The passages that I have just quoted are those which are most frequently used to support the pretension that human societies have begun with promiscuity; they are at once the most ancient, most authentic, and most explicit. We may add to them the assertion of Varro, quoted by Saint Augustine,[73] according to which the Greeks, prior to the time of Cecrops, lived in promiscuity. But how is it possible not to be struck with the weakness of these historical proofs? Some of them are mere general assertions, while others plainly relate either to social anomalies or to cases of polyandry. There is no doubt as to this in regard to the ancient Arabs of whom Strabo speaks, and also to the Protohellenes of Varro. This last instance certainly relates to the matriarchal family, of which I shall have to speak again at some length. In fact, after having stated that the Protohellenes had no marriage, Varro adds that the children only knew their mother and bore her name. The proof is decisive, for the matriarchate does not in the least exclude marriage, as we shall see later, and in the case of the Lydians it lasted until the time of Herodotus. In order to complete this review of ancient texts, I will mention further the passage of the _Timæus_ in which Socrates speaks of the community of wives:--“On the subject of the procreation of children we established a community of wives and children; and we devised means that no one should ever know his own child. They were to imagine that they were of one family, and to regard those who were within a certain limit of age as brothers and sisters; and again, those who were of an elder generation as parents and grandparents, and those who were of a younger generation as children and grandchildren.” But Plato had a lively imagination. He was a “great dreamer,” as Voltaire said of him, and this passage evidently describes a purely utopian society. Traditions relative to a very ancient epoch of promiscuity are found here and there outside the Greco-Roman world. In China, for example, the women are said to have been common until the reign of Fouhi.[74] A tradition of the same kind, but more explicitly stated, is mentioned in the Mahabharata (i. 503): “Formerly it was not a crime to be faithless to a husband; it was even a duty.... This custom is observed in our own days among the Kourous of the north.... The females of all classes are common on the earth; as are the cows, so are the women; each one has her caste.... It is Civéta-Ketou who has established a limit for the men and women on the earth.”[75] This assertion is vague, and has not the least proof to support it. If, continuing our inquiry, we attempt to correct these historical documents by ethnographical information, we shall hardly find, on this side of the subject more than on the other, anything but simple assertions, which are either too vague or too brief, or evidently open to dispute.