The rite at Byblus must, however, be distinguished from those we are considering. They were performed by every woman without alternative, but they were performed only once. If they were an expiation for marriage we should expect to find them described as part of the marriage rites. The Balearic islanders, the Nasamonians and the Auziles in antiquity had, as well as many modern savages, such rites, whether or not they can be properly explained as an expiation for marriage. But at the most the rites with which we are now concerned were a preliminary to marriage—a necessary preliminary, perhaps, but one that might have been accomplished at any period before it. Indeed, so far as appears from Herodotus, the victim, if we may call her so, of the Babylonian rite was not necessarily unmarried. But comparison of the accounts of the practice at Heliopolis, in Lydia and in Cyprus renders it fairly certain that it was only unmarried women who were subjected to it, and that it was essentially a sacrifice of maidenhood. A passing reference by Eusebius has been interpreted to imply that at Heliopolis both married women and girls were prostituted in the service of the goddess.[271.1] But Eusebius says nothing about the goddess. His reference must be construed in the light of Socrates’ statement that women were by the law of the country required to be common, and hence the offspring was doubtful, for there was no distinction between fathers and children.[271.2] Whatever else those phrases may mean, they entirely negative the theory of expiation for marriage. But they do not refer to the custom of prostituting virgins to strangers, which the historian expressly distinguishes.

It may be objected to this reading of Herodotus that while he uses the generic term women (γυναῖκες) in speaking of the victims, on the other hand, in a previous chapter referring to the Babylonian marriage customs, he reports that once a year in every village the marriageable maidens (παρθένοι) were all put up to auction, the respective purchasers being required to give security that they would marry them; and it was unlawful to give them in marriage in any other way. The objection is of little weight. It is needless to consider whether we are to understand the specific term παρθένοι literally. Even if so, there would doubtless be ample time for the performance of the rite at the temple of Mylitta between the auction and the marriage. It does not appear that marriage followed the auction immediately. Had that been contemplated, security would hardly have been necessary. When the anniversary came round all the maidens who had during the preceding year attained puberty and thus become ripe for marriage (γάμων ὡραῖαι) were probably put up. Those who had not previously undergone the rite would, if my interpretation be correct, be required to submit to it before marriage.

It is superfluous to discuss other and obvious objections to the theory of expiation for marriage. But the appearance of prostitution which the rite presents demands further consideration. At Babylon, although a piece of money passed, the payment seems to have been merely pro forma. It mattered not how small the coin was, it could not be refused. Whatever it was, Strabo tells us it was considered as consecrated to the goddess. Lastly, the rite once performed, no gift, were it ever so great, would be accepted to repeat it. The details of the rite at Heliopolis and among the Lydians have not been preserved to us; but we may with probability infer that they were similar. In Lydia, indeed, if we are to trust both Ælian and Herodotus, two distinct customs are traceable, namely, the sacrifice of virginity and the life of prostitution to earn a dowry. A Greek inscription of the second century A.D., found at Tralles and referred to by Dr Frazer, discloses also the existence of religious prostitution by girls expressly chosen by the god and set apart for that end.[273.1] This is a similar custom to that of the Armenian girls already mentioned, and is not to be confounded with the prostitution mentioned by the Father of History as practised by all the daughters of the common people. Whatever may have been the origin of the latter, the other two in the time of Ælian were connected with religion. On the island of Cyprus we seem to find much the same state of things. If we may believe Justin, the maidens earned their dowry by prostitution. From other sources we learn that there were mysteries of the Cypriote Aphrodite, which were said to have been instituted by Cinyras, king of Paphos and father of Adonis. Into these mysteries there was a regular initiation. Sexual matters no doubt formed their staple teaching; and what classical and especially apologetic writers would call prostitution would be practised. The legend ran that the daughters of Cinyras, through the wrath of Aphrodite, united themselves with strangers.[273.2] Probably it was believed to be in imitation of them that the maidens of Cyprus sought prostitution on the seashore. In any case the story indicates, as Dr Frazer has pointed out, “that the princesses of Paphos had to conform to the custom as well as the women of humble birth.” But if this be so, the object of the harlotry alleged by Justin falls to the ground, since it would be unnecessary for princesses to earn their dowry. It may be suspected, therefore, that Justin or his authority has confounded two disparate customs, that of earning the dowry by prostitution, and that of a religious sacrifice of virginity in connection with the mysteries of Aphrodite, in which the other party to the rite was a stranger. Only thus can we satisfactorily explain the limitation of the practice to stated days, probably festivals of Aphrodite, and the phrase about paying the offerings to her for future chastity.

The money payment, whether large or small, was in the Byblian rite, as in the Babylonian and (if I interpret correctly) in the Cypriote rites, consecrated to the goddess. We may infer that the same was the case wherever else the rite was performed. At Byblus it was the alternative to the consecration of the woman’s hair. Prostitution—that is, sexual intercourse for hire—is not a primitive practice. The appearance of prostitution in connection with religion may be accounted for by the influence upon the religious practice of the general practice of harlots. Analogy would suggest that intercourse other than conjugal or the satisfaction of the genuine passion of love demanded a monetary consideration. But when that intercourse was the performance of a religious duty the money was not kept as gain by the woman. It was not earned for herself, but devoted to the goddess. Where bands of “harlots” were attached to a temple their earnings probably went to swell the temple funds out of which they were supported.[274.1] It may accordingly be suggested that the hire was not an essential part of the rite, but merely an aftergrowth in the process of adapting an older custom to the changing manners and religious ideas of a growing civilization.[274.2]

Assuming, therefore, that the rite was a sacrifice of virginity to which every woman was subjected, it would probably be performed either on the attainment of puberty or as a preliminary to the marriage ceremonies. But we gather from the historian’s account of the sale of the village maidens around Babylon that the auction followed almost immediately after the attainment of puberty, or within (say) a year at the furthest. The practice of most ancient nations, as of nearly all barbarous and savage peoples, and indeed of many in a high stage of civilization, would lead us to expect that marriage would be entered into within a very short time of the bride’s puberty. Sometimes marriage even precedes puberty. Where, as more usual, it follows that epoch of life, the rites incident to puberty must first be completed. Among such rites defloration is not infrequently found. In this respect the Australian tribes are notorious. In the Boulia district of Northern Queensland the girl is compelled to intercourse with a number of men.[275.1] Among the Dieri of South Australia a ceremony called Wilpadrina is performed on the young women when they come to maturity, in which the elder men claim and exercise a right to them, and that in the presence of the other women.[275.2] The Arunta and Ilpirra tribes in the centre of the continent perform a ceremony on every girl when she arrives at a marriageable age, but before she has been taken over by the husband to whom she has been allotted. As part of that ceremony a number of men have access to her in ritual order; and the intercourse is often repeated the following day.[276.1] Analogous proceedings are known in other parts of the world. The central tribes of New Ireland have a women’s house in every village. When a girl attains puberty she withdraws into a small house, called mbak, built inside it. There it is said she has to remain for ten months, only going outside at night. During this period she is waited upon by the old women, and through their intervention every man who chooses has access to her. On leaving the mbak she belongs only to the husband to whom she has probably been betrothed since infancy.[276.2] In the west of the island of Serang between Celebes and New Guinea, a girl after ceremonial bathing goes round clothed with a sarong woven of the fibre of the Pandanus repens, at the service of every man until her family have collected the necessary materials for a feast. In certain districts, however, before actual puberty the teeth are filed. When this operation is completed, a feast is prepared of which the novice must taste everything. Further, an earthen pot filled with spring-water is covered with a fresh pisang-leaf. One of the old women then taking the index-finger of the girl’s right hand thrusts it through the leaf as “a symbol of the rupture of the hymen, or to show that the possession of virginity means nothing for her.” The leaf is then displayed on the ridge of the roof. This done, the women fall to eating and drinking. When they have finished they begin singing to the accompaniment of drums. The men are then admitted to the house. In some villages the old men have free access that evening to the room of the girl in whose honour the feast is given, while the other guests amuse themselves with singing outside. After this celebration the girl is entitled to free intercourse with men, even before puberty.[277.1] In East-Central Africa the Azimba maiden is artificially deflowered during a period of retirement and instruction in the forest. When the retirement is over she celebrates her attainment of puberty by a dance in which only women take part. That night a man, hired by her father for the purpose, sleeps with her, and once this is done she is supposed to have no further intercourse with him. Often, however, she is already married before puberty, and consequently no longer a maiden. None the less is she taken from her husband that the puberty customs may be performed. When she is brought back he himself sleeps with her apparently as a ritual act, without the necessity of hiring a man for the purpose.[277.2] Among the Wanyasa, or Mang’anja, at the southern end of Lake Nyasa, ceremonies are performed similar to those of the Intonjane (girls’ puberty ceremonies) of South Africa, and every girl on her return after the initiation must find some man “to be with her,” otherwise she will die.[277.3] The Intonjane among the Kaffirs is well known to be an occasion of sexual indulgence. It may be surmised that the ceremonies of the Suahili on the east coast were originally similar to those just mentioned. But the Suahili have become partially Arabized, though their Mohammedanism is little more than a veneer over their heathen customs and belief. Among them now a girl returns from her seclusion in silence and gives her hand to every man she meets, receiving from him in return a few small coins.[277.4] It is said that the girls of the Wamegi, also a tribe near the coast, are artificially deflowered at puberty by certain old women.[278.1] Artificial defloration at puberty is also practised by the Sawu Islanders. The Sakalava girls in Madagascar perform it on themselves in case their parents have not previously taken the trouble.[278.2] Other examples could be cited, but the subject need not be pursued.

I would venture to suggest then that the Babylonian rite was a puberty rite, and that a maiden was not admitted to the status and privileges of adult life until she had thus been ceremonially deflowered. Among those privileges, and the chief of them, was the gratification of the sexual instinct. It was, therefore, a prerequisite to marriage. Ceremonial defloration of the bride by others than her husband has prevailed in many places. When marriage follows closely after puberty it is difficult to determine whether the custom really belongs to the puberty rites, or to those of marriage. I am not concerned here to deny that among many peoples who practise it as part of the marriage rites it may have been such ab initio. The determination of this question would involve an examination of marriage customs extending far beyond the space at my disposal. But it will be admitted that as puberty rites gradually became simplified or altogether obsolete such a custom could only maintain existence as part of the marriage rites. It is then usually (but, as we shall see, not always) performed by one or more of the bridegroom’s friends or by an appointed official, and ultimately degenerates into the jus primæ noctis vested in some powerful personage, as a lord or priest. Nothing of the sort appears in the accounts which have come down to us of the ancient rite in Western Asia. In all of them (save among the Lydians) emphasis is laid on the performance by a stranger. At Babylon our information does not connect the rite with marriage at all. Elsewhere it is referred to not as part of the marriage rites, but as a preliminary to marriage.

That such a rite should be found annexed to the temple and worship of a luxurious goddess causes no surprise; on the contrary, it is what might have been anticipated. Every reader will call to mind numerous examples of archaic rites which have become attached to Christian festivals, and of Christian shrines which are simply shrines of an earlier religion adapted and consecrated afresh under Christian names. The difficulty of uprooting old customs, and their consequent incorporation and adaptation by advancing culture or a new religion, are phenomena too well known to be insisted on here. Probably the Cypriote mysteries were adapted to the worship of Aphrodite from a ruder stage in which no divinity was invoked. And if this could happen once in Western Asia, it might have happened also at Babylon, at Heliopolis and elsewhere. It is possible that other practices, such as the prostitution of the Armenian girls at the temple of Anaitis, or that of the Lydian and Paphian girls to earn their dowries, are no more than the adaptation of a custom common enough in the lower barbarism, by which unmarried girls have unfettered liberty in their sexual relations. The Armenian maidens, at all events, though spoken of as harlots by Strabo, do not seem to have exercised their calling for money, nor to have admitted indiscriminately to their favours all who offered. They reserved themselves for their equals in rank, and entertained them in their dwellings with more hospitality than in a spinsters’ house in the Pacific Islands. The surmise may be indulged that it was in fact originally, if not in later times, their way of choosing husbands. The Lydian girls are expressly said to have bestowed themselves in marriage.

Mannhardt contended (and his opinion is so far endorsed by Dr Frazer) that the maidens who surrendered their virginity in connection with the cult of a goddess like Aphrodite did so in imitation of their divinity, as her representatives, the human players of her part.[280.1] This may have been the mode by which the ancient custom was adapted to the newer order of things. But it is submitted that it is a very insufficient account of it. The custom must have been older than any definite belief in the goddess’s habits or any story of her various intrigues. Are we then to suppose that it was a magical rite designed to promote the fertility of animal and vegetable life? Such rites are known in both hemispheres. The great goddess worshipped under different names throughout Western Asia personified, we may concede, the reproductive energies of Nature. Many of the rites employed in her cult are in the last analysis magical, and had for their purpose to assist those energies. By a well-known mental process magical efficacy is often ascribed to acts and usages not essentially of a magical, nor indeed of a ritual, character. Thus the general prostitution of young girls to earn their dowries, and that of widows—customs which are probably of quite a different origin—are among certain tribes of Morocco held to be not without their effect on the abundance of the crops.[281.1] Such a belief may have consecrated lives of habitual harlotry in Armenia, in Lydia, and in Cyprus. It by no means follows that every rite performed in the name of the goddess acquired that meaning, still less that that was its primitive meaning. Many such rites would be wholly personal. They would be intended to secure personal blessings to the worshipper, and nothing more, though everyone might have been required to perform them. It is needless to suppose without express evidence that the rite described by Herodotus as taking place at the temple of Mylitta had more than a personal reference. The most obvious personal blessing to be secured from such a goddess would be fertility. It is possible that this was the intention here. Puberty customs are doubtless performed for the good of the individual, and of the tribe or nationality through the individual. We must not infer, however, that the personal blessing of fertility was held to be the natural and direct outcome of the sacrifice of virginity. That was not the way in which it would be envisaged, however logical such an outcome may seem to us. The savage would not expect the natural result, but one that we call magical. At the stage represented by the Babylonian custom the blessing was invoked from the goddess, and was her gift. In other words, it was not that specific ritual act of coition, but future acts rendered licit and consecrated by it which were expected to bear fruit. This interpretation is perhaps strengthened by a rite said to have been regularly performed in the Troad. “Every maiden on the approach of her marriage was required to go and bathe in the Skamandros, and standing in the water to pronounce the sacred formula: ‘Skamandros, take my maidenhood as a gift.’” Dr Farnell, citing the fictitious letter of Aischines on which the evidence for the custom rests, observes: “The letter narrates how a mortal assumed the human form of the god and took a treacherous advantage; but originally, we may suppose, the rite of consecration was not associated with any anthropomorphic divinity, but was performed in the hope that the spirit of the river might enter into the maiden, and that the child she might afterwards bear to her wedded husband might thus be mystically akin to the guardian of the land. The many early myths concerning heroines and princesses being made pregnant by river-gods suggest that the ritual just described was once prevalent in primitive Greece; for such myths could arise naturally from such a custom.”[282.1] But the ritual itself suggests that it is a relic of a still older rite parallel with the rite at the temple of Mylitta.

