SAPIR] NASS RIVER TERMS OF RELATIONSHIP 267
(no. 10). A natural consequence of this distribution of terms is the classification of cousins into "siblings" (nos. n, 12, 13) and "cross cousins" (no. 14). Whether these facts are explainable on the basis of the exogamic phratry organization of the Tsimshian tribes, as would be currently assumed, or of the levirate, is not clear. Perhaps neither factor is the historically primary cause.
6. The distribution of terms for "uncle's or aunt's spouse" and, reciprocally, for "spouse's niece or nephew" is somewhat curious. The maternal aunt's husband is classed with the father (no. 5); reciprocally, the wife's sister's child with one's own child (no. 7). The paternal aunt's husband is designated by a descriptive term, "husband of paternal aunt" (no. 26), to which corresponds, as reciprocal, a descriptive term, " cross-sibling's-child of wife" (no. 27). The maternal uncle's wife is classed with the paternal aunt (no. 9) ; reciprocally, the husband's sister's child is classed with one's cross-sibling's child (no. 10). Finally, the paternal uncle's wife is classed with the mother (no. 25) ; the corresponding reciprocal term for the husband's brother's child is classed with one's own child (no. 7).
7. Somewhat unexpected is the distribution of terms for "sibling's child-in-law " and, reciprocally, for "parent-in-law's sibling." The child-in-law of the brother or sister is consistently designated by purely descriptive terms (nos. 20, 21, 22). The reciprocals, however, are only partly analogous. The sister of the parent-in-law is descriptively defined (nos. 23, 24), but the brother of the parent-in-law is merged with the parent-in-law (no. 16).
8. The fairly extended use of transparent descriptive terms (cf. English "father-in-law") is noteworthy (nos. 20, 21, 22, 23, 24, 26, 27). Analogous formations occur further south in Washo and in certain Shoshonean systems. That these terms are to be looked upon as genuine terms of relationship, not merely as for- mations ad hoc, seems to be indicated by the fact that their range of actual significance is more restricted than that of their etymological significance (see note to no. 23; also supplementary note 3, p. 261). Thus gimxdd la'mc-C "my father-in-law's sister" (no. 23) has a far wider etymological significance, as it might also refer to "my
18
�� �
