to one body of 
 relatives ; while at the same time marriage carried 
 women out of the circle of relationship in which they 
 were born, and to which their male kinsmen belonged, 
 into another circle of relationship. 
 
 It is easy to see, too, that the two conditions we have 
 now arrived at are capable of accounting fully for the 
 main features of Agnation, and that without any aid 
 from Patria Potestas. 
 
 For, through the operation of the latter condition 
 and that without any aid from the husband's paternal 
 power, and whatever its degree the children of a 
 married woman would belong to the relationship or kin- 
 dred into which she had married ; and in consequence 
 of the former without any effect of their father's 
 paternal power, and whatever its degree they could 
 belong to that relationship or kindred only. Through 
 the joint operation of those two conditions, therefore 
 without the help of the paternal power, and whatever 
 the paternal power might be children would be cut 
 off from the relationship to which their mother belonged 
 by birth. A practice of not marrying within the circle 
 of acknowledged relatives is, of course, a practice of 
 exogamy. 
 
 We discover by the help of Ancient Law, there- 
 fore, that those two conditions seem likely to give a 
 full and satisfactory account of the exclusions of Agna- 
 tion. Given a body of people acknowledging kinship 
 through males, and acknowledging relationship to one 
 another, among whom no relationship was acknowledged
 
 200 THE PATRIARCHAL THEORY. 
 
 outside the body (clan or gens), and who did not marry 
 within it whose marriage custom, that is, was exo- 
 gamy then there appears to be the strongest likelihood 
 that their system of relationship would in time exhibit 
 every feature of Agnation. 
 
 It would follow that Agnation arose, where it did 
 arise, only among exogamous peoples who acknowledged 
 kinship through the father ; and only among those of 
 them with whom relationship was restricted to the clan 
 or gens. 
 
 And as to exogamy the custom which restrains 
 from intermarriage men and women who are counted to 
 be of the same kindred, of the same acknowledged 
 blood-relationship it may here be pointed out that it 
 is, in fact, a far-reaching law of incest ; and that it is 
 founded upon relationship, and that therefore relation- 
 ship in some form was older than it. It can have had 
 nothing to do with the beginnings of relationship. On 
 the contrary, ties of blood of one sort or another 
 must have been thoroughly acknowledged before a 
 custom so extremely inconvenient (to say nothing more) 
 could have been founded upon them. 
 
 Agnation, therefore, cannot, if exogamy was a factor 
 in producing it, date from the beginnings of relationship. 
 
 If we go on to inquire whether that which has been 
 suggested to us as the other condition precedent to Agna- 
 tion limitation of relationship to a body or group is 
 likely to have arisen in early stages of relationship or in 
 some relatively late stage, there is one well-known fact 
 which can help us. The most obvious relationships of all,
 
 '.^ PATEIA POTESTA8 AND AGNATION. 201 
 
 and those which, therefore, had the first chance of being 
 noticed, are those which arise through the mother.* And 
 relationship, where it is counted through women only, is 
 not limited to a group. It connects together persons in 
 different groups. It follows the related person wherever 
 he is. This shows a tendency in early relationship 
 in primary relationship not to admit of restrictions ; 
 a capacity of transcending the group where necessary ; 
 a tendency to reach to every person who was within its 
 conditions and could be traced, and to include such 
 persons wherever they might be. 
 
 Now, what could there be to counteract this ten- 
 dency where relationships through the father were ac- 
 knowledged as well as the more obvious relationships 
 through the mother ? Unless there appears to have 
 been something which could do so, the probable opinion 
 is, that strict limitation of relationship to the group does 
 not belong to the earlier history of relationship. 
 
 Could the father's relation to his family have had a 
 limiting influence upon relationship ? Here we have 
 to consider that, man being a social creature, there must 
 have been bodies of people living together before there 
 was any thought of relationship. This being so, the 
 father's power over his family could have no chance 
 of restricting relationship to a group unless the intro- 
 duction of kinship through males at once dissolved the 
 group into which it was introduced, and made every 
 
 * See Studies in Ancient History, by J. F. M'Lennan. London, 
 1876 : " Primitive Marriage," chapter viii., page 121 et seq.
 
 202 THE PATRIAECHAL THEORY. 
 
 man in it the nucleus of a new group. That seems an 
 extravagant supposition. But, even with it, it is most 
 unlikely that the father's power, which could only for a 
 brief period extend to the whole of a group, should per- 
 manently limit relationship to the group. Nor is it in 
 the least intelligible why the father's power should cause 
 denial of any actually perceived relationship. 
 
 A more promising supposition that which appears 
 indeed to be the most promising is that the cohesion of 
 the family, of the body of known relatives, of the whole 
 group, of which kinship through males permits, would 
 tend to make the group relationship the practically im- 
 portant part of relationship ; while, on the other hand, 
 relationship outside the group might even be not traceable 
 beyond the nearer degrees. But, between that and the 
 denial of all outside relationship there is a great differ- 
 ence. And, when we ask why this should follow, there 
 appears to be again a want of reason. 
 
 On the other hand, this limitation can be understood 
 if recognition of kinship through males did not take 
 place until relationships previously acknowledged had 
 had time to grow into a system, and the system had 
 operated so that people were divided by it into kindreds, 
 and that a person could belong to only one of them. 
 The system of kinship only through females has thus 
 operated. It has counted as one kindred all persons 
 related to one another through women all the persons 
 whom it could show to be each other's blood relations. 
 At first there was no relationship acknowledged outside
 
 PATEIA POTESTAS AND AGNATION. 203 
 
 this kindred, because there was none perceived. 
 And, by the time that other relationships were per- 
 ceivable, it had hardened into an institution, with con- 
 ditions which united those who were born of it to one 
 another and cut them off from other people ; it was 
 fully established as the social unit, and the primary 
 social obligations depended upon it obligations which 
 no person could throw off without becoming an outcast, 
 and which bound every person to one kindred ex- 
 clusively, and to his own kindred as against all other 
 kindreds ; and thus though, as time went on, with in- 
 dications of approaching change acknowledgment of 
 relationship continued to be limited to it. 
 
 If the limitation of relationship to the group can 
 be more probably explained in the way just indicated 
 than in any other, we should have to conclude that 
 Agnation, where it happened to be produced, was 
 arrived at through a progress, set in motion by the 
 recognition and assertion of paternity, from the system 
 which counts kinship through females only. 
 
 Exogamy is constantly found with the system of 
 kinship through females only, and, therefore, on this 
 view, it would be antecedent to the limitation of 
 relationship to the group. As it is found with kinship 
 through females only, the father's power can have had 
 no share in producing it. Reason has already been 
 shown for the opinion that the father's power cannot 
 have limited relationship to the group. Nevertheless, on 
 the view of the history of that limitation here suggested,
 
 204 THE PATRIARCHAL THEORY. 
 
 it can be made probable that the conditions which 
 would favour Agnation would also favour the power of 
 fathers over their wives and children. 
 
 It should be added that, while exogamy has been 
 common among peoples who acknowledge kinship 
 through males, the known cases of Agnation have 
 been exceedingly few ; and unless, notwithstanding 
 this, there appear reason to conclude that it has been 
 common or, as Sir Henry Maine supposes, universal 
 if the limitation of relationship to the group was a 
 condition precedent to Agnation, it would follow that 
 this limitation has been rare. That both should 
 have been exceptional and there seems no reason for 
 doubting that they were so is intelligible on the view 
 that kinship through males had to supplant in kindreds 
 already established a pre-existing system of kinship. 
 When this had been done, people might stop in general 
 a good way short of complete denial of the relationships 
 which had been supplanted. 
 
 It is perhaps unwise to attempt what the limits 
 and objects of the present work considered cannot be 
 done satisfactorily ; but it seems worth while to develop 
 more fully, even though in an incomplete way, the view 
 of the origin of Agnation which has just been shadowed 
 forth.
 
 CHAPTER XIII. 
 
 THE ORIGIN OF AGNATION. 
 
 ROMAN jurists are agreed that, at first, the Patrician 
 familia was identical with the Patrician gens, and 
 agnatio the same thing with gentilitas; though after- 
 wards when the gentile bond had grown feeble and the 
 family bond had grown strong, and connections had 
 come to be formed between gentiles and clients, the 
 term familia came to be restricted to bodies of kindred 
 within the gens, and it became necessary to dis- 
 criminate between gentilitas and agnatio.* If this 
 view be accepted and there appears to be no reason 
 against accepting it since gentilitas and agnatio were 
 originally the same, it is obvious that it is the gentile 
 bond we ought to look to when seeking an explanation 
 of the peculiarities of Agnation, 
 
 Of the gens thus much is certainly known it was 
 a body of persons having a common name and common 
 religious rites, and considered to be related to one 
 
 Q ' 
 
 another. And very little besides is certainly known of 
 
 * " Family " is the common term nearly all the world over for a 
 " tribe of descent," or stock-group indicated by a common totem.
 
 206 THE PATRIARCHAL THEORY. 
 
 it. It was generally believed to consist of persons 
 descended from a common ancestor ; but this was only 
 a theory to account for the connection in the same body 
 of persons who were considered relatives and had a 
 common name and worship. Children belonged to the 
 gens of their father ; and it was natural therefore 
 the gentiles being considered to be unquestionably 
 relatives to suppose that there was a first father 
 from whom they all were descended. But relationship 
 throughout the gens could not be traced, and, though 
 eponyms had been formed for most of the gentes, it is 
 undoubted that the common ancestor was in every case 
 mythical. The gens was really of unknown origin, of 
 unknown history. 
 
 A common name usually taken from an animal or 
 plant a common worship, or common object of reve- 
 rential regard, and a bond of inherited relationship 
 denoted by these are the marks of totem kinship (of 
 which more by-and-by). And an alternative theory of 
 the origin of the gens is that it was a body of totem 
 kindred in which gentile connection had come to be 
 taken from the father. It can be shown of the majority 
 of the Eoman gentes that they have every mark of 
 having been groups of this type. 
 
 There are indications that the gentes were exo- 
 gamous, but it must be owned that the evidence to 
 that effect is not direct and not distinct. Plutarch 
 tells us what, with the knowledge of early custom we 
 now possess, can be easily received, that in early times 
 the prohibition of marriage extended as far as the tie of
 
 THE ORIGIN OF AGNATION. 207 
 
 blood ;* and, if this be received, it involves since the 
 gentiles considered themselves to be of the same blood 
 that there could not be marriage between persons of 
 the same gens. That, among the early Roman marriages 
 recorded, there is an absence of marriages between 
 persons of the same gens (this seems to be the fact), 
 though much is not to be made of it, is consonant with 
 this ; for it tends to show that the gentes were exo- 
 gamous in practice, whether required by rule to be so 
 or not. Exogamy, however, in its widest sense of a 
 prohibition of marriage between persons of the same 
 stock-connection or kindred, has been so common that we 
 may, upon imperfect evidence, venture to think that the 
 gentes were exogamous. Even the direct assumption that 
 they were exogamous would not be a violent assump- 
 tion. That there was anciently extensive prohibition of 
 marriage between relatives is clear enough, and not- 
 withstanding that we have been told that, in ancient 
 times, people could not be relatives of their mother's 
 relatives at Rome prohibition extended to relatives on 
 the mother's side. The first marriage of a Patrician at 
 Rome within the seventh degree of cognation excited a 
 sedition in the city.f 
 
 * Plutarch, " Koman Questions," vi. 
 
 t See "Anecdoton Livianum," in Hermes, vol. iv., pp. 371-2, 
 Berlin, 1870. "Livius libro vicesimo. P. Celius patricius primus 
 adversus veterem morem intra septimum cognationis gradum duxit 
 uxorem. Ob hoc M. Kutilius plebeius sponsam sibi prseripi novo 
 exemplo nuptiarum dicens seditionem populi concitavit adeo ut patres 
 territi in Capitolium perfugerent." 
 
 Paul Ivrueger and Th. Mormnsen append to this short articles
 
 208 THE PATRIARCHAL THEORY. 
 
 Let us take it, then, hypothetically that the gens was 
 anciently exogamous, and that the gens and the familia, 
 gentilitas and agnatio, were anciently identical. And now 
 let us see what has happened in the numerous cases in 
 which, with exogamy for marriage law, kindred or family 
 (gentile) connection has been taken, not from the father, 
 as among the Romans, but from the mother. 
 
 In these cases, relationship being transmitted only 
 through women, those persons only were considered to 
 be relatives who were connected (or believed to be 
 connected) with one another through female descents ; 
 and, in addition to the prohibition of marriage, which 
 extended as far as relationship extended, there arose out 
 of relationship another important consequence. The 
 blood-feud united those who were considered to be 
 relatives to one another, for mutual assistance and 
 defence, and for revenge of injuries ; and it helped 
 powerfully to bind them to one another, and to sever 
 them from other people. A man could belong even 
 after circumstances allowed of the fact of paternity 
 being perceived and contemplated only to that one 
 kindred of which he was born, to which he was bound 
 by the blood-feud, within the limits of which marriage 
 was forbidden to him. 
 
 And, since a man could marry only a woman of a 
 different kindred from his own, and children were of the 
 kindred of their mother, a man's children were never of 
 his kindred they were "lost to his family." There 
 
 containing references to corroborative authorities regarding forbidden 
 degrees of cognation.
 
 THE ORIGIN OF AGNATION. 209 
 
 could be no extension of a kindred through a male ; it 
 was the wife's kindred that was extended by marriage. 
 We find in these cases the family in the sense of 
 stock or kindred being, however, the unit spoken of 
 everything that Sir Henry Maine has said of Agnation 
 reversed as between the sexes that it has all to be 
 altered by the substitution in proper places of male for 
 female and female for male. Instead of a woman being 
 the " terminus of the family " (mulier finis familice), 
 a man was the terminus of the family. A male name 
 " closed the branch or twig of the genealogy in which it 
 occurred." None of the descendants of a male were 
 included in the notion of this relationship. It is clear 
 that the system of counting kinship through females 
 only the custom which made children be of the 
 kindred (gentile) connection of their mother and the 
 prohibition of marriage between persons of the same 
 kindred produced those consequences without any 
 aid from the powers which mothers possessed over 
 their children. 
 
 And is it not suggested to us that all other things 
 remaining the same a change in kinship which would 
 make children of the kin or gentile connection of their 
 father would yield for the body of kindred (familia or 
 gens) the statements that would be denoted by Agnation, 
 when equivalent to gentilitas, without aid from the 
 powers which fathers possessed over their children ? 
 
 To proceed, however, a man's nearest acknowledged 
 relatives are everywhere his nearest heirs ; and in general 
 acknowledged relatives only are anywhere a man's heirs.
 
 210 THE PATEIAEGHAL THEOEY, 
 
 We find, accordingly, in the more advanced communities 
 in which the system of counting kinship only through 
 females, and exogamy have prevailed, that a man's own 
 children, not being considered his relatives, did not 
 inherit from him or from his family. Rights of in- 
 heritance arose to children in the families of their 
 mothers. A man's nearest relatives were, first his 
 brothers by the same mother, and then the children 
 of his sisters by the same mother ; and they were his 
 heirs in that order.* 
 
 Here we have for the family proper a rule of 
 inheritance which is the counterpart of Agnation. 
 
 And is it not suggested to us that change sufficient 
 to make children of the gentile connection of their 
 father, would in time yield the rule of inheritance 
 which was the substantial thing in Agnation ? 
 
 It is now, indeed, easy to see how, it being taken 
 that in early times the familia or gens was exogamous, 
 and agnatio and gentilitas indifferently expressed the 
 gentile relationship, every feature of Agnation can be 
 fully and satisfactorily accounted for, without any other 
 condition than that relationship was limited to the 
 
 * Among the Ashantees, according to Bowdich, a man's son might 
 inherit on the failure of sisters' children. Failing a son, the heir was 
 the chief slave. The son, that is, on the failure of the proper heirs, 
 was preferred to the chief slave. But in the Fantee country the chief 
 slave excluded the son. The Fantees and the Ashantees were really 
 the same people ; and the place given to the son in succession among 
 the latter must be taken as indicative of a transition in kinship. It 
 should be said that we have no direct evidence of exogamy among the 
 Ashantees. But there are many circumstances from which it can be 
 inferred.
 
 THE ORIGIN OF AGNATION. 211 
 
 familia or gens as, with the system of kinship through 
 females only, it is limited to the kindred. 
 
 For by effect of this restriction of the gentile rela- 
 tionship (gentilitas or agnatio), a person could belong to 
 one gens or familia only, and could not have relation- 
 ship with persons in any other. And as, by effect of 
 exogamy, a woman could not marry a man of her own 
 gens, and as children belonged to the gens of their father, 
 the children of a married woman would not belong to 
 the gens of her birth, and therefore could have no rela- 
 tionships in it. They would be lost to the familia or 
 gens of her birth, and lost also to the sub-family in it, 
 or family proper, to which she belonged by birth. A 
 married woman would thus be finis families the 
 terminus of the family, both in the larger and in the 
 restricted sense of the word ; neither the body of 
 kindred nor the family proper could receive any exten- 
 sion through its women ; the children of a married 
 woman would have no right of inheritance in the gens 
 of their mother's birth, or in the family in it to which 
 she belonged being by their own birth strangers to 
 both and all this without any reference to the powers, 
 greater or smaller, which male parents might possess 
 over their descendants. When bodies of relatives who 
 could trace their relationship to one another (familiw in 
 the later sense of the word) grew up into distinctness 
 and importance within the gens, their relationships also 
 would be subject to the limitations of the gentile rela- 
 tionship would be what the gentile relationship deter- 
 mined them to be ; and the children of a married 
 
 p 2
 
 212 THE PATRIARCHAL THEORY. 
 
 woman would be cut off from relationship with the 
 familia to which their mother belonged by birth, and of 
 course since only persons who are considered to be a 
 man's relatives can be his heirs from all rights of 
 inheritance in it. 
 
 And, since a man's nearest relatives are everywhere 
 his nearest heirs, and children, when acknowledged as 
 relatives, are evidently nearer than brothers, a man's 
 heirs would be first his sons and possibly his daughters 
 but not married daughters if marriage passed women 
 into another gens, or if women were settled with on 
 marriage ; next, his brothers by the same father ; and 
 after these the more distant members of his familia. It 
 would not be surprising, but might rather be expected, 
 that the body of the kindred there being between all 
 within it admitted though untraceable relationship 
 should come in as ultimus hceres. 
 
 Rules as to relationship and inheritance would thus 
 arise which would present all the features of Agnation. 
 They would be established while the gentile system was 
 in full vigour, but there would be nothing to disturb 
 them, as rules for inheritance, while it was gradually 
 waning, and, as such, they would then be gaining in 
 prescriptive force. Nor would it be possible to question 
 them as rules for inheritance until what was most 
 distinctive of the gentile system had disappeared, until, 
 gentile connection having ceased to be of importance, 
 the limitation of relationship to the gens had become 
 obsolete. Even then, that they had immemorially 
 formed the settled usage for the distribution of in-
 
 THE ORIGIN OF AGNATION. 213 
 
 heritances would give them a hold upon men which 
 might not be easily shaken. They could not per- 
 manently survive the conditions which produced them ; 
 they were sure to give way under the dissolving influ- 
 ence of admitted relationship. But they might survive 
 those conditions long. 
 
 This account of Agnation is consistent with the 
 history of the breakdown of Agnation at Rome, which 
 the account of it which derives it from Patria Potestas 
 is not. The gentile system had ceased to be of import- 
 ance before Agnation was seriously attacked ; while 
 Patria Potestas survived Agnation. And, considering 
 that the conditions from which Agnation is here derived, 
 with the difference of gentile connection being taken 
 through the mother instead of through the father, un- 
 questionably account for the exclusion of persons con- 
 nected with a family or a kindred through males only 
 from relationship with it, it seems difficult seriously 
 to doubt that they give us so far the true explana- 
 tion of Agnation. It is scarcely necessary to recall that 
 there is authority for believing that gentilitas and agnatio 
 were in early times coincident, and that there is some 
 evidence to show that the gens was in early times 
 exogamous. These points admitted, it seems clear that, 
 to account for Agnation, we must further believe and 
 there is supplied us, in the parallel operation of kin- 
 ship through females only, sufficient reason for believing 
 that it was preceded by relationship strictly limited 
 to the gens, just as with kinship through females only, 
 relationship is limited to the kindred.
 
 214 THE PATRIARCHAL THEORY. 
 
 When we go on to consider the conditions which 
 appear to have produced Agnation, it may be well to 
 point out that, if we try to show that Agnation was 
 arrived at by a progress from relationship through 
 female descents, exogamy, as a factor in accounting 
 for it, is nothing but an aid. For exogamy is found 
 with kinship through females only, and it is clear 
 that, with that system, it took the leading part in 
 marking kindreds off from one. another. That the 
 feeling against marriage between persons considered to 
 be of the same kindred thus established maintained 
 itself when new relationships had come to be acknow- 
 ledged, accounts for exogamy being found with kin- 
 ship through males. On the other hand, if we try 
 to explain Agnation as a primary system of kinship, 
 we have as a preliminary to account for exogamy to 
 show how it would arise where kinship was acknowledged 
 through males. 
 
 As to the limitation of relationship to a group, 
 some reasons have already been given for thinking 
 that it did not belong to the earliest period of rela- 
 tionship. But that is beyond doubt if it has to be 
 admitted that Agnation arose in exceptional cases only, 
 and if it be as an exceptional form of kinship that 
 Agnation has to be accounted for. 
 
 Now, as a fact, whether in ancient times or in 
 modern, cases of Agnation arc scarcely to be met with. 
 It will appear hereafter that Sir Henry Maine, though he 
 thinks it has been almost universal, has never been able 
 to find it anywhere except in Rome. And there is no
 
 THE ORIGIN OF AGNATION. 215 
 
 reason to think that it ever was common. On the con- 
 trary, there is reason to think that it always has been 
 uncommon ; for, to judge by the lasting power shown 
 by the system of kinship through females only, to judge 
 by the lasting power shown by Agnation itself at Eome, 
 it would have lasted well wherever it prevailed, and, 
 had it once been common, would now and then have 
 been heard of. An approach to certain effects of it 
 could not but be made wherever the clan or kindred 
 had anything like coherence. That they stood together 
 for mutual defence and for revenge was enough to bind 
 the men of a kindred to one another, and to give the 
 group relationship the first importance ; and, even apart 
 from that, it was inevitable that men should cleave to the 
 body in which they were born, in which they lived, and 
 with which most of their rights, duties, and interests were 
 bound up. It is intelligible, too, that, where a kindred 
 was settled upon land, it should deny to the outside 
 relations of individuals any interest in its land ; and 
 that, in so far as the interests of its members were con- 
 joined or common, it should deny to the outside rela- 
 tions of individuals any right to become sharers in them. 
 But all this and similar things have happened, without 
 denial of outside relationship, and with the recognition 
 of effects attaching to it without it being at all in- 
 volved that relationship could be extended only through 
 male descents in other words, without Agnation. 
 
 And if it be as an exceptional form of kinship 
 that Agnation has to be accounted for, whatever its 
 early history, it cannot have been primary. If, while
 
 216 THE PATRIARCHAL THEORY. 
 
 relationship was still young, kinship through males, when 
 it began to grow into a system, had confined relation- 
 ship to male descents through some influence emanating 
 from it which acted immediately or speedily, Agnation, 
 instead of being exceptional, would at first have been 
 as common as kinship through males. And we may be 
 sure that, in that case, it would long have continued 
 common. For there would have been the influence 
 which produced it to maintain and establish it ; it 
 would have had prescription in its favour ; and the 
 circumstances of early tribes would have tended to save 
 it from serious attack. 
 
 Though not primary, however, it may conceivably 
 have been arrived at otherwise than by a progress from 
 the system of kinship through females only. It may have 
 been produced among peoples who from the first had a 
 system of kinship through males by influences which 
 were not general, or which, at any rate, did not often 
 prevail, which came into operation among them as time 
 went on. But there are facts and considerations which 
 favour the view that male kinship in general has been 
 arrived at by a progress from relationships through 
 female descents. The more important may be indicated. 
 
 That relationship happened to be reflected upon 
 when the fact of paternity was obscure or uncertain 
 seems to be the only possible explanation of kinship 
 being in any case counted through women only ; and it 
 seems to be the only possible explanation of this kinship 
 being anywhere developed into a system that, where it 
 was so developed, the fact of paternity continued long
 
 THE OEIGIN OF AGNATION. 217 
 
 to be obscure or uncertain. If this be so, what might 
 be expected to happen when, where a system of kinship 
 through women was established, the fact of paternity 
 became clearly observable, so that it might be the sub- 
 ject of reflection ? May we not feel sure that it would be 
 reflected upon, and that, sooner or later, by one process 
 or another, effect would be given to its suggestions of 
 relationship ? That the establishment of a system of 
 kinship upon which kindreds had been moulded would 
 retard the acknowledgment to practical effects of the 
 new relationships need not be questioned ; on the con- 
 trary, one might be prepared to find that, among peoples 
 governed by custom, devices would have to be hit upon 
 whereby to procure for them practical recognition. But 
 when reflection, which had previously established a 
 system of kinship through the weaker parent, had 
 shown that there was kinship through the stronger, we 
 need not doubt that means would in general be found 
 of ensuring the recognition of this kinship. And, once 
 recognised, it would, almost as a matter of course, 
 become forthwith the more important of the two. 
 
 It is probable, in short, that, wherever a system of 
 kinship only through females had been established, there 
 would be, sooner or later, a progress from it to a system 
 of kinship through males, and that the latter would 
 thereafter become the predominating kinship. 
 
 It is obvious, too, that this progress may have 
 taken place frequently, commonly, even universally, 
 and evidence of it having taken place be wanting, or 
 not readily recognisable for what it is while the facts
 
 218 THE PATRIARCHAL THEORY. 
 
 about kinship, superficially regarded, might seem to 
 support a totally different view of its history. Transi- 
 tion facts disappear. In general, we cannot expect to 
 have any record of them ; and, even when we have, 
 their significance is not always above dispute. 
 
 If transition facts can be found, however, and their 
 significance be established, they will form a class of 
 facts which really give us light as to what the history of 
 kinship has been in the cases which exhibit the condi- 
 tions which are present in them in cases in which the 
 factors which appear in them have been concerned in 
 the moulding of society. And there will be nothing to 
 make against the indication they give us as to what the 
 early history of kinship has been in such cases. They 
 will be evidence and the best evidence there is upon 
 the matter to show that, in similar cases, there has 
 been a progress from kinship through females to kinship 
 through males. And probability will incline to the 
 conclusion which they suggest. 
 
 Now, in a few a very few cases, what appear to 
 be unmistakably transitional stages of kinship, or steps 
 on the way to a change of kinship, are found among 
 exogamous peoples. They occur among peoples who 
 have advanced very slowly, but it is only among such 
 peoples we could hope to find them. If, however, they 
 show that transition has taken place, this is enough, as 
 the matter stands, to raise a probability that transition 
 has taken place wherever we find exogamy with male 
 kinship. 
 
 There is a more numerous class of facts better
 
 THE ORIGIN OF AGNATION. 219 
 
 known, too, than those just referred to which can be 
 understood on the view that there has been transition 
 from a system of kinship through females, and appear 
 inexplicable upon any other view. That, for example, 
 early Hebrew custom permitted marriage between such 
 relations as brother and sister german, and uncle and 
 niece (brother's daughter), brother and sister uterine not 
 being permitted to marry one another, is intelligible on 
 the view that, relationship through the mother being 
 recognised, relationship through the father was not to 
 all effects established. And it seems as if nothing else 
 could be made of it. It need scarcely be said that rela- 
 tionship through the father was most fully recognised 
 among the Hebrews at a later period. Hindoo marriage 
 custom (according to the commentator, Kulluka, at any 
 rate) also presents a fact which seems explicable only on 
 the view of a transition having taken place, while show- 
 ing conclusively as to Agnation that it was unknown 
 among the early Hindoos. Among the Hindoos a man 
 could not marry in the primitive stock of his father nor 
 (on the authority of Kulluka) in the primitive stock 01 
 his mother. The gotra or primitive stock, it should be 
 said, has been in times that are known to us large and 
 scattered, connection with it being indicated only by a 
 common family name. That marriage within it was 
 forbidden on account of it being the primitive stock 
 that is, on the score of relationship is quite clear. 
 And it is not to be doubted that the prohibition of 
 marriage within it could not have been introduced in 
 times when the gotra kinship was regarded as very
 
 220 THE PATRIARCHAL THEORY. 
 
 remote not to be doubted that it draws back to the 
 primitive period of clanship, when the gotra was com- 
 paratively small and comparatively coherent. Now 
 Agnation denies the relationship of a woman's children 
 to her brothers. And here among the Hindoos it being 
 to the last degree unlikely that the fact should have been 
 invented, and certain that, if it was fact, it must have 
 been of most ancient origin we find the relationship of 
 a woman's children to every person of her stock ad- 
 mitted, so that they were prohibited from marrying any 
 person of that stock. This rule is absolutely irrecon- 
 cilable with Agnation Agnation cannot have prevailed 
 where it came to be established. On the other hand, it 
 is easily understood on the view that there had been a 
 progress from the system of kinship through females 
 only. On that view there would have been prohibition 
 of marriage within the mother's gotra before connection 
 through male descents was recognised and prohibition 
 of marriage extended to those so connected. And it is 
 an obvious explanation of this rule that the prohibition 
 of marriage was continued for the old kindred, that of 
 the mother, when it was established for the new kindred, 
 the kindred of the father.* 
 
 * The text of Manu (Manu iii. 5) prohibits marriage only in the 
 gotra of the father. As stated above, it is in the gloss of the com- 
 mentator, Kulluka, that the prohibition is extended to the gotra of 
 the mother. Kulluka, however, is a very great authority. It is not 
 to be supposed that he made so sweeping an addition to the text with- 
 out ample reason, nor without being sure that the validity of the pro- 
 hibition would be acknowledged as readily in the one case as in the 
 other. The prohibition of marriage in the mother's gotra, it may be
 
 THE ORIGIN OF AGNATION. 221 
 
 But, further, it appears that the family system 
 amongst many peoples which have been exogamous is 
 far more intelligible upon the view that it is the issue 
 of a progress from a system of kinship through females 
 only to a system of kinship through males than it is 
 upon any other view. And there are single facts which 
 seem almost to prove this for example, the Levirate. 
 The Levirate has prevailed far and wide. And, if of 
 polyandrous origin, it carries us far back towards the 
 social condition in which kinship could be counted 
 through females only. 
 
 It seems worth while to add that we find in marriage 
 among some of the greatest of ancient nations a custom 
 which while no doubt serving other uses also would 
 be effectual as a device for reconciling an ancient system 
 which made children of the same kindred with their 
 mother with the resolve that children should be counted 
 of the family and stock of their father. The wife, at 
 marriage, was taken into the family and stock (gens, 
 gotra) of her husband, and admitted to his sacra, to 
 participation in his religious rites ; and she ceased 
 thereafter to belong to the kindred and worship of her 
 birth. Being, by this device, of her husband's family 
 and stock, her children even if they took gentile and 
 family connection from her would be of the gentile 
 and family connection of their father. 
 
 said, is quite consistent with all we know about the Hindoos. It has 
 been remarked already that, where Manu mentions father and mother 
 together, he often mentions the mother first. And, in some of the 
 most important passages in Manu relating to parent and children the 
 text speaks of the mother only.
 
 222 THE PATRIARCHAL THEORY. 
 
 If the transition did take place, the problem for 
 those by whose means it was carried out would have 
 been : Given kindreds already formed, bound together 
 by the blood-feud and the prohibition of marriage 
 within the kindred kindreds connection with which 
 had been transmitted through female descents only 
 how to secure that connection with the kindred should 
 come to depend upon male descents ; in other words, 
 how to secure that children should be counted of the 
 same kindred with their father instead of being counted 
 of the kindred of their mother. Men would have no 
 reason for shaking themselves free of the existing 
 kindreds ; nor is it intelligible how they could do it if 
 they wished. For the kindred would be the basis of 
 society as it existed ; acknowledged relationship would 
 be limited to it, and the rights and duties connected 
 with relationship also ; every man would belong to one 
 kindred, to which he was bound by the strongest ties 
 at the time acknowledged. To shake it off could hardly 
 even be thought of; and all that men would have a motive 
 for doing, would be to change the kindred or gentile con- 
 nection of their children who, as matters stood, would 
 not be the relatives of their own fathers to wrest 
 children from the kindred of their mothers, and to win 
 them for the kindred of their fathers. That anyhow 
 accomplished, there would be no obstacle to the full 
 affiliation of children to their father. 
 
 It might be expected that various methods or devices 
 for effecting that would be employed and there are 
 indications of this having been the case. Being of the
 
 THE ORIGIN OF AGNATION. 223 
 
 nature of devices, they could succeed only because men 
 would generally be favourable to the object for which 
 they were used. And, though an effect upon the kindred 
 of the children would be necessary to their succeeding, 
 or would necessarily take place if they succeeded, it 
 would be the individual members of kindreds by whom 
 they would be employed. It is easy to see that it 
 would be indispensable to them all alike that a woman's 
 relatives should cease to have rights over her children. 
 Perhaps it might be expected that they would fre- 
 quently involve that the woman's relatives should 
 give up to the husband the woman and her children, 
 and cease to have to do with them. This could 
 not be accomplished without contract. And it appears 
 to have been in many cases accomplished by means 
 of contract alone. 
 
 The most effectual method, no doubt, would be for 
 the man to be able, in virtue of the contract he had 
 made, to bring the woman into his own clan; and where 
 religion was influential, probably this was always, in the 
 long run, found necessary. This being done, a man's 
 children though taking gentile connection from their 
 mother would be, by birth, of his own kindred, among 
 his acknowledged relatives, and there would be nothing 
 to prevent him from having them for his children in the 
 fullest sense. A device by means of which they would 
 be, by birth, of his own gentile connection would make 
 them his children. They would become his children 
 through being by birth of his clan. 
 
 It is in a class of cases in which religious influence
 
 224 THE PATRIARCHAL THEORY. 
 
 was slight, and made little or no difficulty, that the 
 connection with the mother's kindred seems to have 
 often been got rid of by means of contract alone. The 
 mother's relatives having been satisfied, their non- 
 interference having been secured, the husband got the 
 children for his own, and was free to count them of his 
 own family and clan. 
 
 Whatever method was used, however, the gentile 
 system being what it was, when a man won his children 
 for his clan he would win them for himself, and when he 
 won them for himself, he would win them for his clan. 
 And, as the change went on, the gentile bond would 
 apply the old restraints and obligations to the kindred 
 as now recruited. What wonder, nevertheless, if, for a 
 time at least after the change was generally accom- 
 plished, the restraint on marriage should extend to the 
 mother's kindred as well as the father's ? We see, in- 
 deed (by such marriages as those of Abraham and Nahor), 
 that, in the case of the Hebrews, the restraint as to the 
 mother's kindred was alone effective for a time. And, 
 though children would be now of the clan of their 
 father, and not of the clan of their mother, what wonder 
 if, in general, relationship to the mother's relatives were 
 to continue to be to some effects admitted ? The gentile 
 bond would still tend (so to speak) to deny relationship 
 to those who were outside it and, therefore, to lead up 
 to Agnation. But the associations of the old relation- 
 ship would resist it. The restraint on marriage, where 
 that continued for the mother's kindred, would, of course, 
 be an admission of relationship outside it.
 
 THE ORIGIN OF AGNATION. 225 
 
 The more completely children were severed, by the 
 circumstances surrounding marriage, from the mother's 
 kindred, however and, marriage everywhere tending to 
 fall into some convenient uniform fashion, the kindred 
 would be recruited through all the men contained in it 
 under much the same conditions the more tightly 
 could the gentile bond be drawn. Where the resistance 
 to it was slight it would succeed, and the kindred would 
 be on the way to Agnation. What the circumstances 
 were which would favour it may perhaps be gathered if 
 we bear in mind what Roman marriage was. 
 
 It is obvious that the elaboration within the present 
 work of the view now presented is not to be thought 
 of. But some further consideration of the conditions 
 which may be believed to have affected early kindreds 
 seems almost necessary, and may perhaps be ventured 
 upon. What, in a general way, these conditions were 
 may be gathered by observing the conditions affecting 
 those rude societies which are known to us. It will be 
 convenient to take for examples societies which have 
 been observed in modern times. 
 
 Wherever, with exogamy, kinship is counted 
 through females only, there are at least two kindreds 
 represented in every household the husband is of one 
 family or kindred and his wife and children are of 
 another. And, if he has more than one wife, there may 
 be represented in the household an additional kindred 
 for every additional wife. An interfusion of kindreds 
 in every household one represented by the husband 
 only is in all cases inevitable.
 
 226 THE PATRIAECHAL THEORY. 
 
 In the simpler or ruder examples of this kind, how- 
 ever those, for instance, which occur among the 
 Australians or North American Indians the degree 
 of interfusion is very remarkable. It commonly happens 
 that all the separate communities local tribes, or, to 
 use the North American term, nations which occupy 
 an extensive district, are, in the main, similarly com- 
 posed, each being made up of persons belonging to the 
 same four or six or more separate kindreds. The local 
 tribe or "nation" the body of people who live, and 
 hunt, and make war together is not itself a kindred 
 group ; it is anything but that. It is an agglomeration 
 of persons belonging to several separate families or 
 kindreds. Persons belonging to different kindreds are 
 contained in every household. Persons belonging to four 
 or six or more kindreds make up all the households in 
 the community. Even a single household, when there 
 are several wives, or when it contains three or four gene- 
 rations every man marrying into another kindred than 
 his own, and his children being of the kindred of their 
 mother may contain representatives of every kindred 
 included in the " nation." 
 
 And, on the other hand, as has been said already, 
 the kindreds which may be all represented in a single 
 household in one community are commonly found to be 
 the components of all the communities in an extensive 
 district. These communities, moreover the local tribes 
 or nations are not fortuitous combinations, or of 
 recent origin. They believe themselves immemorially 
 to have been composed much as they are at present.
 
 THE ORIGIN OF AGNATION. 227 
 
 The families or kindreds, which make up one and all of 
 them, consist each of persons related to each other 
 through female descents. Of course, their relationship 
 is mostly untraceable. All of a kindred, nevertheless, 
 acknowledge relationship, and the restraints and obli- 
 gations of common blood. Whether in the same com- 
 munity or not, they do not intermarry with each other. 
 And the blood-feud unites them for mutual assistance, 
 reparation of injuries, and revenge. The latter obli- 
 gation, however, is frequently confined to those of the 
 kindred who are of the same community or local tribe. 
 
 How then are the members of a kindred so diffused 
 recognisable by each other as belonging to the kindred ? 
 They have what serves them for a common name, but is 
 much more than that. To use a convenient (though in 
 some respects unfortunate) American term, they are of the 
 same totem, and their relationship is indicated to them by 
 the totem. The totem is some natural object commonly 
 an animal or plant, sometimes a heavenly body ; and they 
 figure it upon their bodies (whence tatooing), and call 
 themselves by its name. All of one kindred are Bears ; 
 all of another Wolves ; all of a third Suns, and so on ; 
 and the totem being taken from the mother, a stranger 
 is at once recognised by those who bear the same totem 
 mark and totem name with him as their relation con- 
 nected with them (though in an unknown way) by 
 female descents. The totem or kindred name is the 
 more prominent because personal names are usually 
 carefully concealed. It more than serves the uses of a 
 family name. It is the totem mark alone that is placed 
 
 Q 2
 
 228 TEE PATRIARCHAL THEORY. 
 
 upon a grave. It is with it alone that, in intercourse 
 with Europeans, the representative of a kindred signs a 
 treaty. 
 
 But, further, the kindred believes itself to be some- 
 how descended from that (the animal let us say) which 
 is its totem, and to be still mysteriously connected with 
 it. They reverence it as their protector and friend. 
 They have what must be called a religious regard for 
 it. They consider an individual of the species as their 
 relative, and call it brother. And they will on no 
 account hurt, or kill, or eat it. Totem kindreds, at a 
 somewhat more advanced stage for it may be said that 
 the totem can be traced almost everywhere often try, 
 by filing, distorting, or pulling out teeth, by arrange- 
 ment of the hair, and other devices, to indicate by out- 
 ward resemblances their connection with the animal 
 which is their totem. Sometimes they believe that it 
 gives them counsel in dreams. They are constantly 
 found believing that they will take its animal form 
 after death ; and varieties of what may be called the 
 were-wolf superstition also testify to their conviction 
 that their human form is interchangeable with the 
 animal form of the totem. When people become specu- 
 lative, a great totem is often made out to have been the 
 creator of all things. And then, strong as the objection 
 to killing or eating it is, we find that frequently its body 
 and blood are partaken of eucharistically. In the very 
 rudest cases, however, the kindred besides getting from 
 the totem a common name and a test of inherited re- 
 lationship have in it the nucleus of a common worship.
 
 THE ORIGIN OF AGNATION. 229 
 
 Totemism is not our subject ; but it cannot be amiss 
 to recall that a hypothesis has been framed to account 
 for the animal and plant worship of ancient nations 
 on the view that they all passed through the totem 
 stage ;* it being shown, among other things, that that 
 worship, in numerous cases, exhibited the leading notes 
 of totemism ; that animals, for example, were worshipped 
 by tribes of men who were named after them and 
 believed to be of their breed, and never eaten unless 
 eucharistically. And on the other hand, it may be 
 pointed out that the totem unquestionably carries us 
 back at least to the earliest days of relationship to a 
 period while there was yet no prejudice against the 
 marriage of kindred persons. 
 
 Scattered throughout various separate communities 
 as the kindreds we have been noticing are, it is almost 
 impossible to resist the suggestion that there was once 
 a separate coherent nucleus for each. The communities 
 are just what the interfusion by exogamy (or capture) 
 and the system of kinship of so many originally sepa- 
 rate, or at least coherent, groups, each of which con- 
 sidered itself a kindred, would make them. And it 
 seems absolutely impossible that the totem bond should 
 unite a kindred scattered throughout different com- 
 munities, as it now does, unless it had marked it off from 
 other kindreds before systematic interfusion of kindreds 
 
 * " The Worship of Animals and Plants." By J. F. M'Lennan. 
 Fortnightly Review, 1869-70. See also "Animal Worship and 
 Animal Tribes among the Arabs, and in the Old Testament." By 
 Professor W. Kobertson Smith, Journal of Philology, vol. ix.
 
 230 THE PATRIARCHAL THEORY. 
 
 had begun. Without it all relationship, except what 
 memory could take note of, would now be lost sight of. 
 While it yet was not, apart from memory, all relationship 
 must have been lost sight of. It is practically all there is 
 to preserve the tradition of relationship. We may take 
 it, therefore, that such relationship as preceded it, if any, 
 was lost sight of. And to follow the blood as it does 
 now, it must have marked it at a common source. The 
 totem therefore was precedent to that by which the 
 interfusion of kindreds is regularly caused, exogamy ; 
 and there can be no trace left of any interfusion made by 
 the capture of women or otherwise which preceded it. 
 We are thus carried back to a time when there were 
 bodies of kindred having each a common totem, which 
 were necessarily bound together for common defence 
 that is, essentially by the blood-feud which were con- 
 nected by that system of relationship which first was 
 formed, and which had no objection to the intermarriage 
 of related persons. 
 
 Among the North Americans, and even among the 
 Australians not unfrequently among the former, only 
 rarely among the latter we find also totem kindreds in 
 which kinship through males is acknowledged, and the 
 totem or gentile connection is taken from the father. 
 Where information about the marriage law is accessible, 
 it is that the kindreds are exogamous that is, that 
 persons of the same totem may not marry ; but we 
 come also upon arrangements to prevent marriage 
 between near relations who are not of the same totem. 
 Of such cases, when there is nothing to suggest that
 
 THE ORIGIN OF AGNATION. 231 
 
 there has been a recent change of kinship, the probable 
 explanation is that a general practice of polyandry in 
 the Tibetan form with which there is certainty as to 
 the father's blood, though not as to the father at an 
 early stage made a transition to kinship through males 
 natural and easy. That there should anywhere be 
 kinship only through the mother can only be explained 
 by there having long been uncertainty about fatherhood ; 
 and in " Primitive Marriage," reason was shown for 
 believing that that kinship must in numerous cases have 
 given way before the certainty of the father's .blood 
 obtained with Tibetan polyandry. * 
 
 In some of the cases now being spoken of, however, 
 there are indications that there has been a change of 
 kinship at a period not remote. In America, for ex- 
 ample, the Algonquins, on the whole, have kinship 
 through males ; but there are detached portions of that 
 people not very long separated from the main body 
 among whom the totem is said to be still taken from the 
 mother. In America, at least, the circumstances of the 
 family have long been not unfavourable to the acknow- 
 ledgment of relationship between father and children. 
 To prevent it there has been the force of custom the 
 fact that children have always been of the totem of their 
 mother. But, once the desire to have their children for 
 themselves and for their own totem became general 
 among men, some way of giving effect to it was sure 
 to be found ; and we may be content with knowing 
 
 * Studies in Ancient History. By J. F. M'Lennan. London, 
 Quaritch, 1876, " Primitive Marriage," chapter viii.
 
 232 THE PATRIARCHAL THEORY. 
 
 this where we find circumstances suggestive of a recent 
 change. 
 
 One thing we can be confident about. It has been 
 shown that the totem is older than exogamy. That, 
 like the totem kindreds previously noticed, the kindreds 
 now spoken of have acquired the prejudice against 
 marriage within the kindred is enough to show that, up 
 to a certain point at least until after exogamy had been 
 established the circumstances and history of both were 
 substantially the same. 
 
 The change in kinship, supposing it to have taken 
 place in the cases just noticed, was sure to add greatly 
 to the cohesion of the kindred, and proportionally to 
 weaken the cohesion of the local tribe or "nation." 
 
 With kinship through females, in a household which 
 contained three or four generations every totem in the 
 community might be represented. The kindreds were 
 so interfused that close observation was needed to show 
 that they were not all one body of relatives. Family 
 ties mitigated the virulence of the chief disruptive 
 force, the blood-feud. And, indeed, the community 
 could not be broken up without a break-up of every 
 household included in it. With the new kinship all of 
 a household, except the wives, became of one totem, and 
 were one for the purposes of the blood-feud. It was 
 now natural for the households that were bound together 
 by the blood -feud to draw together and form separate 
 bodies within the community. That done, they were 
 totem groups again proper clans or tribes of descent, too 
 with nothing except the conditions found to be indis-
 
 THE ORIGIN OF AGNATION. 233 
 
 I 
 
 pensable for living with one another to check the dis- 
 ruptive influence of the blood-feud. And this is what 
 we usually find with clanship where, gentile connection 
 being taken from the father, the community or local 
 tribe is made up of a number of separate clans. 
 
 Since we have seen that, with the earlier kinship, 
 the same kindreds often enter into adjacent com- 
 munities, the view that there has been a change of 
 kinship is greatly favoured by our finding, as may 
 often be done, that the same clan is a component of 
 more communities than one is supported by it in 
 proportion to the difficulty of otherwise accounting 
 for such a fact. And it is favoured also by our 
 finding as may be done in Australia the segregation 
 of kindreds from each other within the same community, 
 though far advanced, not yet complete. In the latter 
 case it may be inferred that a change of kinship took 
 place at no very remote date. 
 
 The kindreds in which a change in kinship took 
 place within the totem stage, when they had thought 
 their totem the mysterious parent, protector, and 
 friend into a god, would, with their common name 
 and common worship and tradition of relationship, 
 greatly resemble the totem kindreds in which countries 
 like ancient Italy abounded. In very many cases, how- 
 ever, the totem seems to have developed into a god 
 before the change in kinship took place. Then, each 
 kindred holding stiffly to its own worship, it can be 
 seen that it interposed a formidable obstacle in the way 
 of the change (which, on the other hand, it supplied
 
 234 THE PATRIARCHAL THEORY. 
 
 a new motive for making) an obstacle which it taxed 
 the ingenuity of mankind to surmount. We can see 
 very well among peoples known to us how great the 
 obstacle was, and we can discern among them also what 
 were some of the expedients by which it was surmounted. 
 
 In Western Africa we find, with circumstances 
 seemingly favourable to the acknowledgment of relation- 
 ship between father and children the men masters, 
 severe laws against adultery, and a certain accumulation 
 of wealth the only drawback being an unusual fre- 
 quency of polygamy the system of kinship through 
 females only still firmly rooted ; and, though polygamy 
 may favour it, it is religion chiefly which keeps it up. 
 The kindred to which a person belongs determines what 
 god he shall serve, what religious observances he is 
 bound to, what is to him forbidden meat, and almost 
 certainly what women he may not marry. And, in 
 marriage between equals, or when the conditions are 
 of the usual sort, the force of religion sustains ancient 
 custom so that children belong by birth to the kindred, 
 and to the god and worship, of their mother. Wives, 
 it may be said, are purchased ; but, if they have 
 property, when they marry the husband does not get 
 it it remains the wife's and goes to her children, 
 who are heirs, not of their father and his relatives, 
 but in the family of their mother. 
 
 In Africa, contract that is purchase appears to 
 have been found, in a class of cases which will be 
 looked at hereafter,* of itself sufficient to carry children 
 * See Chapter xvii., on " Sonship among the Hindoos."
 
 THE ORIGIN OF AGNATION. 235 
 
 to their father from the kindred of their mother. In 
 the cases now spoken of, though used, it was not found 
 sufficient. But it is very curious and instructive to find 
 among the people of Guinea a custom which shows that 
 there, unaided from without, men were in a good way 
 for getting over, by the help of religion, what religion 
 had made a formidable difficulty. It has been shown 
 that the custom found among ancient nations of admit- 
 ting the wife at marriage into the kindred of her 
 husband, and to participation in the sacred rites to 
 which he was bound, would be effectual as a device 
 for reconciling the ancient system of kinship through 
 females with the resolve that children should be of the 
 kindred of their father. Now we find this very device 
 in common use in Guinea; and, though the purpose 
 for which it was used is not clearly stated, and it may 
 have been intended to serve more purposes than one, 
 there can be no doubt as to one effect it was sure to 
 produce. It could not fail to make children in the 
 cases in which it was employed of the kindred and 
 worship of their father ; for, as it was used, there was 
 no other kindred or worship they could belong to. 
 
 In general, Bosman tells us,* marriage carried the 
 consequences already described. In the case of his 
 chief wife, an equal, the husband perhaps could not 
 hope to escape from them ; and, even in his ordinary 
 marriages he appears to have had to put up with them. 
 It was customary, however, for a man to buy and take 
 
 * Bosnian's Description of Guinea. Translated from the Dutch. 
 Second edition. 1721. Page 169.
 
 236 THE PATRIARCHAL THEORY. 
 
 to wife a slave, a friendless person with whom he could 
 deal at pleasure, who had no kindred that could interfere 
 for her, and to consecrate her to his Bossum or god. The 
 Bossum wife, slave as she had been, ranked next to the 
 chief wife, and was, like her, exceptionally treated. 
 She alone was very jealously guarded. She alone was 
 sacrificed at the husband's death. She was, in fact, 
 wife in a peculiar sense. And, having by consecration 
 been made of the kindred and worship of her husband, 
 her children would be born of his kindred and worship. 
 
 Whether intended or not, therefore, we find here 
 a beginning made of kinship through males, or, at 
 any rate, a great advance towards it at least, an ex- 
 pedient found which could make children of the 
 kindred of their father, the use of which, if its effect 
 was welcome, was likely to spread quickly. It is now 
 easier to believe that, in the other cases in which we 
 know the same custom to have prevailed, one reason 
 at least for it coming into general use was the effect 
 which, by means of it, could be produced upon kinship. 
 
 It need not be supposed that, in those other cases, 
 the expedient was first used in the case of slaves 
 though it is not unlikely that that was sometimes, or 
 even often, the case. However that may have been, in 
 a progress from the Bossum system as it was in Guinea, 
 once a feeling in favour of the new kinship had become 
 general and strong among men, Bossum marriage would 
 cease to be confined to slave women ; gradually men 
 would get into the way of giving their own women in 
 marriage on the new conditions. Once these had
 
 THE ORIGIN OF AGNATION. 237 
 
 become the common conditions, the contract for mar- 
 riage would commonly surrender to the husband (as we 
 know that it has constantly done) all rights of the 
 woman's family over her and her issue ; while conse- 
 cration to the husband's god, making her and her 
 children of the husband's clan and worship, would free 
 her family from gentile connection with and gentile 
 duties towards them. 
 
 With Bossum marriage, as it was at first, moreover, 
 children would have been, by the circumstances of the 
 case, cut off from the relations of their mother. And 
 with this form of marriage spread as has been sup- 
 posed, there would even be present the circumstances 
 which favour the rise of Agnation the circumstances 
 which would dissever a married woman's children from 
 her kindred so as to oppose the least resistance to the 
 old view of the gentile bond, which confined relation- 
 ship to those who were within that bond. Never- 
 theless, the rise of Agnation would, in such cases, be 
 far from certain. Much, it would seem, would depend 
 upon whether the transition from the old kinship to 
 the new was quick or slow upon the length of time 
 for which the old system maintained itself beside the 
 new. For, with the transition protracted and men's 
 minds long kept fixed on the fact that there was rela- 
 tionship through the mother there would be greater 
 likelihood that some acknowledgment of that relation- 
 ship might be thought consistent with the prevalence 
 of the other. 
 
 Polygamy, therefore, which would obviously lengthen
 
 238 THE PATRIARCHAL THEORY. 
 
 and graduate the process of change since it would 
 be more difficult for a man to get all of many wives on 
 the new conditions than to get one which, moreover, 
 by dividing a man's children into groups each specially 
 connected with a mother, the persons in each being 
 specially connected with each other because of their 
 mother, in all cases emphasises the connection between 
 child and mother and the connection which arises 
 through the mother and which would in both ways 
 keep up the tradition of kinship through the mother 
 alongside of the new kinship through the father, might, 
 where it prevailed to any considerable extent, counteract 
 the circumstances which make for Agnation, with the 
 result that, when the latter kinship generally prevailed, 
 there would still remain acknowledgment of kinship 
 through the mother. With monogamy the change 
 would be more rapid, and much more likely to be 
 complete. It is scarcely necessary to say that Koman 
 gentile marriage was monogamous, and that the bride 
 exchanged the gens and sacra of her birth for the gens 
 and sacra of her husband. 
 
 It may be well to add that where as among the 
 Hindoos after male kinship had been established, 
 prohibition of marriage remained for men as regards all 
 women of the name of their mother, Agnation in any- 
 thing like strictness could not be looked for. And, 
 among the Hindoos, we find that a man, when he had 
 no son, could just as if kinship had been through the 
 mother only reserve to himself a son of his married 
 daughter necessarily by a man of another stock
 
 THE ORIGIN OF AGNATION. 239 
 
 (gotra), and take him to be of his own name and stock, 
 to be his heir and the continue! of his family. 
 
 Of the power which a husband possessed over his 
 wife and children when they had been made fully over 
 to him by the woman's kindred, and the protection of 
 that kindred as a duty, at any rate had been with- 
 drawn from them, it may be said that custom would 
 everywhere settle what it was ; and that it is only by 
 a knowledge of custom we can know what it has been. 
 Naturally it would somewhat vary. It would tend to 
 be great. But, were it unlimited and though we 
 can find the paternal power recognised by law as 
 unlimited in one case only, it would be very rash to 
 say it never was unlimited in any other case power 
 possessed as a consequence of purchase could not be 
 primitive. We should be put upon inquiry as to the 
 history which had led up to it. 
 
 It should be noticed, too, with reference to the Patria 
 Potestas theory of Agnation, that it was not the power 
 (whatever it might be) which men acquired over their 
 wives and children, but the fact that wife and children 
 were cut off from the woman's kindred, that would 
 favour the rise of Agnation. 
 
 In another class of cases which may have been 
 common we find the transition from the one kinship 
 to the other going on under circumstances less favour- 
 able to the rise of Agnation. These are cases in which, 
 for a time, the gentile connection of children was not 
 taken regularly from either father or mother, but some- 
 times from the one and sometimes from the other it
 
 240 THE PATRIARCHAL THEORY. 
 
 being settled, in some cases by the relatives before 
 marriage, in others by the parents when the birth of the 
 child was at hand, which clan a child should belong to. 
 In all such cases we find the totem grown into a god, 
 but with its totem origin clearly recognisable ; we find 
 the kindreds interfused much as they are when gentile 
 connection is taken from the mother ; and they all 
 involve that the persons who married each other 
 belonged to different clans. 
 
 We come upon a few cases of this sort in Africa, with 
 evidence that, just as in marriages between Protestants 
 and Catholics among ourselves, the boys belong to 
 the clan of the father, and the girls to the clan of the 
 mother. But the Hervey Islands furnish the best case 
 of the kind.* In the Island of Mangaia, parents, 
 at the birth of a child, arranged between themselves 
 whether it should be dedicated to the father's god or to 
 the mother's. The dedication took place forthwith, and 
 it finally determined the child's gentile connection. And 
 gentile connection carried with it obligations and 
 liabilities such as arise with it when it is determined by 
 the blood of the mother or the blood of the father. 
 There is evidence enough to make very probable, what 
 in such a case could hardly be doubted, that there had 
 previously been kinship only through the mother. On 
 the other hand, we are told that, on the whole, the 
 father's clan and god got the preference ; and as the 
 father carried off the child to be dedicated immediately 
 
 * See "Myths and Songs from the South Pacific." By the 
 Rev. W. Wyatt Gill. London : Kegan Paul & Co. 1878.
 
 THE ORIGIN OF AGNATION. 241 
 
 after its birth, and dedication was final, he always if 
 he chose to break the compact made, which sometimes 
 happened had the matter in his own hands. The 
 Mangaians therefore were passing through a transition 
 towards kinship and gentile connection being taken 
 from the father. A little later, and all children would 
 have been of the father's clan and worship. Later still, 
 the wife, when she married, would probably have ex- 
 changed the clan and worship of her birth for the clan 
 and worship of her husband. 
 
 In these cases, children of the same parents belong- 
 ing to different clans, no clan containing all who were 
 of the same blood on either father's side or mother's 
 side, and every clan being made up of persons who, 
 though attachable to it in right either of their fathers 
 or their mothers, were bound together only by their 
 common worship, and the obligations and liabilities 
 which followed upon it, the clan which had ceased to 
 consist of persons believed to be connected through 
 female descents, and was on the way to being composed 
 of persons who would believe themselves to be con- 
 nected through male descents was, for the time, more 
 like a religious body than a body of kindred.' The sense 
 of relationship on the side of both father and mother 
 though carrying no gentile effects could subsist and 
 develop with it. And, when finally the father's kinship 
 and worship prevailed, the sense of the other kinship 
 might well be strong enough to prevent clanship from 
 becoming agnatic. 
 
 Enough has now been done to indicate how the cir- 
 
 R 

 
 242 THE PATRIARCHAL THEORY. 
 
 cumstances of early peoples appear to have affected the 
 tendency of clanship to become agnatic. Another class 
 of cases in which there seems to have been a transition 
 from kinship through females to kinship through males 
 effected apparently by means of contract alone will 
 be more conveniently considered in a future chapter.* 
 
 It scarcely need be added that all that has here been 
 said has been said of cases in which exogamy, that is 
 prohibition of marriage between all persons considered 
 to be of the same kindred, became established as 
 marriage law. That exogamy has prevailed most 
 extensively is well known ; it may be believed to have 
 been general, and nearly universal. An extensive 
 practice of polyandry might however (as has been 
 reasoned in "Primitive Marriage") prevent the rise 
 of exogamy.t Of such cases, in which the exogamous 
 restriction never was established, all that need be said 
 is that the religious and quasi-religious difficulties which 
 have had to be noticed could not have occurred, or were 
 very unlikely to occur, in them. Husband and wife 
 might be of the same kindred ; and, as soon as certainty 
 of fatherhood permitted, no obstacles would lie in the 
 way of gentile connection being taken from the father, 
 and full allowance made for all ties of blood, whether 
 deduced through males or through females. 
 
 * Chapter xvii., on " Sonslrip among the Hindoos." 
 
 t Studies in Ancient History. " Primitive Marriage," chapter viii.
 
 CHAPTER XIV. 
 
 EXAMPLES OF AGNATION. 
 
 IT seems unnecessary to repeat all that has been said 
 in chapter v. to show that Agnation did not prevail 
 among the early Hebrews, and that, among them, 
 relations on the mother's side were kindred in the 
 fullest sense. It seems enough to mention Laban's 
 family, of which his married daughters, and their 
 children, and their husband too, formed a part ; and 
 the case of Abimelech, who is acknowledged by his 
 mother's kindred as " their bone and their flesh," and 
 aided by them on the ground of relationship.* It may 
 be well, however, to recall that we have met with what 
 seems clear Scriptural indication of beenah marriage as 
 prevailing at a very early time a form of marriage 
 with which Agnation is impossible, and which, on the 
 contrary, cuts off the husband himself, as well as his 
 children, from the connections of the husband's birth. 
 
 * Prof. W. K. Smith points out that in Lev. xxv. 49 (Heb.), "of 
 his flesh" is explained to mean "of his clan." Thus it would seem 
 that his mother's kin recognise Abimelech as their clansman. So, 
 too, we find among the Australians that people of the same kindred 
 
 may not intermarry because they are " of the same flesh." 
 
 K 2
 
 244 THE PATRIARCHAL THEORY. 
 
 And it should be added that marriages like those of 
 Abraham and Nahor could not occur except at a 
 time when relationship through males was not itself 
 thoroughly enough acknowledged for a bar to marriage 
 to have been raised upon it.* 
 
 To show the absence of Agnation among the earliest 
 Hebrews known to us is all we are concerned to do 
 though, no doubt, if Agnation could be shown to have 
 prevailed among later Hebrews, we should have proof 
 that, as regards the Hebrews, Agnation was not a 
 primary kinship ; and also proof since the Hebrews 
 never had Patria Potestas that Agnation could arise 
 independently of Patria Potestas. But Agnation was 
 unknown among the later Hebrews. 
 
 We learn from Numbers xxvii. 1-11 that, after the 
 Israelites had become a people, when they had made 
 conquest of territory and were just about to divide it, 
 the daughters of a man who was dead put in a claim to 
 share, as representing their deceased father, in the land 
 to be allotted to the men of their tribe ; and that the 
 claim was at once admitted, and, in the first instance, 
 
 * With those cases in view, it seems as if there can be no reason- 
 able doubt that the Athenian custom which, in the case of an heiress 
 allowed brother and sister german to marry, became established at a 
 time when, at Athens, the law of incest made no real difficulty for 
 such a marriage. And with this it agrees that, according to a tradition 
 preserved by Athenrcus from Clearchus of Soli, a disciple of Aristotle 
 (Miiller, Fr. Hist. Gr. ii. 319, Athen. xiii. 2), the Athenians practised 
 Nair polyandry before the time of Cecrops, and that no man knew who 
 was his father. The story of King David's children, Amnon and 
 Tamar, shows that marriage between brother and sister german was 
 allowed among the Hebrews at the time to which it refers without 
 there being any property reason for it.
 
 EXAMPLES OF AGNATION. 245 
 
 without any conditions it being at the same time 
 prescribed as a rule that the daughter should succeed 
 her father in land, failing a son. A later passage 
 (Numbers xxxiii. 1-9) tells us that a condition was 
 subsequently imposed upon the admission of this claim, 
 and upon the law as thus established ; and this passage 
 is extremely interesting and instructive as disclosing the 
 motives for those rules upon which, though they disprove 
 Agnation, a belief in the general prevalence of Agnation 
 has mainly been founded. The chiefs of the tribe came 
 forward and pointed out that the effect of allowing 
 women to become sharers in the land of the tribe might 
 be since they might marry men of other tribes that 
 men of other tribes would acquire a settlement upon the 
 land of the tribe 'and after a time carry it off to the 
 tribes to which they themselves belonged. It was 
 thereupon laid down that the claimants, as a condition 
 of getting the share they had asked for, must marry 
 within " the family of the tribe of their father." The 
 law prescribed for the like cases was correspondingly 
 modified, and. as so modified, it continued to be the law 
 for Israel. A desire to prevent confusion of tribal lands 
 to keep the land of each tribe intact for the tribe 
 was the motive for the condition imposed upon the 
 daughter's right of inheriting. Had the law of incest 
 been so strict as to make the marriage of the daughter 
 within the paternal kindred a thing not to be thought 
 of, the same motive might have been of itself 
 enough to account for her being excluded from inherit- 
 ance altogether. In this case, be it observed, the
 
 246 THE PATRIARCHAL THEORY. 
 
 right of a daughter to inherit is not said to have been 
 questioned though the inconvenience that might attend 
 it in a new case is said to have led to its being subjected 
 to a condition. When the son of an heiress came 
 to inherit he obtained the inheritance in right of his 
 mother. 
 
 It may here be said that we see from 1 Chron. 
 ii. 34-36, that, among the Israelites after they 
 had become a people, the family could be con- 
 tinued through a daughter. It is there said that 
 Sheshan, having no son, gave his daughter to wife 
 to Jarha, his Egyptian servant, and that she bare 
 him a son named Attai. And through this Attai 
 the family genealogy is continued just as if he had 
 been Sheshan's son or son's son. The names of women, 
 as can easily be seen, are of very frequent occurrence 
 in the Hebrew genealogies. It may perhaps be worth 
 mentioning also that there seems to have been no case 
 for the Levirate when a deceased elder brother, though 
 without a son, had left a daughter. 
 
 The case of the Hebrews, apart from its direct 
 bearing upon the prevalence of Agnation, and the 
 place of Agnation in the history of relationship, is of 
 the utmost value, as showing beyond question that, 
 from finding rules of inheritance which postpone 
 daughters and their descendants to sons and their 
 descendants, we cannot infer the prevalence at an 
 earlier time of Agnation where such rules are found. 
 For the Hebrews* when they had become a settled 
 people, postponed daughters to sons in inheritance;
 
 EXAMPLES OF AGNATION. 247 
 
 and their earliest records clearly show that the earlier 
 Hebrews were not agnatic. The importance of this 
 will appear immediately. The case of the Hebrews 
 is also valuable, of course, for the light it throws 
 upon the origin of rules of inheritance unfavourable to 
 women. A clan which acknowledged kinship through 
 males was sure to be continued chiefly through its 
 males ; and, when marriage within the clan was strictly 
 prohibited, it could, apart from special devices, only 
 be so continued. Inevitably, therefore, when a clan 
 had settled upon land its rule for inheritance was 
 such as to keep the land for the men of the clan 
 so that daughters could have no chance of inheriting 
 land if there were sons, and, apart from special devices, 
 possibly no chance even failing sons unless they could 
 marry among their own kinsfolk. 
 
 To pass now from the Hebrews, although Sir Henry 
 Maine has told us that agnatic relationship is discover- 
 able almost everywhere, his particular statements as to 
 the existence of Agnation have always been very few. 
 And, in his latest work,* his earlier statements have 
 been most seriously modified. Let us briefly show 
 what they were, and what they are, and what is 
 implied in them as they now stand. 
 
 The Hindoos, with the Romans and Slavonians, are 
 said in Ancient Law to furnish the greater part of the 
 legal testimony to the truth of the Patriarchal Theory. 
 And it is stated in that work (p. 150) that "in Hindoo 
 law, kinship is entirely agnatic." In Early Law and 
 * Early Law and Custom, passim.
 
 248 THE PATRIARCHAL THEORY. 
 
 Custom it appears, however, that, from the earliest times 
 known to us, a Hindoo who had no son might be suc- 
 ceeded, and was customarily succeeded, by the son of 
 his daughter (the son of an appointed daughter); and 
 we are told that customs akin to this appear to have 
 been very widely diffused over the ancient world * the 
 daughter in all such cases, as Sir Henry Maine observes, 
 becoming a channel through which the father's blood 
 passed to a male child. 
 
 The modification made in the later work obviously 
 involves an admission that, to judge by the evidence 
 to which Comparative Jurisprudence usually confines 
 itself, the early Hindoos and also the numerous early 
 peoples who were in the same case with them- were 
 not agnatic. The two leading notes of Agnation are 
 that a married woman is the " terminus of the family " 
 that the family cannot be continued through her ; 
 and that there can be no right of succession for any 
 descendant of a woman. And, among the Hindoos, 
 as Sir Henry Maine aptly says, a daughter could be, 
 as it were, a channel through which the father's blood 
 passed to a male child that is, the family could be 
 continued through her ; and her child was her father's 
 heir, and was in every respect as a son to him. What 
 we know of the Hebrews makes it clear that from 
 finding the right to inherit of a daughter or her de- 
 scendants limited as it was among the Hindoos, there 
 can be no inference of a prior prevalence of Agnation. 
 The attempt made in Early Law and Custom, upon 
 
 * Pp. 90-92.
 
 EXAMPLES OF AGNATION. 249 
 
 the basis of Ancestor-worship, to reconcile the several 
 cases of Hindoo sonship with the Patriarchal Theory 
 will, however, be discussed at length in a separate 
 chapter. 
 
 As for the Slavonians, it is unnecessary to do more 
 than repeat what has been shown of them in previous 
 chapters.* Their earlier history, so far as it is known 
 to us, affords no evidence in favour of the Patriarchal 
 Theory, and furnishes not a little evidence against it, 
 and especially against Agnation. Sir Henry Maine's 
 views about them, on the other hand, have been formed 
 on contemporary evidence ; and all the evidence that 
 could be quoted for those views is of comparatively 
 modern date. The Slavs have been good Christians 
 for many centuries beyond the date of the oldest part 
 of such evidence ; and no inference as to the insti- 
 tutions of the Primeval Family can be made from what 
 it discloses to us of their social condition. The con- 
 stitution of their House-Communities and their family 
 usage, nevertheless, in no way countenance on the con- 
 trary, they decidedly discountenance the Patriarchal 
 Theory. And it is certain that the Slavs fully acknow- 
 ledge relationships through women ; and that they give 
 women and their descendants rights of inheritance, 
 though postponing them to sons. It is needless to do 
 more than mention again the warning conveyed by 
 the case of the Hebrews as to inference from a limi- 
 tation upon the succession of women such as is found 
 among the Slavs. 
 
 * Chapters vii. and viii.
 
 250 THE PATRIARCHAL THEORY. 
 
 The Hindoos and the Slavs once considered, the 
 weightiest part of Sir Henry Maine's evidence as to 
 the prevalence of Agnation has been dealt with. In- 
 deed, of particular statements as to Agnation made in 
 Ancient Law to support the general statement that it 
 is discoverable almost everywhere, only two remain to 
 be noticed ; and Sir Henry Maine has fully disposed 
 of both in his later work. It is said in Ancient Law 
 (p. 151), of the rule of royal succession known as the 
 Salic law, that it has certainly an agnatic origin " being 
 descended from the ancient German rule of succession 
 to allodial property ; " and the rule in the Customs of 
 Normandy which excluded uterine brothers from suc- 
 ceeding to one another in property derived from their 
 fathers is spoken of as " a strict deduction from the 
 system of Agnation under which uterine brothers are 
 no relations at all to one another." As to the latter, 
 however, Early Law and Custom discloses to us,* that 
 the rule in the Customs of Normandy did not exclude 
 uterine brothers from succeeding each other in property 
 derived from their common parent, their mother so 
 that, though connected only through the mother, they 
 were admitted to be each other's relations, and could 
 inherit from each other according to their relationship ; 
 and there was no " strict deduction from the system of 
 Agnation," or rather there was no question of Agnation 
 at all. And for the Salic law another origin is found 
 in Early Law and Custom. Instead of it being of 
 agnatic origin because " descended from the ancient 
 * Note on p. 151.
 
 EXAMPLES OF AGNATION. 251 
 
 German rule of succession to allodial property," the 
 later view of its history declares the notion that it 
 could be derived from any law of succession to pro- 
 perty to be utterly mistaken. It was, according to 
 this view, when the question of succession to the 
 throne came to be mooted in France, a doubtful 
 point whether a claimant deriving his right through a 
 female could succeed to the throne of France or not 
 there was an absence, that is, of any known rule for 
 the case which leaned towards Agnation ; and the 
 French decision against the right of our Edward III., 
 who claimed through a female, was determined by 
 feelings which are explained conjecturally, and which, 
 as so explained, arose out of a natural misapprehension 
 as to earlier practice. Something has been said of this 
 view of the origin of the Salic law already,* and it 
 appears to be erroneous. That rule of royal succes- 
 sion may have had, and had so far as anybody knows, 
 nothing to do with the land law of the ancient 
 Germans ; but it seems beyond question to have 
 been, in the apprehension of those who discussed it 
 and those who adopted it, founded upon a provision 
 of the Salic Code as to the succession to Salic land. 
 What concerns us here, however, is to point out that 
 the view of its origin propounded in Early Law and 
 Custom makes it unnecessary to say more of the earlier 
 view of its origin which derived it from a land law said 
 by implication to have been agnatic. 
 
 It appears then that, of particular statements made 
 * See JSTote to Chapter vii.
 
 252 THE PATRIARCHAL THEORY. 
 
 by Sir Henry Maine to show that Agnation is discover- 
 able almost everywhere, there are none now remaining 
 except those relating to the Hindoos and Slavonians. 
 There are some general statements which have yet to 
 be noticed. But, before noticing them the ancient 
 Germans having been mentioned it is impossible not 
 to have in mind what Tacitus has handed down to us as 
 to the relationships of the ancient Germans. A reference 
 to their rule of succession to allodial property is be- 
 wildering. Montesquieu, no doubt, believed he knew 
 what it was, because he knew what the Frank rule was, 
 and had a hypothesis in which he firmly believed 
 that the laws of the Salian Franks had come from the 
 ancient Germans. But Montesquieu had nothing more 
 or less than conjecture for this. And it need scarcely 
 be said that what Tacitus has told us as to succession 
 among the Germans, though, like many similar state- 
 ments of modern travellers, not definite enough to be 
 of use* is, so far as it is worth anything, unfavour- 
 able to Agnation. What Tacitus has told us of the 
 relationships of the ancient Germans, on the other hand, 
 is perfectly intelligible. What it involves does not 
 admit of the shadow of a doubt. 
 
 " A sister's sons," he says, " are as highly esteemed 
 by their uncle as by their father. Some, indeed, look 
 on this as the closer and more sacred bond of blood, 
 and, in taking hostages, prefer to choose them on 
 this principle, holding that such hostages give a surer 
 
 * "Failing sons, the next in succession are brothers, paternal 
 uncles, maternal uncles." Tacitus, Germania, ch. xx.
 
 EXAMPLES OF AGNATION. 253 
 
 hold on the individual as well as a broader hold on his 
 house." * 
 
 This shows, to begin, that the ancient Germans were 
 very far from being in the agnatic state of relationship. 
 On the agnatic view, a child and his uncle on the 
 mother's side would not have been each other's relations 
 whereas, among all Germans, the uncle on the mother's 
 side was as close to a child as its own father; and 
 among some, he was even closer, so that, in taking 
 hostages, a man's sister's son was taken from him in 
 preference to his own son, as being the nearer and 
 dearer to him of the two. The sister's son was preferred 
 as a hostage, too, as affording a broader hold on the 
 house of the giver from which it appears that the 
 house (domus) included kindred through women .t 
 May we not ask in passing whether it may not have 
 done so among the ancient Slavs also ? 
 
 But the question now occurs whether there is any 
 form of kinship known in which such a relation between 
 a man and his sister's child as Tacitus describes to us is 
 found to subsist. And the answer is that it is regularly 
 found in communities in which kinship is acknowledged 
 only through women. What Tacitus has told us of 
 uncle and nephew among the Germans is true of the 
 
 * Tacitus, Germania, ch. xx. 
 
 t From a subsequent passage (Germania, ch. xxi.) it appears that 
 compensation for blood was paid to the whole Domus. It follows 
 that kindred through women shared in the compensation for blood, 
 and that they were under the obligations of the blood-feud. It 
 should be said that the suggestion as to the composition of the ancient 
 German Domus made above has come from Professor Robertson Smith.
 
 254 THE PATRIARCHAL THEORY. 
 
 Nairs of Malabar among whom paternity is unknown 
 or unheeded ; and is true of peoples like those American 
 Indians (of whom something has been said in a previous 
 chapter) by whom, though paternity may not be very 
 uncertain, kinship is traced through women only. 
 Among such peoples (where there is a father) the son is 
 not counted of the same blood as his father, and is not 
 his acknowledged relative whereas he and his mother's 
 brother are counted to be of the same blood, and are 
 consequently each other's nearest male relations. The 
 maternal uncle therefore is a child's natural protector 
 among such peoples rather than the father ; and it is not 
 surprising to find him bestowing upon his nephew much 
 of that affection which, where kinship through males has 
 long been thoroughly established, people reserve for 
 their own children. 
 
 Now it is not intelligible how such a relation between 
 a man and his sister's son could arise where fathers had 
 always been fully acknowledged. It is known that it 
 exists where fathers are not fully acknowledged. And 
 it is intelligible that, once formed, it should subsist, 
 and even subsist long, after full recognition had been 
 obtained for fatherhood. When we find it then, as Tacitus 
 discloses it to us, among a people who acknowledge 
 kinship through males, there seems to be room for only 
 one conclusion about it. That is, that kinship through 
 males was not always acknowledged among that people. 
 And the statement of Tacitus appears therefore to con- 
 tain convincing evidence that the Germans anciently 
 acknowledged relationship through women only.
 
 EXAMPLES OF AGNATION. 255 
 
 The ancient Germans then have to be classed with 
 the Hebrews among peoples who, when they first become 
 known to us, were clearly not agnatic. And, moreover, 
 they have to be classed with the Hebrews among peoples 
 who acknowledged kinship only through women before 
 they acknowledged kinship through males. There 
 appears to be no doubt, it may be said, that the early 
 Greeks have to be placed with them in the former of 
 these classes, and it seems difficult to escape the con- 
 clusion that they should stand, with Hebrews and 
 Germans, in the second also.* 
 
 Of the more general, or less particular, statements 
 as to the prevalence of Agnation which have been 
 referred to as remaining to be noticed, one occurs in 
 Ancient Law (p. 151), and is to the effect that the 
 agnatic view of relationship pervades so much of the 
 laws of the races who overran the Koman Empire as 
 appear to have really formed part of their primitive 
 usage. It would have been of the greatest interest 
 and value to have had, for even one of the races alluded 
 to, the native and ancient element in its laws sifted 
 from the recent and extraneous in the manner suggested 
 in this sentence, and the former shown to be pervaded 
 by the agnatic view of relationship. But Ancient Law 
 attempted nothing of this sort. And the sweeping 
 generality it gave us instead seems to be effectually 
 
 * As to this, see " Kinship in Ancient Greece " in Studies in 
 Ancient History. London : Quaritch, 1876. The succession law dis- 
 closed in the story of Meleager is the law peculiar to the system of 
 kinship through women only. A man's heir, failing a brother, was his 
 sister's son, and, conversely, the heir of a woman's son was her brother.
 
 256 THE PATRIARCHAL THEORY. 
 
 neutralised by other statements of a general sort which 
 we have to notice, and which are to be found in Early 
 Law and Custom. 
 
 It has already appeared from that work that the 
 succession of a daughter's son, to her father, failing 
 sons of his own, was " very widely diffused over the 
 ancient world." This of course can mean nothing 
 less than that the ancient world, very, generally, was 
 not under the influence of the agnatic view of relation- 
 ship. But Early Law and Custom tells us further (p. 151) 
 that there is "a strong probability " that "all sorts of 
 ideas about succession to property " (and not agnatic 
 ideas only) prevailed among the Aryan barbarians who 
 overran Western and Southern Europe ; and (the reason 
 for that statement) it has to admit (p. 149) that many 
 of the greatest races of mankind, when they first appear 
 to us, allowed rights of succession, more or less, to 
 women and the descendants of women so that they 
 were not agnatic. Early Law and Custom here seems 
 to go directly counter to the general statement in 
 Ancient Law. And so no more need be said of it. 
 
 Having now gone through all the evidence bearing 
 upon the prevalence of Agnation which Sir Henry 
 Maine has ever spoken of, we find that, after deducting 
 what has in one way or another been given up, there 
 remains to prove that Agnation is discoverable almost 
 everywhere in addition to the prevalence of Agnation 
 at Rome of course (l) the chance of Sir Henry Maine 
 making something of the Hindoos notwithstanding 
 that the Hindoos, when we first come upon them, were
 
 EXAMPLES OF AGNATION. 257 
 
 certainly not agnatic (as to which, in a future chapter) ; 
 (2) such evidence for Agnation as can be got from the 
 social condition of Nineteenth Century Slavonians 
 who are not, and, it must be admitted, could not possibly 
 be, agnatic; and, along with these, (3) distinct state- 
 ments that the succession of daughters' sons was very 
 widely diffused over the ancient world, and that many 
 of the greatest races of mankind, when they first appear 
 to us, allowed some rights of succession to women and 
 the descendants of women so that relationship with 
 the descendants of women must have been fully 
 acknowledged among them. 
 
 Sir Henry Maine has, in a passage in Early Law 
 and Custom (p. 149), given a modified form of his 
 statement as to the prevalence of Agnation or rather 
 a variant form, for it does not seem really intended 
 to assert less. " The greatest races of mankind " this 
 is the new statement " when they first appear to us, 
 show themselves at or near a stage of development in 
 which relationship or kinship is reckoned exclusively 
 through the males." " They are in this stage," he adds ; 
 or (this is said to show that two views can be taken of 
 the ancient facts which are unfavourable to Agnation) 
 "they are tending to reach it; or they are retreating 
 from it." 
 
 This new statement prompts the question, Which of 
 the greatest races of mankind, when they first appear to 
 us, were in the agnatic stage ? Sir Henry Maine has 
 not told us. And it is impossible not to see that he 
 has never adduced a single real instance of Agnation.
 
 258 THE PATRIARCHAL THEORY. 
 
 Sir Henry Maine has not even offered any argument 
 to show that something of common occurrence which 
 is not itself Agnation say a preference of males over 
 females in succession, which clearly is not itself 
 Agnation implies a former prevalence of Agnation. 
 Had he done so, no doubt, the argument for the Patri- 
 archal Theory would have taken a rather complicated 
 form. Patria Potestas, having disappeared, would have 
 had to be inferred from Agnation; and Agnation, 
 having also disappeared, would have had in its turn 
 to be inferred from something else. He has trusted 
 entirely to the frequency of Agnation. And yet he 
 has never mentioned one clear case of Agnation other 
 than the Roman case. 
 
 If what has been advanced in this chapter as to 
 the Hebrews and others be well founded, such an argu- 
 ment as has just been spoken of could not succeed. 
 Apart from this, it seems reasonable to hold that that 
 which is very rare is exceptional, rather than to assume 
 that it is normal, and the original of radically different 
 forms which are actually met with.
 
 CHAPTER XV. 
 
 AGNATION CONCLUDED . 
 
 IN the preceding chapters, in which Agnation has been 
 treated of, it has been maintained that Sir Henry Maine 
 has not enabled himself to recover lost cases of Patria 
 Potestas by means of Agnation. An attempt has been 
 made to show that his account of Agnation does not 
 explain how Patria Potestas can have produced, or have 
 been the foundation of, Agnation ; to show that all we 
 know of Patria Potestas discountenances the supposition 
 that it can have had the effect attributed to it ; and to 
 show that, even on Sir Henry Maine's own statement, 
 Patria Potestas could not have originated Agnation, or 
 have been the foundation of it, unless people had syste- 
 matically married outside the circle of their acknow- 
 ledged relatives so that this system of marrying, which 
 is nothing other than exogamy, appeared, on his own 
 statement, to be the real determinant of Agnation. 
 These objections to his account of Agnation, of course, 
 involve that it is a hypothesis which must be rejected, 
 and, indeed, that, on his own statement, it is manifestly 
 
 8 2
 
 260 THE PATRIARCHAL THEORY. 
 
 a bad hypothesis. It has further been maintained 
 against this theory of Agnation that it so connects 
 Patria Potestas and Agnation that it would be neces- 
 sary, were it a hypothesis which otherwise seemed 
 sufficient, to show how Agnation could survive Patria 
 Potestas, and be found where no trace or memory of 
 Patria Potestas was remaining while this has not been 
 attempted ; and that therefore the occurrence of cases of 
 Agnation, independently of Patria Potestas, were they 
 forthcoming, would only tend to discredit and disprove 
 the hypothesis. It has been pointed out that, while Sir 
 Henry Maine's account of Agnation assumes that Agna- 
 tion could survive Patria Potestas indefinitely, at Rome 
 Patria Potestas survived Agnation. 
 
 Of course, a hypothesis, even if in appearance suf- 
 ficient, cannot be received for more than a hypothesis 
 much less can it rightly be the subject of confident 
 belief unless it can be in some way sufficiently verified. 
 And Sir Henry Maine's hypothesis, whatever may be 
 thought of it, has been in no way verified. To make 
 us believe that Patria Potestas and Agnation always 
 occurred together in the early history of man he can 
 only show us that they occurred together once, and that 
 in a community very far from primitive. 
 
 Of the theory of Agnation propounded in this work 
 all that need now be said is that the main points of it 
 are deducible from Sir Henry Maine's attempt to show 
 that Patria Potestas was the foundation of Agnation. 
 
 If what has been urged against Sir Henry Maine's 
 account of Agnation or even a part of it be thought
 
 AGNATION CONCLUDED. 261 
 
 maintainable, then, were cases of Agnation ever so 
 plentiful, Sir Henry Maine would be left so far as they 
 were concerned, with one case of Patria Potestas found, 
 so to speak, in a middle-aged community, by means of 
 which to convince us that Patria Potestas was a uni- 
 versal institution among primeval men. 
 
 But when we go on to inquire as to the prevalence 
 of Agnation, it appears that, so far as Sir Henry Maine 
 has the means of showing us, Agnation is just as rare as 
 Patria Potestas itself. If we are to believe in the uni- 
 versal prevalence of Agnation we must do so, it would 
 seem, upon inference actual cases of Agnation being, 
 at any rate, not produced ; and as Sir Henry Maine has 
 never offered any argument to connect Agnation with 
 forms of kinship which are found but which are not 
 Agnation since they show beyond all question the 
 fullest admission of relationship for the descendants of 
 women we are left, so far as he is concerned, to make 
 the inference without any reason. It would have been 
 simpler, and quite as legitimate, for him to have inferred 
 Patria Potestas from the existence of fathers, and then 
 on the strength of the theory connecting the two 
 to have inferred Agnation from Patria Potestas. 
 
 What we know of the Hebrews, however, to say 
 nothing of other peoples^ seems to make it clear that 
 there can be no inference of Agnation in the cases upon 
 which a belief in its prevalence would have to be 
 formed. Granted that there have been cases in which 
 Agnation existed or broke down, admission of relation- 
 ships through women following, we appear to be even
 
 262 THE PATRIARCHAL THEORY. 
 
 without means of recognising such cases without 
 means of distinguishing them from cases of which the 
 precedent history was different. They would be in 
 appearance identical with the case of the Hebrews, in 
 which it is plain there was no Agnation. 
 
 And what is there to make us think they have been 
 universal ? What to make us think that they ever 
 were numerous ? What to make us think that, where 
 they did occur, they dated from the beginnings of 
 human history ? 
 
 The sum of the matter so far is that Sir Henry Maine 
 has got a single clear instance of Agnation whereby 
 to make it probable that Agnation has prevailed every- 
 where; and with it a single instance of Patria Potestas 
 both occurring together in a community by no means 
 barbarous, and neither of them short-lived in that com- 
 munity whereby to show that Patria Potestas and 
 Agnation were universal in the primeval family. And on 
 the other hand, the researches into which we have been 
 led in examining his statements, have disclosed to us 
 that the early Hebrews and other races, when they 
 first appear to us, were certainly without Agnation ; 
 while they show almost unmistakable signs of having 
 acknowledged kinship through women only before they 
 acknowledged kinship through men. 
 
 The instances which came naturally in our way in 
 the preceding chapter, of ancient nations Hebrews, 
 Germans, Greeks whose customs exclude the notion of 
 their having been agnatic, are, of themselves, enough 
 even to raise a strong probability that the kinship
 
 AGNATION CONCLUDED. 263 
 
 actually found among other early nations was, in 
 general, not preceded by Agnation whereas, from the 
 occurrence of Agnation at Rome, we cannot get the 
 inference that Agnation was primeval even among 
 Roman tribes, or get any light as to the extent of 
 its prevalence elsewhere. 
 
 And were all the instances in which actual Agnation 
 has been found by observers, or even suspected, added 
 to the Roman instance, the amount of reason there 
 would be for thinking that Agnation commonly preceded 
 among ancient nations states of kinship that were not 
 Agnation would not be appreciably increased there 
 would still, that is, be none at all. There are, as has 
 been said already, exceedingly few known cases of 
 Agnation, and they all occur among non- Aryan peoples 
 who may without rashness be suspected of having made 
 a progress from the system of kinship through females 
 only. 
 
 On the other hand, there are so very few known 
 cases of Agnation that, Agnation appearing to be a 
 stable system, and having shown good powers of 
 endurance at Rome, we seem to be forced to conclude 
 that it can never have been frequent, that it occurred 
 in exceptional cases only, and therefore, that it cannot 
 have been anywhere a primary form of kinship. 
 
 That the gentile bond, in those cases in which it 
 became, with kinship through males, the limit of 
 acknowledged relationship, appears to have been a factor 
 in producing Agnation is, indeed, the best reason there 
 is for believing that Agnation, though so little known
 
 264 THE PATRIARCHAL THEORY. 
 
 to us, must have occurred to some extent. It appears to 
 be a reason for believing this, at any rate, if we hold 
 that kinship through males has, in general, been arrived 
 at by a progress from a previous system of kinship 
 through females only. Of the slight essay to describe 
 the conditions of such a progress made in this work, a 
 great part has been taken up with inquiry as to the cir- 
 cumstances which would favour Agnation, and there 
 appeared reason to think that more or less of Agnation 
 was to be looked for. It may be added that Agnation 
 was, on the view submitted, likely to have occurred 
 more rarely than a paternal power resembling Patria 
 Potestas in its degree. The abandonment by a woman's 
 relatives of her and her children to the man to whom 
 they gave her in marriage, without which, it seemed, 
 there would be no possibility of Agnation, might have 
 sufficed for the establishment, for a time at least, of a 
 high paternal power for it would leave the man, so far 
 as the woman's kindred were concerned, uncontrolled 
 master of his family ; while other conditions besides this 
 seemed to be necessary for Agnation. Where, on the 
 other hand, wife and children were not thus abandoned 
 to the husband where they were under the protection 
 of the woman's relatives on the view submitted, neither 
 Agnation nor anything like uncontrolled paternal power 
 were to be looked for; and it is consonant with this that 
 we learn from Tacitus* that the German father was 
 required to respect the lives of his children. 
 
 * Germania, ch. xix.
 
 AGNATION GONCLUDED. 265 
 
 What has been said in this chapter is, of course, 
 said subject to the result of the inquiry as to 
 sonship among the Hindoos, which has now to be 
 undertaken.
 
 CHAPTER XVI. 
 
 SONSHIP AMONG THE HINDOOS. 
 
 IN a previous chapter* we had occasion to describe the 
 Levirate as it existed among the Hindoos and among 
 the Jews, and to discuss the conditions on which it 
 could be believed that the Levirate was derived from 
 Tibetan polyandry. Referring the reader to what has 
 been said in that chapter, it seems sufficient for our 
 present purpose to repeat that when, among the early 
 Hindoos, a man died leaving a widow and no children, 
 it was the duty of his surviving brother to beget a son 
 for him upon the widow; and that the child was 
 counted in every respect equivalent to a legitimate son 
 of the deceased, and continued his family, and was his 
 heir excluding his actual father from inheriting. 
 Failing a brother, a kinsman, the nearest available, 
 might fill what was normally the brother's part. It 
 may be added that the Brahminical writers on Hindoo 
 custom, regarding the Levirate with disgust and loath- 
 ing, insisted that the connection between the widow 
 
 * Chapter xi. Page 156 et seq.
 
 80N8HIP AMONG THE HINDOOS. 267 
 
 and her husband's brother ought not to subsist after one 
 son, or two at most, had been born ; and that they 
 attempted in other ways to put what they thought a 
 filthy practice under regulation. We do not know to 
 what extent their injunctions were deferred to. 
 
 We are now to consider, among other matters, 
 whether the Levirate has been accounted for inde- 
 pendently of polyandry. 
 
 There is still, it may be said, a good deal of polyandry 
 in India, and it is found both in the higher Tibetan form, 
 in which the co-husbands are brothers, and in the lower 
 form, in which the co- husbands are not brothers, of 
 which indeed the typical example is furnished by the 
 Nairs of Malabar. Nor is the British polyandry un- 
 known. The evidence for Tibetan polyandry among 
 the early Aryans, while small in quantity is excellent 
 in quality indeed convincing ; not the less that it 
 may be said it shows us polyandry not as a prevail- 
 ing practice but surviving as an ancient custom 
 amid a prevailing practice of monandry. Of the more 
 direct and striking portions of that evidence a brief 
 account will be found in an article on " The Levirate 
 and Polyandry," which has already been referred to, 
 or in Mr. J. D. Mayne's work on Hindu Law and 
 Usage.* 
 
 It is evident, at any rate, that, in point of gross- 
 ness, there is nothing to choose between the Levirate 
 and polyandry. And the Hindoo family presents other 
 
 * "The Levirate and Polyandry." Fortnightly Review, 1877. 
 Hindu Law and Usage. Madras and London, 1877.
 
 268 THE PATRIARCHAL THEORY. 
 
 facts which appear to indicate practices as bad as either 
 or worse. 
 
 The list of sons enumerated by the early law 
 writers, besides the legitimate son and the son 
 of an appointed daughter a daughter appointed to 
 give an heir to her father because he had no son 
 includes the following : (1) the son begotten on the wife 
 (by a man appointed for the purpose by the husband) ; 
 
 (2) the son born secretly (as in the husband's absence) ; 
 
 (3) the damsel's son (born before marriage, and who, if the 
 woman did not marry, was among her father's heirs) ; 
 
 (4) the son taken with the bride (unborn at the time of 
 marriage) ; and (5) the son of a twice-married woman 
 (not to be described briefly, and of whom it suffices to 
 say that he also might be the child of a stranger). This 
 discloses a view of the matrimonial relation which might 
 be thought to show affinity to the Nair rather than to 
 the Tibetan polyandry. 
 
 Sir Henry Maine has, in Early Law and Custom, 
 offered an explanation of the Levirate and also of the 
 peculiar forms of sonship just mentioned as having been 
 acknowledged among the Hindoos. Mr. J. D. Mayne, a 
 thorough -going supporter of his theories, had previously 
 propounded an explanation of the same facts. As the 
 Levirate is crucial for the Patriarchal Theory, it is 
 proper that the views of both writers about it should be 
 noticed here ; and Mr. Mayne's view, as being the earlier 
 and the easier to deal with, may be taken first. 
 
 Mr. Mayne* thinks he has accounted for the Levi- 
 * See his Hindu Law and Usage, pp. 48 et seq.
 
 SONSHIP AMONG THE HINDOOS. 269 
 
 rate among the Hindoos independently of polyandry, 
 by saying that, with them, it was an extension of the 
 Niyoga the appointment by a husband in his lifetime 
 of a man to beget a son for him. 
 
 From a consideration of the facts relating to sonship 
 among the Hindoos which have been mentioned, he 
 deduces that, in the theory of the law, or rather on the 
 old Hindoo way of regarding the matter, physical pater- 
 nity was not material, was indeed of no importance, in 
 connection with sonship ; and that a man counted his 
 wife's son as his son, whoever the father might be, 
 because the wife belonged to him.* On this view, the 
 Niyoga child, like all other children, was counted the 
 husband's child because it was the child of his wife 
 and he was owner of her ; and it ranked high among 
 children (this it certainly did), as having been begotten 
 by a man selected by the husband. Ultimately the 
 man proper to be appointed was a brother or, failing a 
 brother, a near kinsman ; but in all the early examples 
 recorded, as Mr. Mayne points out, he was a stranger 
 so that there is no doubt that in early times a man 
 might get a child through a stranger. 
 
 When Mr. Mayne goes on to show how, among the 
 Hindoos, the Levirate (it may be so called for dis- 
 tinction's sake) was an extension of the Niyoga, that 
 
 * The only case (apart from the Levirate) in which he has found 
 some difficulty in making this view square with the fact is that 
 which, for a perfectly sound theory, might have been expected to 
 be the easiest case of all that of the son of an appointed daughter, 
 in which a man who has no son takes as a son his daughter's eldest 
 son.
 
 270 THE PATRIARCHAL THEORY. 
 
 there was in this case no husband living either to make 
 an appointment or to own the wife and child is felt by 
 him to be a difficulty. And so it is. The Levirate 
 did not depend, as the practice from which he seeks 
 to derive it did, upon a husband's will ; and, with the 
 Levirate, the affiliation of the child could not result, as 
 with that practice it might do, from actual ownership 
 of the mother. From Mr. Mayne's point of view, there- 
 fore, the Levirate plainly is not in the same position 
 with the Niyoga. And Mr. Mayne perceives this ; and 
 perceives, too, that it cannot be brought under the theory 
 of sonship of which he found in the Niyoga an apt 
 enough illustration. Indeed, in trying to show it to be 
 an extension of the Niyoga, he gives up his theory of 
 sonship. He says that, with the Levirate, "the ele- 
 ment of fiction was introduced," and he throws out the 
 suggestion that the fiction may have been that the 
 husband was regarded as surviving in the wife which, 
 whether a fiction likely to be resorted to or not, 
 would be a fiction consistent with his theory. But, 
 merely mentioning this, so far as he commits himself 
 to any meaning, his view is that there was introduced 
 with the Levirate a fiction, or rather a false pretence, 
 of paternity of the paternity, that is, of the putative 
 father that one thing which, in the Niyoga, was 
 treated as absolutely of no importance. 
 
 Mr. Mayne does not attempt to show how, in ex- 
 tending the Niyoga, a fiction so inconsistent with it 
 came to be used. And this, of course, means that he 
 has failed, on his own view of the Niyoga, to account
 
 SONSHIP AMONG THE HINDOOS. 271 
 
 for the Levirate as an extension of the Niyoga. The 
 Levirate child could not be reckoned a son of his 
 putative father in virtue of those facts which, on Mr. 
 Mayne's view, gave the Niyoga child the position of a 
 son ; and (according to Mr. Mayne) he was so reckoned 
 in virtue of a fiction which is not derivable from the 
 Niyoga. If there was a fiction of paternity with the 
 Levirate, Mr. Mayne leaves us to seek for ourselves an 
 explanation of it and therefore leaves it as open a 
 question as ever whether it was derived from Tibetan 
 polyandry or not. 
 
 All this, however, must seem very like trifling ; 
 seeing that, beyond the inquiry whether the Levirate 
 was an extension of the Niyoga, there is the question, 
 what originated the Niyoga itself? The early cases 
 in which the person called in was a stranger clearly 
 do not belong to a regulated practice, and appointment 
 is not the word to apply to them husbands could do 
 what they pleased, and it is more accurate to say that 
 they selected or requested than that they appointed. 
 When such cases could occur, there can have been no 
 regulated practice though what afterwards was the 
 regulated practice may have been customary practice 
 even then. At any rate, in the Niyoga as embodied in 
 the law, we find the choice of appointment restricted 
 exactly as it was in the Levirate ; and this raises in con- 
 nection with it the very question raised by the Levirate. 
 Strange as it may seem, Mr. Mayne has not noticed this. 
 The element in the Levirate which can be taken as 
 indicative of Tibetan polyandry is the substitution for
 
 272 TEE PATRIARCHAL THEORY. 
 
 the husband of his brother, or, brothers failing, of 
 his kinsman. The same substitution occurred in the 
 Niyoga and Mr. Mayne has not given this a thought. 
 If we grant to him, however, that every child of a 
 married woman was counted her husband's child, it 
 is quite clear that there remains the question, why 
 was it required with the Niyoga (as with the Levi- 
 rate), that the father should be the husband's brother 
 or kinsman ? To overlook this question is to miss 
 what is the problem of both Levirate and Niyoga 
 altogether. 
 
 Mr. Mayne has, moreover, not thought of con- 
 sidering why, among the early Hindoos, physical 
 paternity was, as regards paternal right, so immaterial 
 as he takes it (justly, we think) to have been, or what 
 may be involved in that. He appears to have taken 
 this to be absolutely unimportant. Having found (or 
 thought he had found) among the early Hindoos wives 
 owned, paternity of no account, and maternity so 
 far the determinant of sonship, he concluded forthwith 
 that he had got to the primitive state, and had found 
 it quite consistent with the Patriarchal Theory. 
 
 It remains to be added that Mr. Mayne has tried 
 to account for the Levirate independently of polyandry 
 in the Hindoo case only, and by circumstances which, 
 so far as he discloses, are peculiar to that case. He 
 has not troubled himself about other cases. But the 
 Levirate has been widely spread ; and all cases of it, 
 whatever minor differences there may be between them, 
 agree at least in this, that a man " raises up seed " to
 
 SONSHIP AMONG THE HINDOOS. 273 
 
 his brother or kinsman deceased. Is it likely that men 
 were led to that in one place by one set of circum- 
 stances, in another place by another set and that there 
 may be as many explanations of it, therefore, as there 
 are instances of it ? On the contrary, are we not 
 forced, or almost forced, to think that its origin was 
 everywhere the same ; that what would be a good 
 account of it in one case would be a good account of 
 it in every case ; and that an account of it evolved 
 from one case which does not suffice for the others 
 must be bad ? If this be conceded, Mr. Mayne's view, 
 could it be entertained, would involve rather startling 
 consequences. Had the early Hebrews the Niyoga, 
 and were they as indifferent as the early Hindoos 
 who were the fathers of their children ? Perhaps so ; 
 but this has never been laid to their charge ; and 
 it imputes to them practices in comparison with which 
 Tibetan polyandry may be deemed respectable.* 
 
 * We do find among the early Hebrews a practice analogous to the 
 Niyoga, but one which indicates a different state of social relations 
 from that which is being considered among the Hindoos that a wife 
 could appoint her female slave to bear a child for her, which should 
 be counted her child. 
 
 Upon this Professor Robertson Smith contributes the following 
 note, which, besides its bearing upon the matter in hand, contains 
 additional Scriptural evidence of weight against the Patriarchal 
 Theory. 
 
 " The slave in this case," he says, " is the slave of the wife, and 
 does not cease to be so, Gen. xvi. 6, 9 (a chapter which is by a different 
 and older hand from Gen. xxi., in which, at v. 12, Hagar is Abraham's 
 bondwoman). It appears, then, that the case would most naturally 
 arise in a state of society where the child belonged to the mother 
 rather than to the father, and of this we have a striking piece of
 
 274 THE PATRIARCHAL THEORY. 
 
 If it be not conceded that there must be a common 
 explanation for all cases of the Levirate, a theory re- 
 stricted like Mr. Mayne's to a single case could not, even 
 if accepted, appreciably lessen the difficulty which the 
 Levirate makes for the Patriarchal Theory. 
 
 While, as we have seen, a careful study of Indian 
 facts has impressed upon Mr. Mayne albeit a supporter 
 of the Patriarchal Theory that physical paternity was, 
 in the theory of ancient Hindoo law, not material to 
 sonship, Sir Henry Maine's account of the Levirate sets 
 out from a view of an entirely different sort. Putting- 
 aside the law-books (though it is to such writings that 
 Comparative Jurisprudence makes its appeal), making 
 no appeal to the Hindoo writings that are more ancient 
 than them, Sir Henry Maine assumes the Patriarchal 
 Theory, and makes his start from that. * 
 
 evidence in 1 Sam. i., where the mother has power to give her child 
 out of her family by devoting, or, as she calls it, "giving " or "lending " 
 him to Jehovah, so that he leaves his own people and is incorporated 
 among the ministers of the Sanctuary. By the later law (Num. xxx. 
 3-9) the vow of a daughter or a wife is not valid if her husband dis- 
 allows it ; but even if this held in the time of Samuel, and Elkanah 
 had to confirm his wife's vow, it is clear that she could not make a 
 vow of a son that was not hers but the father's. Note, by the way, 
 that as only daughters in the father's house, and not sons, require the 
 consent of the father to their vows, Num. xxx. proves that sons were 
 not in manu." 
 
 * In his Early Law and Custom, London, Murray, 1883. This 
 assumption cannot be fully treated of until the Hindoo facts have been 
 set forth. But something may be said of it at once. Sir Henry Maine 
 begins, then, with some observations on Ancestor- Worship a large 
 and very difficult subject, of which he does not profess to have made 
 a special study. He concludes (at p. 75) that, when ancestor-
 
 SONSHIP AMONG THE HINDOOS. 275 
 
 This assumption gives him, for the continuation of 
 the Hindoo family, to begin, a basis of legitimate sons. 
 His theory always involved that these could be added 
 to through the fiction of adoption ; and finding that 
 the ancient Hindoo reckoned as his sons the (as we 
 should say) illegitimate son of his wife born before 
 or after marriage and the illegitimate son of his 
 daughter, he suggests that such persons also were 
 assumed as sons in virtue of some fiction, and dubs 
 them fictitious sons. A man who had a daughter, and 
 no son, could "appoint" his daughter to bear a son for 
 him ; but, instead of throwing doubt on the Patriarchal 
 Theory, this only appears to him extremely interesting 
 as probably marking one of the points at which the 
 right of women to inherit made its way into the strict 
 
 worship arose, paternity was fully recognised. He proceeds (at p. 86) 
 to a vastly stronger proposition that, for an explanation of ancestor- 
 worship and its legal consequences, we must assume that when 
 ancestor-worship arose, " the Father of each family appeared to them 
 in the form in which he constantly shows himself on the threshold of 
 jurisprudence" that is, with the position and powers of a Koman 
 Paterfamilias. Here, as a reason for the assumption of the Pater- 
 familias and the Patriarchal Theory, is the statement which reiteration 
 has made so familiar. But we have seen in what number of cases Sir 
 Henry Maine can show us the Paterfamilias " on the threshold of 
 jurisprudence." "We had been led to look for him on the threshold of 
 Hindoo jurisprudence. But it is just because he cannot find him there 
 that Sir Henry has gone to Hindoo ancestor-worship in search of him. 
 The reason given is a bad reason not good in fact and the assump- 
 tion, not being justified by it, is the merest assumption. 
 
 The " legal consequences " to which Sir Henry Maine makes vague 
 allusion are those very facts which impressed upon Mr. J. D. Mayne 
 that paternity was not the determinant of sonship in the theory of 
 Hindoo law. 
 
 T 2
 
 276 THE PATRIARCHAL THEORY. 
 
 agnatic system of kinship and succession which (he 
 assumes to have) prevailed among the early Hindoos. 
 When he comes to the Levirate, he begins by setting 
 forth what the facts were. He tells us (Early Law and 
 Custom, p. 100) that, "in the opinion of some of the 
 Hindu doctors," a man who had died leaving no children, 
 real or fictitious, might have an heir born to him by means 
 of the Levirate; and that "here and there" (p. 102), the 
 practice of substituted begetting received an extension 
 even more revolting to modern delicacy in the appoint- 
 ment by a husband of another man to beget a child for 
 him. It thus appears that, instead of regarding the 
 Levirate (so to call it still for convenience sake) as an 
 extension of the Niyoga, as Mr. Mayne does, he regards 
 the Niyoga as an extension of the Levirate. Both 
 practices, however, being now exhibited as of paltry 
 dimensions, he proceeds to account for them as 
 follows : 
 
 " Let us suppose," he says, " that in a par- 
 ticular society an intense desire has arisen for male 
 issue, whether through its worship of ancestors or 
 otherwise. Let us assume that in a particular case 
 actual issue of the father's loins is impossible. There 
 are no daughters. The accepted fictions, by which sons 
 are created for the sacrifice, cannot be made serviceable. 
 What is to be done, that the name of the aged or dead 
 man be not put out on earth nor his lot placed in 
 jeopardy beyond the grave ? Now all ancient opinion, 
 religious or legal, is strongly influenced by analogies, 
 and the child born through the Niyoga is very like a
 
 SON SHIP AMONG THE HINDOOS. 277 
 
 real son. Like a real son, he is born of the wife or 
 the widow; and, though he has not in him the blood of 
 the husband, he has in him the blood of the husband's 
 race. The blood of the individual cannot be continued, 
 but the blood of the household flows on. It seems to 
 me very natural for an ancient authority on customary 
 law to hold that under such circumstances the family 
 was properly continued, and for a priest or sacerdotal 
 lawyer to suppose that the funeral rites would be per- 
 formed by the son of the widow or of the wife with a 
 reasonable prospect of ensuring their object. The very 
 differences of opinion which arose on the subject in the 
 most ancient Brahmanical law-schools seem to me 
 exactly those which would be provoked by a plausible 
 and yet non- natural contrivance." * 
 
 The first sentence of this passage shows a desire to 
 make the explanation apply to all cases of the Levirate. 
 But it is plain that the passage has been written with 
 Indian facts only in view. 
 
 The son born through the Niyoga, we are told in 
 this passage, was " very like a real son." But it is 
 scarcely necessary to point out that he could not appear 
 so to a people in whose theory paternity was of the first 
 importance. A married woman's child by another man 
 than her husband does not to modern people seem 
 " very like " a child by her husband. That the man is 
 her husband's brother does not make it more like that 
 " the blood of the household flows on " in the child 
 makes no difference. And the reason is that we think 
 * Early Law and Custom, pp. 106-7.
 
 278 THE PATRIARCHAL THEORY. 
 
 paternity as important as maternity. That being so, 
 the blood of the household does not satisfy us any more 
 than the blood of a stranger. And it could not have 
 satisfied the Hindoos had they had the modern view 
 about paternity. It seems to follow that they had a 
 lower or less exigent view, consistently with which the 
 child could pass for the son of the husband. How 
 much lower then how much different from our view ? 
 Sir Henry Maine has not assisted us here, as he might 
 have been expected to do ; seeing that it is clear that 
 the Niyoga child could not to people of all views alike 
 appear " very like a real son." 
 
 Fortunately, we are not dependent upon con- 
 jecture in the matter. The view on which the Niyoga 
 was practised is clearly disclosed to us by Apastamba, 
 and it most amply accounts for the practice. " They 
 declare," says Apastamba, " that a bride is given to 
 the family (of her husband and not to the husband 
 alone)." * To people holding that view of marriage, a 
 Niyoga child would seem not merely " very like " a real 
 son, but quite a good son it was enough for them that 
 he was the child of the wife by one of the family 
 their theory did not exact that the husband should be 
 the actual father. Need it be said that it is a view of 
 marriage that might have come straight out of Tibetan 
 polyandry ? 
 
 Apastamba is by Sir Henry Maine taken to be one 
 of the most ancient and one of the greatest of autho- 
 
 * Apastamba II., 10, 27, 3. Sacred Books of the East, Vol. II, 
 p. 164. Oxford, 1879.
 
 SON SHIP AMONG THE HINDOOS. 279 
 
 rities ; but in this instance, and in another that will 
 have to be referred to, he ignores Apastamba's evidence 
 as to the views or ways of his countrymen. What 
 Apastamba discloses, however, supersedes conjecture as 
 to the view on which in his time the Niyoga was prac- 
 tised ; and will explain perfectly its being an established 
 custom. Sir Henry Maine, nevertheless, has felt free 
 to set it aside and to conjecture for himself. And his 
 conjecture is that we have in the Niyoga " a plausible 
 and yet non-natural contrivance" for continuing 
 the family. 
 
 He had to account for the Niyoga, however, not 
 as a means of continuing the family it was not that 
 which was in danger of becoming extinct but as a 
 means of continuing the line of a particular member 
 of the family. He has therefore not faced his problem. 
 And he has really not even taken a step towards 
 accounting for the Levirate independently of poly- 
 androus views. 
 
 It must be said, moreover, that the whole view of 
 Hindoo facts which culminates in this account of the 
 Levirate appears to be untenable. 
 
 Sir Henry Maine, as we have said, has assumed 
 not by way of hypothesis that the Hindoos started 
 with the Patriarchal Family. He has accordingly had to 
 treat the facts available for judging what their early 
 family life really was as showing aberrations from this 
 primitive model (because they are ex facie irreconcilable 
 with it). It then became his task to account for such 
 aberrations no easy matter. His preliminary assump-
 
 280 THE PATRIARCHAL THEORY. 
 
 tion seems to have here led him into underrating the 
 importance of the supposed aberrations, and into 
 assigning for them a time of origin which would allow 
 something like a breathing space for itself. Facts 
 which look as if they must be very early are accordingly 
 presented to us as late. Facts which used to be thought 
 important dwindle until it scarcely matters whether a 
 good account is given of them or not. And other 
 facts which should have been reckoned with have been 
 neglected altogether. 
 
 What has now been said will, we venture to think, 
 be justified in course of this chapter. 
 
 It might be added that Sir Henry Maine has, with 
 the utmost confidence, treated as the oldest and most 
 authoritative of Hindoo law writers certain authors who 
 have generally been regarded as neither the oldest nor 
 of the highest authority, to whom of late a great 
 antiquity has been assigned upon reasoning the feeblest 
 and most inconclusive and which, on the best view of 
 it, could only justify a tentative opinion. But this need 
 not be urged. For those authors do not contribute 
 to the establishment of the Patriarchal Theory. Apas- 
 tamba is one of them. We have already had occasion to 
 make use of him, and we shall have again to refer to 
 him by-and-by. 
 
 We go on now to consider how Sir Henry Maine, 
 upon his assumption of the Patriarchal Theory, accounts 
 for the sonship of the illegitimate, or adulterine 
 offspring whom Hindoo custom classed as sons. Instead 
 of their being first so classed when the Hindoos were
 
 SONSHIP AMONG THE HINDOOS. 281 
 
 uncivilised and not too nice, he takes it that they came 
 to be counted in some degree as sons only when the 
 Hindoos had fallen off from their original strictness 
 under the influence of religion, which made them 
 earnestly desire to have male offspring. At times, as 
 has been said already, he seems to think such persons 
 were reckoned as sous in virtue of fictions comparable 
 to the fiction employed in adoption, and he calls them 
 fictitious sons. He has made no suggestion as to what 
 the fictions were ; and it is idle to speak of fictions 
 unless the fictions can be indicated. But this need 
 not be insisted upon ; for he goes on to explain how, 
 consistently with Patriarchal custom, such persons might 
 have been counted as sons, and here he dispenses 
 with fiction altogether. 
 
 " They are all," he says of the illegitimate and 
 adulterine sons, " the offspring of women who are under 
 the shelter of the household, or who are brought under 
 it. These women are under the protection of its head ; 
 they belong to him, and the status of their children is 
 settled by the well-known rule which, in Eoman law, 
 would settle the status of a slave." Paternal power 
 and protective power, he proceeds, are inextricably 
 blended ; even the slave was in some sense a member 
 of the family ; and those children were " permitted to 
 rank as in some remote degree sons" because their mothers 
 were protected by the head of the household, and 
 because they were themselves protected by him.* This 
 is not so precisely expressed as could be desired ; but 
 * Early Law and Custom, pp. 95-99.
 
 282 THE PATRIARCHAL THEORY. 
 
 the effect of the whole is that the child was ranked as 
 " in some remote degree " a son because both his mother 
 and himself were under the protection of the man who 
 was counted " in some remote degree " his father. He 
 ranked as a son, as far as he did so, because he was his 
 mother's son and both were protected by the head of the 
 household who had what was equivalent to paternal 
 power over him. 
 
 The use of a fiction, be it observed at any rate, 
 if it operated like the fiction of adoption would have 
 been more effective. Adoption would have made the 
 child of the adopter's stock and family a son in the 
 fullest sense. The protection of mother and of child, 
 however, according to Sir Henry Maine, made the child 
 rank as a son only in some imperfect sense. 
 
 What he has here tried to show is something less 
 than the case required. But that may be passed over 
 for the moment. It is proper first to point out that he 
 has not kept in view what being a son involved among 
 the Hindoos. It at least involved that son and father 
 should be of the same stock (gotra). And it is evident 
 that mere protection could not introduce a man into a 
 stock to which he did not belong. This explanation 
 therefore is at any rate defective. And next, there 
 is something more serious to say of it. The Hindoo 
 facts appear to exclude it. There are several facts, 
 indeed, which, taken singly, are enough to put it out of 
 the field. 
 
 Considering the passage above referred to, one may 
 perhaps discern an argumentative reason for defining a
 
 SONSHIP AMONG THE HINDOOS. 283 
 
 wife or a daughter for the heirship of the child of an 
 unmarried daughter is among the facts to be reckoned 
 with as " a woman protected by the head of the house- 
 hold." But it is a very odd definition, and very far 
 from being exhaustive it would apply to a female 
 slave. It takes no account of the relation of husband 
 and wife, of father and child, which, in the one case and 
 in the other connected the man with the woman. And 
 yet both of these relations generally involve consequences 
 of importance. When a relationship to the man arose 
 through the woman, surely it is necessary to inquire 
 how far it was attributable to the relationship previously 
 subsisting. It would seem that a theory which in such 
 cases carefully excludes consideration of that relation- 
 ship has been planned so as to give it every chance of 
 being wrong. 
 
 Let us look at the case of an unmarried woman's 
 child. So long as the woman remained unmarried the 
 child belonged to her father, and was among his heirs. 
 In virtue of what ? Is it absurd to say that it was 
 simply in virtue of the tie which connected the man and 
 the woman ? On the contrary, there is no need in this 
 case even to consider whether Sir Henry Maine's sug- 
 gestion can account for the child's position. There is 
 plainly no room for it. For the relationship which was 
 precedent to protection of the child and laid the founda- 
 tion for it the relation between father and daughter 
 accounts for the child's position. An unmarried woman 
 was of the stock of her father, and he had a father's 
 rights over her. Her child might have belonged to its
 
 284 TEE PATRIARCHAL THEORY. 
 
 own father. But by Hindoo custom it never did. It 
 might have been of its father's stock. But by Hindoo 
 custom it never was. The alternative was that it should 
 " follow the mother" and belong to her father who had 
 rights over her. And by Hindoo custom it was her 
 father's and that because its mother was his. More- 
 over, it was of his stock which was also its mother's. 
 That is shown by its having been his heir. Being his 
 daughter's child, and of his own stock, it was among his 
 heirs. Here we see that parental right over the mother 
 was the foundation of parental right over her child. 
 And parental right over the child was the basis of the 
 protection given by the woman's father, and not in any 
 way an effect of it. 
 
 That without adoption for of that nothing is any- 
 where said the child of an unmarried woman was of her 
 father's stock, which was also his mother's, yields, it 
 would seem, irresistibly the inference that the stock 
 could be transmitted through the mother that the child 
 in this case took blood or stock connection from the 
 mother ; an inference to be noted, because of its relation 
 to the familiar facts which first raised doubts about the 
 Patriarchal Theory. 
 
 To proceed with this case when the woman married, 
 the Hindoos being exogamous, she married a man of 
 another stock than her own, and she passed into his 
 stock. The child perhaps could be retained in her 
 father's family and if the father had no son he would 
 be retained as a matter of course. But if he were taken 
 over with his mother and that "the damsel's son"
 
 SONSHIP AMONG TEH HINDOOS. 285 
 
 was the descriptive title of a class of sons seems to show 
 that this happened commonly and in pursuance of a 
 customary arrangement what founded the connection 
 between him and his mother's husband ? Surely it was 
 marriage his mother's marriage. Protection came after- 
 wards as a consequence of the connection so founded. 
 He was assigned to her husband along with her. That he 
 was now counted her husband's son shows that he was 
 now counted of her husband's stock ; and that shows very 
 clearly that he "followed the mother" in the change 
 of stock. Mother and child passing (by admission to 
 his sacred rites, no doubt) into the husband's stock, 
 the man got a wife and a child at the same time a 
 thing not perhaps " patriarchal," but, nevertheless, by 
 no means unexampled. 
 
 Here we find in marriage, as in the previous case we 
 found in the relation of father and child those elements 
 in creating relationships which Sir Henry Maine has 
 forgotten the explanation which he has sought in mere 
 protective power. The problems of this case, however 
 how to account for the child being first affiliated to the 
 stock and family of the mother's father, and afterwards 
 to the stock and family of her husband; and howtoexplain 
 this happening so systematically as to give rise to a 
 legal description of a class of sons seem scarcely to 
 have had full consideration from him. 
 
 As to the adulterine offspring of a married woman, 
 while the law always recognised their sonship, and the 
 most ancient custom gave them to the husband, it 
 appears to be quite clear that the husband's right
 
 286 THE PATRIARCHAL THEORY. 
 
 to them was founded upon marriage and nothing 
 else. 
 
 It is in the salient or peculiar features of any body 
 of custom at the points in which it differs from other 
 bodies of custom or law in what is distinctive in the 
 practices permitted by it that we can discern the basis, 
 if there be any, on which the whole has been built 
 up. And if all that is distinctive is found to be con- 
 sistent, and referable to the same basis, and that which 
 is not distinctive opposes no difficulty, we cannot hesi- 
 tate to believe that we have found the basis of the 
 whole. Let us search then the quest will not occupy 
 us long in the differentiating points of Hindoo practice 
 for the basis of Hindoo family law. 
 
 We have seen that, in very early times, a man 
 could appoint, or call in, a stranger to beget a child 
 for him, and that the child was his own and not the 
 actual father's. The child was born to be his son ; 
 and that by itself is enough to dispose in this case 
 of Sir Henry Maine's protective power theory and 
 enough, indeed, to make an end of it altogether as an 
 
 O ' * o 
 
 account of the putative sonship of the Hindoos. More- 
 over, it was not " in some remote degree " a son, but 
 a highly-valued son ; and it was not brought into 
 existence for lack of children but because the qualities 
 of the sire gave promise of a child of superior excellence. 
 Next, it is plain that at this early period a child did 
 not belong to the begetter, or at least that it need 
 not belong to him. It is plain, too, that at this 
 period a Hindoo need not have been the actual father
 
 SON SHIP AMONG THE HINDOOS. 287 
 
 of his own son. And since the child was certainly not 
 the father's child in virtue of physical paternity, the 
 foundation of his right to it can only have been that 
 he was the husband of the mother. It was his in 
 virtue of his marriage. It was his because it was the 
 child of his wife. 
 
 The case, which has already been dwelt upon, of the 
 son of an unmarried woman, carries us a step further. 
 The child "followed the mother," and was of her 
 father's family, the actual father having no right to him. 
 Here there was nothing which gave the father of her 
 child a right over the woman, and we see that the 
 paternity gave him no right to the child that is, that 
 the child did not belong to the begetter. We find 
 also that the child belonged to the woman's father, 
 to whom she herself belonged that is, that it belonged 
 to him who possessed rights over the mother. Bearing 
 in mind that, when a woman married, the rights of 
 her father were transferred to her husband, and 
 recurring to the case last dealt with, it becomes quite 
 clear that the appointed person could not in that 
 case, as being actual father (even had there not been an 
 implied contract to the contrary), have any right to his 
 own child ; and also that the rights over the mother 
 which marriage had passed from her father to her 
 husband were enough to explain how the child was 
 counted the husband's. The child was his because he 
 alone possessed rights over the mother. 
 
 The case of an appointed daughter carries us further 
 still. In that case, when the woman married and her
 
 288 THE PATRIARCHAL THEORY. 
 
 father had no son, the right to her first son was reserved 
 by her father. Hence we see that, though the husband 
 might be, and in all probability was, the father, that 
 fact gave him no right to his child by his own wife. The 
 son did not belong to the begetter even with marriage. 
 It did not even belong to the husband of the mother. 
 It went back to the mother's father. Why ? Because 
 the father, who had given the woman to the husband, 
 had not given her unconditionally, but with reservation 
 of one of her offspring to himself. That which he could 
 reserve must have been part of that which in ordinary 
 cases was given he could reserve nothing else. And 
 we see that he could reserve one of his daughter's off- 
 spring even when it was also the offspring of her hus- 
 band. We know that a woman's offspring, while she was 
 unmarried, belonged to her father. It follows that a 
 husband's right even to his own children by his wife 
 sprung out of the right which at marriage passed to him 
 from her father, and that he got no right to them merely 
 as being actually their father.* And while he got the 
 
 * Apastamba, the only Hindoo writer who gives full logical effect 
 to the doctrine that a son belongs to the begetter, consistently oinits 
 mention of the appointment of a daughter. He prefers to allow to a 
 daughter rights of inheritance, failing sons. It will be seen hereafter that, 
 in giving his cherished dogma full play, he has had either to forbid or 
 to omit everything that was distinctive in Hindoo family custom. But 
 he stands alone among early authors ; and his work, so far as it 
 relates to family custom, is throughout a protest against the existing 
 customs and not an account of them. This appears beyond doubt 
 from disclosures made by himself. It may be mentioned that while, 
 as stated above, he admits daughters to inheritance, he does not 
 recognise adoption. The Patriarchal Family would not have pleased 
 him.
 
 80N8HIP AMONG THE HINDOOS. 289 
 
 right to have children by his wife who should be his 
 own children, if he chose it, he got more. The first case 
 that has been noticed is enough to show that anciently 
 Hindoos chose to get more. 
 
 How far the right passed to the husband at marriage 
 might extend is to go on to another case to be 
 gathered from the Levirate ; from which we see that a 
 man might have issue of his wife counted his even 
 after he was dead. Children so counted to him were 
 children by his brother. But it was not any children of 
 his brother who could be thus counted to the deceased. 
 It was necessary that they should also be children of 
 the woman to whose issue he had acquired right by 
 marriage. The Levirate shows us once more that 
 physical paternity by itself gave no right, and that 
 Hindoo fatherhood was not dependent upon it. 
 
 It may be said by the way that the facts considered 
 above seem to point beyond serious question to marriage 
 among the Hindoos having, in early times, commonly 
 proceeded upon contract and that, in the ordinary 
 case, a contract for the transference from father to 
 husband of a woman and with her, her issue. A right 
 a substantial, a highly valued one passed from the 
 father to the husband. On occasion, a part of it could 
 be reserved. Out of what could the custom of reserving 
 this part come if not out of a system of bargaining ? 
 The Hindoos were exogamous, and among early exo- 
 gamous peoples, who had got beyond capture, purchase 
 was the only way of getting a wife. A man who had 
 to look for his wife to people who, being strangers to 
 
 v
 
 290 THE PATRIARCHAL THEORY. 
 
 him in blood, owed him no duty, could get a wife 
 (putting capture aside) in no other way. 
 
 The Hindoos retained in historic times a re- 
 mainder of marriage by capture. Marriage by the 
 Asura-rite that is, by purchase, though not approved 
 of by the authorities, was always recognised as among 
 the forms of marriage. Moreover, it was a custom, 
 coming down from very early times that, at the time of 
 marriage, the bridegroom should make a gift and a 
 very substantial one to the bride's father. This it was 
 the duty of the latter to return. But why was the gift 
 made ? How arose the custom of making it ? Was 
 there not a time when the father did not return it ? 
 Apastamba protests that the use of the word sale, in 
 reference to these transactions (which occurs in some 
 Smritis), is metaphorical. But, plainly, we have here a 
 symbolism of purchase ; and the word "sale" would have 
 literally described that which was symbolised.* We 
 need not hesitate then to conclude that marriage was 
 commonly among early Hindoos, as their affiliations of 
 children indicate it to have been, an affair of sale and 
 purchase. And it may further be concluded since one 
 
 * Apastamba II., 6, 13, 12. The passage is as follows: "It is 
 declared in the Veda that at the time of the marriage a gift for (the 
 fulfilment of) his wishes should be made (by the bridegroom) to the 
 father of the bride, in order to fulfil the law. ' Therefore he should 
 give a hundred (cows) besides a chariot; that (gift) he should make 
 bootless (by returning it to the giver).' In reference to those 
 (marriage-rites) the word ' sale ' (which occurs in some Smritis is 
 only used as) a metaphorical expression, for the union (of the hus- 
 band and wife) is effected through the law."
 
 SONSHIP AMONG THE HINDOOS. 291 
 
 child could be reserved that the contract did not in 
 early times of necessity whether commonly it did so 
 or not carry to the husband the right to the children. 
 
 It may be added that Hindoo custom always allowed a 
 husband to communicate to another man by contract the 
 right to have a son by his wife. The " son of two fathers ' 
 was the child of a married woman by a stranger, to whom 
 her husband had made her over upon contract for 
 a price to let him have a son by her. Where such 
 a contract was made, the child might, if the contract 
 were to that effect, belong entirely to his actual father 
 of course, in virtue of the contract. But, in general, he 
 was to be the son of both, and so was called the son of 
 two fathers. The husband of the mother, though not 
 his father, had him as a son in virtue of his right to the 
 woman's issue. The actual father had him as a son in 
 virtue of the contract. He could perform the usual 
 rites for both his fathers. But we learn from Gautama 
 that the actual father's right was rather precarious. For 
 he might lose it by leaving the rearing of the child to 
 the husband. The case is curiously illustrative of the 
 ideas connected with Hindoo family right.* 
 
 Having considered the salient or differentiating 
 features of Hindoo practice, we have now found that, 
 by Hindoo custom, (1) the son certainly did not belong 
 
 * The place of contract in connection with early marriage is one 
 of many facts which show that the acceptance of the famous maxim 
 that the movement of progressive societies has been a movement from 
 status to contract will have to be reconsidered. As regards marriage 
 and the family it is clear there has been a movement from contract 
 
 to status. 
 
 u 2
 
 21)2 THE PATRIARCHAL THEORY. 
 
 to the begetter (it has seemed necessary to show this 
 clearly though we may be sure the doctrine negatived 
 has never been embodied in the usage of any people) ; 
 (2) that a husband even did not acquire right to 
 his wife's child in virtue of his being actually its 
 father; but that (3) that was included in the right which 
 passed to him from his wife's father at marriage ; and 
 (4) that, in the ordinary case, the right then passed 
 extended to all the woman's issue, whether actual issue 
 of the husband or not. There is nothing inconsistent 
 with these deductions in any part of Hindoo family 
 custom. 
 
 It is in full harmony with them that adulterine 
 children and the child " taken with the bride " 
 were counted among sons. They are so counted by 
 every old Hindoo law-book except that of Apastamba 
 which omits them. But convincing evidence as to the 
 sonship of adulterine children is to be got even from 
 Apastamba. 
 
 As matter of theory, or rather of metaphysic 
 the general practice (that is, as we should say, the 
 actual law) unaffected the sonship of adulterine 
 children seems to have been more or less in con- 
 troversy among speculative Hindoos from an early 
 period. The opinion hostile to it was the dogma 
 that the son belonged to the begetter ; and the in- 
 genuity of its supporters put it in a form in which, 
 its antagonism to custom being admitted, it was not 
 unlikely to have much influence upon practice. The 
 ways of this world being against them, they appealed
 
 SONSHIP AMONG THE HINDOOS. 293 
 
 to the ways of the next. They represented that a son 
 (though belonging to the husband of the mother in 
 this world) belonged to him who had begotten him 
 in the world of Yama so that no spiritual benefit was 
 to be got by the reputed father through him. And we 
 learn this very clearly from Apastamba. 
 
 " They quote also," says Apastamba " (the following 
 Gatha from the Veda) : ' (Having considered myself) 
 formerly a father, I shall not now allow (any longer) 
 my wives (to be approached by other men), since they 
 have declared that a son belongs to the begetter in 
 the world of Yama. The giver of the seed carries off 
 the son after death in Yama's world, therefore they 
 guard their wives, fearing the seed of strangers. Care- 
 fully watch over (the procreation of) your children, lest 
 stranger seed be sown on your soil. In the next world 
 the son belongs to the begetter, an (imprudent) hus- 
 band makes the (begetting of) children vain (for 
 himself).'"* 
 
 Here we find a careless husband, who had been 
 content with the children his wives gave him, because 
 he regarded their children as his own, alarmed at the 
 suggestion represented as new to him that they 
 would not be his in the world of Yama. That they 
 could belong to anybody else in this world the new 
 doctrine did not suggest. It threatened careless 
 Hindoos with what would happen in the world of 
 Yama to make them cease from relying upon what 
 was settled among the people they were living with. 
 * Apastamba II., 6, 13, 7.
 
 294 THE PATRIARCHAL THEORY. 
 
 If proof were needed, there could not be clearer proof 
 that a child did not belong to the begetter in this 
 world. There seems to be no doubt entertained, says 
 Mr. J. D. Mayne, commenting upon this passage, that 
 the son begotten on a married woman by a stranger 
 belonged in this world to her husband. 
 
 It is Hindoo practice we are concerned with, and 
 especially ancient practice. And here we have it 
 disclosed by Apastamba himself an extreme supporter 
 of the reforming doctrine that, in ancient times at 
 any rate, Hindoo practice did not give the son to the 
 begetter but to the husband of the mother. 
 
 Apastamba has given effect so thoroughly to the 
 doctrine that the son belongs to the begetter that he 
 has either omitted or forbidden everything that was 
 distinctive in Hindoo custom discarding among other 
 things, that sonship by adoption of which the 
 Patriarchal Theory has made lavish use. But he 
 stands quite alone. And as we learn from himself 
 besides having other conclusive evidence that ancient 
 practice was against his view, and as his authority 
 (if he really was a very early writer) did not affect 
 subsequent practice, his opinion, however interesting, 
 seems only to prove that there were Hindoos in his 
 time (whatever it may have been) who strongly dis- 
 liked the ways of their countrymen. He does not at 
 any point say a single word which can suggest that 
 he was making a stand against corruptions of recent 
 introduction. On the contrary (as will be seen from 
 the passage quoted on the next page) he admits that
 
 SON8HIP AMONG THE HINDOOS. 295 
 
 precedents the most ancient, and seemingly the most 
 weighty, could be quoted against him besides disclosing 
 incidentally that the general practice of antiquity was 
 against him. A thorough-going dogmatist, he laid 
 down the law as, on his dogma, it ought to be. And 
 it was with spiritual consequences he threatened those 
 who did not conform to the law so laid down.* 
 
 * See Apastamba II, 6, 13, 6-11. The full text is as follows: 
 " A Brahmana (says) ' The son belongs to the begetter.' Now they 
 quote also the following Gatha from the Veda : ' (Having considered 
 myself) formerly a father, I shall not now allow (any longer) my 
 wives (to be approached by other men), since they have declared that 
 a son belongs to the begetter in the world of Yama. The giver of 
 the seed carries off the son after death in Yama's world ; therefore 
 they guard their wives, fearing the seed of strangers. Carefully watch 
 over (the procreation of) your children, lest stranger seed be sown on 
 your soil. In the next world the son belongs to the begetter, an 
 (imprudent) husband makes the (begetting of) children vain (for 
 himself).' Transgression of the law and violence are found among the 
 ancient (sages). They committed no sin on account of the greatness 
 of their lustre. A man of later times who seeing their (deeds) follows 
 them, falls. The gift (or acceptance of a child) and the right to sell 
 (or buy) a child are not recognised." 
 
 The fact stated about the ancient sages is what is important. It 
 would be mere silliness to ask us to take the explanation along with 
 it. A Hindoo jurist had to maintain that the law had been from the 
 beginning as he was declaring it. And he had a simple way of 
 accounting for the sages having conformed to a ruder sort of law. 
 
 Gautama, more practical than Apastamba, though of the same 
 school, lays down that " a child belongs to him who begat it, except 
 if an agreement (to the contrary has been made)." (Gautama, 18, 9, 
 10). Apastamba admits no qualification of his principle that " a son 
 belongs to the begetter" and rightly. To qualify a view of that 
 sort is to give it up. But Gautama wished to state actual law, and 
 to do that and not to qualify his principle was impossible. It is 
 plain too that he desired to put on a thoroughly good basis the posi- 
 tion of a Myoga child. Notwithstanding his general rule, he includes
 
 296 THE PATRIARCHAL THEORY. 
 
 That so high (and in truth, for this world, so im- 
 practicable) a view about paternity should have been 
 broached among the Hindoos, doubtless at a rather 
 early time, seems to show that, among metaphysical 
 Hindoos, from an early time thinking had been going 
 on about the connection between parent and child. But 
 customs are not made by metaphysicians though meta- 
 physicians may try to amend them* nor are they 
 based on metaphysical theories. It appears from Manu, 
 too,t that the metaphysicians were very far from being 
 agreed among themselves. Manu, though holding the 
 connection of father and child superior to the con- 
 nection of mother and child, distinctly and strongly 
 sustains the established usage which assigned to the 
 husband as children all offspring of his wife. He held 
 that the man who had rights over the woman was her 
 " lord " ; and was entitled to her child by a stranger, just 
 as the owner of land in which seed had been sown by a 
 stranger without agreement, would be entitled to the 
 resulting crop and apparently for the same reason, 
 because the receptacle, as he says, is more important 
 than the seed.J 
 
 the classes of adulterine and illegitimate sons in the family, and gives 
 one of them the " son born secretly " equal rights of inheritance 
 with a legitimate son. The others he admits to inheritance only on 
 the failure of preferable heirs and then only in a fourth of the 
 estate. 
 
 * Were views like Apastamba's most likely to occur where practice 
 did not countenance them ? We find them among the Greeks as well 
 as among the Hindoos. At Home, where there was Agnation, thinking 
 broke it down. 
 
 f Manu ix. 32; x. 70. J Manu ix. 48-55.
 
 SONSHIP AMONG THE HINDOOS. 297 
 
 Writers on the laws learned to discriminate between 
 the various sorts of persons who were by usage regarded 
 as sons, to arrange them in classes, and to make or 
 suggest some distinctions as to their relative rights. But 
 they had, in deference to usage, to count them all as sons. 
 And their writings show that two principal views had been 
 mooted as to the foundation of paternal right views 
 which of course came into actual conflict only over those 
 cases in which, while custom gave sonship, actual pater- 
 nity was admittedly wanting ; one of these views being 
 intended to sustain and account for actual usage, while 
 the other was a protest against usage, and intended 
 to revolutionise it. 
 
 It is perhaps scarcely worth while to add that the 
 family of the Patriarchal Theory would not have satisfied 
 Apastamba. But the use of adoption is as inconsistent 
 with the dogma that the son belongs to the begetter as 
 the sonship of adulterine children, of Niyoga children, 
 of Levirate children, or any other distinctive feature of 
 Hindoo family law ; and accordingly he has not recognised 
 it.* 
 
 The Hindoo customs have come down to us from an 
 extremely remote period, while the Irish laws took the 
 shape in which we have them at a time comparatively 
 recent, and certainly have not escaped Christian influence. 
 
 * Sir Henry Maine has himself noticed (Early Law and Custom, 
 pp. 97-8) that, among early Hindoos, the place of adoption was not 
 an important one. It afterwards became a very important one. It 
 does not seem to have occurred to him that this, so far as it goes, 
 makes against his theory which supposes that free use was made of 
 adoption in the primitive family.
 
 298 THE PATRIARCHAL THEORY. 
 
 Nevertheless, as to some of the matters now being dealt 
 with, there is a surprising amount of coincidence between 
 the two enough to show that the more modern has been 
 arrived at by a movement onwards from the more ancient, 
 or rather that both have been arrived at by a movement 
 onwards from the same system of contracting for marriage. 
 
 From the Book of Aicill (Ancient Laws of Ireland, 
 Vol. III., p. 311) we learn that, among the Irish, the 
 husband of a married woman had a right to her adul- 
 terine son till he was purchased from him by the actual 
 father. The actual father, a man who could prove his 
 paternity, had in virtue of his paternity, a right to 
 purchase. But, until he purchased, and if he did not 
 purchase, the child was the husband's. The father, 
 having proved his paternity by the methods in vogue, 
 and having made full payment body-price, honour- 
 price, the price of fosterage, and what had been paid 
 on account of delicts got the child for his own, and 
 thereafter it was on a par with his legitimate children. 
 Here there was the concession, in favour of paternity, 
 of a right to purchase that was the amount of difference 
 between the Irish custom and the Hindoo. Paternity 
 had gained the right to overcome the ancient effect of 
 contract by means of contract. 
 
 From the same book (Ibid. p. 541), under the head 
 of " Abduction without Leave," we learn that the child of 
 a woman who had been abducted without leave had from 
 her family, unless begotten more than a month after the 
 abduction, belonged not to the abductor but to the 
 mother's family. And if the mother had been forcibly
 
 SQNSHIP AMONG THE HINDOOS. 299 
 
 abducted it belonged to them absolutely they might 
 refuse to sell it to the abductor. It was a moot point 
 whether, if they wished to sell, he could refuse to buy. 
 And if, as the result of such a state of circumstances, he 
 in the end got the child gratis, the law obliged him to 
 educate it, " because it is a good contract for him " 
 (Ibid. p. 543). If the mother had consented to the 
 abduction, the abductor could force her family to sell. 
 In either case the abductor had to pay for his own child 
 "the full price of its life" (Ibid. p. 541). It was, 
 after purchase, considered as " the child of a first wife 
 of contract, or of an adaltrach woman of contract." 
 When there had been an abduction without leave, the 
 woman's family were allowed a month "to force the 
 man to the law " (Ibid. p. 403) to bring him to terms 
 about her or to reclaim her. When she remained with 
 him longer, if there had not been a contract there ought 
 to have been, and her family, for their neglect, lost the 
 right to the offspring (Ibid. pp. 543, 403). 
 
 Here we see that contract about the mother was 
 necessary to take her children from her family and give 
 them to their father ; that children begotten while there 
 was no contract (if the mother's family were not con- 
 sidered to be in fault) belonged to the mother's family 
 and had to be bought from them by the father to whom, 
 if he were altogether without excuse, they could refuse 
 to sell ; but that custom had fixed a time within which 
 a contract ought to be come to, or the woman brought 
 back. That the woman's family lost their right if they 
 did not come to terms about her, or get her back, within
 
 300 THE PATRIARCHAL THEORY. 
 
 a customary period, is the amount of difference between 
 the Irish custom and the Hindoo. Abduction was 
 common among the Irish, and people had settled that it 
 was well to make the best of it. 
 
 That the right of the woman's family to her offspring 
 was originally unlimited where there had been no 
 contract about her that it was older than the limita- 
 tions on it is not to be doubted. These limitations 
 are only intelligible as having been made in the interest 
 of social peace though, no doubt, there was also a con- 
 cession to paternity. But when the abductor was wholly 
 in the wrong, and so long as the woman's family were 
 considered still to be in the right, notwithstanding the 
 man's paternity, the child belonged absolutely^ to the 
 woman's family. 
 
 Among the predecessors of the Irish, therefore, as 
 among the early Hindoos, the children of a woman who 
 had not been parted with by her family always belonged 
 to her family ; and, on the same reasoning, the adulterine 
 child of a woman who had been made over to a husband 
 always belonged to her husband. 
 
 Sir Henry Maine only tried to show that, consis- 
 tently with the Patriarchal Theory, the " fictitious " 
 sons might have been sons "in some remote degree." 
 His reasoning has not gone far enough to explain to 
 us how they could have been sons in any degree ; and 
 it has been shown that the protective power theory on 
 which it rests is simply excluded by Hindoo facts. 
 In fact they were recognised as sons from the earliest 
 times known to us ; and that they were sons is
 
 SONSHIP AMONG THE HINDOOS. 301 
 
 accounted for by the principle laid down by Manu that 
 a child belonged to the man who was "lord" of the 
 mother. 
 
 That principle too has been shown to have under- 
 lain all Hindoo family custom. It gave the child of 
 an unmarried woman to her father. It gave the ille- 
 gitimate child of a married woman to her husband. 
 It gave the legitimate child of a married woman to her 
 husband. And it took his legitimate child by his 
 wife from him, and gave it to his wife's father, in a 
 case in which full right was not conveyed. Eight to 
 the woman then was the foundation of paternal 
 right. And it is the principle that a child belonged 
 to the "lord" of the mother, or rather the whole 
 series of practices in which it was embodied, which 
 had to be derived from the Patriarchal Theory. 
 
 It is however to a view of the growth of society 
 widely different from the Patriarchal Theory that the 
 Hindoo practices lend countenance to reconcile them 
 with that seems utterly impossible. Until the contrary 
 of this is shown, it would not be open to Sir Henry 
 Maine, even if other obstacles to his doing so were 
 wanting, to say that the heirship of the son of an 
 appointed daughter marks an initiatory step in the 
 breakdown of Agnation. No case, indeed, presents 
 more formidable difficulties to an upholder of the 
 Patriarchal Theory than this one. For, besides that, 
 when the daughter married, the husband, in virtue 
 of the reservation made, lost the paternity of his 
 legitimate child, when she did not marry (and there
 
 302 THE PATRIARCHAL THEORY. 
 
 was no need that she should marry), her child was her 
 father's because she was his. And it has been shown 
 that this proves that the father's stock could be trans- 
 mitted through her though a woman. So serious 
 a divergence from the Patriarchal view could scarcely 
 denote an early stage in the breakdown of the Patri- 
 archal Family. There was no difference in the world 
 between a son, or the son of a son, and the son of an 
 appointed daughter. The daughter's son was as good 
 an heir as either, and delivered his grandfather in the 
 next world as well as either. In Manu, too, the con- 
 tinuation of the family through daughters is represented 
 as of immemorial antiquity among the Hindoos. " In 
 this manner," it is stated (Manu ix., 128), "Dacsha 
 himself, Lord of created beings, appointed all his fifty 
 daughters to raise up sons to him for the sake of 
 multiplying his race." 
 
 Going back now to the substituted begetting on 
 widow or wife, it must be said that Sir Henry Maine 
 appears seriously to underrate the facts. He tells us, 
 the Levirate was permissible in the opinion of " some 
 of the Hindoo doctors." And, "here and there," there 
 was an extension of the same thing in what to modern 
 thinking appears a grosser form. It is Hindoo practice 
 that concerns us rather than opinions about it. But 
 that Apastamba, as required by his dogma that the 
 "son belongs to the begetter," disapproved of the 
 Niyoga, seems to be all there is to say for the limita- 
 tion in either of those statements. 
 
 Every other early writer includes the son " begotten
 
 SON8HIP AMONG THE HINDOOS. 303 
 
 on the wife " in his list of persons recognised as sons, 
 and most of those writers give him one of the highest 
 places in the list. Generally he comes after the legi- 
 timate son and either after or before the son of an 
 appointed daughter (to whom Mann, holding him as 
 equivalent to a legitimate son does not give a separate 
 place in his list) ; and his rights as an heir are care- 
 fully defined. Not one of those writers suggests a doubt 
 as to his right to inherit, and when there was no legi- 
 timate son he was held fully equivalent to a legitimate 
 son. Gautama, though he had a strong leaning to 
 Apastamba's doctrine, compromised the principle that 
 a son belonged to the begetter (or rather he gave it up) 
 because he felt it necessary to leave beyond question 
 the position of a Niyoga son. This shows that, instead 
 of being a doubtful practice of exceptional occurrence, 
 the Niyoga was thoroughly established. What better 
 evidence of its recognition could one desire ? 
 
 The recognition of the Levirate child was no less 
 general than that of the son " begotten on the wife." 
 And he had in every respect the position of a legi- 
 timate son he was indeed born to be the heir of the 
 dead man who was counted his father. 
 
 It is true that Manu regarded the practice of sub- 
 stituted begetting with disfavour, and talks of it as 
 " only fit for cattle." But he prescribes the rule for 
 carrying it out enjoining at the same time that only 
 one child or two at most should be begotten on the 
 widow or childless wire. And, while contemning the 
 practice, he discloses that even the " twice- born classes "
 
 304 THE PATRIARCHAL THEORY. 
 
 had had it formerly. He speaks of it as of an ancient 
 practice, which had died out or was dying out among 
 the better sort of people. But he fully recognised it 
 no doubt because it was still too general not to be 
 recognised. 
 
 But even what Apastamba says of the Niyoga bears 
 out what has been said of it above. What he says is as 
 follows : 
 
 " (A husband) shall not make over his (wife), who 
 occupies the position of a 'gentilis' to others (than to 
 his ' gentiles ') in order to cause children to be begot for 
 himself. For they declare that a bride is given to the 
 family (of her husband and not to the husband alone). 
 That is (at present) forbidden on account of the weak- 
 ness of (men's) senses. The hand (of a gentilis is con- 
 sidered in law to be) that of a stranger, as well as 
 (that of any other person except the husband). If the 
 (marriage vow) is transgressed, both (husband and wife) 
 certainly go to hell. The reward (in the next world) 
 resulting from observing the restrictions of the law is 
 preferable to offspring obtained in this manner (by 
 means of Niyoga). "' 
 
 "Were there nothing else known about the Niyoga 
 this passage would be enough to show that it was an 
 established custom in the time of Apastamba ; that the 
 Niyoga child was then a good heir ; that it was then an 
 old-established custom ; and that the customary explana- 
 tion of it was that a wife was given to the family of her 
 
 * Apastamba II., 10, 27, 2-7. Sacred Books of the East, Vol. II., 
 p. 164.
 
 SONSHIP AMONG THE HINDOOS. 305 
 
 husband and not to the husband alone. It also shows 
 (which is less important) that, as his dogma required 
 him to do, Apastamba disliked the Niyoga very much. 
 He suggests that those who concurred in arranging for 
 it would "certainly go to hell." But the last sentence 
 of the passage admits that offspring, that is heirs, could 
 be got by it in this world. Eewards in the next world 
 are promised to those who resolve to forego such offspring. 
 And it will be seen that, in the first sentence, the 
 commentary overrides the text, changing the meaning 
 and stultifying all that follows obviously because what 
 Apastamba had enjoined was so contradictory of all 
 that was known about usage that nobody could take 
 it as a statement of actual law. Apastamba is made 
 by the commentator not to deny (or rather to admit) 
 that the explanation made for the Niyoga had in 
 earlier ages been a sufficient reason for it. And he 
 sets it aside, not as being historically groundless, but as 
 embodying a view of marriage unsuited to the weak- 
 ness of his contemporaries, or, let us say, to human 
 weakness. 
 
 That it was on this view of marriage the Niyoga 
 was justified is the most valuable thing we get from 
 Apastamba. He shows us that a practice which, so 
 far as it went, was equivalent to Tibetan polyandry 
 proceeded upon a polyandrous theory. 
 
 Sir Henry Maine ought, it would seem, to have 
 reckoned with that theory. But after all, it is the 
 account of the Levirate which he has himself pro- 
 pounded that concerns us here. And that has failed
 
 306 THE PATRIARCHAL THEORY. 
 
 completely. He has, to speak plainly, shirked the 
 difficulties of the matter. He has tried to show how 
 it might have been held that " the family " was properly 
 continued by the Niyoga child. What he ought to 
 have shown was how, among a people who had had 
 a strict view about paternity, it could be held that the 
 line of a particular member of the family was properly 
 continued by the Niyoga child. It being plain that 
 that could not be held except where there was some 
 degree of indifference about paternity, the question, 
 what degree of laxity about it whether much or little 
 would make the Niyoga child look " very like a real 
 son," presented itself. But he has not considered it. 
 The consequence is that he has reasoned the matter 
 much as a Conservative Hindoo might have argued it 
 against Apastamba. What he says would be of some 
 weight from a polyandrous point of view. It has no 
 weight at all from a Patriarchal point of view. 
 
 It may be well to add as to the Levirate among 
 the Hebrews since an explanation offered for the 
 Hindoo institution has that to reckon with that 
 there can be no question raised as to its having been 
 an established and an ancient custom. The story of 
 Judah and Tamar shows that the childless widow of 
 an elder brother could claim to have his brothers in 
 succession for her husband ; and that her child by 
 her new husband took the place in the family of 
 the husband who was gone, and was counted to 
 him as a son. And the Levirate was retained in the 
 law of Israel the duty of a husband's brother to the
 
 SONSHIP AMONG THE HINDOOS. 307 
 
 childless widow was prescribed by the law at a 
 much later time when marriages between persons 
 related by blood or affinity were in other cases strictly 
 prohibited. The law took his legitimate child from 
 the new husband and assigned it to the brother de- 
 ceased, that it might be his heir. It was not every 
 child of his that would suffice for the purpose. By 
 itself the blood of the family was useless for this 
 purpose. Among both Hebrews and Hindoos the 
 child who could be assigned to the deceased brother 
 was also a child of the woman who, had that 
 brother lived, would have been the mother of his 
 children. 
 
 It may by this time be permissible to consider it 
 quite hopeless to connect the body of family custom 
 found among the Hindoos with the Patriarchal Family. 
 The interval between the two seems too great to be 
 bridged by any effort of conjecture. For the Patriarchal 
 Theory, as a hypothesis to account for the Hindoo facts, 
 there is, it would seem, nothing to be said. And we 
 are now in a position to feel the full force of the 
 reasons against allowing Sir Henry Maine's assumption 
 that the Hindoos, at an earlier time than any we 
 know about, had the Patriarchal Family as well as 
 to see the weakness of the reason for making it. Let 
 us state briefly how the matter stands. 
 
 The facts of the Hindoo family tell very strongly 
 against the Patriarchal view. It would take very 
 powerful reasons to bear down their evidence against 
 that view, so far as the Hindoos are concerned. And 
 
 x 2
 
 308 THE PATRIARCHAL THEORY. 
 
 to bear them down there is only a conjecture pro- 
 pounded in connection with a subject of the greatest 
 difficulty, the history of which has still to be properly 
 investigated, and is at present much more obscure than 
 the history of the family on which it is sought to make 
 it shed light; but got in reality by means of the 
 averment which there are no means of supporting 
 that we constantly meet with the Paterfamilias " on the 
 threshold of jurisprudence." To bear them down, that 
 is, there is a conjecture for which there is really no 
 basis, and which, though made in course of a disquisition 
 on ancestor- worship, might as legitimately have been 
 made without any reference to ancestor- worship. 
 
 Observe, too, the nature of this conjecture, as 
 related to ancestor- worship. It is that the Patriarchal 
 Family was necessary to give birth to ancestor-worship, 
 and yet could not live with it. The Patriarchal Family 
 was necessary for ancestor-worship to begin with, but 
 too much for it to go on with. It would take some 
 reasoning to show that it is not inconsistent with itself. 
 What is perhaps more important is that it is so framed 
 as to elude investigation. 
 
 Since it is made in connection with ancestor-worship 
 it must be said that, before founding upon that, one 
 would have to inquire whether ancestor-worship had 
 not its rude beginnings and, if so, what the family 
 has been where these have been found ; what the family 
 has been also where it is found in full development; and 
 whether it, in fact, appears to be specially connected 
 with any one set of family relations and powers.
 
 SONSHIP AMONG THE HINDOOS. 309 
 
 Without this having been done, there is no excuse for 
 sweeping inferences from ancestor- worship. 
 
 In inferring the Patriarchal Family from (or in 
 connection with) ancestor- worship, Sir Henry Maine 
 has, of course, assumed that the family underwent no 
 change all the time that ancestor- worship was growing 
 up. He seems to imply that, after that, it deteriorated 
 rapidly the reason being that the Patriarchal Family 
 did not yield children enough for the requirements of 
 ancestor- worship. 
 
 His conjecture having been made specially for the 
 Hindoos, let us see what difficulties there are in its way 
 in the case of the Hindoos. 
 
 First, then, he asks us to believe that paternity 
 was fully recognised when ancestor-worship arose. 
 This is his first step towards the assumption of 
 the Patriarchal Theory ; and is meant to give him the 
 Father, with a position undefined, but founded upon his 
 paternity. He means by paternity physical paternity ; 
 and he means that it was recognised as the basis, or, at 
 any rate, as a basis, of paternal right. Less than this 
 would avail him nothing. 
 
 If we are to conjecture, however, is it not the 
 reasonable thing to conjecture that the paternity which, 
 among the Hindoos, was enough for ancestor-worship 
 to go on with, was enough to set it going ? The Hindoo 
 father, such as he was head of the household, protector, 
 possessor of a father's customary rights, often, no doubt 
 commonly, father in fact would have made, it would 
 seem, as good a basis for ghost-worship (if that is to be
 
 310 TEE PATRIARCHAL THEORY. 
 
 taken to have been the foundation of ancestor- worship) 
 as any father could do. It is true that the legal basis 
 (so to speak) of his parental right was not actual pater- 
 nity, but the fact of his being " lord " of the mother. 
 But it is not easy to see how, as regards that matter, 
 this could make any difference. Why, at any rate, 
 should we think it would have made a difficulty for 
 ancestor-worship at the beginning, when we know that 
 it made no difficulty afterwards ? 
 
 But there is very strong reason against carrying 
 conjecture further. We know that the high doctrine 
 about paternity maintained by Apastamba was a novelty 
 among early Hindoos, and that it found a practice hostile 
 to it so firmly established that for ages it made no 
 sensible impression upon custom ancestor- worship 
 going on all the time. We learn, however, that it 
 began at once to influence conduct; and probably it 
 influenced conduct greatly, even among those who did 
 not accept it as a doctrine. 
 
 And what appeared to the supporters of this doc- 
 trine to give it a means of acting upon conduct ? 
 We have seen that it was ancestor-worship that 
 they should represent that the sonship of custom 
 was insufficient for the uses of ancestor-worship. The 
 new doctrine sought support from ancestor-worship and 
 received it though not enough for the overthrow of 
 ancient customs which had long co-existed with ancestor- 
 worship without any thought of their being unfit for its 
 requirements. 
 
 Can we believe that a system which, as thus appears,
 
 SONSHIP AMONG THE HINDOOS. 311 
 
 tended, when it was mature, to support paternity, under- 
 mined paternity when it was of newer growth making 
 men expect the benefits of sonship from persons whom 
 (if physical paternity were thought important) they well 
 knew to have no claim to be counted their sons ? 
 Then, can we also believe that, having done this, 
 it replaced the system of sonship connected with 
 paternity, not by a bundle of unconnected customs, 
 but (this is the fact to be reckoned with) by a system of 
 sonship consistently based upon maternity ? This, too, 
 while the high doctrine about paternity just spoken of 
 appears to us among the Hindoos as a new doctrine, 
 which protested against the ancient custom which gave 
 children to the husband of their mother. 
 
 Are we to think, too, that, after ancestors began to 
 be worshipped, descendants became scarcer ? Did women 
 become more immoral ? Did men become more careless 
 of their wives ? There appears to be no reason for 
 answering such questions in the affirmative ; but, unless 
 reason can be shown for answering them affirmatively, 
 they all tend, and strongly too, to make us believe that, 
 when ancestor- worship arose, the Hindoo family was no 
 better than it was when we first come upon it. 
 
 More than enough has surely been now said to show 
 that it is only while the facts are neglected that Sir 
 Henry Maine's suggestion can be listened to. 
 
 What he seeks to get by it, as has been said already, 
 is the Father, with a position arising out of his paternity. 
 But, even were this conceded to him, it would leave him 
 a long way short of that to which he moves on the
 
 312 THE PATRIARCHAL THEORY. 
 
 Patriarchal Family, with Patria Potestas in the Father, 
 and Agnation as the family bond. To this immense 
 enlargement of his assumption ancestor-worship does 
 not contribute anything ; nay what by itself would 
 exclude the Patriarchal Theory the Hindoo obser- 
 vances spoken of as ancestor-worship were paid not 
 only to male ancestors but to female.* 
 
 * There remains a question as to the relation between ancestor- 
 worship and the observances paid by Hindoos to deceased ancestors. 
 These observances were, according to the books, intended to deliver 
 the deceased from the hell called Put.
 
 CHAPTER XVII. 
 
 SONSHIP AMONG THE HINDOOS. 
 
 HAVING been led so far into discussion of the right 
 to children among the Hindoos, it seems worth while 
 to go further, and inquiry to what early system of 
 counting kindred the Hindoo customs are really re- 
 lated. Let us first recall the facts as they have 
 appeared to us. 
 
 Among the early Hindoos the child did not belong 
 to the begetter. A man had no right to a child merely 
 as being its father. There was affiliation of children 
 to the mother ; and children, following the mother, 
 were also affiliated to the man who was "lord" of 
 the mother. Sonship was not founded upon paternity ; 
 and the family did not include the illegitimate children 
 of the husband (as we have seen that it might 
 do among the Irish), although it did include the 
 illegitimate children of the wife. It was founded 
 upon maternity ; and, through the mother, a child 
 belonged to the man who possessed rights over the 
 mother. The " lord " of an unmarried woman was
 
 314 TEE PATRIARCHAL THEORY. 
 
 her father; and she was affiliated to her father 
 and of his stock and family. Her issue also were 
 affiliated to her father, and of his stock and family. 
 Marriage passed a woman, and with her usually, it 
 would seem, her issue, if she had any (no doubt by 
 some process like dedication), into the stock and family 
 of her husband. It transferred her father's rights over 
 her to her husband, except that, in one case, something 
 was usually reserved. Her husband was now her 
 "lord," and, saving the reservation, her issue was 
 his of his stock and family and affiliated to him. 
 The reservation of a son of the marriage, customarily 
 made in the case of a brotherless daughter, kept in the 
 family of the woman's birth a child which, but for the 
 transference of rights made at marriage, would have 
 belonged to it. It suggests that to part with the 
 woman was not necessarily to part with her issue. 
 
 In general, however, as has just been said, the hus- 
 band not only got the woman's issue born after marriage 
 but, it would seem, her issue born before. Her child 
 born before marriage before her husband's " lordship " 
 had begun passed with her out of her father's stock 
 and family into her husband's, and was thereafter 
 affiliated to her husband. And, on the other hand, 
 her issue born after the husband's death after actual 
 " lordship " had ceased could be counted his with the 
 help of the Levirate. Custom, in times comparatively 
 late, retained a mode of replacing him for one case 
 that of his dying childless ; and, substituting for him 
 his brother, assigned to him, though dead, as his heir
 
 SONSHIP AMONG THE HINDOOS. 315 
 
 the child of the woman who had been made over to him 
 in marriage.* 
 
 It may be added that, incidentally, reason has been 
 shown for concluding that marriage was commonly 
 among early Hindoos (as among other peoples in a 
 rude state) an affair of sale and purchase ; and that 
 the peculiarities of Hindoo family law sprung out of 
 an early system of contracts for marriage made on that 
 footing. The facts indicate that the ordinary contract 
 transferred to the husband the whole right of the 
 woman's family the right to her and her issue, in- 
 cluding any child there was at the time of transfer ; 
 but that her father usually reserved the right to one 
 child when he had no son and perhaps the right to 
 claim one in the event of his afterwards finding himself 
 sonless. 
 
