The Huron coat, however, which must have belonged to some chief, is rather exuberantly ornamented with animal figures, horse heads, human heads, and birds on the back, sides, and sleeves. I was told by the Huron that such figures were quite commonly used, although they can hardly be as typical as the simpler flower figures. One of the animal figures, evidently a beaver, the emblem of the Huron,1 shows very well the solidly embroidered areas (pl. III). The leggings are similarly decorated down the outside, the designs here being cat's paws, balsam fir, flox, stumps, and roots, besides animals, crescents, and ellipses. On the whole this costume is the oldest, and in many respects the most typical specimen showing the moose hair technique, that has so far come under observation.

It appears, in conclusion, that the moose ha'r appliquée embroidery of this tribe forms an integral part of a widespread northern technique, similar in many respects to quill work. As to the flower designs, about the origin of which at present very little in general is known, it seems that with the Huron at least three are native. From this it may be expected that, upon investigation, certain elements at least of the flower patterns in the art of other northern and eastern tribes may likewise be found to be native.2 As regards the relationship between the moose hair technique and the realistic names, I think the latter are purely secondary in origin, the design having acquired names from their resemblance, in the native eye, to certain familiar objects. To judge by the appearance of these non-geometrical designs, plant and flower likenesses would be most naturally suggested. The occurrence of the evergreen balsam fir as the chief name and figure corresponds to what is found among

1Cf. also J. A. Maurault, Histoire des Abenakis (Sorel, 1866), p. 227.

2 Evidences of the early use of flowers, as decorative motives among the New England Indians are to be found in some of the colonial accounts; cf. Willoughby "Textile Fabrics of the New England Indians," Amer. Anth., vol. 7, p. 88, quoting Gookin; also "Dress and Ornaments of the New England Indians," ibid., pp. 502, 508. Also as an example of apparently indigenous flower designs among the central Algonkian, mention might be made of the conventional but quasi-realistic trailing arbutus figure in its quill decorations on birch bark boxes of the Michigan Potawatomi. With these people the arbutus is known as the "tribal flower," according to the late chief Simon Pokagon. Cf. "Ogimawkwe mitigwaki" Queen of the Woods, S. Pokagon, p. 155-9 (Hartford, Mich., 1899).

Indians of other regions, where the pine, for instance, occurs as a motive both in California and on the Plains. From an early beginning with only three or four plant names the Huron appear to have developed their art by inventing certain new flower figures and modifying some of the old conventional ones and giving them new plant names. So, it is claimed, representations of exotic plants such as forget-me-not, flox, and others, have crept in as motives in modern times making the technique less prominent than the attempt at realism so far as the decorative idea in the mind of the native artist is concerned.

As an early result of investigations now being carried on among the northeastern Algonkian tribes, I find incidentally that the first three figures of the Huron set, fig. 6 (a, balsam fir; b, star; c, cat's paw) also occur in the incised designs of the St Francis Abenaki, Penobscot, Malisit, and Micmac. Furthermore the figure known as balsam fir by the Huron goes under the name of tree among the Penobscot where it is exceedingly common in their exquisite decorative wood carving. Identical figures are to be seen on the skirts of painted buckskin coats from the Naskapi Indians of the Labrador interior,1 and another figure identical with the Huron barberry occurs on another Naskapi coat in the American Museum of Natural History, N. Y. It is evident that these figures had a wide distribution among the northeastern Algonkians with whom they were shared by the Huron, if not altogether borrowed by them.2

UNIVERSITY OF PENNSYLVANIA MUSEUM,

PHILADELPHIA, PA.

1Cf. Turner, "Ethnology of the Ungava District," 11th Annual Report Bur. Amer. Ethnology, p. 282, fig. 88.

2 I have recently seen two specimens of the moose hair embroidery among the Penobscot, the figures and technique of which are identical with those of the Huron. Moreover, both Mr Mechling and myself have independently encountered the same technique and designs among the Malisit of St John River.

SOME ASPECTS OF NOOTKA LANGUAGE AND CULTURE1

BY EDWARD SAPIR

INTRODUCTION

HE two brief sketches that follow are based on linguistic and

ΤΗ

ethnological material collected during September to December of 1910 among two tribes of the Northern Nootka of Barkley Sound and Alberni Canal, the Tslicya" atн" and Hopátclas at H,2 now living at the head of Alberni Canal and on the banks of Somass River, in the neighborhood of the present town of Alberni. During the time spent among the Nootka Indians a few points were studied with relative fulness, though at best only a beginning was made even for these, while many other points of great importance were only touched upon. In the present paper two matters of considerable interest in regard to the linguistic and cultural affiliations of the Nootka and Kwakiutl are somewhat hurriedly discussed. The full presentation of the facts involved is reserved for future publications. It should be added that the wolf ritual was witnessed by the writer.

I.

THE LINGUISTIC RELATIONSHIP OF KWAKIUTL AND NOOTKA The Wakashan linguistic stock is divided into two main branches, the Kwakiutl and the Nootka or Aht; the former embraces Kwakiutl proper, Xaisla, and He'ftsa q", the latter Northern Nootka (from about Cape Beale north to Cape Cook on the west coast of Vancouver Island) and Southern Nootka or Nitinat (south of Cape Beale to Cape Flattery). By careful comparison of the two Wakashan branches one can in part reconstruct a Wakashan "Ursprache," but the actual differences between Kwakiutl and Nootka are in fact very great; they differ perhaps as much as Slavic and Latin.

