Front Page Titles (by Subject) LECTURE VII.: ANCIENT DIVISIONS OF THE FAMILY. - Lectures on the Early History of Institutions
The Online Library of Liberty
A project of Liberty Fund, Inc.
LECTURE VII.: ANCIENT DIVISIONS OF THE FAMILY. - Sir Henry Sumner Maine, Lectures on the Early History of Institutions 
Lectures on the Early History of Institutions, 7th edition (London: John Murray, 1914).
About Liberty Fund:
The text is in the public domain.
Fair use statement:
ANCIENT DIVISIONS OF THE FAMILY.
Before the establishment of the (English) common law, all the possessions within the Irish territories ran either in course of Tanistry or in course of Gavelkind. Every Signory or Chiefry with the portion of land which passed with it went without partition to the Tanist, who always came in by election or with the strong hand, and not by descent; but all inferior tenanties were partible between males in Gavelkind.’ (Sir J. Davis’ Reports, ‘Le Cas de Gavelkind,’ Hil. 3, Jac. 1., before all the Judges.)
This passage occurs in one of the famous cases in which the Anglo-Irish Judges affirmed the illegality of the native Irish tenures of land. They declared the English common law to be in force in Ireland, and thenceforward the eldest son succeeded, as heir-at-law, both to lands which were attached to a Signory and to estates which had been divided according to the peculiar Irish custom here called Gavelkind. The Judges thoroughly knew that they were making a revolution, and they probably thought that they were substituting a civilised institution for a set of mischievous usages proper only for barbarians. Yet there is strong reason for thinking that Tanistry is the form of succession from which Primogeniture descended, and that the Irish Gavelkind, which they sharply distinguished from the Gavelkind of Kent, was nothing more than an archaic form of this same institution, of which Courts in England have always taken judicial notice, and which prevailed far more widely on the European Continent than succession by Primogeniture.
It will be convenient that we should first consider the so-called Gavelkind of Ireland, which is thus described by Sir John Davis: ‘By the Irish custom of Gavelkind, the inferior tenanties were partible among all the males of the Sept, both Bastards and Legitimate; and, after partition made, if any one of the Sept had died, his portion was not divided among his sonnes, but the Chief of the Sept made a new partition of all the lands belonging to that Sept, and gave every one his part according to his antiquity.’
This statement occasions some perplexity, which does not, however, arise from its being antecedently incredible. It is made, you will observe, not of the Clan or Tribe in its largest extension, but of the Sept. The first was a large and miscellaneous body, composed in great part of men whose relationship of blood with the Chief and the mass of free tribesmen, was a mere fiction. The last was a much smaller body, whose proximity to a common ancestor was close enough to admit of their kinship either being a fact or being believed to be a fact. It apparently corresponded to the small Highland communities observed in Scotland, by an English officer of Engineers about 1730. ‘They (the Highlanders) are divided into tribes or clans under chiefs or chieftains, and each clan is again divided into branches from the main stock, who have chieftains over them. These are subdivided into smaller branches, of fifty or sixty men, who deduce their original from their particular chieftains. (Quoted by Skene, ‘Highlanders,’ i. p. 156.) Such a body, as I have already stated, seems to be the Joint Family well known to the Hindoos, but continued as a corporate unit (which is very rarely the case in India), through several successive generations. There is no difference in principle, and little in practical effect, between the mode of succession described by Davis and the way in which a Hindoo Joint Family is affected by the death of one of its members. All the property being held in common, and all earnings being brought into the ‘common chest or purse,’ the lapse of any one life would have the effect, potentially if not actually, of distributing the dead man’s share among all the kindred united in the family group. And if, on a dissolution of the Joint Family, the distribution of its effects were not per capita but per stirpes, this would correspond to what Davis probably means when he describes the Chief as giving to each man ‘according to his antiquity.’
The special novelty of the information supplied to us by the ancient Irish law consists in its revealing to us a society of Aryan race, settled, indeed, on the land, and much influenced by its settlement, but preserving an exceptional number of the ideas and rules belonging to the time when kinship and not the land is the basis of social union. There is, therefore, nothing extraordinary in our finding, among the ancient usages of the Irish, an institution savouring so much of the ‘natural communism’ of the primitive forms of property as this Irish Gavelkind. This ‘natural communism,’ I have repeatedly urged, does not arise from any theory or à priori assumption as to the best or justest mode of dividing the land of a community, but from the simple impossibility, according to primitive notions, of making a distinction between a number of kinsmen solely connected by their real or assumed descent from a common ancestor. The natural solvent of this communism is the land itself upon which the kindred are settled. As the common ancestry fades away into indistinctness, and the community gets to consider itself less an assemblage of blood-relations than a body of co-villagers, each household clings with increasing tenacity to the allotment which it has once obtained, and re-divisions of the land among the whole community, whether at fixed periods or at a death, become rarer and rarer, and at last cease altogether, or survive only as a tradition. In this way the widely diffused but modified form of tribal succession, which in England is called Gavelkind, is at last established; the descendants of the latest holder take his property, to the exclusion of everybody else, and the rights of the portion of the community outside the family dwindle to a veto on sales, or to a right of controlling the modes of cultivation. Nevertheless, surveying the Aryan world as a whole, and looking to societies in which some fragments of the ancient social organisation still survive, we can discover forms of succession or property which come surprisingly near to the Irish Gavelkind described by Davis. The best example of this occurs in a practice which existed down to our own day over a large part of Russia. The principle was that each household of the village was entitled to a share of the village-lands proportioned to the number of adult males it contained. Every death, therefore, of a grown-up man diminished pro tanto the share of the household, and every member of it grown to manhood increased its lot in the cultivated area. There was a fixed unit of acreage corresponding to the extent of soil cultivable by one man’s labour, and at the periodical division each household obtained just as much land as answered to its number of adult labouring men. The principal distinction between this system and that which seemed so monstrous and unnatural to Sir John Davis is, that under the first the re-division took place, not as each death occurred, but at stated intervals. I must not, indeed, be understood to say that I think the distinction unimportant. It is very possible that re-distributions at deaths of a common fund may mark a more advanced stage in the history of Property than periodical redistribution, and that the recognition of interests for an entire life may have preceded and paved the way for the final allotment of permanent shares to separate households. Until, however, this last point has been reached, all the modes of re-division known to us are plainly referable to the same principle.
The difficulty suggested by the recital in the ‘Case of Gavelkind’ is thus not a difficulty in believing it if it stood by itself, or if it were made with less generality. But it is distinctly stated that all the lands in Ireland which did not descend by the rule of Tanistry descended by the rule of Gavelkind. The indications of the state of law or custom furnished by the Brehon tracts certainly seem to me inconsistent with this assertion. They show us proprietary rights defined with a sharpness and guarded with a jealousy which is hard to reconcile with the degree of ‘natural communism’ implied in the language of Davis’s Report. The Corus Bescna, of which I said something before, and which deals with rights over tribal lands, implies that under certain circumstances they might be permanently alienated, at all events to the Church; and we shall presently have to discuss some very singular rules of succession, which, however they may affect the Family, certainly seem to exclude the Sept. Dr. Sullivan, who appears to have consulted many more original authorities than have been translated or given to the world, expresses himself as if he thought that the general law of succession in Ireland was nearly analogous to the Gavelkind of Kent. ‘According to the Irish custom, property descended at first only to the male heirs of the body, each son receiving an equal share. . . . . . Ultimately, however, daughters appear to have become entitled to inherit all, if there were no sons’ (Introd., p. clxx).
I do not expect that the apparent contradiction between the Brehon tracts and the language of Davis and his contemporaries respecting the Irish law of succession to land will be fully accounted for till the whole of the ancient legal literature is before the world; but meanwhile it is a plausible explanation of the discrepancy that the Irish and the English writers attended to different sets of phenomena. I cannot doubt that the so-called Irish Gavelkind was found over a great part of the country. The statements of English authorities on the point are extremely precise. They affirm that ‘no civil habitations were erected, and no enclosure or improvement was made of land where Gavelkind was in use,’ and they say that this was especially the case in Ulster, ‘which was all one wilderness.’ Nevertheless it is extremely probable that another set of facts justified the indications given by the Brehon tracts, and that there were other modes of succession known besides succession by Tanistry on the one hand, and besides on the other hand the peculiarly archaic system under which each lapsed share was at once divided between all the members of the Sept. Such an institution as the last, though exceptional circumstances may keep it alive, contains within itself a principle of decay. Each household included in the Joint Family gains a firmer hold on its share of the lands as the distance increases from the common ancestor; and finally appropriates it, transmitting it exclusively to offshoots from its own branch. Nothing is more likely than that there were frequent examples of Irish septs with their land-customs in this condition; and it is still more probable that usages of a similarly modern stamp prevailed in estates permanently severed or ‘booked off’ from tribal possession or established at a distance from the main seat of the tribe. It is true that, in society based on kinship, each family separated from the rest tends itself to expand into a joint family or sept; but in these severed estates custom would be apt to be enfeebled and to abate something of its tyranny. Thus, putting the rule of Tanistry aside, I can quite conceive that the Irish Gavelkind, the modern Gavelkind known to Kent, and many forms of succession intermediate between the two, co-existed in Ireland. Both the English and the Irish authorities on law had prejudices of their own which might lead them to confine their attention to particular usages. The Brehon writers seem to me distinctly biassed in favour of the descent of property in individual families, which commended itself to them as lawyers, as friends of the Church, and (it may be) as well-wishers to their country. On the other, the strange ancient form of ownership which he called Gavelkind would fascinate the observation of an Englishman resident in Ireland. He would assuredly have none of the curiosity about it which we feel nowadays, but surprise and dislike would fix his attention upon it, and perhaps prevent his recognising the comparatively wide diffusion of institutions of the opposite type.
This interpretation of the seeming contradiction between our authorities is consistent with the very little we know respecting actual divisions of land in ancient Ireland. It constantly happened both in Ireland and the Scottish Highlands that a Chief, besides the domain which appertained to his office, had a great estate held under what the English lawyers deemed the inferior tenure. There are two cases on record in which Irish Chiefs of considerable dignity distributed such estates among their kindred. In the fourteenth century Connor More O’Brien, a chief who had children of his own, is stated to have divided his land on principles which must have more or less corresponded to those condemned by the Anglo-Irish Judges. The bulk of the estate he assigned to the various families of the Sept formed by his own relatives. To himself he reserved only one-sixth of one-half of one-third, and even this sixth he divided between his three sons, reserving only a rent to himself. But at the end of the fifteenth century Donogh O’Brien, son of Brien Duff, son of Connor, King of Thomond, divided all his lands between his eleven sons, reserving to himself only the mansion and the demesne in its vicinity. The difference between the two cases, which (it is instructive to observe) are separated by at least a century, appears to me sufficiently plain. In the first the land had remained in a state of indivision during several generations; in the second it had been periodically divided. Connor More O’Brien was distributing the inheritance of a joint family; Donogh O’Brien that of a family (Vallancey, ‘Collectanea de Rebus Hibernicis,’ i. 264, 265.)
It is worthy of observation that in the more ancient example Connor More O’Brien appears to have paid regard to the various stirpes or stocks into which the descendants of the original founder of his family had branched out. The principle he followed I suppose to be the same as that pointed out by Davis when he speaks of the chief dividing a lapsed share between the members of a sept ‘according to their antiquity.’ The proceeding deserves to be noted, as showing an advance on the oldest known tribal customs. In the most archaic forms of the Joint Family, and of the institution which grew out of it, the Village-Community, these distributions are per capita; no one person who is entitled takes more than another, whether the whole estate or a portion is divided, and no respect is paid to the particular way in which a given individual has descended from the common ancestor. Under a more advanced system the distribution is per stirpes; careful attention is paid to the lines into which the descendants of the ancestor of the joint-family have separated, and separate rights are reserved to them. Finally, the stocks themselves escape from the sort of shell constituted by the Joint Family; each man’s share of the property, now periodically divided, is distributed among his direct descendants at his death. At this point, property in its modern form has been established; but the Joint Family has not wholly ceased to influence successions. When direct descendants fail it is even now the rules of the Joint Family which determine the taking of the inheritance. Collateral successions, when they are distant, follow the more primitive form of the old institution, and are per capita; when they are those of the nearer kindred they are adjusted to its more modern shape, and are per stirpes.
The remark has further to be made that both Connor O’Brien and Donogh O’Brien divided their own land among their sons or kindred during their own lifetime. Like Laertes in the Odyssee and like Lear in the tragedy of Shakespeare, the old Chief, in the decay of his vigour, parts with his power and retains but a fraction of the property he had administered; and the poorer freeman becomes one of those ‘senior’ pensioners of the tribe so often referred to in the tracts. Precisely the same practice is recognised, and even (as some think) enjoined, by the more archaic bodies of Hindoo jurisprudence. The principle is that the right of each member of a family accrues at his birth; and, as the family has in theory a perpetual existence, there is no particular reason why, if the property is divided at all, it should be exclusively divided at a death. The power of distributing inheritances vested in the Celtic chiefs has been made the basis of some very doubtful theories, but I have no doubt it is essentially the same institution as the humble privilege which is reserved to the Hindoo father by the Mitakshara. It is part of the prerogative belonging to the representative of the purest blood in the joint family; but in proportion as the Joint Family, Sept, or Clan becomes more artificial, the power of distribution tends more and more to look like mere administrative authority.
Under some systems of Hindoo law, the father, when making a distribution of property during his lifetime, is entitled to retain a double share, and by some Indian customs the eldest son, when dividing the patrimony with his brothers, takes twice as much as the others. There are a good many traces of the usage in this last form in a variety of communities. It is, for instance, the ‘birthright’ of the Hebrew patriarchal history. I mention it particularly because it seems to me to be sometimes improperly confounded with the right conferred by what we call the rule of Primogeniture. But the double share is rather given as the reward or (perhaps we should say) the security for impartial distribution, and we find it often coupled with the right to take exclusively such things as are deemed incapable of partition, the family house, for instance, and certain utensils. The proof that it is not essentially a privilege of the eldest son, we find in the circumstances that it is sometimes enjoyed by the father and sometimes by the youngest of the sons, and in this way it is connected with our own custom of Borough English, of which I shall have more to say presently. There is a difference of historical origin between this kind of privileged succession and that which we call Primogeniture. The first is descended from a custom of the Tribe; the last, to which I now pass, seems to me traceable to the special position of the Chief.
The Brehon tracts at present translated do not add much to the knowledge which we possessed of the Irish customs corresponding to the usage of exclusive succession by the eldest son; and Primogeniture remains what I called it thirteen years ago (‘Ancient Law,’ p. 227), ‘one of the most difficult problems of historical jurisprudence.’ The first of the difficulties which surround it is the total absence, before a particular epoch in history, of recorded precedents for any such mode of succession to property. It was unknown to the Hellenic world. It was unknown to the Roman world. It was unknown to the Jews, and apparently to the whole Semitic world. In the records of all these societies there are vestiges of great differences between the succession of males and the succession of females; but there was nothing like the exclusive succession of a single son to property, although the descent of sovereignties to the eldest son of the last reigning king was a familiar fact, and though the Greek philosophers had conjectured that, in an earlier state of society than theirs, the smaller groups of men—families and villages—had been governed by eldest son after eldest son.
Even when the Teutonic races spread over Western Europe they did not bring with them Primogeniture as their ordinary rule of succession. The allodial property of the Teutonic freeman, that share which he had theoretically received at the original settlement of the brotherhood to which he belonged on their domain, was divided at his death, when it was divided at all, equally between his sons or equally between his sons and daughters. It is quite certain, however, that the appearance of Primogeniture in the West and its rapid diffusion must be connected with the irruption of the barbarians, and with the tribal ideas re-introduced by them into the Roman world. At this point, however, we encounter another difficulty. The Primogeniture which first meets us is not uniformly the Primogeniture with which we are now familiar. The right of the eldest son sometimes gives way to the right of the eldest male relative of the deceased, and occasionally it seems as if neither the succession of the eldest son nor that of the eldest relative could take effect without election or confirmation by the members of the aggregate group to which both belong.
As usual, we have to look for living illustrations of the ancient system to the usages of the Hindoos. The Family, according to the Hindoo theory, is despotically governed by its head; but if he dies and the Family separates at his death, the property is equally divided between the sons. If, however, the Family does not separate, but allows itself to expand into a Joint Family, we have the exact mixture of election and doubtful succession which we find in the early examples of European primogeniture. The eldest son, and after him his eldest son, is ordinarily the manager of the affairs of the Joint Family, but his privileges theoretically depend on election by the brotherhood, and may be set aside by it, and, when they are set aside, it is generally in favour of a brother of the deceased manager, who, on the score of greater age, is assumed to be better qualified than his nephew for administration and business. In ancient Irish society the Joint Family, continued through many generations, has grown first into the Sept and then into the Clan, contracting a greater degree of artificiality in proportion to its enlargement. The importance, meanwhile, of the Chief to the Tribe has rather increased than diminished, since he is no longer merely administrator of its civil affairs but its leader in war. The system produced from these elements appears to me sufficiently intelligible. The veneration of the Tribe is not attracted by individuals of the Chieftain’s family, but by the family itself, as representing the purest blood of the entire brotherhood. It chooses its head and leader (save on the very rarest occasions) from this family, and there are instances of the choice being systematically made from two families in alternation. But the necessity of having a military leader in the vigour of his physical and mental powers is much too imperious to admit of his choice being invariably deferred to the death of the ruling Chief, or to allow of the election falling universally or even generally on his son. ‘It is a custom among all the Irish,’ says Spenser, ‘that presently after the death of any of their chief lords or captains, they do presently assemble themselves to a place generally appointed and known unto them to choose another in his stead, where they do nominate and elect for the most part, not the eldest son, nor any of the children of the lord deceased, but the next to him of blood that is eldest and worthiest, as commonly the next brother if he have any, or the next cousin, and so forth, as any is elder in that kindred or sept; and then, next to him, they choose the next of the blood to be Tanaist, who shall succeed him in the said Captaincy if he live thereunto. . . . For when their Captain dieth, if the Signory should descend to his child, and he perhaps an infant, another might peradventure step in between or thrust him out by strong hand being then unable to defend his right and to withstand the force of a forreiner; and therefore they do appoint the eldest of the kin to have the Signory, for that commonly he is a man of stronger years and better experience to maintain the inheritance and to defend the country. . . . And to this end the Tanaist is always ready known, if it should happen to the Captain suddenly to die, or to be slain in battle, or to be out of the country, to defend and keep it from all such dangers.’ (Spenser’s ‘View of the State of Ireland.’)
Primogeniture, therefore, considered as a rule of succession to property, appears to me to be a product of tribal leadership in its decay. Some such system as that represented by the Irish Tanistry belonged probably at one time to all the tribal communities which overran the Roman Empire, but no precise assertion can be made as to the stage in their history at which it began to be modified, especially since Sohm’s investigations (in his ‘Fränkische Reichs-und Gerichtsverfassung’) have shown us how considerably the social organisation of some of these communities had been affected by central or royal authority in the interval between the observations of Tacitus and the writing of the Salic Law. But I think we may safely conjecture that the transition from the older to the newer Primogeniture took place everywhere under circumstances nearly the reverse of those which kept Tanistry so long alive in Ireland. Wherever some degree of internal peace was maintained during tolerably long periods of time, wherever an approach was made to the formation of societies of the distinctive modern type, wherever military and civil institutions began to group themselves round the central authority of a king, the value of strategical capacity in the humbler chiefs would diminish, and in the smaller brotherhoods the respect for purity of blood would have unchecked play. The most natural object of this respect is he who most directly derives his blood from the last ruler, and thus the eldest son, even though a minor, comes to be preferred in the succession to his uncle; and, in default of sons, the succession may even devolve on a woman. There are not a few indications that the transformation of ideas was gradual. The disputes among great Highland families about the title to the chieftaincy of particular clans appear to date from a period when there was still a conflict between the old principle of succession and the new; and at a relatively later period, when throughout most of Western Europe tribal customs have been replaced by feudal rules, there is a visible uncertainty about such of these rules as affect succession. Glanville, writing of English military tenures in the later part of the reign of Henry the Second, observes: ‘When anyone dies, leaving a younger son and a grandson, the child of his eldest son, great doubt exists as to which of the two the law prefers in the succession to the other, whether the son or the grandson. Some think the younger son has more right to the inheritance than the grandson . . . . but others incline to think that the grandson ought to be preferred to his uncle.’ (Glanville, vii. 7.) This ancient doubt has left traces of itself on literature no less than on history, since it manifestly affects the plot of Shakespeare’s Hamlet; but the very question of principle arose between the descendants of daughters in the controversy between Bruce and Baliol. The succession to the Crown of Scotland was ultimately settled, as it would have been in earlier times, by what amounted to national election, but the decision of Edward the First in favour of Baliol was undoubtedly in accordance with principles which were gaining ground everywhere, and I quite agree with Mr. Burton (ii. 249) that the celebrity of the dispute and the full consideration given to it did much to settle the rule which prevailed in the end, that the whole of the descendants of an elder child must be exhausted before those of the younger had a title. When, however, the eldest son had once taken the place of his uncle as the heir to the humbler chieftaincies, he doubtless also obtained that ‘portion of land attached to the Signory or Chiefry which went without partition to the Tanaist;’ and, as each community gradually settled down into comparative peace under royal or central authority, this demesne, as it was afterwards called, must have assumed more and more the character of mere property descending according to the rule of primogeniture. It may be believed that in this way a principle of inheritance was formed which first of all extended from the demesne to all the estates of the holder of the Signory, however acquired, and ultimately determined the law of succession for the privileged classes throughout feudalised Europe. One vestige of this later course of change may perhaps be traced in the noble tenure once widely extended on the Continent, and called in French ‘Parage,’ under which the near kinsmen of the eldest son still took an interest in the family property, but held it of him as his Peers. There were, however, other causes than those just stated which led to the great development of Primogeniture in the early part of the Middle Ages, but for an examination of them I may be allowed to refer to the work of mine which I mentioned above. (‘Ancient Law,’ pp. 232 et seq.)
I do not think that the disaffirmation of the legality of Tanistry, and the substitution for it of the rule of Primogeniture, can justly be reckoned among the mistakes or crimes of the English in Ireland. The practice had been perpetuated in the country by its disorders, which preserved little groups of kinsmen and their petty chiefs in an unnatural vitality; and probably Sir John Davis does not speak too harshly of it when he charges it with ‘making all possessions uncertain, and bringing confusion, barbarism, and incivility.’ The decision against the Irish Gavelkind was far less justifiable. Even if the institution were exactly what Davis supposed it to be, there was injustice in suddenly disappointing the expectations of the distant kindred who formed the sept of the last holder; but it is probable that several different modes of succession were confounded under the name of Gavelkind, and that in many cases a number of children were unjustifiably deprived of their inheritance for the advantage of one. All that can be said for the authors of the revolution is that they seem to have sincerely believed the mischievousness of the institutions they were destroying; and it is some evidence of this that, when their descendants a century later really wished to inflict an injury on the majority of Irishmen, they re-introduced Gavelkind, though not in its most ancient shape. They ‘gavelled’ the lands of Papists and made them descendible to all the children alike. There seems to me a melancholy resemblance between some of the mistakes which, at two widely distant epochs, were committed by Englishmen, apparently with the very best intentions, when they were brought into contact with stages in the development of institutions earlier than that which their own civilisation had reached. Sir John Davis’s language on the subject of the Irish custom of Gavelkind might be that of an Anglo-Indian lawyer who should violently censure the Brahminical jurists for not confounding families with joint undivided families. I do not know that any such mistake has been made in India, though undoubtedly the dissolution of the Joint Family was in the early days of our government unduly encouraged by our Courts. But there is a closer and more unfortunate similarity between some of the English experiments in Ireland and those tried in India. Under an Act of the twelfth year of Queen Elizabeth the Lord Deputy was empowered to take surrenders and regrant estates to the Irishry. The Irish lords, says Davis, ‘made surrenders of entire countries, and obtained grants of the whole again to themselves only, and none other, and all in demesne. In passing of which grants, there was no care taken of the inferior septs of people. . . . So that upon every such surrender or grant, there was but one freeholder made in a whole country, which was the lord himself; all the rest were but tenants at will, or rather tenants in villenage.’ There are believed to be many Indian joint-families or septs which, in their later form of village-communities, had the whole of their lands similarly conferred on a single family out of their number, or on a royal taxgatherer outside them, under the earliest Indian settlements. The error was not in introducing absolute ownership into Ireland or India, but in the apportionment of the rights of which property is made up. How, indeed, this apportionment shall be wisely and justly made, when the time has fully come for putting individual property in the place of collective property by a conscious act of the State, is a problem which taxes to the utmost the statesmanship of the most advanced era, when animated by the highest benevolence and informed with the widest knowledge. It has been reserved for our own generation to witness the least unsatisfactory approach which has hitherto been made towards the settlement of this grave question in the great measures collectively known as the enfranchisement of the Russian serfs.
The Irish practice of Tanistry connects itself with the rule of Primogeniture, and the Irish Gavelkind with the rules of succession most widely followed among both the Eastern and Western branches of the Aryan race; but there are some passages in the Brehon tracts which describe an internal division of the Irish Family, a classification of its members and a corresponding system of succession to property, extremely unlike any arrangement which we, with our ideas, can conceive as growing out of blood-relationship. Possibly, only a few years ago, these passages would have been regarded as possessing too little interest in proportion to their difficulty for it to be worth anybody’s while to bestow much thought upon their interpretation. But some reasons may be given why we cannot wholly neglect them.
The distribution of the Irish Family into the Geilfine, the Deirbhfine, the Iarfine, and the Indfine—of which expressions the three last are translated the True, the After, and the End Families—is obscurely pointed at in several texts of the earlier volumes of the translations; but the Book of Aicill, in the Third Volume, supplies us for the first time with statements concerning it having some approach to precision. The learned Editor of this volume, who has carefully examined them, describes their effect in the following language: ‘Within the Family, seventeen members were organised in four divisions, of which the junior class, known as the Geilfine division, consisted of five persons; the Deirbhfine, 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 Deirbhfine, the eldest member of the Deirbhfine 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 organisation altogether. It would appear that this transition from a lower to a higher grade took place upon the introduction of a new member into the Geilfine division, and therefore depended upon the introduction of new members, not upon the death of the seniors.’
