measured, not by his wealth or power, not even by what is written about him, but by the number of things he may or may not do. A family on its promotion practises the most rigid abstinence from particular kinds of food and drink, abstains from all sorts of actions, is scrupulously careful about the marriage of its daughters, and goes daily through a punctilious ceremony of domestic worship. It engages a Brahman chaplain and a Brahman cook; and thus the entire Brahman priesthood of the country will perhaps be led to countenance its pretensions to high-caste extraction. Once taken under the shelter of Brahmanism, the fiction can hardly be distinguished from a fact.

The effect of these remarkable observations is to suggest a theory of the origin and growth of society among the higher races of mankind, which differs in some material respects from any hitherto propounded, though it is much more consistent with some of the current theories than with others. Sir Alfred Lyall follows Mr. Carlyle in saying that the perplexed jungle of primitive society springs from many roots; but the Hero is the taproot from which in a great degree all the rest are nourished and grown.' mighty man of valour, with his kinsmen and retainers, founds a clan. Through the very fact of success, this clan is saved from the first from the calamities which arise from an unequal balance of the sexesthe real secret, as I believe, of those unhappy usages

A

which have been saddled by recent theories upon all mankind. It becomes therefore a pure clan, having a genuine pedigree, in which certainty of paternal descent from the famous founder or founders is assumed from the outset. It may also be exogamous, either through the number of female captives which always formed part of its spoil, or simply because the practice of taking its wives from a distance, however this came about, increased its physical vigour and caused it to prevail in the struggle for existence. The formation of such a clan might be a fact by itself, and, so far as we have gone, it would be a plausible objection that the wholesale formation of such clans was highly improbable. But now we see how such a clan acts on the masses of men around it. It starts a process of ferment and crystallisation by which all tribes and assemblages in its neighbourhood or within its influence group themselves in circles as nearly as possible adjusted to the heroic model. The original communities of men may have taken all sorts of forms in the present state of these inquiries it is impossible not to suspect that no statements can be hazarded on the subject which are at once safe and very general. But evidence of many different kinds suggests that this 'miscellany' of primitive society was brought into shape by the influence of dominant types, acting on the faculty of imitation which must have always belonged to mankind. The communi

ties which were destined to civilisation seem to have experienced an attraction which drew them towards one exemplar, the pure clan, generally exogamous among the Aryans, generally endogamous among the Semites, but always believing in purity of paternal descent, and always looking back to some god or hero as the first of the race.

NOTES AND ILLUSTRATIONS.

NOTE A.

THE GENS.

THE passage in the text respecting the ancient groups more or less answering to the still extant House Community has been somewhat altered since it was first printed in the Nineteenth Century.'

It will be seen that, in the present state of these inquiries, I do not accept the account of the origin of the Gens given either by Mr. McLennan or by Mr. Morgan as universally true. I do not, for example,

venture to dissent from the view which the Romans themselves took of the history of this peculiar group as known to themselves. What this view was may be inferred from a passage in Varro ('De Linguâ Latinâ,' viii. 4) which has been often quoted. The grammarian observes that there is a certain' agnation and gentility' among words. All the cases of the nounÆmilius' are descended from the nominative, just as all the members of the Gens Emilia, all the Emilii, are descended from a single original Æmilius. The Romans, therefore, regarded gentility' as a kinship among men not essentially different from agnation.' The Agnati were a group of actual or adoptive descendants, through males, from a known and remembered ancestor; the Gentiles were a similar group of descendants from an ancestor long since forgotten. It is

6

true that some learned Romans seem to have perceived or thought that this Gentile relationship was to some extent fictitious; but, on the whole, they figured it to themselves as a form of kinship arising from descent, through males, from a common male ancestor. For reasons given in Chapter VII., I think that the Roman theory of the origin of the Roman Gens was at least probable. I see no reason to doubt that, though some of the gentes may have been fictitious, and others partially fictitious, there was a real core of agnatic consanguinity in most of them from the very first. The probable character of the fictions which clustered round this core may be gathered from the latter part of the preceding chapter. As Sir Alfred Lyall's description of the mode in which groups simulating true tribal groups are formed is now printed in his Asiatic Studies,' I might have omitted my abridgment of it; but I retain it, because nothing seems to me to have more affected primitive society, and yet to have been more neglected by those who have theorised on it, than the imitative faculty which man has always possessed and which Sir Alfred Lyall has witnessed in actual employment by barbarous men.

On superficial consideration, we are apt to think that man's mimetic faculty confines itself to matters of taste and personal habit. But, in truth, there is no successful, or conspicuous, or simply fashionable model which men, in the various stages of their progress, will not endeavour to imitate. The habit of political imitation, which has always been strong, still survives. Make us a King to judge us, like all the nations,' said the Israelites to Samuel (1 Sam. viii. 5). Give us a Constitution to regulate our liberty, like that of one particular nation,' is the corresponding modern and Western demand. If any

6