1 Published by permission of the Geological Survey of Canada.

* Boas' and Swanton's phonetic system is used with some modifications for the vowels. i, e, and o are short and open, thus corresponding to their î, ê, and ô; į, ẹ, and ọ, are short and close, and correspond to their i, e, and o; î, ê, and ô are long and open, a being equivalent to Boas' d. Superior vowels used with H denote vocalic timbre.

As regards phonetics, Kwakiutl and Nootka, while both showing characteristic Northwest Coast features, differ rather considerably. The sonant or intermediate stop series of Kwakiutl is absent in Nootka, Kwakiutl p and b for instance being replaced by Nootka p. Besides the s-series, which Kwakiutl and Nootka possess in common, Nootka has a c-series, which is doubtless derived from the Kwakiutl and Wakashan k'-series, which in turn Nootka lacks; thus Kwakiutl g and k are cognate with Nootka tc, k'! with tc!, and x with c. There is no l in Nootka, n corresponding to both Kwakiutl and n. The velars q! and x, while somewhat infrequently found in Nootka, are not the regular Nootka representatives of Kwakiutl q! and x; q! has developed into a peculiarly harsh and choky glottal stop, which I write, x into a strangulated-sounding h which I write H, these two consonants respectively resembling Arabic 'ain and hâ; ordinary and h are also frequently found in Nootka. As regards phonetic processes, Kwakiutl and Nootka agree in allowing no initial consonant clusters in words; initial Kwakiutl and Nootka Em, En, w, y, and Kwakiutl 1 are undoubtedly related to ordinary Kwakiutl and Nootka m, n, w, y, and Kwakiutl / as are Kwakiutl and Nootka p!, t!, k!, L!, ts!, q!, Kwakiutl k'!, and Nootka tc! to non-fortis Kwakiutl and Nootka p, t, k, L, ts, q, Kwakiutl k', and Nootka tc. In both Kwakiutl and Nootka certain derivative suffixes "harden" the final consonant of the stem; thus p, q, and 7, become Kwakiutl p, q, and 1, Nootka p!, , and y. The "softening" of Kwakiutl seems to be represented in Nootka by but a few stray phonetic processes. Syllabically final glottal stops and glottally affected consonants-such as - and -p!,-which are common in Kwakiutl, are entirely absent in Nootka. Medial and final consonant clusters are not as freely allowed in Nootka as in Kwakiutl, i often serving in Nootka to lighten them (cf. Nootka -qEmił, "round thing," with Kwakiutl -gEmł "mask"). All final vowels and stopped consonants in Nootka are aspirated. Peculiar to Kwakiutl is the change of k-stops to spirants (x, x", x') before consonants, whereas in Nootka they remain; in this point Nootka seems more archaic than Kwakiutl.

ទ

In general morphology Kwakiutl and Nootka are quite similar,

despite numerous differences of detail. In both the stem is, as far as its meaning allows, indifferently verbal or nominal and one or more suffixes are required to give rise to definitely verbal or nominal complexes; in Nootka a suffixed - is often used to substantivize a verb form. Both Kwakiutl and Nootka are absolutely devoid of prefixes, most of the elaborate grammatical mechanism being carried on by means of suffixes, to a lesser exent by means of initial reduplication, and, in Nootka, consonantal changes. The suffixes of Nootka and Kwakiutl express similar ideas and are used in more or less parallel fashion, though the number of suffixes that are etymologically related form but a small percentage of those found in either; so far about ninety Nootka suffixes have been discovered that are entirely or in part cognate to Kwakiutl suffixes. Examples of local suffixes shared by Kwakiutl and Nootka are: Kwakiutl -ōyō “in the middle," Nootka -win'; Kwakiutl -nēq" "in the corner," Nootka -nikw-; Kwakiutl -atūs "down river," Nootka -atis; Kwakiutl -ts!ō "in," Nootka -ts!o'; Kwakiutl -k-E "top of a box," Nootka -tci "full"; Kwakiutl -!a1 "on the rocks," Nootka -la a'; Kwakiutl -es "on the beach," Nootka -is; Kwakiutl -it"in the house," Nootka -it; Kwakiutl -xs "in a canoe," Nootka -qs, -lанs. A few examples of body-part suffixes are: Kwakiutl -!ōs "cheek," Nootka -as; Kwakiutl -xō "neck," Nootka -as-Haul "chest"; Kwakiutl -āp! "neck," Nootka -aplał "back." Important temporal elements held in common are: Kwakiutl -L "future," Nootka -āq-L, -3¿L; Kwakiutl -xid "inceptive," Nootka -ci-L. There are some striking agreements in verbifying derivative suffixes, as: Kwakiutl -lexst "to desire," Nootka -!iH" "to try to get," -st!iH" "to have as goal"; Kwakiutl -la "to go in order to," Nootka -las; Kwakiutl -klāla "to make a noise," Nootka en' (= Wakashan *-q!Ela); Kwakiutl -g at "beginning of a noise," Nootka -aL (= Wakashan *-qlat); Kwakiutl -q/Es "to eat," Nootka -lis; Kwakiutl -nuk" "to have," Nootka -nak'. Examples of nominal suffixes are: Kwakiutl -aanō "rope," Nootka -anut "long"; Kwakiutl -gas "woman,” Nootka -aqs; Kwakiutl -asdē "meat," Nootka -act" "dried meat"; Kwakiutl -mis "useless," Nootka -mis "mass"; Kwakiutl -p!ē-q

'I denotes a "strengthening" of the preceding consonant.

AM. ANTH. N. S., 13-2