singke
(singke)
#1
238 WILLIAM H. JACOBSEN, JR.
2. The data
In order to get an idea of the relative representation in Nootka of these dif
ferent patterns of clause combining, I have examined the first seven texts
(pp. 14-41) in Sapir and Swadesh's Nootka Texts (1939). These are
mythological tales, five of which are presented with both interlinear and
free translations. They contain 72 passages that are direct quotes of the
characters (along with eight brief songs), giving us a chance to compare the
styles of the third-person narrative proper and of the conversational
interactions, and as a bonus, ways of indicating the transitions from the one
to the other. My citations from this source are identified as to page and
line. A few references to other texts in this volume and to the text analyzed
by Sapir (1924) are also made.
Nootka is spoken along the central west coast of Vancouver Island,
which is off the coast of southern British Columbia, Canada. It is closely
related to two languages to its south, Nitinat and Makah, to form the Noot-
kan family. These in turn are more distantly related to the Kwakiutlan lan
guages to their north to make up the Wakashan family (cf. Jacobsen
1969:140-141, 1979a:83-84, 1979b; Embleton 1985:52-56). These texts were
obtained in 1910 or 1913-14 by Edward Sapir from three informants,
Hamilton George, Frank Williams, and Big Fred. They represent a south
erly form of this dialectally differentiated language, that of Barkley Sound
and Alberai Canal.^8
3. Survey of findings
The general picture that emerges is as follows. There is no explicitly
marked coordination above the level of noun phrases. Although some
sequences of predications sharing the same subject can readily be construed
this way, nothing overtly distinguishes them from cosubordinate or
unlinked clauses (cf. similar findings in Mithun 1984a:502-503). Sequences
of unlinked clauses not sharing an argument are also quite common.
Marked subordination of both clauses and cores is reasonably common,
occurring in a proportion midway between those of Kathlamet and Tlingit.
This is indicated mostly by suffixes on their predicates, but there are also a
few subordinating particles that precede their clauses. When subordinated,
both cores and clauses function as core arguments most of the time,