It seems an inference from all the passages bearing on the subject that any member of the Joint-family or Sept might be selected as the starting-point, and might become a root from which sprung as many of these groups of seventeen men 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 Deirbhfine. A succession of such births completed in time the Deirbhfine division, and went on to form the Iarfine and the Indfine, the After and the End Families. The essential principle of the system seems to me a distribution into fours. The fifth person in the Geilfine division I take to be the parent from whom the sixteen descendants spring, and it will be seen, from the proviso which I inserted above, that I do not consider his place in the organisation to have been ever changed. He appears to be referred to in the tracts as the Geilfine Chief.
The interest of this distribution of the kinsmen consists in this: whatever else it is, it is not a classification of the members of the family founded on degrees of consanguinity, as we understand them. And, even if we went no farther than this, the fact would suggest the general reflection which often occurs to the student of the history of law, that many matters which seem to us altogether simple, natural, and therefore probably universal, are in reality artificial and confined to limited spheres of application. When one of us opens his Prayer-book and glances at the Table of Prohibited Degrees, or when the law-student turns to his Blackstone and examines the Table of Descents, he possibly knows that disputes have arisen about the rights and duties proper to be adjusted to these scales of relationship, but it perhaps has never occurred to him that any other view of the nature of relationship than that upon which they are based could possibly be entertained. Yet here in the Book of Aicill is a conception of kinship and of the rights flowing from it altogether different from that which appears in the Tables of Degrees and of Descents. The groups are not formed upon the same principles, nor distinguished from one another on the same principles. The English Tables are based upon a classification by degrees, upon identity in the number of descents by which a given class of persons are removed from a given person. But the ancient Irish classification obviously turns upon nothing of the sort. A Geilfine class may consist of a father and four sons who are not in the same degree, and the Brehon writers even speak of its consisting of a father, son, grandson, great-grandson, and great-great-grandson, which is a conceivable case of Geilfine relationship, though it can scarcely have been a common one. Now, each of these relatives is in a different degree from the others. Yet this distribution of the family undoubtedly affected the law of inheritance, and the Geilfine class, to our eyes so anomalous, might succeed in certain eventualities to the property of the other classes, of which the composition is in our eyes equally arbitrary.
This singular family organisation suggests, however, a question which, in the present state of enquiry on the subject which occupies us, cannot fairly be avoided. I have spoken before of a volume on ‘Systems of Consanguinity and Affinity in the Human Family,’ published by the Smithsonian Institute at Washington. The author, Mr. Lewis Morgan, is one of the comparatively few Americans who have perceived that, if only on the score of the plain extant evidences of the civilisation which was once enjoyed and lost by some branches of their stock, the customs and ideas of the Red Indians deserve intelligent study. In prosecuting his researches Mr. Morgan was struck with the fact that the conception of Kinship entertained by the Indians, though extremely clear and precise, and regarded by them as of much importance, was extremely unlike that which prevails among the now civilised races. He then commenced a laborious investigation of the whole subject, chiefly through communications with correspondents in all parts of the world. The result at which he arrived was that the ideas on the subject of relationship entertained by the human family as a whole were extraordinarily various, but that a generalisation was possible, and that these ideas could be referred to one or other of two distinct systems, which Mr. Morgan calls respectively the Descriptive and the Classificatory system. The time at our command will only allow me to explain his meaning very briefly. The Descriptive system is that to which we are accustomed. It has come to us from the Canon law, or else from the Roman law, more particularly as declared in the 118th Novel of Justinian, but it is not at all confined to societies deeply affected by Civil and Canon law. Its essence consists in the giving of separate names to the classes of relatives which are formed by the members of the family who are removed by the same number of descents from yourself, the ego or propositus, or from some common ancestor. Thus, your uncle stands to you in the third degree, there being one degree or step from yourself to your father or mother, a second from your father or mother to their parents, a third from those parents to their other children, among whom are your uncles. And ‘uncle’ is a general name for all male relatives standing to you in this third degree. The other names employed under the Descriptive system are among the words in most common use; yet it is to be noted that the system cannot in practice be carried very far. We speak of uncle, aunt, nephew, niece, cousin; but then we get to great-uncle, grand-nephew, and so forth, and at length lose our way amid complications of ‘great’ and ‘grand’ until we cease to distinguish our distant kindred by particular designations. The Roman technical law went considerably farther than we do with the specific nomenclature of relatives; yet there is reason to think that the popular dialects of Latin were more barren, and no Descriptive system can go on indefinitely with the process. On the other hand, the Classificatory system groups the relatives in classes, often large ones, which have no necessary connection with degrees. Under it a man’s father and his uncles are grouped together, sometimes his uncles on his father’s side, sometimes on the mother’s side, sometimes on both; and perhaps they are all indifferently called his fathers. Similarly, a man’s brothers and all his male cousins may be classed together and called his brothers. The effect of the system is in general to bring within your mental grasp a much greater number of your kindred than is possible under the system to which we are accustomed. This advantage is gained, it is true, at the expense of the power of discriminating between the members of the several classes, but still it may be very important in certain states of society, since each of the classes usually stands under some sort of conjoint responsibility.
I am not now concerned with the explanation of the Classificatory system of Kinship. Mr. Morgan and the school to which he belongs find it, as I said before, in a state of sexual relations, alleged to have once prevailed universally throughout the human race, and known now to occur in some obscure fragments of it. The fullest account of the condition of society in which these views of relationship are believed to have grown up may be read in Mr. McLennan’s most original work on Primitive Marriage. The point before us, however, is whether we have a trace of the Classificatory system in the Irish division of the Family into four small groups, no one of which is necessarily composed of relatives of the same degree, and each of which has distinct rights of its own, and stands under definite responsibilities. Undoubtedly, the Descriptive system was that which the ancient Irish generally followed; but still it would be an interesting, and, in the opinion of pre-historic writers, an important fact, if a distribution of the Family only intelligible as a relic of the Classificatory system remained as a ‘survival’ among the institutions reflected by the Brehon Laws. My own opinion, which I will state at once, is that the resemblance between the Irish classification of kindred and the modes of classification described by Mr. Morgan is only superficial and accidental. The last explanation Mr. Morgan would admit of the remarkable ideas concerning kinship which form the subject of his book would be that they are connected with the Patria Potestas, that famous institution which held together what he and his school consider to be a relatively modern form of the Family. I think, however, I can assign some at least plausible reasons for believing that this perplexing four-fold division of the Celtic Family is neither a mere survival from immemorial barbarism nor, as most persons who have noticed it have supposed, a purely arbitrary arrangement, but a monument of that Power of the Father which is the first and greatest land-mark in the course of legal history.
Let me repeat that the Irish Family is assumed to consist of three groups of four persons and one group of five persons. I have already stated that I consider the fifth person in the group of five to be the parent from whom all the other members of the four divisions spring, or with whom they are connected by adoptive descent. Thus, the whole of the natural or adoptive descendants are distributed into four groups of four persons each, their rank in the Family being in the inverse order of their seniority. The Geilfine group is several times stated by the Brehon lawyers to be at once the highest and the youngest.
Now, Mr. Whitley Stokes has conveyed to me his opinion that ‘Geilfine’ means ‘hand-family.’ As I have reason to believe that a different version of the term has been adopted by eminent authority, I will give the reasons for Mr. Stokes’s view. ‘Gil’ means ‘hand’—this was also the rendering of O’Curry—and it is, in fact, the Greek word χείρ. In several Aryan languages the term signifying ‘hand’ is an expressive equivalent for Power, and specially for Family or Patriarchal Power. Thus, in Greek we have ὑποχείριος and χέρης, for the person under the hand. In Latin we have herus ‘master,’ from an old word, cognate to χείρ; and we have also one of the cardinal terms of ancient Roman Family Law, manus, or hand, in the sense of Patriarchal authority. In Roman legal phraseology, the wife who has become in law her husband’s daughter by marriage is in manu. The son discharged from Paternal Power is emancipated. The free person who has undergone mancipation is in mancipio. In the Celtic languages we have, with other words, ‘Gilla,’ a servant, a word familiar to sportsmen and travellers in the Highlands and to readers of Scott in its Anglicised shape, ‘Gillie.’
My suggestion, then, is that the key to the Irish distribution of the Family, as to so many other things in ancient law, must be sought in the Patria Potestas. It seems to me to be founded on the order of emancipation from Paternal authority. The Geilfine, the Hand-family, consists of the parent and the four natural or adoptive sons immediately under his power. The other groups consist of emancipated descendants, diminishing in dignity in proportion to their distance from the group which, according to archaic notions, constitutes the true or representative family.
The remains which we possess of the oldest Roman law point to a range of ideas very similar to that which appears to have produced the Irish institution. The Family under Patria Potestas was, with the Pater-Familias, the true Roman Family. The children who were emancipated from Paternal Power may have gained a practical advantage, but they undoubtedly lost in theoretical dignity. They underwent that loss of status which in ancient legal phraseology was called a capitis deminutio. We know too that, according to primitive Roman law, they lost all rights of inheritance, and these were only gradually restored to them by a relatively modern institution, the Equity of the Roman Prætor. Nevertheless there are hints on all sides that, as a general rule, sons as they advanced in years were enfranchished from Paternal Power, and no doubt this practice supplies a partial explanation of the durability of the Patria Potestas as a Roman institution. The statements, therefore, which we find concerning the Celtic Family would not be very untrue of the Roman. The youngest children were first in dignity.
Of course I am not contending for an exact resemblance between the ancient Roman and ancient Celtic Family. We have no trace of any systematised discharge of the sons from the Roman Patria Potestas; their enfranchisement seems always to have been dependent on the will of the Pater-Familias. The divisions of the Celtic Family seem, on the other hand, to have been determined by a self-acting principle. An even more remarkable distinction is suggested by passages in the Book of Aicill which seem to show that the parent, who retained his place in the Geilfine group, might himself have a father alive. The peculiarity, which has no analogy in ancient Roman law, may possibly have its explanation in usages which many allusions in the Brehon law show to have been followed by the Celts, as they were by several other ancient societies. The older members of the Family or Joint Family seem in advanced age to have become pensioners on it, and, like Laertes in the Odyssee, to have vacated their privileges of ownership or of authority. On such points, however, it is safest to suspend the judgment till the Brehon law has been more thoroughly and critically examined.
At the date at which the Book of Aicill was put together the Irish division of the Family seems only to have had importance in the law of succession after death. This, however, is the rule in all societies. When the ancient constitution of the Family has ceased to affect anything else, it affects inheritance. All laws of inheritance are, in fact, made up of the débris of the various forms which the Family has assumed. Our system of succession to personalty, and the whole French law of inheritance, are derived from Roman law, which in its latest condition is a mixture of rules having their origin in successive ascertainable stages of the Roman Family, and is a sort of compromise between them.
The authors of the Brehon Law Tracts frequently compare the Geilfine division of the Family to the human hand, but with them the comparison has at first sight the air of being purely fanciful. The Geilfine group has five members, and the hand has five fingers. Dr. Sullivan—who, however, conceives the Geilfine in a way materially different from the authorities whom I follow—tells us that ‘as they represented the roots of the spreading branches of the Family, they were called the cuic mera na Fine, or the ‘five fingers of the Fine.’ If the explanation of ‘Geilfine’ which I have partly taken from Mr. Whitley Stokes be correct, we must suppose that, at the time at which the Brehon tracts were thrown into their present form, the Patria Potestas of the ancient Irish, though frequently referred to in the tracts as the father’s power of ‘judgment, proof, and witness’ over his sons, had nevertheless considerably decayed, as it is apt to do in all societies under unfavourable circumstances, and that with this decay the association of the Geilfine group with ‘hand’ in the sense of Paternal Power had also become faint. There is, however, a real connection of another kind between the Geilfine group and the five fingers of the hand. If you ask why in a large number of ancient societies Five is the representative number, no answer can be given except that there are five fingers on the human hand. I commend to your attention on this point Mr. Tylor’s most instructive chapter on the infancy of the Art of Counting, in the first volume of his ‘Primitive Culture.’ ‘Finger-counting,’ he observes, ‘is not only found among savages and uneducated men, carrying on a part of their mental operations where language is only partly able to follow it, but it also retains a place and an undoubted use among the most cultured nations as a preparation and means of acquiring higher arithmetical methods’ (I. 246.) Five is thus a primitive natural maximum number. You will recollect that the early English Township was represented by the Reeve and the four men. The Council of an Indian Village Community most commonly consists of five persons, and throughout the East the normal number of a Jury or Board of arbitrators is always five—the punchayet familiar to all who have the smallest knowledge of India. The Geilfine, the representative group of the Irish Family, consisting of the Parent and the four descendants still retained under his Patria Potestas, falls in with this widely extended conception of representation.
The Patria Potestas seems to me the most probable source of a well-known English custom which has occasioned no little surprise to students of our law. ‘Borough English,’ under which the youngest son and not the eldest succeeds to the burgage-tenements of his father, has from time immemorial being recognised as a widely diffused usage of which it is the duty of our Courts to take judicial notice, and many writers on our real property laws, from Littleton downwards, have attempted to account for it. Littleton thought he saw its origin in the tender age of the youngest son, who was not so well able to help himself as the rest of the brethren. Other authors, as Blackstone tells us, explained it by a supposed right of the Seigneur or lord, now very generally regarded as apocryphal, which raised a presumption of the eldest son’s illegitimacy. Blackstone himself goes as far a-field as North-Eastern Asia for an explanation. He quotes from Duhalde the statement that the custom of descent to the youngest son prevails among the Tartars. ‘That nation,’ he says, ‘is composed totally of shepherds and herdsmen; and the elder sons, as soon as they are capable of leading a pastoral life, migrate from their father with a certain allotment of cattle, and go to seek a new habitation. The youngest son, therefore, who continues longest with the father, is naturally the heir of his house, the rest being already provided for. And thus we find that, among many other Northern nations, it was the custom for all the sons but one to migrate from the father, which one now became his heir.’ The explanation was really the best which could be given in Blackstone’s day, but it was not necessary to go for it so far from home. It is a remarkable circumstance that an institution closely resembling Borough English is found in the Laws of Wales, giving the rule of descent for all cultivating villeins. ‘Cum fratres inter se dividant hæreditatem,’ says a rule of that portion of the Welsh Law which has survived in Latin; ‘junior debet habere tygdyn, i.e. ædificia patris sui, et octo acras de terrâ, si habuerint’ (L. Wall., vol. ii. p. 780). And, when the youngest son has had the paternal dwelling-house, eight acres of land and certain tools and utensils, the other sons are to divide what remains. It appears to me that the institution is founded on the same ideas as those which gave a preference to the Geilfine division of the Celtic family. The home-staying, unemancipated son, still retained under Patria Potestas, is preferred to the others. If this be so, there is no room for the surprise which the custom of Borough English has excited, and which arises from contrasting it with the rule of Primogeniture. But the two institutions have a different origin. Primogeniture is not a natural outgrowth of the family. It is a political not a tribal institution, and comes to us not from the clansmen but from the Chief. But the rule of Borough English, like the privileges of the Geilfine, is closely connected with the ancient conception of the Family as linked together by Patria Potestas. Those who are most emphatically part of the Family when it is dissolved by the death of its head are preferred in the inheritance according to ideas which appear to have been once common to the primitive Romans, to the Irish and Welsh Celts, and to the original observers, whoever they were, of the English custom.