If our knowledge of the Indo-European vocabulary of kinship has not been noticeably advanced since the study of Indo-European kinship by Delbrück (1890), ethnological research, for its part, has made great progress, and this is what today provokes the linguist to revise the traditional interpretation of certain lexical “anomalies.”

The terms relating to kinship are among the most stable and securely established items of the Indo-European vocabulary, because they are represented in nearly all languages and emerge from clear correspondences. All the conditions favorable for an exhaustive study are fulfilled. In spite of this, no advance has been made in this problem since 1890, the date of publication of Delbrück’s work, entitled Indogermanische Verwandtschaftsverhältnisse, where the two principal conclusions which can be drawn from these correspondences are set out. On the one hand, the structure of the family implicit in the vocabulary is that of a patriarchal society, resting on descent in the paternal line and representing the type of “Grossfamilie” (still observed in Serbia in the nineteenth century) with an ancestor, around whom are grouped the male descendants and their immediate families; on the other hand, the terms of kinship concern the man; those which relate to the woman are not very numerous, are uncertain and often variable forms.

However, the progress made in the last seven or eight decades has not merely consisted of the assembly of a greater mass of data derived from a greater number of societies, but also and more particularly of a better interpretation in the light of a progressively refined general theory of kinship.

The systems which have been studied outside the Indo-European world sometimes make use of identical terms for degrees of relationship which are distinguished in modern western societies: those, for instance, for “brother” and “cousin,” or for “father” and “paternal uncle.” Inversely, they distinguish relationships which we confuse, e.g. “mother’s brother” and “father’s brother” (for us “uncle”), “sister’s son” and “brother’s son” (for us “nephew”), etc.

But relationships which are strange to us nowadays sometimes have their equivalents in the ancient Indo-European world, in which we must try to discern, as with all systems of kinship, certain principles of classification.

The Indo-European vocabulary of kinship in fact presents a certain number of anomalies which can perhaps be better defined in the light of other systems. For instance, the Lycian people, according to Herodotus (I, 173) have matronymic names: “they call themselves after their mothers and not after their fathers”—and he adds: “If a female citizen marries a slave the children are considered to be of good stock; but if a citizen, even if he were the first citizen, has a foreign wife or concubine, the children are of no account.” Thus in Lycia we have matrilineal descent. But Herodotus’ assertion seems not to be confirmed by the personal names of Lycian inscriptions. However, Herodotus has not invented this peculiarity. He gives us other information which has since been confirmed, for instance that the indigenous name of the Lycians was Termilai. We can sense the importance of women in Lycia already in the legend of Bellerophon, as it is told in Homer: (Il. 6, 192-195). The king of Lycia gives his daughter to the Argive Bellerophon, as well as half his royal prerogatives, making him both his son-in-law and his successor. Thus Bellerophon acquired royal rank by his marriage. Now, from the inscriptions we can get an idea of the system of kinship among the Lycians. In a bilingual dedication of the fourth century B.C. on the base of a statue we read: Πόρπαξ Θρύψιος Πυριβάτους ἀδελφιδοῦς Τλωεὺς ἑαυτὸν καὶ τὴν γυναῖκα Τισευσέμβραν … Ὀρτακία θυγατέρα Πριανόβα ἀδελφιδῆν ‘Porpax, son of Thrypsis, nephew of Pyribates, citizen of Tlos, himself and his wife Tiseusembra, daughter of Ortakias, niece of Prianobas…’ The same text is given in the Lycian language. We have the name of the person with his paternal descent (assuming Thrypsis to be the name of a man, which is not certain); but it says also “nephew of…,” his wife is called “daughter of…” and also “niece of…” This wording is found also in many other Lycian inscriptions and quite often the sole description is even “nephew of…” What is the sense of “nephew” in this case?

In a system which prescribes marriage between cross-cousins, a man may marry the daughter of his father’s sister, or his mother’s brother, but never the daughter of his father’s brother or his mother’s sister—and this for a classificatory reason: the brother of the father is called “father”; the sister of the mother is called “mother.” Consequently, the son of the father’s brother or the mother’s sister is called “brother” and the daughter “sister.” We understand now the impossibility of marriage with “sisters and “brothers.” No less clear are, inversely, the conditions of kinship which permit a legitimate marriage: the father’s sister, the mother’s brother belong to other clans, as do their children. The relationship of uncle to nephew is defined as follows: the “uncle” is for the nephew his mother’s brother, the “nephew” is for the uncle his sister’s son. The word “nephew” in many societies means only “sister’s son.” In our Lycian inscription, Pyribates is the maternal uncle of Porpax, and Prianobas the maternal uncle of Tiseusembra. Thus we have here a mixed system where the paternal descendance is indicated as well as the maternal clan.

There is another fact which we have to account for. Why is the Indo-European vocabulary so poor in expressions for female kinship? This has been explained by the predominance of masculine functions in the family. This may be true, but male preponderance could have maintained itself without provoking the same lexical consequences: the legal conditions of the woman had changed little in Europe until the eighteenth century, but that does not prevent our vocabulary from being strictly reciprocal (e.g. father-in-law/mother-in-law), etc. The explanation must be rather that the wife leaves her clan to enter that of the husband and this institutes relations between her and the family of her husband which demand expression. Now, this family being a “Grossfamilie” of the type known from Homeric society, these relations are manifold: the newcomer enters into special relations with the father, the mother, the brothers and their wives. On the other hand, for the man, there is no necessity to distinguish relatives of his wife by specific terms since he does not co-habit with them. To characterize them he contents himself with the general term “related, allied,” which refers to them indiscriminately.

A third fact must be noted : the frequent variations in the designation of certain degrees of kinship. The terms for “father” and “mother,” “brother” and “sister” are clear and constant, but for “son” there is considerable variety of terms, with frequent innovations. Similarly, the term for “uncle, aunt; nephew, niece” are ambiguous and present much diversity from language to language (Latin nepos means both “nephew” and “grandson”). Finally it would appear that we are unable to reconstruct even partially an Indo-European designation for “cousin.” These variations raise serious problems on various planes.

If we consider merely the particular systems in each separate language, some strange correlations come to light: thus avunculus ‘uncle’ in Latin is the diminutive of avus ‘grandfather’. Here are some of the problems which arise at all levels, some of which concern the sense of the terms, others their distribution, and still others their evolution.