BOAS] SOCIAL ORGANIZATION OF THE KWAKIUTL 115
might well disappear. The union of L !aL !asiqwala and of NaqEm- g'ilisala in one village has not yet led to their fusion, but externally at least they form a single tribe. The stability of tribes is primarily due to the fact that the tribal units have fairly definite functions distinct from the functions of the tribal divisions. These appear particularly in formal gatherings in which the tribes are arranged in rank and in which, furthermore, definite tribes are matched. Thus in northern Vancouver island, we find the following parallel arrangement of Kwag'ul tribes and of the tribes further to the east.
GwetEla matched with Mamaleleqala
Q !omoya e ye matched with Qweq u sot !enox u
e walas Kwag'ul matched with Nimkish
Q !omk' lotEs matched with Lawits !es
Notwithstanding the relative stability of the tribes, the tribal divisions must be considered as the fundamental units. In previous writings I have used the terms "gens" and "clan" according to the varying impression of prevalence of maternal and paternal descent, both of which are important. After much hesitation I have decided to use the native term numaym ( e nE e me'm) because the characteristics of the unit are so peculiar that the terms "gens" or "clan" or even "sib" 1 would be misleading. We have to recog- nize first of all that positions in a numaym, or at least the ranking positions must be filled and that their disappearance, according to the ideas of the Indians, would be a misfortune. A position is defined by the name attaching to it and by a number of privileges (k'!es 6). I prefer the term "privilege" to the term "crest," because the privileges are quite varied in character, although not so much varied in form as among the Nootka.
A clear understanding of the constitution of the numaym is made very difficult by the fact that the number of positions is at present greater than the number of members of the tribe, so that many individuals hold more than one position in more than one numaym. It may be that even in early times, important per- sonages had the right to do so, but the present extension of this
1 See Robert H. Lowie, " The Matrilineal Complex," University of California Pub- lications in American Archaeology and Ethnology, vol. VI, vol. II, p. 30.
�� �
