Sex is a matter in which modern society manifests an interest that is half-scientific and half-morbid; so that it is perhaps more than an accident that the psychology which has largely helped to stimulate this interest should be concerned with the pathological side of the mental life. Or, again, when we allow our minds to dwell on the subject, those of us who are least inclined to gloat over the indiscretions of the clinic may yet be indulging at second hand a not unpardonable taste for romance. The realities of the situation, on the other hand, are not to be grasped unless marriage be viewed in the light of its biological outcome, namely, the reproduction of the species. So far as human life implies survival, childlessness stands at but one remove from racial death.

Not only biologically, then, but even socially—since society has no significance, at any rate for history, except as a vital continuum—mating and educating are not distinct functions so much as parts of a single process. Indeed, were it possible to credit an indifferent Nature with purpose in any real sense, one might even say that mating was the subordinate element in the complex, standing as it does for a means, and not the only means, whereby the end of propagation is achieved under the myriad forms which life may assume. Thus we must be careful, in submitting it to moral valuation, not to divorce it from its context as an event in that organic linkage of lives which is actually a prior condition of morality, since the humanities themselves are contingent on human existence.

Such, then, being the ultimate bearing of sexual selection on the welfare of the species, as bound up with its perpetuation, it is rather remarkable how little the responsibilities of parenthood would appear to colour the anticipations of those about to marry. The happy pair—happy at all events on the principle that ignorance is bliss—have become partners in a gamble with a Nature who is given to concealing a card up her sleeve. Her game is to deal out pleasant illusions, and then to trump them with the family. Meanwhile the losers as a rule pay up good-humouredly enough, having a dim conviction that somehow it is all for the good of the house. It is with such a more or less instinctive blindness to consequences that the supporter of an organized eugenics contends with so little success. Prudence was never an attribute of Aphrodite, though even so her cult has always appeared more attractive than that of any Goddess of Wisdom.

Yet in this very recklessness which is inherent in sexual appetite there are the makings of something noble, because, in its association with the onward thrust of the evolutionary process, it is hardly distinguishable from courage and hope. The human race was fairly caught in the stream of life long before it could give thought to the direction of its going; while, even as regards its manner of keeping afloat, it cannot afford to be particular so long as the miracle happens. Such a metaphor indeed will be found to fit the historical facts well enough, since matrimony in any form—and mankind has been fertile in experiments—has proved a somewhat crazy craft; though one that nevertheless has hitherto safeguarded down the rapids its crew of adventurers, inured to discomforts, and exhilarated rather than cowed by the risks. Here, then, is constant crisis, such as is ever the opportunity of religion. Natural science, which stops short at the mechanization of life, may introduce sundry improvements in navigation; but the real struggle is between two elemental forces—the torrent and the will to ride it, an impetuous instinct and a moral resolution to make it serve as a means of spiritual advance.

In attempting, then, to frame a working hypothesis of the origins of marriage law, we may safely rule out conscious concern for the offspring from the list of primary motives. Indeed, though we are obviously dealing with an institution that is immensely old, there survive to this day traces of an age of innocence when the arrival of babies was explained in terms of storks and gooseberry bushes; for such precisely are the fancies which Sir James Frazer has grouped together under the name of ‘conceptual totemism’. Thus any love-affair claimed attention as an interesting event quite unconnected with the other that is apt to be so described. Moreover, it needed regulation on its own account, if society was not to be convulsed by the fury of the most individualistic of the passions.

Hence those theorists would seem to be wide of the mark who suppose the primeval legislator to have been alive to the physiological hazards of inbreeding. It has been suggested, for instance, that near marriages were to a noticeable extent more sterile than others—a matter about which science is not prepared even now to make a definite statement. How absurd, then, to attribute observations and inferences of so puzzling a kind to a time when in all probability childbirth was viewed as exactly on a par with the budding of trees, namely, as a more or less miraculous act of spontaneous generation. It is, perhaps hardly necessary to add that, if the rules of exogamy were founded on any such a train of reasoning, it would have been impossible to reach the conclusion actually authorized by world-wide custom that, whereas it is always a deadly sin for the children of two brothers or of two sisters to marry, it is permissible for brother's child to unite with sister's child, and may even be the proper thing to do. That primitive society, having thought out the principle of the alleged infertility of near relatives, should then proceed to make such a hash of its application is altogether beyond belief.

Meanwhile, other theorists, who are prepared to deny to the savage all foresight of consequences as concerns the numbers or quality of the offspring, go on quite arbitrarily to assume that therefore he must have tabood incest regardless of any consequences whatsoever. In other words, they fall back on instinct, that last refuge of the psychologically destitute. But all the evidence goes to show that Man has been trading on his intelligence for many thousands of years. Moreover, as for his matrimonial observances in particular, there are many signs that an animal condition corresponding to the predominance of instinct has long been left behind. Thus a far more regular waxing and waning of sexual interest must have once taken place in conformity with the periodicities of temperature and food-supply; and there can be little doubt that the development of culture has somehow disturbed this rhythm. The same thing has happened with our domesticated animals, which display perverted instincts that must ruin their chances of survival, did they not lean on our intelligence for protection.

Similarly, then, a like intelligence applied to our own attairs has enabled us to cope with what might fairly be called an artificial state of perennial amorosity, involving a perilous lowering of the threshold of erethism in the male, while for the female impregnation has been rendered more occasional, the risks of parturition being thereby probably increased. Causes at which we can only guess such as the appropriation of rock-shelters thanks to the use of fire—already known to Sinanthropus—or improved methods of obtaining a steady supply of food, might bring about a habit of herding liable to put a great strain on instincts previously adapted to a rather solitary kind of existence, analogous to that of the anthropoid apes. A conflict between old and new tendencies gives intelligence its chance to establish a compromise, which, so far as Nature has to give way, is felt as a discipline. Whatever the working arrangement that may be reached, instinct will be partly, though never wholly, against it. Thus, although it will always be easy enough to prove that exogamy follows the lead of spontaneous impulse up to a point, it would be quite unscientific to ignore the inhibitions which on all sides restrain and redirect the force of the current. Exogamy, and everything else that we do, is instinct with a plus; and for the student of religion, or of human culture in any of its forms, to account for that plus is the sum and substance of his task.

How, then, are we going to find a social reason for those widely prevailing restrictions on marriage which, from a physiological point of view, are so arbitrary as to be virtually ineffective? Perhaps the fundamental consideration in building up a society of moderate size that can hold together as an economic unit is to secure peace at any price. Now mating, on the other hand, is thoroughly individualistic in tendency. The only observable experiments in the direction of a sexual communism have always been of the nature of proprietary concessions of a limited kind, and indeed would never appear to have worked particularly well. Thus, whereas a number of males living side by side in daily contact with their several wives would, at a mental stage when appetite is dominant, be sure to quarrel, an association of men and women sexually taboo to each other would command a far greater degree of social cohesion. The latter condition, however, could not arise, or at any rate could not maintain itself for long, unless the women bore children in sufficient quantity; and a prevailing theory of parthenogenesis would not alter the fact that actually they must have had access to spouses offering themselves from outside. Since any expansion of numbers implies a favourable environment, there would presumably be other groups, offshoots in most cases of the same stock, whose wandering males would be ready to play the part of tolerated, if furtive, lovers; the more sedentary females in the meantime forming the nucleus of each separate camp.

On such a principle a matrilocal kin might grow in strength up to the point at which matrimonial give-and-take had established such friendly relations with their neighbours that two or more groups—a dual organization being the simplest and therefore possibly the earliest type of combination—could once for all amalgamate. Henceforth, locality having ceased to count, the husband would gradually oust the mother's brother as protector of those children which he now has a better chance of claiming as his own, even if causa uxoris; though it will be a long time before unilateral claims of kinship, whether matrilineal or patrilineal, cease to be all-important. Yet only then can the bilateral or individualized family, consisting of the married partners and their offspring, become economically and socially free of all collective status, short of membership of the community at large.

Now this reconstruction of a whole chapter of exceedingly crepuscular history must be taken for whatever it may be worth; its immediate purpose being merely to provide a background for a singular fact concerning primitive religion, namely, that its attitude towards marriage is almost entirely negative. It puts all its strength into the commination: Those whom God has put asunder, let no man join. Incest is the abominable sin, even when the conception is stretched to cover relationships which are so slight as to have no eugenical significance at all; being in fact nominal, that is, due to an extended and non-physiological use of names such as mother or sister for classificatory purposes within the kin. This is not the moment at which to investigate the origin of the religious sanction which bans with such utter detestation the commingling of blood in what is actually or putatively one line of descent, without regard to that complementary line of descent which from a biological point of view is of a strictly equivalent importance. On the very plausible assumption, however, of the priority of mother-right, we can ascribe to the mother's blood as such, that seemingly self-sufficing source of human life, an inherent sanctity compelling a respectful avoidance under penalty of a conditional curse.

Thereupon congress with a female of the blood would come to be placed in the same category as conflict with a male of the blood, both alike being desecrations of the principle of life which might pervert and annul its wonderworking power. Whereas, then, there is little likelihood that any a posteriori reason existed for believing that inbreeding makes for sterility—if indeed it be true in fact at all—there might well have been an a priori conviction to much the same effect based simply on a fear that to tamper with the fountain-head of human vitality would somehow cause it to dry up. It would be quite a mistake, however, to regard this as a dim groping after some law of physical causation. Rather the argument moves from first to last on the moral plane, and the blood is thought of as resenting its own violation with disgust and indignation. The mystic substance cries unto God for vengeance when there is a God other than it and above it; and, when there is none, protests none the less with an autonomous power to make its wrath felt by those who have dared to sully and impugn its mana.

Meanwhile, the obvious ministers of its will are the mothers of the people, the sole vehicles of the common life of the kin. Theirs it is, therefore, to guard the mystery, to emphasize the horror of the curse. To these dread high-priestesses of the blood, then, I venture to attribute the original formulation of the law of exogamy, with its essentially religious basis in the most binding of taboos. But exogamy is only marriage law in the sense that it explains how not to marry. On illegitimate love is pronounced a terrible imprecation of which the echoes haunt us still; so that no science would dare to suggest as applicable to human beings those possibilities of selective inbreeding that have been tried with no little success in the case of certain domesticated animals. But what of legitimate love? How is religion going to deal with a natural function which, if cut off in one direction, is bound to find an outlet in another. Here is something which, if it cannot be made to yield good, will have to be tolerated as a necessary evil. Sex stands for the continuity of life, and celibacy is but a dead end. It is clear, then, that religion, whatever its antipathy to the lusts of the flesh, must achieve some compromise with Nature; unless it is prepared to offer to poor humanity the impossible alternative of either being born in sin or ceasing to be born altogether.

Yet, despite the inexpugnable validity of the union of the sexes as a law incumbent on the natural man, a religious sanction in respect to its earlier forms is on the whole conspicuous by its absence. Thus Dr. Lowie, the author of an excellent summary of the most authentic data relating to the first beginnings of social organization, can declare: ‘Generalizations about primitive tribes are dangerous, but few exceptions will be found to the statement that matrimony with them is not so much a sacrament as a civil institution.’1 Since instinct is in this case definitely enlisted on the other side, there must be some historical reason for the fact that religion, confronted with the elemental call of sex, finds it so much easier to curse than to bless. Now, if we start from the hypothesis of the matrilocal kin, or homogalactic group, we can see that the status of the woman will chiefly depend on the discharge of two functions. The first, which might be called incipiently religious, since it is surrounded with mystery, is concerned with her duties as a mother. The second, which by contrast might be termed civil, relates to her usefulness in providing such food as is obtained by gathering as distinguished from hunting—probably the more dependable, if less dainty, part of the daily bill of fare. Apart from this double service to her group she indulges in flirtations with strangers that are her private affair and of comparatively little social interest, inasmuch as husbands remain something of a luxury, so long as their claim to rank likewise as fathers is still in abeyance. At most, it would slowly dawn upon the neighbouring groups that mutual courting made for peace and, eventually, for co-operation. The earliest form of commerce may well have been of the sexual kind.

Thus there could be no mistaking of intentions when, as occasionally happens among unsophisticated folk, even at the present day, the visiting party consists of young women, looking, though it may be hardly wearing, their best; for be it noted that, however individualistic sexual passion may be, a collective inspiration, perhaps ultimately due to seasonal influences, may be needed to lead up to what may be described as a plural pairing. Given, then, such an incursion of philanderers of the gentler sex, the resulting encounters, even if a certain amount of horseplay is mingled with the dalliance, cannot possibly be confused with the brawls likely to attend any meeting between the males of the rival camps. If on the other hand as also occurs, the young men take the initiative in paying a call, the women go forth as official peacemakers to discover whether they offer love or bring war.

Thus in addition to and apart from the dual capacity in which she serves her group as mother and as food-gatherer, woman gradually comes to fulfil a third function as wife in the restricted sense of a go-between, a link with the opposition. She is the predestined means whereby the stranger is to be converted from an adversary into an ally; but it is by no means easy for her to subordinate her other tasks to this single one, and to adjust her feelings accordingly. Let full allowance be made for the fact that sexual selection will tend to create enduring partnerships, so far as male rivalry does not interfere to break them up. Even so, this individual attachment will have to contend with a group-loyalty belonging in the first instance to the motherhood, with its associate brotherhood; for herein the woman was reared in a complete solidarity of social sentiment, until puberty threatened to disturb the harmony. Should she and her chosen mate continue to occupy separate environments, there can be little more than the satisfaction of desire between them. But neither can permanently abide with the other without raising between the two groups concerned the whole question of ownership as regards alike offspring and economic services.

Of the two the services can be divided more easily than the offspring, though it is not unknown to primitive law that children born before full compensation is paid are retained by the wife's people, whereas the husband's group is entitled to the aftermath. For the moment, however, it is enough to note how considerable are the difficulties that arise over the matter of the transference of kin-allegiance, if husband and wife are to become one in spirit. No wonder that religion finds it hard to sanction marriage at all when it amounts to an attempt to fuse two discordant faiths into one. A union of twin souls is one thing, and a treaty between rival communities is quite another. The way of the world has been to discover an instrument of social consolidation in the marriage of convenience rather than in the marriage of affection; and it may be that religion is rightly suspicious of a business of which the political is more obvious than the moral advantage.

Politically, then, even if there are incidental drawbacks from a moral point of view, a decisive point in the evolution of society was reached when, drawn together by the attractions of each other's womenfolk, two hitherto independent kins were led to make common cause as a single cohabiting and co-operating group, a duality in unity. It must be admitted that this step has already been taken by virtually all those savages who offer themselves to direct observation as practising exogamy within recent times. Quite exceptionally are there rumours, as from New Britain, of matrilineal groups who find no inconsistency in being at once on marrying and on sparring terms. On the other hand, even when amalgamation has gone so far, as in Australia, that the two intermarrying divisions composing each local group live in harmony, though continuing to camp a little apart, Spencer and Gillen note that, when one local group gets to loggerheads with another, what starts as a quarrel between group and group is apt to develop into one in which the men of each phratry take sides together, regardless of the difference between friend and stranger.

Thus there is at least a certain amount of evidence pointing back to a fusion, due to intermarriage, as the historical explanation of the dual organization. On the other hand, the opposite theory that it came about by way of a fission, deliberately engineered in order to avoid the evils of inbreeding, has neither facts nor probabilities in its favour; while it cannot even be shown that the system attains its alleged object. Meanwhile every detail of its working arrangements fits in neatly with the theory that the composite or heterophyle group is a permanent association or alliance having a ius conubii as its ultimate foundation. If the present argument is at all sound, however, we are not to suppose with Tylor that exogamy was instituted because wise statesmen perceived the utility of doubling their strength by making it obligatory to marry across. On the contrary, exogamy acted simply as a vis a tergo. It warned them to avoid a curse, but did not indicate how they might find a blessing. No counsel of perfection, but rather a counsel of despair, drove them to seek abroad what was utterly denied them at home. Thus the statesmanship came as an afterthought. Contacts that were but private, casual, and furtive must be transformed into communications that should henceforth be public, stable, and above board.

From first to last, however, during this process of developing relations of friendliness, negotiations would inevitably turn on the marriage question. For it is plain that henceforth, with the duplication of the social unit, the restrictions on promiscuous intercourse making for jealousy and disunion will have to be supplemented. In addition to the old homophylic taboos forbidding pollution of the blood-tie, there must be established heterophylic avoidances, in order to keep apart certain classes of men and women who, though of different blood, are nevertheless held unfit to mate together. The typical rule of this kind is that marriage goes by age-grade, the males of different generations consenting to a policy of ‘Hands off’ as regards one another's coevals among the marriageable women. Such a regulation would easily crystallize out of the custom of forming wooing-parties, connexions thus formed shortly after puberty being likely to become habitual and lasting. In this way father and daughter, or mother-in-law and son-in-law, would, however physically attracted, be aware of a certain social impropriety in overstepping the age-limit, even at the stage when the woman belonged to one horde and her paramour to another.

At the later stage of group-coalescence, when the opportunities of meeting would be of daily occurrence—except in so far as the sexes as such were inclined to keep a good deal to themselves, as is usual in primitive society—a strict etiquette would come into being, obliging potential spouses of the wrong age to keep at a respectful distance from one another. But here religion had no symbol which, like blood in the case of the blood-tie, could convey the suggestion of an elemental abhorrence. It is doubtful whether, in their original form, avoidances as between persons of different blood are to be reckoned as taboos at all. To ignore them would be an outrage on the proprieties, a scandal sure to end in a rumpus; but it would not draw down a curse, would not poison the tree of life at its very roots. So far as totemism can be correlated with a religion of the blood—and perhaps it was never more than its by-product, degenerating into sheer nominalism when readapted to the patrilineal way of reckoning descent—we can see how to comprise avoidances and the taboos against incest within one category would involve a bi-totemic sanction, a thing at which the primitive imagination boggles. As a mystery, the totem can have little or no appeal beyond the circle of its own totemites, which never includes both wife's and husband's relations. Thus the consolidated bilateral society needs a new religion of its own—one that must somehow subordinate totemic pluralism to a system of common sanctities. A composite society must likewise compound its ritual observances.

This process at various stages of its development can be studied to best advantage in Australia, whence we get several trustworthy accounts of those initiation ceremonies which come nearest to filling the part of a tribal religion, and have even, as in the south-east of the continent, brought about the installation of a tribal god. Their function as a means of educating youth in its social duties must be reserved for future discussion. For the moment it is important to fix attention on the fact that in their primary intention they serve as a preparation for marriage. This becomes quite clear if we realize that each phratry makes it its business to put the young men of the other phratry through their paces. Male candidates for matrimony have to satisfy a body of elder brothers of the other side of the alliance, who are deputizing for their group-sisters about to come out. The girls likewise undergo puberty rites, though of these very little is known. We may suspect, however, that they are normally the affair of their own group-mothers rather than that of the women of the other division; though it happens among the Arunta that the men preside over the ceremony of female excision.

Now it is very significant that a tribal meeting, prolonged it may be for months, and serving to promote amicable feelings and dealings in all sorts of ways, should hinge on the regulation of the future affairs of adolescents by those of riper years. The actual focus of social convergence is undoubtedly the marriage-market, and it looks altogether as if the initiation ceremony was evolved out of a hymeneal festival that became less spontaneous and more controlled in proportion as the political advantages of the alliance resulting from it became manifest to the elder statesmen. In a certain sense, then, the initiation ceremony might be regarded as equivalent to the marriagerite itself; since on its mystic no less than its practical side it is an introduction to the married state. Initiation is a confirmation of maturity. Thus at the same time it is a solemn certification of marriageability according to age-grade. It consecrates marriage, not indeed between individuals, but between whole groups of young persons now duly attested as equal to the rights and duties of maintaining the league between the phratries by legitimate sex-communion.

This, then, may well have been the original object of the initiation rites; since it is hard to say at what point it became clear to those concerned that race-continuance actually depended on providing the mothers of the tribe with suitable mates. Ignorance of paternity on the part of mankind, however, could not prevent Nature from arranging hymeneal festivals after her own irresponsible fashion. Moreover, Man would doubtless step in to organize such proceedings, as soon as it was discovered that an armistice and a season of mutual love-making implied each other. Thereupon, as a sort of proclamation of the collective banns, an inspection by each party of the other's bachelors would take place, before these could as the batch of that academic year be licensed to take the further degree of Magister Amoris. By one external sign or another, by circumcision, for instance, or by the knocking out of a tooth, their new status would henceforth be plain to all; and they were now made free to seek a mate of the opposite phratry and same age-grade with complete social approval, the rest being their private affair.

It would be quite erroneous, however, to suppose that, because the social and religious sanction takes this collective form, marriage at the same time cannot remain in essence an inter-individual relation. There is every reason to believe that the pairing instinct is well developed in the human species. Thus Savigny's aphorism that private property rests on adverse possession ripened by prescription might well apply to that earliest of tenures, the exclusive right to a given woman's favours; even if her services as a mother and as a worker at first belonged to the other social division. Possession, however, was always adverse in the sense that the marriage-relation could not but be potentially triangular as between the mate in being and a possible rival of the same matrimonial group. Thus Howitt reports of certain Australians: ‘If two blacks quarrel over a woman, the tribe does not interfere if they are of the same class, and the stronger of the two keeps the woman.’2 Moreover, as the same authority proves by many examples, native Australia is torn by a perpetual battle between the marriage of convenience and the marriage of love. The relations promise a girl to a man who can offer a sister in exchange, but the lady in the meanwhile bolts into the bush with her heart's desire. On the return from the honeymoon, however, the fortunate lover has to meet the disappointed one in a stand-up fight, and may likewise have to undergo expiation by dodging the spears of the kinsmen whose deal he has spoiled.3 To judge by the popularity of such elopements, the principle that ‘none but the brave deserve the fair’ still holds its own as against a nascent system of bride-purchase; for the tribe as such is not interested in such private barter so long as the proper groups intermarry, and so maintain the communion between its complementary sections.

Moreover, we may expect the religious interest to be almost entirely bound up with the collective at this stage of society; the individual note in religion, apart from the ecstatic experiences of the medicine-man, being as yet hardly sounded. The initiation ceremony, however, is thoroughly tribal in its whole outlook, bidding the consanguine groups to combine in a new and higher kind of unity that might perhaps be termed inter-sanguine, since congress between man and woman of different blood is its seal and symbol. Corresponding, then, to this political synthesis achieved through the medium of marriage there comes into existence what is sufficiently surprising at this level of culture, namely, a supernatural being of a markedly anthropomorphic stamp including ethical attributes and personal individuality—in short, a god. He is as it were the embodiment of tribal headmanship, with all the wonder-working powers claimed by the human leaders of society raised to the nth degree. His chief—one might almost say his sole—function is to preside over the initiation mysteries, that is to say, to promote tribalism in its essential character of an intermarrying alliance. How precisely he may have evolved is not very clear, and even in Howitt's south-eastern region of Australia, from which most of the evidence comes,4 we may suspect a plurality of origins—an apotheosis of the bull-roarer in several cases, and in another the promotion of a phratric totem to tribal, that is, super-phratric, rank. Functionally regarded, however, there can be no doubt that he stands for the moral achievement needed to consecrate a consortium by way of conubium. Children are bonds, declares Aristotle in his Ethics; but, long before humanity could accord a joint interest in their offspring to both parents, or had even grasped the dual nature of parenthood at all, it had become manifest that women are bonds—that, apart from its biological utility as an enticement making for race continuance, the tender passion has for its immediate psychological issue a plighted troth, an engagement of mutual fidelity, which may become the spiritual prototype and emblem of a universal pledge to love one another.

It remains to consider the institution of individual wedlock, the recognition of exclusive conjugal rights as between two, or sometimes more, persons, so far as it comes within the purview of primitive religion. As has been already noted, this aspect of marriage is, ceremonially at all events, less prominent than with us, who have come to identify the sacrament more or less with the leading of the bride to the altar. No need is felt to-day for a previous confirmation service to establish the moral, let alone the physical, fitness of the candidates for matrimony. The one preoccupation of modern ritual would seem to be to anticipate an irregular connexion by invoking a blessing on a union between two persons who in point of health, character, or age may be quite unsuitable for any kind of conjugal task. On the other hand, as soon as the knot is tied, its sacramental nature is held by many persons of strict principles to render it indissoluble, however morally disastrous may be the actual consequences of the match. That the bond, however, has no more than an earthly reference is suggested by the fact that, at any rate within the limits of Western civilization, there is but little prejudice against the remarriage of widows.

