178
��INTERNATIONAL JOURNAL OF AMERICAN LINGUISTICS
��VOL. I
��comprehensive grammatical sketches. Still, his evidence for the probable Algonkin origin of Yurok and Wiyot is of such a character as to take it out of the "purely accidental" class and to justify further researches in this field. It is, therefore, desirable, even imperative that in the near future, the Wiyot language be made the object of a thorough investigation and that this investigation be carried on by some authority on the structure of the Algon- kin languages. To my mind, such a study, whether positive or negative in its results, will contribute far more to the general problem presented by the American Indian languages, than can be said of the continued researches into the minutest details of Algonkin word- formation or into the dialectic differentiations within the Algonkin family.
On the other hand, Sapir's articles establish- ing relationships between Uto-Aztakan and between Haida, Tlingit, and Athapascan are most convincing. The comparative data pre- sented in these two papers are so voluminous and conclusive, covering not only phonetics and lexical material but also morphology and structural correspondences that, to my mind, the unreserved acceptance, by all students of the American Indian languages, of a genetic relationship between the Shoshonean, Sonoran- Piman, and Nahuatl families on one hand, and of a similar affiliation between Athapascan, Haida, and Tlingit on the other hand, is only a question of time. The extreme likelihood of these two reductive theories is undoubtedly due to the fact that all comparisons have been based upon extensive material; and although some of the Nadene correspondence may, upon further investigations, prove to be er- roneous, sufficient correspondences have been found to meet the requirements of even the conservative and exacting scholars.
I have purposely dwelt at such length upon the efforts of my co-workers to reclassify and to reduce a number of so-called independent stocks, so that my own conservative attitude towards a potential genetic relationship be-
��tween the Takelman, Kalapuyan, and Chi- nookan languages may become clear. While carding and indexing my Kalapuya field- material (collected three years ago), prepar- atory to the writing of a grammatical sketch of these languages, I was forcibly struck by some marked correspondences in the lexi- cography of Kalapuya and Takelma, and of Kalapuya and Chinook. The Kalapuyan-Chi- nookan agreements are far less than those be- tween Takelma and Kalapuya ; and I am will- ing to admit that some of these correspond- ences may be due to borrowing or, in part, at least, to unconscious substitutions, by my in- formants, of Chinookan equivalents for Kala- puyan values. Such a possibility must by no means be disregarded, in view of the close proximity and long contact that has existed between the peoples speaking these two di- vergent languages. Still, some of these re- semblances are so peculiar as to render the ex- clusive theory of borrowing rather doubtful, especially since all Kalapuya data are not yet available. On the other hand, the resem- blances between Kalapuya and Takelma are much greater and far more numerous, al- though, as has been stated before, only part of the Kalapuya data have thus far been tab- ulated. I am certain that a complete analysis of the lexical material of all Kalapuya dialects will substantially add to the amount of com- parative data. Whether such an analysis will bring forth close morphological and structural correspondences, I am as yet unprepared to say. I am at present working out minutely the morphological structure of the Kalapuya language and will, upon the completion of this work, institute comparisons between the mor- phological elements of Kalapuya and Takelma. I will, however, state that the highly special- ized character of Takelma may prove a serious obstacle in the finding of many positive cor- respondences. However, this statement must be taken as only tentative. There are so many radical agreements and disagreements be- tween the structures of these two languages,
�� �
