The Online Library of Liberty

A project of Liberty Fund, Inc.

• Preface.
• Early Law and Custom.
• Chapter I.: The Sacred Laws of the Hindus.
• Chapter II.: Religion and Law.
• Notes and Illustrations.
• Note A.: Wheel-pictures.
• Chapter III.: Ancestor-worship.
• Chapter IV.: Ancestor-worship and Inheritance.
• Notes and Illustrations.
• Note A.: Hindu Patria Potestas.
• Note B.: Polyandry.
• Chapter V.: Royal Succession and the Salic Law.
• Chapter VI.: The King, In His Relation to Early Civil Justice.
• Chapter VII.: Theories of Primitive Society.
• Notes and Illustrations.
• Note A.: the Andaman Islanders.
• Chapter VIII.: East European House Communities.
• Notes and Illustrations.
• Note A.: the Gens.
• Chapter IX.: The Decay of Feudal Property In France and England.
• Notes and Illustrations.
• Note A.: Village-communities and Manors.
• Chapter X.: Classifications of Property.
• Chapter XI.: Classifications of Legal Rules.

CHAPTER IV.

ANCESTOR-WORSHIP AND INHERITANCE.

The close connection between succession to property after death and the performance of some sort of sacrificial rites in honour of the deceased has long been known to students of classical antiquity. A considerable proportion of the not very plentiful remains of Greek legal argument to be found in the Athenian Orators is occupied with questions of inheritance, and the advocate or litigant frequently speaks of the sacrifices and the succession as inseparable. ‘Decide between us,’ he says, ‘which of us should have the succession and make the sacrifices at the tomb’ (Isæus, ‘In the goods of Philoctemon,’ Or. vi.) ‘I beseech you by the gods and immortal spirits not to allow the dead to be outraged by these men; do not suffer his worst enemies to sacrifice at his grave’ (Or. ii.). In a former work I pointed out the number, costliness, and importance of these ceremonies and oblations among the Romans, and I insisted on their probable significance as the source of the peculiar fictions which cluster round early family law (‘Ancient Law,’ p. 191). The best explanation, I argued, of the facility with which a stranger can be made a son is that, being admitted to the religious observances, he is not distinguishable from a son under his religious aspect. The later experience of the world may show us that in the mere blending of the ideas of inheritance and offering there is nothing to surprise us. It is natural enough. Wherever it has been matter of belief that the surviving members of a dead man’s family could do anything to better his lot in the world after death, it has been thought their duty to do it before they entered upon his possessions. The mediæval Christian Church held this view of personal or movable property; it was primarily a fund for the celebration of masses to deliver the soul of the owner from purgatory. Upon this doctrine was founded the jurisdiction of our Ecclesiastical Courts, in which all property of this kind vested in the first instance before it could be distributed; and this jurisdiction, coupled with the necessary powers over Executors in the case of Wills, and of Administrators in the case of Intestacies, has descended to the modern Court of Probate. The new light which we owe to the author of ‘La Cité Antique’ is his determination of the nature of the divine beings to whom the oblations, which exercised so powerful an influence on Athenian and Roman heritages, were devoted. They were of course not offered to any one Supreme God. But neither were they offered to the greater deities of the local Pantheon. ‘Le culte des dieux de l’Olympe et celui des Héros et des Mânes n’eurent jamais entre eux rien de commun,’ says M. Fustel de Coulanges. The worship was given to the dead, chiefly to the remembered dead who had just passed away into a life not further removed from their late existence than a sleep from reality.

I will note in passing that the excessive expensiveness of the Roman sacra privata, which is the burden of Cicero’s complaints in his private letters, seems to be a feature of still surviving ancestor-worship. The writer of a paper I have before quoted (‘Ningpo and Buddhist Temples’) gives a curious calculation, upon what is probably American missionary authority, of the expenses to which the Chinese are put by worshipping their forefathers. ‘One well entitled to know what he spoke of said that fully thirty millions of dollars are annually expended in China at the three great festivals in honour of the dead, and, with the average expenditure of each family, fully 150 millions of dollars are spent in quieting the spirits.’ There is no doubt also that funeral rites and oblations are extremely expensive in India, and I have heard their heavy cost seriously urged as a reason against imposing a duty on legacies and successions. The expensiveness of religious observance among Hindus arises from the necessity which it involves of feasting Brahmans, sometimes in multitudes, and making them gifts. The oldest of the law-books strongly inculcate the duty of employing the ministry of Brahmans. Three are especially to be in places of honour at all funeral celebrations, who are to represent the three ancestors worshipped—the father, grandfather, and great-grandfather. But many more are to be entertained. ‘Pure, with composed mind, and full of ardour,’ says Apastamba, ‘he shall feed Brahmans who know the Vedas.’ ‘He shall feed an uneven number of Brahmans, at least nine,’ is the rule of Gautama, ‘or let him feed as many as he is able to entertain.’ Some singular but very intelligible texts forbid the worshipper to make these sacred feasts a pretext for entertaining his own relatives. ‘The food eaten at a sacrifice by persons related to the giver is a gift offered to goblins. It reaches neither the Manes nor the Gods. Losing its power to procure heaven, it wanders about in this world, as a cow that has lost her calf runs into a strange stable’ (Apastamba, ii. vii. 17. 8).

We have now to consider some of the ways in which the law and custom of ancestor-worshipping societies have been affected by their peculiar faith and religious practice. The first instance of a transformation in law which I will give is one nearly connected with the growing costliness of the ritual. By this ritual a religious and intellectual aristocracy lived. One of the commonest popular errors about the Brahmans even now current in England is that, because they are spiritually the highest, they are therefore the wealthiest and most powerful of the Hindu castes. They count among them some opulent and powerful families, and one Royal house in India is Brahman, but, on the whole, they are not specially wealthy. It would be more accurate to describe them as a serving and ministering class, their occupations varying from the high administrative duties which they once monopolised in the Mahratta States to such humble functions as those of the cook, whose service is a luxury, because no impurity can be contracted at his hands. The Brahman of the old law-books is still a priest and spiritual director more than anything else, though with a visible tendency to become a lawyer, a judge, or an administrative counsellor. He lives, however, by the bounty of others, by their charitable and pious gifts, more particularly those given to him on the great occasions of sacrifice. It is strongly said by a modern writer (J. D. Mayne, ‘Hindu Law and Usage,’ p. 205) that the modern law, as promulgated by Manu, might be described as a law of gifts to Brahmans. ‘Every step of a man’s life from his birth to his death required gifts to Brahmans. Every sin which he committed might be expiated by gifts to Brahmans. The huge endowments for religious purposes which are found in every part of India show that these precepts were not a dead letter.’ Now one chief impediment to pious liberality is that system of joint ownership by groups larger than families which is still common in India, especially in that earliest home of the Aryan race, the Punjab. Every man’s rights in such a group are more or less limited by the rights of everybody else; and, as a rule, the assent of the entire group is necessary before any part of its property can be alienated. Hence the sacerdotal system, of which the rudiments are to be seen in the law-books, is most manifestly adverse to joint forms of property. The writers frankly avow their motives for this dislike of co-ownership and for their partisanship of partition. ‘In partition,’ says Gautama, ‘there is increase of spiritual merit’ (x. viii. 4). The principle is still more plainly put by Manu (ix. iii.): ‘Either let them live together, or, if they desire separately to perform religious rites, let them live apart; since religious duties are multiplied in separate houses, their separation is therefore right and even laudable.’ The more separate households, the more occasion for domestic sacrifice, the more opportunities for pious largesse to the sacred order.

The modern writer I have quoted (J. D. Mayne, p. 204) cites my own opinion, expressed in a former work (‘Early History of Institutions,’ p. 104) that the Christian Church, when engaged in proselytism among barbarous societies, exerted a similarly dissolving force upon tribal ownership. The Church certainly introduced its barbarous converts to the Testament or Will; it strove to strengthen their reverence for Contracts; and the Irish evidence seems to prove that it largely extended Separate, as distinguished from Tribal, ownership. In reading the Brehon tracts, you remain in doubt whether the writer means to lay down that tribal land under certain circumstances may be parted with generally and in favour of anybody, or whether it is only to be alienated in favour of the Church. The strong probability is that he intended to sanction gifts to the Church primarily; and that a generally enlarged power of separate alienation was the consequence of such rulings. But it has always to be remembered that there is a radical difference between the Brahman encouragement of charitable profusion and the enlargement of legal facilities for pious endowment by the mediæval Christian Church. Charity with the Brahman began strictly at home; he was wedded to it, because he lived by it. But the Church, although it certainly desired to fortify by endowments every asylum and stronghold which it planted amid barbarism, had other classes under its protection besides its own servants and clergy. It fed the poor and needy at its gates. It was ever careful for the orphan and the widow. But the Brahman law-books, with much elevation and some tenderness of feeling, are constantly offensive in the contempt, sometimes amounting to loathing, which they express for all classes except the sacred caste and other castes powerful enough to pretend to equality with it, or proximity to it.1

We come now to some results of Ancestor-worship which are of the highest interest as throwing light on a number of perplexing questions which embarrass our first steps in the examination of very ancient societies. It seems clear that, according to the most ancient ideas, not only must the ancestor worshipped be a male ancestor, but the worshipper must be the male child or other male descendant. Under the conditions of thought we have been supposing, it will have been seen that the verge of life and death was very easily overstepped. The dead man was he who had been the living dreamer, only that he had now passed permanently into the life of dreams. It thus seemed proper that the sacrifice should be offered by a person who one day would take his place in the chain of deified ancestors, and this could only be a male descendant. Hence there arose, among the ancestor-worshipping peoples, a most intense desire for male offspring and, as a consequence of this desire, a remarkable set of ideas about paternity, sonship, and inheritance, which must have been widely diffused of old among all the more powerful races of mankind, and specially those of the Aryan stock. There are manifest traces either of these ideas, or of the customs with which they were intermixed, in the legal antiquities of the Athenians and of the Spartans, of the Romans, of the Celtic Irish, of the Hebrews, and of the Chinese. As is natural, from the deeply sacerdotalised character of their legal literature, the fullest and most detailed account of a family system shaped and interpenetrated by ancestor-worship is to be found in the ancient books of the Hindus. It cannot, of course, be taken for granted that this system, in its integrity, once existed everywhere. One feature of it is found here, another there. But there does seem to have been a general likeness between the deductions which the priests and lawyers of a large number of ancient societies drew from the principle that sacrifice and worship were due, under severe supernatural penalties, from male children to their dead forefathers.

We cannot, as it appears to me, frame in our minds any reasonable explanation of Ancestor-worship and its legal consequences, unless we assume that, when it first arose among men, the Father of each family appeared to them in the form in which he constantly shows himself on the threshold of jurisprudence, and which he probably wore2 when the human race began. He is the Pater Familias. The physical paternity is fully recognised, but it is blended with protective power. Most of the males subject to him are really his children, but, even if they have not sprung from him, they are subject to him, they form part of his household, they (if a word coloured by later notions be used) belong to him. On the other hand, under the religious aspect of these relations, from the point of view of a sacerdotal lawyer, the son is simply the person who can efficaciously offer the sacrifices. Dr. Bühler (Preface to vol. ii. of ‘Sacred Books of the East,’ p. xix.) writes thus of Baudhâyana, whom Sanscritists, Indian and European, generally regard as one of the oldest of the law-writers. ‘Like many other ancient teachers, Baudhâyana permits childless Aryans to satisfy their craving for representatives bearing their name, and to allay their fears of falling after death into the regions of torment through a failure of the funeral oblations, by the affiliation of eleven kinds of substitutes for a legitimate son. Illegitimate sons, the illegitimate sons of wives, the legitimate and illegitimate offspring of daughters, and the children of relatives and even of strangers, who may be solemnly adopted or received as members of the family without any ceremony, are all allowed to take the place and the rights of legitimate sons.’ I will proceed to examine this system of artificial affiliation at some length, together with its bearing on Inheritance in the embryo. Until the ideas upon which it rests have been carefully sifted, it seems to me doubtful whether all investigation of the primitive forms of society is not likely to be imperfect and premature.

First of all, then, the person who offers the sacrifices with the best chance of efficacy in the world beyond the grave is the legitimate son, the son who is physically the offspring of his father and who, by preference, is born of a marriage blessed by Brahmans and contracted under all the conditions which their sacred law prescribes. And, among all sons, the eldest son is most likely to confer spiritual benefit on his father. Here, however, we come upon one of the most remarkable of the extensions which the sacerdotal lawyers give to their doctrine in order to prevent its miscarriage in particular cases. With the purpose of increasing the chance of there being legitimate sons to present the oblation, some among the oldest of these Brahmanical teachers relax the conditions of marriage, and show leniency to strange forms of wedlock, so numerous as almost to include all possible unions of the sexes. Some of these marriages are very strongly condemned by the later Hindu law-writers; for example, marriage with a purchased bride; others, where they occur, are effectuated by violence or by fraud. Still, it is to be observed that the children of all these unions would be physically the children of the husband; and the father and mother, however barbarous their connection, are probably understood to have been unmarried before it.

Among all the sons sprung from the father, the eldest is preferred. The sacerdotal reason assigned by the Hindu lawyers is that, by his birth, the religious obligation to have a son who can continue the chain which binds together the living and the dead has been satisfied. But the privileges of inheritance corresponding to this spiritual primacy are very variously defined in the law-books, and, even when they approach somewhat to modern primogeniture, they are still very unlike it. Sometimes the eldest son is spoken of as taking the whole inheritance of his father and supporting the rest of the family, and this is very probably the secular custom for which the priestly lawyers invented a religious reason. More often, the best portion, or some similar advantage, is assigned to the eldest of the sons, and sometimes alternative modes of providing for him are stated. Of two ancient authorities, Gautama defines his privileges in ambiguous and indeed contradictory language (see chap. xxviii.), while Apastamba, while admitting that Primogeniture gives advantages in certain countries, argues strongly for equal division among all the sons (ii. vi. 14. 14). It seems to me that, at the epoch of these law-treatises, the ancient Primogeniture was decaying among the Hindus, as we know that it decayed in the barbarous world generally. Under the original usage, the eldest son may have taken everything and maintained his brethren; but the Brahmans, as I have explained, were strong partisans of multiplied households, and this feeling must have militated powerfully against the privileges of the eldest. On the whole, the doctrine which tends to prevail is that the division should be equal among sons, with a small advantage to the eldest as the divider of the inheritance, which may have been meant as an inducement to fairness.

Next to the legitimate sons, as proper vehicles for spiritual blessing, the greater number of the ancient Hindu law-writers place the son of the wife, born during her marriage but not necessarily of her husband. At first sight this looks like an application of the long-descended legal maxim, ‘pater est quem nuptiæ demonstrant,’ but all the ancient texts taken together suggest a different explanation, and I will consider the ‘son of the wife’ again when I come to the son of the widow.

The person who, on failure of all the inheritors I have mentioned, can next in order offer the sacrifices for the deceased and claim his succession, is the son of his ‘appointed’ daughter. It is an interesting case for a variety of reasons. The son of a daughter, regarded by himself, would not satisfy the requirements either of a successor or of a worshipper. From the secular point of view, he is, in Roman phrase, a ‘cognate,’ a kinsman through women only, who, according to the usage prevailing among all the more powerful races of mankind either from the first or at a certain stage of their development, cannot continue the family. The religious theory of ancestor-worship would not take any notice of him, for the parent to whom he would sacrifice would be a woman, and women could not in the earliest times be objects of worship, and never at any time by themselves. But the ancient law allowed the father, who had no prospect of having legitimate sons, to ‘appoint’ or nominate a daughter who should bear a son to himself and not to her own husband. Apparently this appointment could be made against the husband’s will, for one of our oldest authorities warns the Hindu against marrying a girl who has no brothers, because her father may ‘appoint’ her, and her husband may have his own naturally-born son converted into the son of the maternal grandfather. The sacerdotal formula of appointment is given in Gautama, xxviii. 19. 18: ‘A father who has no male offspring may appoint his daughter to raise up a son for him, presenting burnt offerings to Agni and to Pragâpati, the lord of creatures, and addressing the bridegroom with these words, “For me be thy male offspring.” ’ ‘Some declare,’ adds the writer, ‘that a daughter becomes an appointed daughter solely by the intention of the father.’

Some customs near akin to the Hindu usage of ‘appointing’ a daughter appear to have been very widely diffused over the ancient world, and traces of them are found far down in history. The daughter here becomes neither the true successor of her father nor the priestess of his worship, but a channel through which his blood passes to a male child, capable, according to the oldest notions, of sacrificing to him; and, according to the newer ideas, of taking his property and preserving the continuity of the household. At first there was always, I should imagine, some expression of the father’s will, coupled with some religious ceremony. Among the Athenians, when our knowledge of their law begins, the Testament or Will has appeared, though its operation is much limited. An Athenian father, fearing sonlessness, might have a son raised up to him by a daughter; and the commonest mode of effecting this object was by devising his property—or, to speak more strictly, the property and the daughter together—to a person selected by himself on condition of marrying her. The son born of the marriage was, on coming of age, transferred to the family of his maternal grandfather—it would seem, with some of the forms of adoption—and took his name, becoming at the same time the legal representative (κύριος) of his own mother. This is essentially the same method of obtaining a male child which was anciently in use with the Hindus. But some such practice must have been followed, and some such ideas must have prevailed among a certain portion of the barbarous communities which contributed their usages to the enormous body of rules finally consolidated as the Feudal Law. According to some systems of mediæval customary law, daughters succeeded, either in order of primogeniture or in a group, when sons had failed. According to others, they were excluded altogether. But between these doctrines there was an intermediate view, that a daughter, though she could not succeed herself, could transmit a right of succession to her male children. Hereafter, I shall have occasion to point out that this was the rule on which our Edward III. based his claim to the throne of France; he admitted that the French princess, who was his mother, could not succeed, but he contended that he himself, as her son, was entitled to succeed his maternal grandfather. This argument did not prevail either in the forum of arms or in the opinion of the feudal lawyers; but it seems to be clearly connected with the range of legal notions before us.

The chief interest of the Hindu ‘appointment,’ and of the counterparts of it in the law of other races, lies in their probably marking one of the points at which the right of women to inherit made its way into the strict agnatic systems of kinship and succession which prevailed among the more advanced of the barbarous societies. The Brahman compiler of Manu, while speaking of the appointed daughter, uses language which seems to show the natural growth of feeling: ‘The son of a man is even as himself, and as the son so is the daughter (‘thus appointed’ adds the commentator): how then, if he have no son, can any inherit his property but a daughter who is closely united with his own soul?’ As the law developed itself, the most general result finally attained was that daughters inherited when sons had failed. But it was not reached at once. Among the ancient Hindu writers, Baudhâyana seems to have wholly denied the right of women to inherit: Apastamba places the daughter at the very end of the list of inheritors, but the more modern Vishnu introduces both mother and daughter immediately after the sons. In works treating of the Athenian law, it is usually stated that when there were no sons daughters succeeded. But this is not an adequate statement of the rule. The daughter of a man who left property but no sons, was not in strictness his heiress. She was, as her Greek name (ἐπίκληρος) indicates, a ‘person who went with the property.’3 As I have said above, her father might compel her by testament to marry the devisee of her share; but, if he died intestate, she was subject to another liability—marriage to his nearest kinsman—which connects itself with some singular branches of our subject to be discussed presently. In all these Athenian rules, it is to be observed that, while the ancestral sacrifices are constantly mentioned, the object of special care is the devolution of the estate in the household. The religious basis tends to drop away from the law. Indeed, the wish to prevent daughters from carrying off the patrimony of one household to another is not at all a feature exclusively of sacerdotalised bodies of usage. The secular law of the unsacerdotalised Hindus of the Punjab applies the same principle and exhibits some instructive variants of the Athenian rules (‘Notes on Punjab Customary Law,’ vol. ii. pp. 75, 81, 184, 239). Under some Punjab usages, the daughter, when there are no sons, inherits a limited interest in her father’s property; but she must resign it when she marries. It is usual, however, for the husband of such a daughter to be adopted by his father-in-law.

The legitimate sons, and the son of an ‘appointed’ daughter, have in their veins the blood of the father to whom they sacrifice and succeed. But when there are no sons, and when there has been no appointment of a daughter, we are introduced by the law-books to a number of possible successors whose sonship is altogether fictitious. I know no part of the ancient Hindu law more curious than this, or demanding more imperatively to be taken into careful account by all who investigate the beginnings of organised human society. That ancient family law is entangled with fictions has long been known. (See my ‘Ancient Law,’ p. 130.) One of them has been so long before our eyes as to be comparatively familiar to us. This is Adoption, the engrafting on the family a son from a strange house. Its importance as a private institution at Rome and Athens is of course well known to students; and, among the Romans of the Empire, it became politically important in a high degree as one of the chief expedients for bringing about the peaceable succession of Prince to Prince. It is true that to Englishmen, nowadays, it is little more than a name; to adopt a child is to nurture and educate it, and perhaps to provide for it by Will. But in the French Civil Code (liv. i. 8; tit. 8, c. 1), and other Continental Codes founded on the French, Adoption survives as an institution: a childless man, though under somewhat severely restrictive conditions, may take to himself an adoptive child who will be entitled to succeed to his property. This familiarity with Adoption, during such a length of history, blinds us to the fact that it is one of the most violent of fictions. The faculty of accepting them, strong as it is in ancient communities, must have been strained to the utmost when, for the purpose of taking part in the most solemn of religious ceremonies and earning a consequent right of inheritance, a strange child was transferred to the household, or a man alien in blood was permitted to enter it voluntarily. No doubt, in the more recent practice of the societies accustomed to the adoption of children, the violence of the fiction is somewhat diminished. The theory may be that the child adopted is a stranger, at most of the same order or caste as the person adopting him, but in India he is generally a blood-relation of some kind; and, on looking through a list of known Roman adoptions, the large majority will be seen to be instances of the adoption of ‘cognatic’ kinsmen—that is, of relatives through women. But the ancient feeling on the subject may be inferred from the place which simple adoption occupies in the list of expedients for continuing the family of a childless father as set forth in these early Hindu law-books. ‘There is a singular disproportion,’ says Mr. J. D. Mayne, ‘between the space necessarily devoted to adoption in the English works on Hindu law, and that which it occupies in the early law-books. One might read through all the texts from the Sutra writers down to the Daya-Bhaga without discovering that adoption is a matter of any prominence in the Hindu system’ (‘Hindu Law,’ p. 81). The truth is, that by its side there are a number of fictitious affiliations which were of at least equal antiquity with Adoption, and which, I suspect, served its object even more completely in very ancient times. They are startling or revolting to modern sentiment, but they seemed perhaps simpler and more natural to ancient thought than the admission of a mere stranger to the family.

These fictitious sons are called by Gautama (xxviii. 32) the ‘son born secretly,’ the ‘son of an unmarried damsel,’ the ‘son of a pregnant bride,’ and the son of a ‘twice married woman.’ It is sufficient to say of them that none of them are necessarily the sons of the father whom they are permitted to worship after his death, while some of them cannot possibly be his children. They are all, to use modern words, illegitimate or adulterine offspring, but then they are all 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 wellknown rule which, in Roman law, would settle the status of a slave. Here it is that these strange usages link themselves to familiar phenomena of primitive societies. Paternal power and protective power are inextricably blended together; even the Slave is in some sense a member of the family. We know in fact that at Rome a Slave could perform the family sacrifices on his master’s death; and it was a common contrivance of men who expected to die insolvent to nominate a slave as the heir in the last resort, in order that the bankruptcy of the estate might be declared in his name. Thus, on the secular side, these fictitious sons are permitted to rank as in some remote sense sons, because they are born of women protected by the head of the household, and because they are themselves protected by him. On the religious side, they are permitted to offer the ancestral sacrifices as a desperate expedient for preserving the ancestor from a total failure of male offspring, and from the terrible consequences of entering the world of the dead without the proper oblations and rites.

It must be, however, understood that strong moral repugnance to the fictitious affiliation of these illegitimate and adulterine children begins to show itself among the oldest of the Hindu law-writers whose treatises have survived. A very ancient authority, Apastamba, gives no list of them, protests against the principle, and lays down broadly that ‘the son belongs to the begetter.’ Even the writers who mention them vary greatly as to their place in the order of succession, and Manu aims at them the remark (ix. 161) that ‘such advantage as a man would gain who should attempt to pass deep water in a boat made of woven reeds, that father obtains who passes the gloom of death leaving only contemptible sons.’ I cannot doubt that the growing popularity of Adoption, as a method of obtaining a fictitious son, was due to moral dislike of the other modes of affiliation which was steadily rising among the Brahman teachers in the law-schools.

Let us now suppose the head of the household to have died without having left a son, without having appointed a daughter, without having taken a son in adoption, without male children born in the house who can satisfy the fiction of sonship,—is there any escape possible from the dreaded consequences of failure in the family succession and the ancestral sacrifices? In the opinion of some of the Hindu doctors, these consequences might be averted by an institution which has lately received a great deal of attention, known commonly as the Levirate, but called by the Hindus, in its more general form, the Niyoga. Under it, a son is born to a childless man of his wife or his widow, not from the husband himself but from his brother or nearest kinsman. The practice of so obtaining a son appears to have extended, with various modifications and with or without the religious sanction, over many branches of the human race. We come upon faint but still recognisable traces of it in the law of the Spartans and Athenians, and in one of its forms it was certainly followed by the Hebrews. The Levirate, under which the son is born to the dead man from his brother, ‘that his name be not put out of Israel,’ is best known to Englishmen from the casuistical question of the Sadducees in the twenty-second chapter of St. Matthew (v. 24 et seq.): ‘Master, Moses said, If a man die, having no children, his brother shall marry his wife and raise up seed to his brother. Now there were with us seven brethren; and the first, when he married, deceased and, having no seed, he left his wife unto his brother: Likewise the second also, and the third, unto the seventh. And after them all, the woman died. In the Resurrection, therefore, whose wife shall she be of the seven? for they all had her.’ In the passage here expressly referred to (Deuteronomy xxv. 5) the duty of the husband’s brother is declared to be imperative. ‘If brethren dwell together, and one of them die and have no child, the wife of the dead shall not marry without unto a stranger: her husband’s brother shall go in unto her, and take her to wife, and perform the duty of a husband’s brother unto her. And it shall be that the first-born which she beareth shall succeed in the name of his brother which is dead, that his name be not put out of Israel.’ The verses which succeed describe the procedure which is to be followed when the brother-in-law declines the obligation; and this procedure, consisting chiefly of a symbolic plucking off of the shoe, reappears in the Book of Ruth, where the idyllic beauty of the story sometimes blinds the reader to the fact it is meant to illustrate, a legal rule which was important in its bearing on a passage in the genealogical history of the Royal House of Judah. The most ancient form of the institution appears, however, to be that which is described by the oldest of the Hindu law-writers. ‘A woman whose husband is dead and who desires offspring may bear a son to her brother-in-law. Let her obtain the permission of her Gurus (that is, her spiritual directors). On failure of a brother-in-law, she may obtain offspring by cohabiting with a Sapinda, a Sagotra (a Roman would have said, an ‘Agnatus’ or ‘Gentilis’), a Samânapravara (that is, one of the same literary or sacerdotal clan as her husband), or one who belongs to the same caste. Some declare that she shall cohabit with nobody but a brother-in-law.’ It is to be remarked that Gautama does not appear to contemplate that the widow will necessarily become the wife of the Levir, and that, as in the Book of Ruth, the obligation is extended by him to kinsmen remoter than a brother-in-law, though he notices the opinion that a brother-in-law alone can raise up seed to his brother (Gautama, xviii. 6 et seq.)

But the practice here and there received an extension even more revolting to modern delicacy than the shape which it takes in the Levirate. ‘The child begotten at a living husband’s request, on his wife,’ says Gautama (xviii. 11), ‘belongs to the husband.’ There are several instances of such requests referred to in the Sanscrit literature, but the practice, when defined as an institution by the lawyers, strictly required that the natural father of the child should always be a kinsman. Gautama immediately adds to the passage just quoted, ‘If the natural father of the child was a stranger, that is not of kin to the husband, it belongs to the stranger.’ And, again, in his list of sons, this ancient writer places the ‘son begotten on the wife by a kinsman.’ It would appear, as I shall have to point out presently, that Hindu sacerdotal feeling was divided from the very earliest times on the morality of the Niyoga; but we must bear in mind that its coarser form was not necessarily more repugnant to the old teachers than the form which seems to us somewhat less offensive. No doubt the birth of the son from the widow does not revolt so much as his birth from the wife. But then the ancient law made little difference between the husband’s old age and his death. It is assumed that an old man will quit his house and family and withdraw to spend the residue of his life in asceticism; and the fittest moment for retirement is frequently described as the time at which he becomes incapable of fatherhood.

There are some vestiges of the class of functions assigned by the Niyoga to the nearest kinsman in the records of both the great States of Greece. A well-known story told by Plutarch (‘Pyrrhus,’ 26) of the relations between a brilliant Lacedæmonian officer, Acrotatus, and Chelidonis, the wife of Cleonymus, and of the way in which the old men of Sparta applauded these relations and invoked blessings on the offspring of Chelidonis, does assuredly suggest that, in that old-fashioned and never very delicate society, some institution like that of the ancient Hindus survived till the third century before Christ. Cleonymus was an aged man, and Acrotatus, his grand-nephew, seems to have been his nearest male relative in the flower of life. At Athens, the most nearly corresponding institution differed considerably from the Hindu form. I have stated that an Athenian father might provide, like an Hindu, for the continuance of his family through the son of a daughter; but if, dying sonless and intestate, he allowed his property to descend to a daughter without special arrangement, she became the Orphan Heiress (or ἐπίκληρος), who makes a great figure in Attic law. She had no power of choosing a husband for herself, but it was the right of her nearest kinsman to marry her and his duty to marry or portion her. The right seems in fact to have been keenly disputed; there was a special proceeding (or διαδικασία) for deciding between different claimants, and men often divorced their wives in order to marry the heiress. The same principle was applied to a group of daughters, whom their various kinsmen in order of proximity had to marry or provide with a portion. The object, of course, is to keep the property in the family, and, if possible, to provide that the daughter’s children should derive a stream of its blood from male descents. An even more remarkable application of the principle occurred when the children left were a brother and sister. In such a case the duty of the brother was to portion the sister, but if she were only a half-sister, the strong Athenian feeling against the marriage of brothers and sisters had to give way, and he might marry her and save the portion to the estate. This power could not, however, be exercised, if the sister were uterine, that is, a child by the same mother though not by the same father; and this limitation has been thought a survival of the remote age at which the Athenians counted kinship through females only. But marriage with an uterine sister would have no tendency to promote the object aimed at. She would have no rights over the father’s estate, and marrying her would not help to keep it from diminution and to preserve in its integrity the fund for the ancestral sacrifices.4 Let me repeat that, in most of the Athenian rules about the rights and duties of the nearest kinsman, we have illustrations of the tendency, manifest also in the last chapter of the Book of Ruth, of ancient contrivances for continuing the family to become mere modes of succession to property.

A few words will not be thrown away on the probable origin and meaning of this group of institutions. The Levirate, which is a special case of the Niyoga and under which one brother raises up seed to another, has had a definite place assigned to it by the late Mr. J. F. McLennan in the evolution of society. Originally, I understand him to lay down, there was promiscuity in the relations of the sexes. This promiscuity became limited by Polyandry,5 one wife having several husbands. These plural husbands came in time to be always brothers, and the Levirate is a relic of this form of Polyandry. It would not be quite easy to bring all forms of the Niyoga (of which the Levirate is, as I have said, only a special case) under this ingenious theory; but I will confine myself to saying that the explanation is not the one suggested, to my mind at all events, by the antiquities of Hindu law. Let us suppose that in a particular 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 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 performed 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. There was a division of opinion about the Niyoga, especially in its more offensive shape, from the very first. Apastamba condemns it in the strongest language, while Baudhâyana and Gautama have nothing to say against it. Manu, in a later age, declares it is only fit for cattle (ix. 65. 66), but Narada, a still more recent authority, almost pervaded by the modern spirit, takes it as a matter of course.

I have stated that, in my opinion, the capacity which came to be recognised in daughters, to transmit to a male child the religious quality of sonship to his maternal grandfather, is connected with the ultimate admission of female descendants to a share in the inheritance. It seems to me, further, a plausible conjecture that the capacity of the widow to produce a son to her deceased husband through the Levirate has helped to confer on her the life-interest in her husband’s property which she enjoys in parts of India; and has also led to the power very generally vested in her by Hindu law and usage of taking a son to her deceased husband by simple adoption. My subject, however, is the dependence of inheritance on ancestor-worship, and these topics are too far removed from it to be fitly discussed at present. In any inquiry into the origins of the succession of daughters to their father’s estate, it would be necessary to examine the practice of giving them portions on their marriage which prevailed widely in the ancient world. The gift to a woman or the provision for her on her marriage cannot be separated from her right of succession. Speaking generally, they are alternative modes of providing for her; and the exclusion of daughters from inheritance in ancient systems of law constantly means that they have a right to be portioned, as a rule, out of the movable property of a family. The ancient Hindu writers scarcely mention the daughter’s succession. Baudhâyana, it is thought, held the opinion that no woman could inherit. Apastamba brings in the daughter not only after the male relatives, but after such remote successors as the religious teacher and the fellow-pupil of the deceased. But still these writers implicitly recognise some separate property in married women (Gautama, xxviii. 24). In the ancient legal systems of the Western world there is a visible connection between inheritance and provision upon marriage. Under Athenian law, when sons have failed and the father has died intestate, daughters must be either married to kinsmen, or portioned by them under the system which I have described. The ancient Roman law, at the earliest stage at which we know it, is thought to have allowed some share of their father’s inheritance to daughters. But the Roman law has bequeathed to modern jurisprudence the doctrine6 that, under certain circumstances, a marriage portion is to be deemed an ‘advance’ of a legacy to a daughter, and, conversely, that a covenant to settle a portion is ‘satisfied’ by a legacy. I have always suspected that this doctrine inverted the principle of the oldest law; and that, anciently, the daughter only succeeded when she had not been portioned. In the Joint-Families of modern India, and in the Slavonian House-Communities, though the estate may be regarded as belonging to the male members of the household, the women are entitled to a portion on marriage, generally amounting to some definite fraction of the share which their brothers would receive on a division; and in India, when the property of a joint-family is distributed, the law saddles the shares with a liability to ‘maintain’ the unmarried women and widows. Nowhere, so far as I know, are women left without provision in ancient societies which have made even a slight degree of advance. The real prejudice or reluctance is against allowing them to confer on their husbands, to whom they are generally married in infancy, any rights over the kind of property, such as land, by which the community lives and holds together. But a provision for them by means of property which is actually movable and transferable is thought not merely just and fair, but so imperatively required that it would be a violation of decency and a blot on the family honour to omit or refuse to provide it.

We have now come to the point at which, if there were any close analogy between a modern legal writer and these ancient expositors of the Brahmanical sacred law, they would take up for discussion (1) the succession of ascendants of the deceased, of his male paternal ancestors, if any survived him, and (2) the succession of collaterals—that is, of the descendants of his paternal ancestors. The second of these subjects, Collateral Succession, has attained a vast extent and complexity in the modern7 law of the Hindus; and on the whole its importance has increased rather than diminished in Western Europe. Englishmen are less interested in Collateral Succession than other peoples, and, indeed, it may be said in all succession by law, through their almost universal habit of determining the devolution of their property by marriage-settlements or wills. But on the Continent, principally through the operation of the French Code and of the Codes modelled on it, the practice of testamentary disposition is said to be on the decline. The rights over the father’s property secured to children are indefeasible, and the chief modern object of a Will, the distribution of property among children according to their character and needs, being thus unattainable, Wills fall into disuse and the law is left to settle the succession of more distant relatives. It shows the remoteness of the legal ideas which I am examining from those now prevalent, that the ancient lawyers before us hardly notice collateral succession. They provide for the ultimate succession, on failure of nearer claimants, of spiritual kinsmen, the Brahman teacher, and the fellow-pupil, and for the succession of the King, but they say hardly anything of Inheritance as now understood, save in the direct line of descent or ascent. Their language on the remoter succession of blood relations is brief and obscure, and they do not use technical terms in the same sense, or in the sense of the modern Hindu law.8 They pass rapidly to the spiritual inheritors whom I have named, and to the King; and one of them adds that ‘in cases for which no rule has been given that course must be followed of which at least ten Brahmans, who are well instructed, skilled in reasoning, and free from covetousness, approve’ (Gautama, xxviii. 48).

The brevity and obscurity of the early law-teachers on certain topics have been accounted for by the assumed purpose of their treatises, which is to give a compendious summary of the law in aphoristic language. It is to be observed, however, that they are full and clear enough on all subjects to which they attach importance. It is, I think, impossible not to see that, so far as regards collateral succession, they were little interested in it. The truth seems to me to be that they trusted, for the proper devolution of the inheritance, to their various contrivances for providing a son when legitimate sons had failed, to the appointment of a daughter, to their fictions of sonship, to adoption and to the Niyoga. It is probable that at first an efficacious sacrifice to the dead could only be offered by a descendant in the direct line; and though some of the artificial methods of obtaining a worshipping representative were disapproved of, it is very likely that a collateral relative could not originally sacrifice at all with any prospect of conferring or receiving spiritual benefit. But all the artificial expedients, save one, for providing sons have long since been exploded in India. They are not permitted, says the orthodox Hindu doctor, in the Iron Age in which we now live, because of the hardness of men’s hearts. As a matter of fact, a current of feeling adverse to some or all of them runs through the most ancient of the law-books, and this is the source of the opinion which has ultimately prevailed. Nowadays, if a man has no legitimate sons, he has no resource but adoption, either by himself or his widow, and there are local disputes whether the widow requires his consent or directions to be given before he dies, and if she requires it, in what form it should be given. Such a state of the law adds greatly to the modern importance of collateral succession, and there are facts which co-operate with the law. There is marked infertility among Hindus of high rank, and, though there may be a theoretical preference for adopting a son rather than allow the succession to go to a collateral, yet (as I am informed) there is a great deal of the same superstitious disrelish for effecting an adoption which is known sometimes to prevent in England the making of a will.

The original authorities for the very extensive body of modern rules governing the succession of collateral relatives are far less the ancient law-books than the so-called mediæval Digests, dating approximately from the eleventh to the fourteenth century, of which the most archaic is the ‘Mitakshara.’ The most general feature of this body of rules is thus described by Mr. J. D. Mayne (‘Hindu Law and Usage,’ p. 51): ‘Except in Bengal, agnates, kinsmen connected through male descents, exclude cognates, kinsmen through females, to the fourteenth degree.’ The same preference for males is observable in the rules of succession shown to prevail in the Punjab, where law and usage are ‘essentially unsacerdotal, unsacramental, secular.’ The judicial experience of the Chief Court of the Punjab here coincides with the conclusions of the official inquirers, and establishes that ‘kinship is wholly agnatic.’9 There can be no doubt, therefore, that agnatic succession among collaterals is the general principle of Hindu usage. It was the exclusive principle of the Roman law under the Twelve Tables, and it governed the remoter collateral successions under the law of Athens, which prescribed that agnates should always have precedence over cognates (προτιμα̑σθαι τοὺς ἀπὸ τω̑ν ἄῤῥένων τω̑ν ἀπὸ θηλείων). Indeed, if a comparatively recent writer1 may be trusted, agnatic succession, succession through males exclusively, was, if I may so put it, the common law of Greece.

But one remarkable exception to this general preference for males in India is specified by Mr. Mayne. In the populous province of Bengal Proper, also noticeable for the nearly total disappearance of the Village Community, cognates are largely admitted to succeed, and sometimes in preference to agnates. ‘Heirs in the female line frequently take before very near Sapindas in the direct male line’ (‘Hindu Law and Usage,’ p. 428). Mr. Mayne has very copiously illustrated this peculiarity of Bengal law, and traced it to its causes, in his sixteenth chapter. The relatively modern authorities followed by the Brahmanical lawyers of that province—the Daya-Bhaga and Daya-Krama-Sangraha—are charged with sacerdotal doctrine. They display not only a close connection between ancestor-worship and inheritance, but a complete dependence of the last upon the first. The first question is, What is the exact amount of spiritual benefit received by the ancestor from the sacrifices, and what is the precise amount reflected on the worshipper?—and this is an accurate measure of the place of the worshipper in the table of succession.

The explanation seems to me to be that the original Ancestor-worship transformed itself, and in the course of change helped to modify the law, but did not affect all the stream of legal doctrine in the same degree. Originally, it cannot be doubted, the ancestor worshipped was a male, and the worshipper was his direct male descendant through males. Again, nothing can be stronger than the denials of the right of any woman to offer a sacrifice which we find in the ancient writers. ‘A female shall not offer any burnt oblation’ (Apastamba, ii. vi. 15. 18). But, as I pointed out before, there seems to have arisen in time a practice of associating the ancestor’s wife with the ancestor as an object of worship. ‘A man must fare by himself in the other world,’ say the Hindu doctors. ‘Even were he to die with him, a kinsman cannot follow his dead relative.’2 ‘All, excepting his wife, are forbidden to follow him on the path of Yama’—a passage which in later times became one of the chief authorities for the burning of the widow. Thus in early, but still not in the most ancient, times, men are found worshipping their mother as well as their father, and also their maternal ancestors, though without quite putting them on the same footing as their ancestors through males. One great breach was thus made in the ancient system. Another transformation of religious ideas, which did not perhaps extend beyond particular Brahmanical schools, may be traced in the Daya-Bhaga. The growing moral dissatisfaction with the artificial modes of procuring sons must have increased the chance of childlessness, and therefore of a failure in the sacrifices. Such a prospective result, drawing with it not only supernatural penalties on the dead, but secular losses to the Brahmans, would tend to produce or strengthen the belief that mere collaterals might efficaciously offer sacrificial honour to the dead, and, further, would aid in enlarging the view of collateral relationship as widely as possible. This, in fact, is the religious system shadowed forth in the treatises of authority in Bengal. It is a system aimed, among other things, at bringing as large a number of relatives as possible, including cognates, or kinsfolk through women, within the circle of more or less efficacious worshippers. It is moreover a system full of that minute detail and of those subtle inferences from supposed principles which are characteristic of a highly developed religion which has long since departed from its original simplicity. I must leave the distinctions between the oblation of an entire funeral cake, the offering of the fragments left on the hands and wiped off them, and the mere libation of water, together with the corresponding distinctions between the classes of relatives admitted to the succession, to be studied in the books of professed writers on Hindu law, and especially in the works of Mr. J. D. Mayne and of Professor Rajkumar Sarvadhikari.

I have already stated my belief that at the back of the ancestor-worship practised by Hindus there lay a system of agnation, or kinship through males only, such as now survives in the Punjab. I so far agree with the theory of M. Fustel de Coulanges that I believe this system to have been at first greatly strengthened by ancestor-worship. But it seems to me plain that ancestor-worship in its later growth, acted as a weakening and dissolving force upon the ancient kinship and the ancient family. The secular law followed by Hindus was not, however, equally or universally affected by the religious development. The Mitakshara, which is, on the whole, of more authority in India than the Daya-Bhaga, is manifestly based in the main upon the more ancient conception of kinship. At the same time I do not regard the system of the Daya-Bhaga as simply an after-growth of the system reflected in the more archaic treatise. It is rather a separate development of the ancient sacerdotal law. The ideas which led to it are more or less discernible in the oldest treatises, but they seem to have been carried to their consequences in some law-schools more rapidly and completely than others. Nobody will understand the relatively late collection of rules called after Manu, who does not recognise that it has been materially affected by the religious transformation.