Now this insistence on the private contract, as if it embodied the whole social significance of the act of mating, offers a violent contrast to the savage conception of a public treaty satisfied by a ceremony equivalent to a general issue of licences to those whose class entitles them to pair without offence. The private affairs of the various parties and of the relatives exercising individual authority over them are thus bound to count religiously, because tribally, as of subordinate importance. Among the Kurnai, for instance, a young couple anxious to elope might hire a medicine-man to throw those likely to interfere with their plans into a magic sleep;5 but such an exploitation of the occult comes very near, if it does not actually fall below, the line that separates a respectable supernaturalism from the black art. Indeed, Dr. Lowie goes so far as to say that ‘even in the higher cultures the individual family is a conspicuously unstable unit’; whereas ‘among primitive peoples, who rarely if ever interpose religious scruples against divorce, all sorts of disruptive forces must be reckoned with’.6 Institutions such as the levirate and the sororate, involving the substitution of the husband's brother and the wife's sister respectively, show how social recognition relates rather to the validity of the pairing arrangement as such than to the rights of the individuals who for the time being participate in it.

Very hard to explain are those bride-sharing privileges enjoyed by the men of his own matrimonial class who help the Australian bridegroom to consummate the wedding by what amounts to a sort of mock capture—a ceremony of which Howitt has collected a number of instances. His own view is that it represents an individual expiation for what was formerly a genuinely effective marriage of group with group.7 But, apart from other serious objections to such a theory, it is to be noticed that the sharing is entirely on the part of the male sex, no bridesmaids so to speak appearing on the other side as a counterpoise to the intervention of the husband's ‘best men’ or, as we might perhaps prefer to call them, seconds. For what actually occurs is a sort of duel between the sexes, fighting as it were under rival flags; for the sex-totems of such tribes as the Kurnai would seem to have it as their exclusive function to inspire amorous encounters, with the result that individual attachments are precipitated out of the general mêlée.8 It is to be noted that, even when the girl's parents have already agreed to the match, so that the young man is simply coming to fetch his duly affianced bride, the latter is nevertheless expected by custom to bite and scratch—indeed, she is apt to boast afterwards about the strength of her resistance with much legitimate pride9—while her lady friends on her behalf bring down their digging sticks on male pates with a gusto proportionate to the very real damage.

Whatever, then, be the ultimate origin of these cheerful proceedings, we can probably descry in them a typical instance of van Gennep's rites de passage, marking in forcible fashion a change of status from the condition of free and unappropriated female. If marriage was originally matrilocal, the point of the new status might well be that the bride was torn from her own kinsfolk. Henceforth she must belong to her husband's side of the alliance, at any rate as an economic associate, this being the regular system all over the continent; whichever way the descent of her children might be counted, a matter on which Australian practice varies by about half and half. As for the ceremonial access, is it more than a rite of sex-communion ratifying the bargain between the phratries? For it is all the more necessary to emphasize this, now that male influence has become paramount enough to compel the woman rather than the man to abandon her natural surroundings, and to make the best of it in an alien home.

Further, we may conjecture that it is not until the patrilocal type of marriage has come into vogue that woman's holiness, always an ambivalent attribute, comes to be conceived rather as an uncleanness which in the main is of ill omen to society. ‘Verily ye are mostly of Hell’, is a traditional saying of Mahomet concerning a sex at which Islam has always looked askance. It would be easy to pile up evidence from the savage world to the same effect concerning the general unluckiness of contact with Woman, and more especially with her blood. Yet one is left with the impression, no doubt needing confirmation by a detailed induction none too easy to carry out, that there is less of this feeling where extreme mother-right prevails; whereas, as patriliny develops, the woman tends to lose caste in every way. At home within the circle of her own kin, though untouchable under the severest of taboos, she was nevertheless revered as the very fountain-head of the social life; being the mother pure and simple, and wife in no significant sense at all, since the views of her furtive mate simply did not count.