 It is easy to see that the facts are exactly what we 
 might expect to find were such a system of contract 
 superinduced upon the system of counting kinship 
 through females only.t 
 
 * A woman who, were the man alive, would be the mother of his 
 children was indispensable for the Levitate. If he left no widow, a 
 man could have no Levirate heir. Custom afforded a means whereby, 
 when dead, he could be replaced, and so that the child could be 
 assigned to him as if he were living. With Tibetan Polyandry, 
 brothers are interchangeable with, or equivalent to, one another ; the 
 child of one is the child of every other while every child is reckoned 
 in a special sense the child of the eldest brother. The Xiyoga child 
 is denned by Manu as " the lawful child of the wife." 
 
 t Of sons engrafted upon the family to supply the lack of 
 children, of whom there were five varieties the son given ; the son 
 made (adopted) ; the son rejected by his parents ; the son bought ;
 
 316 TEE PATRIARCHAL THEORY. 
 
 This system of kinship makes children of their 
 mother's stock, and affiliates them to their mother's 
 family whatever be the connection of which they are the 
 offspring. What would happen then if there were super- 
 induced upon it a system of contract in virtue of which a 
 woman with her issue was transferred from her family to a 
 husband, and she was taken into the husband's stock ? 
 The children, being of the stock of their mother, would 
 now be of the stock of her husband so that they 
 could be his children ; and, being his in virtue of the 
 contract, they would be his children. And as his 
 daughter would be his, so would her issue except so 
 far as he had transferred it by contract to another man. 
 Her child born before marriage would be his. If, when 
 she married, he bargained to have a child of the mar- 
 
 and the son self-given no account need be taken here. For the 
 natural family in some form must have been established before there 
 could have been thought of admitting such persons to the family. It 
 is worth while to note here, however with reference to the position 
 of the mother in the Hindoo family a fact not mentioned in its 
 proper place, that, in the statements made about these classes of sons 
 in Manu, it is involved that the mother as well as the father had a 
 right to and a share in disposing of children. The son given, accord- 
 ing to Manu, might be given by the father or by the mother. Kulluka 
 qualifies the statement as to the mother by the words, " with her 
 husband's assent." But, even with this qualification, the mother had 
 such a right over her children that she could take the initiative in 
 giving one away. And, of course, when there was no husband her 
 right was unqualified. Similarly, it appears that the son taken as 
 having been rejected by his parents must have been left without 
 either parent to take care of him ; while the son bought is defined as 
 a child purchased " from his father and mother," and the son self -given 
 as "he who has lost his parents or been abandoned by them." 
 (Manu ix. 168-177.)
 
 SONSHIP AMONG TEE HINDOOS. 317 
 
 riage, that would come back to him. When he gave 
 her over to a husband without reservation, her children 
 would be the husband's. And all her children would 
 be his if he chose to have them whether they were 
 his own offspring or not. 
 
 Contract, following upon kinship through females 
 only, would therefore yield in the first instance precisely 
 the system of parental right and sonship which we have 
 found among the Hindoos. The circumstances seeming 
 favourable to the actual paternity of the husband, a 
 strong preference for legitimate children might have 
 been expected quickly to spring up. And it is, perhaps, 
 surprising to find the wife's illegitimate and adulterine 
 children in plain terms assigned to the husband after 
 Hindoo customs had been long recorded and had become 
 subject to rigid criticism. There is no doubt of the 
 fact, however, and we know the theory which was held 
 to justify it. Combining both with what has just been 
 said, it seems pretty clear that it is not Agnation but 
 the system of counting kinship through females only 
 that is to be descried as operative among the prehistoric 
 Hindoos.* 
 
 * It is proper to state that, though Mr. J. D. Mayne believes that 
 the Hindoos had the Patriarchal Family, his account of sonship among 
 the Hindoos, which has already been referred to, is up to a certain 
 point substantially coincident with that given in the text. For 
 the case of the Levirate son, he has given up his theory of sonship. 
 But apart from that, such difficulty as he has found in accounting 
 for the facts has arisen out of his overlooking that marriage and its 
 consequences must have been founded upon contract, and that the 
 consequences would have been unusual with a condition (express or 
 implied) which became operative in exceptional cases only. The
 
 318 THE PATRIARCHAL THEORY. 
 
 To mention Aryans and Africans together seems 
 shocking to some people ; but the probability of the 
 explanation of Hindoo facts which has just been offered 
 cannot but be increased by our finding that it appears 
 to be contract superinduced upon female kinship which 
 produces in Africa results very like those which so much 
 astonish us among the Hindoos. An explanation which 
 involves that kinship was once counted through females 
 only can be more readily accepted for African facts 
 by themselves than for Hindoo facts by themselves 
 since in Africa it is a common thing to find kinship 
 taken only through the mother. And, given a set of 
 Hindoo facts and a set of African facts which are much 
 alike, it is reasonable to feel our confidence strengthened 
 in an explanation offered for the former, if we find 
 
 case of the son of an appointed daughter is at least as easy of explana- 
 tion as any other; but he has failed to explain it satisfactorily, 
 because his thoughts were fixed upon the husband's right instead of on 
 the contract out of which it sprang. What is more important is that, 
 through treating the husband's right as if it were an ultimate fact, he 
 has been kept from carrying analysis of the facts he was dealing with 
 as far as he might have done, and led into entirely misapprehending 
 their significance. 
 
 It should be noticed, too, that he has influenced apparently by a 
 reminiscence of the "Family founded on Power" used the term 
 ownership to denote a man's rights over his family. A more general 
 term is preferable to one which can seldom be literally applicable 
 except in a transitional state of family relations. But more than that, 
 there was absolutely nothing in the Hindoo facts to press the sug- 
 gestion of ownership upon him, and he did not need the term for any 
 purpose of explanation. It cannot be necessary to illustrate the fact 
 that a wife who has been got by purchase and her children are usually 
 not slaves.
 
 SONSHIP AMONG THE HINDOOS. 319 
 
 that it is probable enough as an explanation of the 
 latter. 
 
 The African cases in which children " follow the 
 mother " belonging to her clan, bound to her religious 
 observances, heirs in her family, and not heirs of their 
 own father are, the principal of them at least, well 
 known, and are too numerous to be here recited. Nor 
 is there any need to repeat what has been said in a 
 former chapter * to make it appear that there has been 
 a change in kinship though there can be no harm in 
 giving a reminder that kinship can be found in stages 
 of transition, and that we have seen in use expedients 
 well fitted to effect a change in it, one of which, with 
 whatever objects, was in use among the most famous 
 of ancient peoples. We may confine ourselves to certain 
 African facts which appear fitted to throw light upon 
 the influence which contract has had upon relationship. 
 Similar facts, it may be said, could be brought together 
 from many other quarters, and the argument to be 
 submitted greatly strengthened by them. 
 
 When, in Africa, a man on marrying enters his 
 wife's family (beenah marriage), he usually gains his 
 footing, as Jacob did his, by service. In that case, the 
 children all belong to the wife's family; and if the 
 husband goes away which he is free to do at any 
 time he can take neither wife nor children along with 
 him. His being father avails him nothing. His having 
 been husband avails him nothing. The children " follow 
 * Chapter xiv., " The Origin of Agnation."
 
 320 THE PATRIARCHAL THEORY. 
 
 the mother " and belong, as she does, to her family. 
 Very often however indeed, it may be said commonly 
 the communities which still keep up this kind of 
 marriage do not practise it exclusively. Families are 
 more or less willing for value received to give a man 
 a wife to take, or let him take his beenah wife, to his own 
 people. The contract, in this case, usually transfers to 
 the husband the woman and her offspring ; it is this which 
 is bargained for ; and where this has not been bought 
 and paid for, even when the woman has been allowed to 
 live with her husband in his own village, we find that 
 the children may be claimed by her family. If payment 
 has not been made, the woman's family have not to 
 use the current phrase " given her up." Here a con- 
 tract which carries away children, with their mother, 
 from the mother's family is seen to be the sole basis of 
 the father's right to his children. 
 
 This need not mean that men are indifferent about 
 the paternity of their children. They may be either 
 careless or very particular about that. And the circum- 
 stances show that men have a strong desire to have their 
 offspring for their own. It means that paternal right 
 is not, as with us, a consequence of a man having 
 children born to him in the married state ; that, more- 
 over, it does not arise out of paternity ; and that it 
 arises out of purchase. And, arising out of purchase, 
 it extends to all issue of the wife. Paternal feeling it 
 is, no doubt and the wish to have heirs in his own 
 household that makes a man desire that his children 
 should belong to himself and not to the family of their
 
 SONSHIP AMONG THE HINDOOS. 321 
 
 mother. But it is through contract alone they come to 
 belong to him. They "follow the mother," and contract 
 apart are of the mother's family. Contract takes them, 
 with their mother, out of the mother's relationship, and 
 makes them over to their mother's husband. It is the 
 means whereby children are separated from the mother's 
 family, and " given up " with her to her husband. 
 
 We often are not told, and cannot be sure, whether 
 the African peoples among which we find beenah mar- 
 riage are exogamous totem kindreds or not though 
 there seldom is serious doubt about the matter ; 
 nor are we often told whether different kindreds are 
 interfused in the same village, or whether the village 
 contains only one kindred. Kindreds of some sort, 
 however, there are among them. And we have seen 
 how they are recruited. A kindred or family is con- 
 tinued through women by means of children born to 
 women about whom no contract has been made. It 
 is continued also through men by means of contracts by 
 which women and their issue are " given up " to them. 
 It can as a matter of course, however, be continued 
 through the women. It is by means of an expedient it 
 can be continued through the men. For men, blood- 
 connection and marriage, by themselves, go absolutely 
 for nothing. There must also be purchase. On the 
 other hand, there is by birth a connection between 
 children and their mother which, if undisturbed, makes 
 them of the kindred to which their mother belongs. 
 The connection between mother and child is acknow- 
 ledged and never ignored. The connection between
 
 322 THE PATBIAEOHAL THEORY. 
 
 father and child is frequently ignored, and the acknow- 
 ledgment of it has always to be paid for. 
 
 Here it is obvious that blood or stock connection 
 
 does not count for so much as in the cases which 
 
 * 
 
 are more numerous in Africa in which, notwith- 
 standing that the wife is purchased, her children are 
 inalienably of her kindred, and of her kindred alone. 
 But, again it is equally clear that when children 
 are alienable from the mother's kindred to her husband 
 the desire of men to have their children to themselves 
 counts for a great deal. And is not this enough to 
 account for all the difference there is between this 
 class of cases and the other class ? The extent of the 
 difference is that instead of children being necessarily 
 of the kindred of their mother, they are taken out of 
 her kindred if the right to children has been purchased 
 by her husband. 
 
 Plainly, where this right can be got by purchase, 
 and arises to men in no other way, the extent to which 
 children do not belong to the kindred of the mother, 
 as it depends entirely upon the willingness and the 
 ability of men to purchase the right, is governed by 
 that which makes men willing to purchase it, and 
 has established a custom of purchasing it and can- 
 not be greater than what that will account for, though 
 (since the ability may be wanting) it may be less. 
 Now that which disposes men to purchase the right is 
 the desire to have their children to themselves. And, 
 in fact, among the peoples we are speaking of that desire 
 is more common than the ability to gratify it.
 
 80N8HIP AMONG THE HINDOOS. 323 
 
 The extent, therefore, to which, where the right to 
 children can be purchased by the husband, children 
 do not belong to the kindred of the mother, is amply 
 accounted for by the desire of men to have their 
 children to themselves. 
 
 Observe now that, if that desire were to cease, 
 purchase would cease, and all children would belong to 
 the kindred of their mothers. And, again, if there was 
 a time when the desire was without the means by which 
 alone it becomes operative, children then belonged in- 
 alienably to the kindred of their mothers. 
 
 It cannot be believed that contract has been an 
 original, or that it is anywhere more than a temporary, 
 basis of kindred- connection. And, therefore, we may 
 conclude that where now fathers can get the right 
 to their children by purchase only, children did 
 originally belong inalienably to the kindred of their 
 mothers. 
 
 Moreover, it is not to be doubted that the connection 
 so clearly acknowledged between mother and child 
 implies a recognition of the blood-connection between 
 them. And, through it, children still accrue to the 
 kindred of their mother unless contract comes in to 
 carry them off to her husband. When they always 
 accrued to the family of their mother, therefore, kindred 
 connection was based upon blood-relationship through 
 the mother. We may advance, indeed, to this conclu- 
 sion without a word provided we are not prepared to 
 hold that blood-relationship has never been thought of 
 at all among the peoples which practise beenah marriage. 
 
 T 2
 
 324: TEE PATRIARCHAL THEORY. 
 
 It scarcely need be pointed out that the mother's 
 blood, supposing it was anciently the only basis of 
 kindred connection, could not but dwindle in importance 
 in relation to it in the proportion in which contract 
 grew into importance in relation to it. It was inevitable 
 that, as the one came into prominence, the other should 
 lose prominence ; and it is not surprising that, when the 
 one encircles and envelopes the other, as it does now 
 ideas derived from contract prevailing the relation of 
 a woman's family, as well as that of her husband, to her, 
 and to her children when young, should sometimes 
 resemble mere ownership. 
 
 It appears then that we are justified in believing that 
 blood-connection through the mother was, until contract 
 was brought in to cut it down, the sole basis of stock- 
 connection in the cases we have been considering ; that, 
 in the affiliations of children which we find in them, we 
 have a sequel to the system, so common in Africa, of 
 counting children of their mother's blood and clan ; and 
 that, in the system of contract in use, we see a means 
 by which the limitations of the old kinship were got 
 rid of and the way was prepared for a new one. 
 
 But even where marriage is not beenah, and never 
 takes place without purchase, where the wife goes 
 to live with her husband, and the children need 
 not belong to the mother's clan, we find in Africa 
 that marriage does not in all circumstances give the 
 children to the husband. Among the Makololo, the 
 price paid on marriage might merely cover the right to 
 have the wife ; and in this case the children belonged to
 
 SONSHIP AMONG TEE HINDOOS. 325 
 
 the family of the wife's father. It might cover a certain 
 right to the children also, if that had been contracted 
 for; but never such a right as completely separated them 
 from the mother's family. To effect this, it was necessary 
 that a further price should be paid at the mother's death. 
 This paid, her family had "given her up," and her 
 children were entirely dissevered from them. 
 
 Here, with a general desire on the part of men to 
 have their children for their own, we see contract over- 
 coming, as it would seem, with difficulty not at a single 
 operation the connection of children with the relatives 
 of their mother. And it appears, if possible, more 
 plainly than in the former class of cases that the hus- 
 band's paternity could not make the children his chil- 
 dren ; and that mere marriage left children of the 
 family of the mother. Again we may conclude even 
 more confidently than before that the affiliation of 
 children, in case that the dissevering process employed 
 had not been carried out, to the family of the mother 
 was a remainder of the system which counted them as 
 necessarily of the mother's blood and clan ; and that 
 we see the father moving towards the full position of 
 an acknowledged parent by means of a system of buying 
 off the mother's family. 
 
 It may here be said that among many African 
 peoples which count kinship only through the mother, 
 the father, who gets his wife by purchase, has a sort of 
 property right in the children while, that is, they 
 are young and helpless. The wife's family also have 
 rights in them ; and their protection, the fear of their
 
 326 THE PATRIARCHAL THEORY. 
 
 resentment that is of the blood feud effectually dis- 
 tinguishes children from slaves. We sometimes find, 
 however, that the husband can pawn the children. We 
 learn in other cases that he may not (or perhaps it may 
 be that he dare not) sell them unless he can get the 
 consent of his wife's relations ; while, if he can arrange 
 with the latter, there is nothing to hinder him from 
 treating them as absolutely his property. From this it 
 may be understood how, when a desire to have the 
 children of their marriages to themselves had become 
 general among men, effect should in many cases be 
 given to it by means of contracts with the wife's family. 
 Where religious observance had become prominent in 
 connection with clanship, the dedication of the wife of 
 which, as we have seen, a beginning was made in Guinea 
 in the system of bossum-wives might also be necessary. 
 When contract alone sufficed, as the first effect of it 
 was to free the father from all claim on the part of his 
 wife's relations to her children, there is no difficulty in 
 understanding how, for a time, ownership should, in 
 appearance, be more prominent in connection with 
 relationship than blood- connection. 
 
 That contract carried to the African husband, as far 
 as could be done, the right of his wife's family the 
 right to all the offspring of his wife might be shown 
 by many examples, but will be sufficiently illustrated by 
 two, taken from peoples with very different arrangements 
 as to kinship. 
 
 The Ashantees, as we learn from Bowdich, counted 
 kinship only through the mother. When, among them,
 
 80N8SIP AMONG THE HINDOOS. 327 
 
 a man had been missing for three years, his wife might 
 be married again, and, if thereafter he returned, the 
 second husband's claim to her was good. But the 
 children of the second marriage were the first husband's 
 as much as if they had been his own children, and 
 accordingly, they might be pawned by him. The right 
 over the woman's children which marriage had given to 
 the first husband extended, therefore, to all the children 
 she might have, and it survived the second marriage of 
 the woman ; so that, though he could not recover her, 
 the sort of property interest he had in her children was 
 unimpaired. 
 
 Equally conclusive as to the right carried to the 
 husband by purchase, and more interesting in relation 
 to the present inquiry, is the passage subjoined, which 
 refers to a branch of the Bechuana people of Southern 
 Africa : 
 
 "Macheng was not born till some years after the 
 death of Khari, his reputed and ' legal ' father. Neither 
 does this affect his title to be chief. Khari, having 
 elevated the mother of Macheng to the dignity of head 
 wife [it is explained in another place that she was the 
 wife of highest rank, and that her children, therefore, 
 had a preferable claim to the succession], and having 
 paid her price in cattle, she and her offspring are to be 
 reckoned to Khari although the children should be born 
 a dozen years after his death. It is not etiquette ever 
 to refer to the man who thus 'raises up seed' to 
 another in connection with such children. They are the 
 children of him who is dead. But when Sekhome
 
 328 THE PATRIARCHAL THEORY. 
 
 [Khari's eldest SOD, who had usurped the chiefship] 
 was vexed, he sometimes sarcastically declared that his 
 rival was the ' child of cattle/ meaning that the price 
 paid for Macheng's mother at her marriage with Khari 
 was her son's only title to the chieftainship. There 
 is never any question among the natives, however, as 
 to the validity of this title. Even the most ardent 
 friends of Sekhome admit that, according to their 
 customs, Macheng is the rightful heir."* 
 
 This passage reads as if it had been written to show 
 the identity of the family law of the Bechuanas and the 
 family law of the early Hindoos of which probably 
 the writer was entirely ignorant the principle of each 
 being that a man was " father " of all the children of 
 his wife. After reading it, there cannot remain a doubt 
 that the peculiarities of the Hindoo law sprang out of a 
 system of contract for marriage that is, a system of 
 purchasing the wife and her issue. And the marriage 
 law of the Bechuanas was in principle identical 
 with the marriage systems already noticed, as to 
 which reason has been shown for concluding that they 
 show a progress from the system of counting kinship 
 only through the mother.t 
 
 We see no appearance of ownership as a title to 
 children among the Bechuanas, however. There is 
 affiliation to the father. And there is even more 
 
 * Ten Years North of the Orange River. By John Mackenzie. 
 Edinburgh. 1871. Page 364. 
 
 t If space permitted, some striking evidence to corroborate this 
 conclusion could be produced from among the Bechuanas.
 
 SONSHIP AMONG THE HINDOOS. 329 
 
 distinctly affiliation to the mother. The connection 
 with the mother carried so much weight that the 
 children of a polygamous husband took rank among 
 themselves from the mother their maternity thus 
 seeming to overshadow their paternity. It comes out 
 very plainly too, that the husband's connection with his 
 children arose out of contract out of his having become 
 by purchase the husband of their mother; and that 
 what the contract carried to him was the right of his 
 wife's family the right to all the woman's offspring. 
 For, in virtue of the contract, his wife's offspring could 
 be counted to him as children even after his death ; 
 and counted to a chief so that the child of his principal 
 wife born "some years " after his death was, by universal 
 admission, his lawful heir. This, too, was not a con- 
 sequence of the deceased being childless, but held good 
 while there were surviving sons of his by inferior wives 
 born in his lifetime. But here some explanation is 
 necessary. In the passage which has been quoted it is 
 hinted that there was a particular man who " raised up 
 seed " to the deceased ; and the writer mentions in 
 another place,* as one of a number of Levitical practices 
 found among the Bechuanas, " the custom of raising up 
 seed to a deceased brother or relative ; " and it may 
 therefore be taken as beyond doubt that it was as the 
 issue, or reputed issue, of a Levirate connection that 
 Macheng was heir, or that other children of the widow 
 would be counted children of her dead husband. 
 
 It was none the less in virtue of the contract which 
 
 * Ten Years North of the Orange River, p. 393.
 
 330 THE PATRIARCHAL THEORY. 
 
 transferred to Khari the right of his wife's family to 
 his wife and her offspring that her child born after his 
 death could be counted his. Macheng, as his rival 
 said, was "the child of cattle." In virtue of the 
 contract, the woman's children were to be counted to 
 the husband as long as it was possible so to do. Some 
 means of making the right conferred by marriage 
 available after the husband's death was, no doubt, 
 indispensable. And such a means there was in the 
 Levirate in the custom in virtue of which the de- 
 ceased could be effectually replaced by his brother or 
 other near kinsman. 
 
 It should not escape remark that, finding the Levi- 
 rate among the Bechuanas in an early form not cut 
 down by the endeavours of generations of priests and 
 lawyers to modify it in the interest of decorum we 
 in this case find it employed not to give a childless man 
 an heir, but to enable a man's head wife to give him a child 
 who should cut out from succession his sons by wives 
 of lower rank. And, purchased woman as she was, it 
 is a right of the wife alone which is disclosed to us in 
 this particular case. It does not appear whether there 
 were any customary restraints upon the Levirate among 
 the Bechuanas, or whether there was any limit to the 
 number of Levirate children. 
 
 As the Levirate is common to the Hindoos and to 
 the Bechuanas, and as if it must be deemed to be of 
 polyandrous origin we may, upon slight indications 
 otherwise obtained, and indeed without any other 
 indication, believe that, where it occurs there must
 
 SONSHIP AMONG THE HINDOOS. 331 
 
 originally have been kinship through females only and 
 may accept accordingly an explanation in which that is 
 involved some further observations may now be made 
 upon the passage quoted above relative to the Bechuanas. 
 It should be borne in mind that the Levirate was 
 practised among the Hindoos (as we learn from Apas- 
 tamba) upon a polyandrous theory; and that the 
 attempts to account for it among the Hindoos inde- 
 pendently of polyandry which we have had to notice 
 have appeared to be entirely unsuccessful. It should 
 be borne in mind, too, that both polyandry and kinship 
 through females only have been very common in Africa ; 
 and that where, with kinship through females only, 
 polygamy is practised to any extent, the two are 
 necessarily found together. Besides this, it may be 
 pointed out that, at any rate the motherhood of a 
 child must have been regarded as more important 
 than the paternity of the " father " wherever the Levi- 
 rate has prevailed. 
 
 Apart, then, from our knowledge that the Bechuanas 
 had the Levirate, finding that among them a child born 
 to a woman years after her husband's death could be 
 counted as beyond question the husband's lawful heir, 
 to the exclusion of his sons by inferior wives, one could 
 not have doubted that it was so counted in virtue 
 of some recognised connection, and was not a mere 
 child of chance. 
 
 The husband had, no doubt, as Mr. Mackenzie says, 
 acquired the right to his wife's offspring. But, in 
 ordinary course, the rights which a man possesses
 
 332 THE PATRIARCHAL THEORY. 
 
 exclusively of other persons either cease at his death 
 or pass then to some other person. Rights that are 
 not exclusive, on the other hand, on the death of one 
 of the sharers go on subsisting for the survivors. We 
 know of a class of cases in which, with monandry, that 
 is, an exclusive right in the husband, the husband's 
 rights over his wife pass at his death to another person. 
 They are found where the wife is regarded as property. 
 In these cases she passes, with the estate of the de- 
 ceased, to his heir. And, of course, we know of cases 
 of marriage in which, the husband's right being ex- 
 clusive, it ceases at his death altogether. This happens 
 in all the more advanced cases of monandry ; and it 
 happens also in many of the ruder cases of monandry, 
 notwithstanding that in these the husband gets, as 
 the Bechuana husband did, rights over all his wife's 
 offspring. 
 
 When, therefore, it had appeared that, among mon- 
 androus Bechuanas, the husband's exclusive right did 
 not pass to his heir at his death, and yet was not 
 extinguished, we should have been ready to conclude 
 that, among the Bechuanas, there was some traditional 
 means of giving effect to it, and, finding effect given 
 to it so thoroughly, to conclude that this was not any- 
 thing of the nature of a contrivance or expedient, but 
 some connection which was not only sanctioned by 
 custom, but which flowed naturally out of ancient 
 practice. The alternative would have been to think 
 that, in virtue of the right the husband had acquired, 
 any child the wife might have after his death could be
 
 SONSHIP AMONG THE HINDOOS. 333 
 
 his heir to suppose, that is, that the husband's right 
 was gone, but that there remained in its place an un- 
 limited right for his widow which, as a supposition, 
 must have been deemed extravagant. 
 
 Having concluded that the husband's right was 
 made available by means of some recognised connection, 
 one might have gone on, not unhopefully, to speculate 
 as to the nature of that connection. For, in there 
 being such a connection, it is involved that there was 
 some man who, in virtue of ancient practice, could take 
 the deceased husband's place, and who filled it to the 
 common apprehension so sufficiently that the widow's 
 children could continue to be deemed the children of 
 the dead man. Since we know the Levirate as a means 
 of raising up children to a dead man, and history dis- 
 closes no other means for that purpose, it might 
 with much probability have been concluded that the 
 Bechuanas had the Levirate. 
 
 Might it not, at any rate, have been inferred since 
 the husband's right did not pass to his heir or cease at 
 his death, but was carried on as if there had been some 
 sharer in it that the ancient practice indicated was one 
 in which men had been sharers in marriage ? Polyandry 
 having been anything but rare, and having, in particular, 
 been quite common in Africa, a polyandrous solution 
 would have been highly probable. And what other 
 would there have been worth mentioning ? 
 
 As the Bechuanas actually had the Levirate, and 
 in connection with it there is disclosed to us, more 
 prominently than usual, a right of the wife a right to
 
 334 THE PATRIARCHAL THEORY. 
 
 have children by her husband's brother or kinsman, who 
 should be counted her husband's children, and take the 
 place in his family they would have had if born to her 
 in his lifetime how, let us ask, is such a right re- 
 concilable with monandry ? How, with monandry, could 
 marriage give a right to a woman over her husband's 
 brother or, for that matter, a right to her husband's 
 brother over her ? And how could her children by the 
 living brother be counted her children by the dead one ? 
 It seems clear that a monandrous solution is, in this case, 
 not to be looked for. The more hopeless a monandrous 
 solution of the L evirate the more confidently may we 
 accept a polyandrous solution. But apart from that 
 taking the Bechuana case by itself every part of 
 Southern Africa being liable to strong suspicion of 
 polyandry, it will be found almost impossible to rebut 
 the suggestion it gives that the reason why one brother 
 could so effectually replace another was that brothers had 
 at one time been co-husbands, so that the death of one 
 left the wife the others that there had been a practice 
 of polyandry in the Tibetan form, in which a man's 
 child is also the child of his brother, while all the 
 children of a brotherhood are counted specially the 
 children of the oldest brother.* 
 
 At any rate, Sir Henry Maine's suggestion that the 
 Levirate was a contrivance for giving a childless man an 
 heir might be held to be excluded on the strength of 
 
 * The childless widow had the right to have her husband's brother 
 for husband among the Jews. The widow had the right of initiative 
 among the Hindoos also.
 
 SONSHIP AMONG THE HINDOOS. 335 
 
 the Bechuana case alone. For all cases of the Levirate 
 are so much alike that it is allowable to make each of 
 them throw what light it can upon the origin and early 
 conditions of the institution. And it was not necessary 
 among the Bechuanas, for bringing the Levirate into 
 play, that the husband should have died childless. In 
 the Bechuana case, what called it into operation, or was 
 the reason for it being put into operation was, in this 
 particular instance, the childlessness of a particular 
 wife. 
 