Among the forces which have caused and directed the progress of human society, one of the most powerful has been the Edict of the Roman Prætor, which gradually brought law into harmony with a set of principles known under their most general designation as Equity. It completely transmuted the Roman jurisprudence; and the system, formed by its infiltration into older rules, is the fountain of nearly all modern Continental law, of some part of the English law, and of the greatest part of the existing Law of Nations. These principles were finally considered by the Roman lawyers to fit in with a Greek philosophical conception, the Law of Nature, which was destined to have a serious influence on human thought down to our own days. At an earlier stage of legal opinion the Prætor’s Edict was thought to embody the Jus Gentium, a supposed generalisation of the usages of a great part of mankind. But of the most ancient history of the Roman Equity we cannot be said to know anything. We have evidence, however, that the Edict was employed in very early times to transform the Roman law of Inheritance, founding it on a view of kinship very faintly recognised previously or not at all. Now I, at all events, cannot read the ancient Hindu law-tracts and compare them with such treatises as the Daya-Bhaga and Daya-Krama-Sangraha without being led to the conclusion that, in the interval between the two states of the law reflected in the older and the later books, a change has taken place among the Hindus extremely like that which has occurred among the Romans when the Agnatic Inheritance of the Twelve Tables had been altered into the Cognatic Succession of the Edict. But the ancient Roman law of inheritance was closely implicated with ancestor-worship. This at all events must be taken as placed beyond doubt by M. Fustel de Coulanges. The ancient Hindu law had undoubtedly the same basis, but it underwent in parts of India very much the same modifications as the Roman law, and became a system of inheritance, allowing kinsmen through females to inherit as well as kinsmen through males. The newer Hindu law, however, carries with it the explanation of its own origin; the religious element in it has been transmuted, and the law with it. I suggest, therefore, that the Roman Equity had its beginning before legal history began, in a modified ancestor-worship and a change in the religious constitution and religious duties of the family. There are no ancient philosophies, and perhaps not many modern philosophies, which may not be suspected of having their roots in a religion. The Athenian law corresponds in some of its rules of collateral succession to the later rather than to the earlier Roman law, and here, too, I suggest that a change was produced by an alteration of religious ideas.

NOTES AND ILLUSTRATIONS.

Note A.

HINDU PATRIA POTESTAS.

It is possible that the ancient sacerdotal writers, besides being led by the dependent position of their order into denying the multiplication of religious observances through the dissolution of tribal and joint family groups, were also desirous that the period at which each household broke up into several families should not be delayed till the death of its head. Their expectation is that the faithful Hindu, the man twice born through the study of the Scriptures, will retire in advanced years from active life and become an ascetic or a hermit. There are a few texts which have been thought to imply that the sons of an aged father could compel his retirement. Gautama (xv. 19), while condemning such a practice, perhaps admits its existence. But, whatever be the meaning of these texts, I cannot allow that they lend any countenance to an opinion that sons could compel a partition of the family property at any time against the will of their father. I regard them as exclusively applying to the case of a father who has reached an age at which it has become a religious duty for him to abandon secular life. The fulness of the ancient Hindu Patria Potestas may be safely inferred from the veneration which even a living father must have inspired under a system of ancestor-worship. At a much later date the law-book of Manu declares that ‘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’ (Manu, viii. 416). A still more recent, but still ancient, authority—Narada (v. 39)—says that a son is ‘of age and independent in case his parents be dead; during their lifetime he is dependent, even though he be grown old.’ And nowadays Mr. Nelson, speaking of the South of India, over which the crust of sacerdotal Hinduism is thin, describes the Patria Potestas, which he knows by observation, as the one great standing institution of the Hindu. ‘It is the undoubted fact that among the so-called Hindus of the Madras province the father is looked upon by all at the present day as the Rajah or absolute sovereign of the family that depends upon him. He is entitled to reverence during life as he is to worship after his death. His word is law, to be obeyed without question or demur. He is really the ‘master of the family,—of his wife, of his sons, of his slaves, and of his wealth’ (‘View of the Hindu Law,’ p. 56). And, at p. 38, ‘Resistance to the will of the father appears monstrous.’

Note B.

POLYANDRY.

I should be sorry to have it supposed that I doubt the existence of Polyandry, and specially in the form of a plurality of husbands who were brothers, as an occasional practice of the ancient world. The much-discussed story, in the Mahabharata, of Draupadi becoming the wife of the five Pandavi princes may be open to various interpretations (see Mayne’s ‘Hindu Law and Usage,’ p. 52), but there is fairly good evidence (Polybius, xii. 7,732, following Timæus) that the Spartans practised polyandry. What I doubt (with Mr. L. H. Morgan) is the importance of the place assigned by Mr. McLennan to polyandry in the evolution of society. It serves as a caution against being too much impressed by the antiquity of the Indian and Greek examples to be reminded that the President de Brosses accused the Venetian aristocracy of practising the polyandry of brothers in the early part of the eighteenth century (‘Lettres Écrites d’Italie, tom. i. p. 157). The Spartan and Venetian aristocracies were both noted for their want of delicacy in sexual relations, and in both cases the cause of the practice seems to have been the levy of public taxation on separate households which did not come into existence without separate marriages. The usage seems to me one which circumstances overpowering morality and decency might at any time call into existence. It is known to have arisen in the native Indian army.

[1 ]See Note A to this chapter.

[2 ]See Chapter VII. (on ‘Modern Theories of Primitive Society’) below.

[3 ]See Jebb, Attic Orators, ii. 318.

[4 ]This is the explanation of M. Fustel de Coulanges (Cité Antique, p. 83), which seems to me conclusive. He observes that an emancipated son did not enjoy the privilege.

[5 ]See Note B, ‘Polyandry.’

[6 ]See Dig. xxx. 84, 6. Cod. vi. 37, 11.

[7 ]The existing Hindu law on the subject, with the principles on which the two rival sets of doctrines depend, is discussed by Mr. J. D. Mayne in a most instructive chapter (xvi.) of his Hindu Law and Usage.

[8 ]The familiar terms of the mature Hindu law indicating classes of inheritors (Sapinda, Sagotra, &c.) occur in these writers, but not apparently in the more modern sense. A text attributed to Baudhâyana defines ‘Sapinda’ as ‘the paternal grandfather, grandfather, the father, the man himself, his uterine brother by a woman of equal caste (that is, the son of his father by the same mother as himself, provided she be of equal caste with her husband), his son, his son’s son, and the son of the grandson.’ But this cannot be the meaning of Sapinda in Gautama (xiv. 13, and xviii. 6). Vishnu seems to employ Sapinda and Bandhu as synonymous (xvii. 10).

[9 ]See Boulnois and Rattigan, Notes on Punjab Law, p. 85.

[1 ]Diodorus Siculus, xii. 14 (commenting on a probably spurious law attributed to Charondas).

[2 ]Vishnu, xx. 39.