The last of the problems connected with the rite is to explain why it must be accomplished with a stranger. If, as has been alleged, the act of defloration of a maiden were held to be in itself dangerous, it is not easy to say why anyone, even a stranger, should undertake it, unless he were strangely ignorant of the risk or strangely careless. In some places, indeed, a maiden who had come to submit to the rite may have been outwardly indistinguishable from one of the hierai; and hence the man may have been unconscious of his risk, or may have been willing to undertake a risk thus diminished. But at Babylon the women who came thus to offer themselves wore a distinctive head-dress of cords, the emblem, perhaps, of their condition of virginity. Moreover, they seem to have been penned in enclosures divided from each other by ropes, which were broken to let them out for the accomplishment of the rite. There was therefore no mistake as to their status or object.[283.1] On the other hand, if the defloration simply involved ritual impurity such as could be removed by the proper ceremonies, it must be asked why the task was left to a stranger. None of our ancient authorities have condescended to define a stranger. We are probably to understand by that term one who was not an inhabitant of the town or who was not a member of the community. The analogy of certain Australian rites already referred to, and of rites of marriage in some other parts of the world, would lead us to suppose that what was really intended in the first instance was one who was not eligible for sexual relations with the woman in the ordinary course. Thus in Peru and New Granada “the nearest relations of the bride and her most intimate friends” are said to have performed the corresponding rite[283.2]; and even her father is credited with the labour among the Orang-Sakai of the Malay Peninsula, the Battas of Sumatra, the Alfoers of Celebes, and on the Island of Ceylon and the eastern Moluccas.[284.1] From this the more developed morality of the Babylonians would recoil. Mr Crawley, commenting on the Australian rite, surmises that in it “initiation” and marriage are one, and that “initiation” ceremonies (that is to say, puberty ceremonies) “of this kind are marriages to the other sex in abstract.”[284.2] The surmise follows from his theory of the danger of human contact, and especially of marriage, and the importance of ceremonies to avert the peril. The theory itself—at all events pushed to the length to which Mr Crawley pushes it—is very questionable. But defloration at puberty, whether natural or artificial, is undoubtedly (whatever else it may be) a formal introduction to sexual life. Such introduction might be the more authoritative and emphatic if given by one (or more) with whom sexual relations would not in future be sustained. It is a ritual act. Ritual acts are acts out of the ordinary course—often clean contrary to the ordinary course. Therein consists their essence, their virtue. But in the growth of civilization, with the emergence of a new religion or different customs, the real meaning of a traditional rite is obscured, the rite itself becomes decadent, and a new meaning is assigned to it. Hence a puberty rite might easily become part of the cult of a goddess like Mylitta.

Moreover, at the stage of decay which the rite had reached at Babylon and elsewhere in Western Asia, the proviso that the person with whom the act was performed must be a stranger might be intended to prevent an assignation. When the act had to be performed as a sacrifice in honour of the goddess it might be regarded as a profanation to perform it as an act of inclination with a favoured lover. The best way to prevent this would be to require that it should be performed with a chance stranger, who might further be looked upon, if Mannhardt’s interpretation be correct, as a representative sent by the goddess to play Adonis to the maiden’s Aphrodite. The rite at Byblus lends countenance to this conjecture. It is supported also by the artificial defloration enacted only in symbol by Roman brides, but in grim earnest at the temples of Siva by brides in Southern India.