Let her, however, be transferred to the camp of another stock, and, whatever be the convenience for her paramour, the rest are brought into unwilling touch with a stranger; whose very strangeness is likely to be catching, quite apart from the damning fact that her sex in itself renders her uncanny. Thus she is at a double disadvantage in her relation with the males, including her own husband, who must carefully avoid her before engaging with his companions in any joint undertaking such as a hunt or a foray. On the other hand, though the females will belong to her phratry, they will belong to different kins, and, having once suffered redistribution as wives, will have lost all that close touch which united them when growing up at home together. Finally, when marriage is not only patrilocal, but involves the patrilineal reckoning of descent as well, her very status as a mother is prejudiced. Even Australia is sophisticated enough to have, sporadically, evolved a theory that the man is sole parent, and but gives the child to the woman to nurse—the same doctrine that turns up in the Eumenides, be it noted, as a justification of matricide. Truly, from the point of view of the woman who leaves her group, to be given falls a long way short of being received, even though her sacrifice be the price of social union.

In these circumstances, then, it is no wonder if the marriage rites of relatively advanced communities prove decidedly piacular in type, as if their main function were to avert disaster rather than to secure positive happiness. Crawley has with great ingenuity elaborated a theory of what he calls the sexual taboo, which in his opinion such ceremonies are designed to neutralize in the interests of both parties alike.10 He thinks that the young man and the young woman are equally shy of one another at the start; whereupon society kindly stages an introduction intended to accustom them to one another's company. Now there may be something in this hypothesis; but, by putting the individual relation so much into the foreground, Crawley hardly does justice to the social nature of the occasion, and to its essential function as an instrument of group-consolidation—just as in Corsica to-day a marriage is a wonted method of patching up a vendetta. Moreover, his evidence is rather thin for the male aspect of the alleged sexual taboo, whereas the weight of testimony as regards the supposed danger from the woman's side is quite overwhelming. It may be suggested, then, that, as at once a woman and a stranger from without, the bride stands in need of special purification. At the same time her social function as the mediating gift and blood-offering in a peace-making act lends further emphasis to her sacred character. On the other hand, it is quite true that, in their individual capacities, bride and bridegroom alike are facing a critical moment in their lives, and need to have their private apprehensions dissipated, and their aspirations adjusted to the new situation. But I suspect that this ceremonial motive is secondary and late, on the general principle that primitive religion reflects the collective interest rather than that of persons.

It must be confessed, however, that a detailed examination of wedding rites, whether regional, as in Dr. Westermarck's Marriage Ceremonies in Morocco, or world-wide, as in Mr. Fallaize's recent collection of curious lore entitled Marriage, presents one with a rich confusion of hopes and fears aroused by this crowning instance of what M. van Gennep calls ‘a social perturbation’.11 Perhaps the most yielded by analysis is that the fears predominate in the complex. Fortunately it is part of the strength as well as of the weakness of all symbolism that it can be interpreted in different ways, and may even provide a common form of expression for more than one original meaning. Thus when the bride is bathed and adorned in new raiment it may be to rid her of mystic evils; but it likewise confers on her the positive blessing of a clean start in her new career. Her veil, again,—does it prevent her from radiating untoward influences, or protect her from these at a moment of natural trepidation? The probable answer is, Either, and sometimes Both. Once more, in Morocco bride and bridegroom alike are pelted with much spirit, but possibly for separate reasons. In the bride's case it may be to drive away the djinn liable to infest her, whether this be done for her own sake or for that of those who have dealings with her; whereas the bridegroom's pelting may be associated with a rite of capture. Meanwhile, the all-important central motive of all this symbolism would seem to be to proclaim and solemnize, whether by ring, knot, eating together, or otherwise, that union implicit in the sexual bond itself, that mysterious fusion of the complementary halves of human nature, out of which the forms of political organization so directly arise.

Thus far, then, does the sacrament of marriage succeed in deepening the import of a vital function; but scarcely further, so as likewise to anticipate the no less sacred obligations of parenthood. It might almost appear as if the human mind was but vaguely aware to this day of the physiology of generation. Though it duly shudders at incest without study of its particular effects, religion is prepared to celebrate the union of mental defectives; or to join together young and old, quite in the style of the Dieri or of the Pentecost Islanders, among whom, though marriage with the next generation is not permitted, a girl may marry the coeval of her grandfather. It may be, then, that a survey of the historical facts will bring out and explain certain anomalies in the modern outlook of religion on the sexual life. Yet be it never forgotten that what science knows about such matters even now amounts to very little, and that religion is practical, having to act while thought hesitates, and trusting less to the head than to the heart—a policy which those who marry can likewise on the whole be recommended to follow.