 Having found that among the Bechuanas, polyandry 
 in the Tibetan form (or possibly in some closely allied 
 form) seems to give the only possible explanation of 
 the Levirate, and gives for that particular case an 
 explanation of it, otherwise highly probable, it seems 
 now worth while to recall that (to say nothing of the 
 direct evidence of polyandry) the Levirate is not the 
 only Hindoo fact which indicates a prevalence of poly- 
 andry in the Tibetan form among the early Hindoos. 
 
 There is a provision in Manu not taken into 
 account in either of the works which have been noticed 
 in the preceding chapter which ought not to escape the 
 attention of any one who attempts to give, consistently 
 with monandry, an explanation of the Levirate. Where 
 any one of several brothers of the full blood had a 
 son, " Manu pronounces them all fathers of a male 
 child by means of that son "* so that, adds the com- 
 mentator, if such a nephew would be heir the uncle had 
 no power to adopt a son. "With Tibetan polyandry, 
 * Manu, 9, 182.
 
 336 THE PATRIARCHAL THEORY. 
 
 the brothers would all have been " fathers " of the child, 
 and he would have been heir to all of them. And this 
 rule regards them as his fathers in such good faith that 
 it cuts the uncles off from getting by adoption sons who 
 might exclude him from inheriting from them. 
 
 Have we here a remnant of ancient custom, that is, 
 a relic of Tibetan polyandry, or a fiction, or how 
 else is the rule to be explained ? 
 
 Perhaps there is nothing Hindoos were more un- 
 likely to do than to resort to a fiction or an arbitrary 
 rule to prevent men from having each his own heir. 
 To multiply heirs was, in the case provided for, 
 to multiply households, and with households to 
 multiply religious observances ; and that was the 
 interest and desire of every Hindoo " authority." And 
 what reason could there be for the introduction of 
 such a rule if, indeed, there was any power capable of 
 introducing it ? 
 
 The rule we may take it then is beyond question 
 old, and derived from ancient practice. It is in- 
 telligible that it should be found among a people 
 governed by custom, among whom, in times past, it 
 was common for a man's wife to be the wife of his 
 brothers, and his child to be the child of his brothers ; 
 and that it should continue to be authoritative after it 
 had ceased to be popular, or after the reason for it had 
 lost substance. It embodies the polyandrous view of 
 fatherhood and sonship. The Levirate embodies the 
 polyandrous view both of sonship and of marriage. 
 Combining the indications of the two, we seem to be
 
 SONSHIP AMONG THE HINDOOS. 337 
 
 very free from risk of error if we conclude that the 
 early Hindoos were polyandrous. 
 
 If it must be concluded, as to both the Hindoos and 
 the Bechuanas, that they were commonly polyandrous 
 before they were commonly monandrous, and if an 
 explanation offered for the family system of either, 
 which appears sufficient as a hypothesis, involves that 
 they at one time acknowledged kinship through females 
 only, it should be very easy, as has been said already, 
 to accept that explanation. For, wherever polyandry 
 can be found, even in the Tibetan form, a pre- 
 valence either with that polyandry or earlier of 
 the conditions which admit of kinship through females 
 only is highly probable. But as to the Bechuanas, there 
 should be no difficulty at all. For, besides their poly- 
 andry, their family system links on to African marriage 
 customs which appear clearly to have been arrived at 
 by a progress from the system of kinship through 
 females only. And an explanation which is good for 
 the Bechuanas, should be held good for the Hindoos. 
 
 We are carried back then to a time at which, 
 among the Hindoos, the bride, obtained by purchase, 
 was really given to the family commonly enough, and 
 during a period which lasted long enough, to enable 
 the arrangement to leave marks upon common practice ; 
 and, beyond that, to a time when kinship being 
 counted among them only through women, their family 
 conditions were such as are found with that kinship 
 marriage which left children of the stock of their 
 mother, beenah marriage, Nair polyandry. It may
 
 338 THE PATRIARCHAL THEORY. 
 
 here be recalled that, in the marriage law of the 
 Hindoos, as stated by Kulluka, we have already seen 
 reason for believing* that children among them, pre- 
 viously to their being counted of the stock of the 
 father, had belonged exclusively to the stock of the 
 mother. And it should not be forgotten how secondary, 
 relatively to maternity, was the part of paternity in 
 creating family relations among the Hindoos. 
 
 How far, if the views submitted in this chapter be 
 deemed maintainable, the inference of a prevalence of 
 kinship only through females can be established by 
 means of them for other peoples besides the Hindoos 
 cannot be here considered. For the present it is enough 
 to say that we may feel sure there is equal reason for 
 that inference whenever it can be shown that the 
 father's right to children has originated in purchase, 
 and we find also the Levirate, or some other custom 
 or institution of equivalent significance. t 
 
 * Chapter xiii., " The Origin of Agnation." 
 
 t It may be well, however, to mention that some of the most 
 characteristic Hindoo facts of which the argument of these chapters 
 has taken account, reappear in other nations where there is independent 
 evidence of the importance attached to kinship through women. As 
 to this, Professor Kobertson Smith supplies the note subjoined : 
 
 "The ancient Arabs," he says, "recognised the right of the 
 husband to choose another man to beget a child for him on his wife. 
 (Bokhari, 'Kitab an-Mkah,' vol. vi. p. 127 of the Bulak vocalised 
 edition), and in this case as among the Hindoos it appears that the 
 object was to obtain a more excellent offspring. Compare Wilken, 
 Das Matriarchat bei den alien Arabern (Leipsic, 1884, translated 
 from the Dutch), p. 27, et seq. It is instructive to observe that this 
 notice is given by Bokhari along with examples of polyandria proper. 
 Among the Germans, again, the husband who was not himself able to
 
 SONSHIP AMONG THE HINDOOS. 339 
 
 The African cases we have noticed have been referred 
 to simply as showing the influence of contract upon 
 relationship. It need not be supposed that the history 
 of the change of kinship among the Hindoos was 
 precisely parallel. In Africa, the process of change has 
 been very protracted, and the position of women is now 
 in many cases very low. On the other hand, there are 
 facts which tend to show that when the change began 
 among the Hindoos the position of women was very 
 high, and that their position continued to be high after 
 the family system among them had become substantially 
 identical with that which is disclosed to us in their 
 law-books. 
 
 have issue by his wife was allowed to seek a substitute, a practice 
 which Jacob Grimm, Deutsche Rechtsalterthiimer, i. 443, et seq., has 
 shown to have long survived in the written customary law of the 
 peasantry in various districts. Grimm calls attention to the fact that 
 the same thing was allowed in Sparta and Athens, among populations 
 where the evidence for primitive polyandry and female kinship is 
 particularly strong and clear." 
 
 That the Germans once had the system of kinship through females 
 only is indicated unambiguously by the evidence noted above, 
 p. 253, et seq. 
 
 z 2
 
 CHAPTER XVIII. 
 
 THE TUTELAGE OF WOMEN : THE HEIRSHIP OP SLAVES. 
 
 IN Ancient Law* Sir Henry Maine says he thinks 
 it can be shown that the family, as held together by 
 the Patria Potestas, is " the nidus out of which the 
 entire law of persons has germinated." He then pro- 
 ceeds to explain the relation to the Patria Potestas of 
 "the institution known to the oldest Roman law as 
 the perpetual tutelage of women ; " and of the capacity 
 which, from the testimony both of ancient law and of 
 many primeval histories, the slave appears to have 
 possessed of being, under certain conditions, the heir, 
 or universal successor, of his master. 
 
 This part of his subject is not treated thoroughly, 
 or with full argument. While he specifies none of 
 the cases in which a slave might be his master's 
 heir, the few examples of systems of female guardian- 
 ship which are referred to he mentions the Hindoos, 
 the Scandinavian nations, and "the invaders of the 
 Western Empire" appear to be those of peoples for 
 
 * Page 152.
 
 TUTELAGE OF WOMEN: HEIRSHIP OF SLAVES. 341 
 
 which he believed himself to have already established 
 his theory by means of Patria Potestas or Agnation. 
 It may almost be doubted then whether this portion 
 of his work was not inserted as being in itself curious 
 and interesting, and consistent with, or corroborative 
 of, what had been sufficiently made out before rather 
 than to furnish fresh evidence of Patria Potestas. And 
 it is quite clear that, if Agnation had been connected 
 with Patria Potestas, as he supposed it to be, and had 
 been, as he said it was, discoverable nearly everywhere, 
 further evidence of the prevalence of Patria Potestas 
 would have been superfluous while, on the other hand, 
 it is easy to see that there could be no hope of esta- 
 blishing the Patriarchal Theory by means of the tutelage 
 of women and the heirship of slaves alone. 
 
 Nevertheless, the connection alleged to exist between 
 Patria Potestas, on the one hand, and the tutelage of 
 women and the heirship of slaves on the other, must 
 here be discussed. But the discussion need not be a 
 long one. For a great part of what has in this work 
 been said of the connection supposed to exist between 
 Patria Potestas and Agnation, applies also to the con- 
 nection alleged between it and the tutelage of women 
 and the heirship of slaves, and it will suffice to refer to 
 that very briefly. 
 
 What Sir Henry Maine has said to show the con- 
 nection of the tutelage of women with Patria Potestas , 
 is not unlike what he had previously said of the connec- 
 tion with it of Agnation. But there is a difference. 
 Agnation, he said, was " as it were a mould," which
 
 342 THE PATRIARCHAL THE OR?. 
 
 retained the imprint of Patria Potestas. And of perpetual 
 guardianship he says that it "is obviously neither more 
 nor less than an artificial prolongation of the Patria 
 Potestas when for other purposes it has been dissolved." 
 Agnation, however, is described to us as a natural growth, 
 as a necessary result of Patria Potestas ; whereas we are 
 told that perpetual guardianship was " a peculiar con- 
 trivance of archaic jurisprudence" for retaining women 
 in the bondage of their families.* Agnation, moreover, 
 is presented to us as a remainder left by Patria Potestas, 
 having grown out of Patria Potestas and survived it ; 
 while of perpetual guardianship we are only told that it 
 was a contrivance by which Patria Potestas, while it still 
 existed, was artificially prolonged. Nothing is said to 
 show that what suggested this contrivance, or made 
 recourse to it natural, was Patria Potestas further than 
 that this is obvious. 
 
 That this is obvious in the sense of appearances 
 proving it seems, however, by no means to be the 
 case. 
 
 There is no need to repeat what the powers of a 
 Paterfamilias were. On the other hand, we do not 
 know that at Rome guardianship at any time extended 
 to a woman's person. So far as we know, it extended 
 to her property only, and even to her property in a 
 very limited degree ; and it might be a sufficient 
 account of it, as it existed at Rome, that it was intended 
 to put a check upon alienation or waste in the interest 
 of the Agnates who were the woman's heirs, and who, 
 * Ancient Law, p. 153.
 
 TUTELAGE OF WOMEN: HEIRSHIP OF SLAVES. 343 
 
 in the absence of testamentary provision, were them- 
 selves the guardians. At any rate, comparison between 
 the guardian's powers and those of a Paterfamilias 
 shows that the former were slight while the latter 
 were unlimited ; and that the powers which were slight 
 were confined to dealings with property, while the 
 unlimited powers applied both to property and person. 
 The one set of powers then was of an entirely different 
 order from the other ; it is not resemblance, but 
 difference between them that is apparent ; and, indeed, 
 the two coincide only in this, that the persons who were 
 subject to the restraints of guardianship were persons 
 who had once like all other persons been subject to, 
 and who had escaped from, the vastly greater powers of 
 the Patria Potestas. That the powers of guardianship 
 and the Patria Potestas both belong to the law of 
 persons, and that, of this branch of the law, the latter 
 was by far the most important title, appears, in fact, to 
 be all that can be said for the suggestion that the 
 tutelage of women, as it prevailed at Rome, was ob- 
 viously a prolongation of the Patria Potestas. 
 
 Sir Henry Maine, indeed, assumes that the powers 
 of guardians at Rome were originally more extensive 
 than they are known to us to have been, and that they 
 had been cut down while the Patria Potestas yet 
 remained intact. Control of the woman's person, he 
 tells us, was " apparently quite obsolete " at Rome. 
 The assumption, be it observed, is not consistent with 
 the guardianship of women having had greater powers 
 of lasting than Patria Potestas.
 
 344 THE PATRIARCHAL THEORY. 
 
 If, however, we suppose the powers of a guardian 
 to have originally approached nearer to those of a 
 Paterfamilias at Rome, no doubt there would be 
 more reason than there actually is for saying that 
 it is obvious that the former were a prolongation 
 of the latter. But what is there to justify the sup- 
 position ? Nothing except that guardianship can be 
 found extended to a woman's person in places where 
 there is no Patria Potestas, and the Paterfamilias is 
 unknown. 
 
 Even in those cases in which they have been 
 greatest, however, the powers of the guardian cannot be 
 likened to Patria Potestas ; and it cannot be said to be 
 obvious in any case that the one set of powers has been 
 fashioned after the other. On the contrary, if it had to 
 be granted that, when a substitute for the paternal 
 control came to be devised for women, the powers 
 actually possessed by fathers at the time furnished the 
 model for it, there would be a complete absence of 
 reason for thinking that those powers were anywhere 
 Patria Potestas. And perhaps in no case is there less 
 appearance of connection between the two sets of 
 powers than in the one case in which Patria Potestas 
 was undoubtedly present. 
 
 This brings us to make a further observation upon 
 Sir Henry Maine's suggestion as to the connection 
 between Patria Potestas and the tutelage of women. 
 It assumes the prevalence of Patria Potestas. It can 
 leave no impression unless it be assumed that Patria 
 Potestas is known to have been at least of common
 
 TUTELAGE OF WOMEN: HEIESHIP OF SLAVES. 345 
 
 occurrence to have prevailed at the very least as com- 
 monly as the tutelage of women. 
 
 Now, to assume the prevalence of Patria Potestas 
 was, no doubt, allowable in a writer who thought he 
 had already sufficiently proved it. Having already 
 proved this, he might legitimately try to show, by the 
 evidence of resemblance, that Patria Potestas was the 
 model after which guardianship had been moulded ; and 
 success in the attempt might even have strengthened 
 the case for the prevalence of Patria Potestas. But if 
 he is reduced to one instance of Patria Potestas, what 
 becomes of the argument connecting Patria Potestas 
 and guardianship ? What, d fortiori, when it appears 
 that the powers of guardians and the Patria Potestas 
 have nowhere a real resemblance to each other, and 
 that where the two undoubtedly co-existed, the law of 
 guardianship, as we know it, exhibited no mark from 
 which the magnitude of the Patria Potestas could pos- 
 sibly be conjectured ? And here it has to be added that 
 at Koine, where alone the two are known to have 
 co-existed, the tutelage of women disappeared long 
 before Patria Potestas. 
 
 When, putting aside assumptions, we inquire whether 
 the pre- existence of Patria Potestas that is, of the 
 paternal power in the degree in which it was allowed 
 by law in Rome can be inferred from the tutelage 
 of women, the following facts appear : (1) that Patria 
 Potestas is known in one instance only, and that the 
 tutelage of women occurred with it in a mild form, and, 
 so far as is known, only in a mild form ; (2) that there
 
 346 THE PATRIARCHAL THEORY. 
 
 have been numerous examples of the tutelage of women 
 occurring with lower forms of the paternal power, and 
 that many of them show women under greater restraints 
 than were known in Rome, although in Rome paternal 
 power was at the highest ; and (3) that in Rome, where 
 alone the Patria Potestas and the tutelage of women 
 appear together, Patria Potestas was the longer-lived. 
 There is one fact more which will be mentioned imme- 
 diately ; but enough has already appeared to show that 
 where systems of female guardianship are found without 
 Patria Potestas there can be no inference from them 
 of the pre- existence of a family system having Patria 
 Potestas for its leading feature. The fact remaining 
 to be mentioned, however, is of itself decisive of the 
 question. It is that guardianship of women is by no 
 means uncommon among peoples which acknowledge 
 kinship through females only, among whom the head 
 of the household, whatever his powers, is anything 
 rather than a Paterfamilias, and whose family system 
 is at all points in contrast with the family of the 
 Patriarchal Theory. 
 
 Systems of female guardianship will undoubtedly 
 help us in forming conclusions as to the condition of 
 women in the. regions in which they prevail. But it 
 would seem that we should not be safe in making 
 inferences from them as to the position and powers of 
 fathers or even as to the system of kinship. 
 
 Before leaving this branch of our subject, it may be 
 well to say, with reference to the history of systems of 
 female guardianship, and by way of caution as to
 
 TUTELAGE OF WOMEN: HEIRSHIP OF SLAVES. 347 
 
 forming conclusions upon them, that, though Hindoo 
 law-books contain passages which enlarge upon the 
 dependence of women, or perhaps rather upon the 
 propriety of keeping them ever dependent, there are 
 ancient Hindoo facts which are of an entirely different 
 tenor. It can be shown, for example, that, in early 
 times, a Hindoo mother might have her children in 
 ward. Even according to Manu, although the doctrine 
 of the dependence of women is eloquently set forth in 
 that work, it was improper for sons, their father being 
 dead, to divide the family estate during the mother's 
 lifetime. And reference has already been made* to an 
 argument founded upon ancient authorities an argu- 
 ment by a supporter of the Patriarchal Theory 
 designed to show that, in early times, among the 
 Hindoos, after the father's death the mother was head 
 of the family, and possessed, with the powers of control 
 a father had, of the family estate. 
 
 It remains for us briefly to examine what Sir Henry 
 Maine has said in Ancient Law of the relation of Patria 
 Potestas to the slave. 
 
 There is a sense, he tells us,t in which an affirmative 
 answer must certainly be given to the question whether 
 the slave was in the early stages of society a recognised 
 member of the family. " It is clear," he says, '* from 
 the testimony both of ancient law and of many 
 primeval histories that the slave might under certain 
 conditions be made the Heir, or Universal Successor, of 
 the Master, and this significant faculty, as I shall explain 
 * At p. 58, foot-note. f Ancient Law, pp. 163-5.
 
 348 THE PATRIARCHAL THEORY. 
 
 in the chapter on Succession, implies that the govern- 
 ment and representation of the Family might, in a 
 particular state of circumstances, devolve on the bond- 
 man." " The Family," he proceeds, " consisted primarily 
 of those who belonged to it by consanguinity and next 
 of those who had been engrafted on it by adoption ; 
 but there was still a third class of persons who were 
 only joined to it by common subjection to its head, and 
 these were the Slaves." The tie which bound the slave 
 to his master, he explains, was " regarded as one of the 
 same general character with that which united every 
 other member of the group to its chieftain;" and so the 
 slave had, " in the beginnings of society, a definite 
 place reserved to him in the empire of the Father." 
 
 In short, the slave might be his master's heir because 
 he was among the persons subject to the Patria Potestas, 
 and the relation of all such persons to their chief was 
 substantially the same. And in the Patria Potestas 
 we have the explanation of the slave's capacity of 
 inheriting. 
 
 The Ashantees, we learn from Bowdich, acknow- 
 ledged kinship through females only, and Bowdich, in 
 his interesting essay on the constitution and laws of 
 Ashantee,* speaks as follows of their law of succession : 
 
 " The most original feature of their law that of 
 succession has been mentioned in the History, with 
 the argument on which it is founded. It is universally 
 binding. The course [of succession] is the brother, the 
 sister's son, the son, the chief vassal or slave to the j 
 
 * Mission from Cape Coast Castle to Ashantee, 1819, p. 254.
 
 TUTELAGE OF WOMEN: HEIRSEIP OF SLAVES. 349 
 
 stool." The stool is found in every house, and is 
 the emblem of authority, answering to throne. 
 Bowdich goes on to state that in the Fantee country 
 the principal slave succeeded to the exclusion of 
 the son, who inherited only his mother's property, 
 which was frequently considerable, and was inherited 
 from her family independently of her husband. 
 
 Here we see that, with a system of kinship which 
 is the very antithesis of Agnation, with a law of suc- 
 cession which preferred the brother and the sister's son 
 to the son, there was room for the slave to be, under 
 certain circumstances, the heir of the family. And, 
 among the Fantees, he was heir while a son was actually 
 excluded. 
 
 Nothing more seems necessary to show that the 
 slave's capacity of being heir of his master does not 
 imply the Patriarchal Family or " the empire of the 
 Father," and that it has no special relation to Patria 
 Potestas.
 
 CHAPTER XIX. 
 
 CONCLUSION. 
 
 IN one of the earlier chapters of this work an attempt 
 was made to show that the evidence for the Patriarchal 
 Family of Sir Henry Maine's theory, with its incidents 
 of Patria Potestas and Agnation, should be exceedingly 
 
 O */ 
 
 clear and strong. 
 
 The unfitness of his theory by itself to account for 
 the actual forms of early societies, the apparent unfitness 
 of the fiction with which he supplemented it to serve the 
 purpose for which he supposed it to have been used, and 
 the certainty that, if used, it had no effect whatever 
 unless, as he imagined, it mysteriously in a way 
 admitted to be now unintelligible enabled early 
 tribes to impose upon themselves, appeared to lay 
 the theory as a whole under the strongest suspicion 
 and would indeed have been enough to destroy any 
 hypothesis which did not possess some powerful 
 elements of popularity. It was suggested that, on 
 account of the doubt surrounding the theory as 
 a whole, good evidence might be demanded for the
 
 CONCLUSION. 351 
 
 particulars involved in it. The complexity of the 
 primeval family, as Sir Henry Maine has described it, 
 which has made it, regarded as a primary social fact, a 
 puzzle and a wonder at times to himself, appeared to be 
 a still better reason why good evidence should be forth- 
 coming of the prevalence of the family described, and a 
 strong case be made out for believing it, if found, to have 
 been primordial. It was suggested that the frequent 
 occurrence of a family system founded upon kinship 
 only through females supplied one reason more why the 
 evidence for Sir Henry Maine's Patriarchal Family 
 should be strong, and the reasons for holding it to have 
 been primordial convincing. 
 
 After considering the evidence to which, in his 
 various writings, Sir Henry Maine has referred us, the 
 conclusion seems to be forced upon us that he has 
 known the family which has the father's power in the 
 degree of Patria Potestas, and in which Agnation gives 
 the rule for kinship and inheritance, nowhere except in 
 Rome. That this form of the family occurred in Rome 
 appears then to be all there is of direct evidence to 
 show that it was once universal. And that it occurred 
 in Rome among a people very far from primitive is all 
 there is of direct evidence to show that it was primeval 
 as well as universal. 
 
 Against this direct evidence, however, we can scarcely 
 help putting the evidence which investigations suggested 
 by Sir Henry Maine's writings have disclosed to us as to 
 the family systems of peoples known to us in a more 
 primitive state than the Romans. But this need not be
 
 352 THE PATRIARCHAL THEORY. 
 
 summed up here. It will suffice to recall that in the 
 early Hebrew family we find neither Patria Potestas 
 nor Agnation and that, on the contrary, we seem to 
 find beenah marriage and the relationships consistent 
 therewith ; while from not the Hebrews only, but from 
 the Germans and from the early Greeks, we appear to 
 get unquestionable indications of a system of kinship 
 only through women having preceded the acknow- 
 ledgment of kinship through the father. It need 
 scarcely be said that the facts here referred to are 
 evidence against the Patriarchal Theory in every 
 form. 
 
 Of indirect evidence for the Patriarchal Family of 
 Sir Henry Maine's theory there is less than of direct. 
 For, putting Koine aside, Sir Henry Maine has never 
 adduced a single good example of either Patria Potestas 
 or Agnation. 
 
 That Patria Potestas was rarely met with had always 
 been conceded. But though we have come upon old 
 forms of the family in which the father's rights and 
 authority were placed high and, indeed, these tend to 
 be great wherever fathers are recognised, and fill the 
 father's place as protector of the family in not one 
 of them had the father the unlimited powers of the 
 Paterfamilias. In none of the cases we have had to 
 examine have we found him endowed by law or custom 
 with powers of life and death and sale, the monopolist 
 of family property, the only person in the family capable 
 of having rights. 
 
 As to Agnation, it might be enough to say that
 
 CONCLUSION. 353 
 
 Agnation excluded from relationship, and especially 
 from inheritance, the descendants of the women of a 
 family ; and that Sir Henry Maine has never specified 
 any people other than the Komans who allowed no 
 rights of inheritance to the descendants of women. 
 This may be taken as showing sufficiently that cases of 
 Agnation are extremely rare. A preference, more or 
 less, for males in succession has been common enough. 
 But that is not Agnation. 
 
 Even had cases of Agnation been numerous, however, 
 Sir Henry Maine appears to have completely failed in 
 trying to connect Patria Potestas with Agnation even 
 by a hypothesis so that cases of Agnation, if he had 
 them, could not be used by him for recovering lost cases 
 of Patria Potestas. 
 
 There seems then to be no indirect evidence for Patria 
 Potestas or for the Patriarchal Family of Sir Henry 
 Maine's theory to be had by means of Agnation. And 
 in the preceding chapter it has appeared that there is 
 really no other source of indirect evidence for them. 
 
 Of indirect evidence against the Patriarchal Family 
 of the theory which has come in our way, it seems 
 necessary to mention that given by the Levirate. An 
 attempt to attenuate the prevalence of the Levirate 
 among a particular people, and then to show how 
 natural it was that priests or lawyers, as well as the 
 multitude, should take a polyandrous view of marriage, 
 and think the Levirate might be tolerated, could not 
 possibly, in the case of so wide-spread an institution, 
 be of any use in showing the institution to be consistent 
 
 2 A
 
 354 THE PATRIARCHAL THEORY. 
 
 with a primeval and universal practice of monandry. 
 And, if the Levirate must be taken to be of polyandrous 
 origin, we have in it again strong evidence against the 
 Patriarchal Theory, not only as Sir Henry Maine has 
 stated it, but in every form. 
 
 The conclusion we are brought to is that, besides 
 the occurrence of Patria Potestas and Agnation in the 
 Roman family within the historic period, there is really 
 no evidence to show that the Patriarchal Family, as 
 Sir Henry Maine has described it, was primeval and 
 universal. 
 
 It need only be added that the failure of Sir 
 Henry Maine's theory to connect Patria Potestas with 
 Agnation destroys its consistency as a theory. Unless 
 some constant relation between these two can be made 
 out, there is an absence of reason for their appearing 
 together in the Patriarchal Family, or in the family of 
 any period even were there evidence to show that in 
 particular cases they did so appear. And, at the same 
 time, that failure leaves relationship unaccounted for. 
 
 So far as to the merely critical side of the present 
 essay which has dealt with an attempt to show that re- 
 lationship began in the family, and was determined by a 
 force operating therein. If, however, the reader has 
 given his assent to the arguments by which that attempt 
 seems to be condemned as a failure, he will also, it 
 is hoped, be ready to admit that the result is not 
 merely negative. For it has appeared at all points, 
 not only that the phenomena dealt with are not in- 
 telligible on the Patriarchal Theory, but that they
 
 CONCLUSION. 355 
 
 carry us back to a stage of society prior to the form 
 of the family which has a father at its head, to the 
 stage of polyandry and to the form of the family 
 founded upon kinship through women only. The 
 argument, therefore, has been throughout constructive 
 as well as critical. And no slight part of the work is 
 purely constructive. The value of the construction 
 must be left to the judgment of the reader. 
 
 THE END. 
 
 CHARLES DICKENS AKD EYA1T8, CSY8TAL PALACE PBKS8.
 

 
 UC SOUTHERN REGIONAL LIBRARY .FACILITY 
 
 A nnn TRF;