The Sapir-Kroeber correspondence Edited by Victor Golla (review)

REVIEWS183 Ross, John R. 1967. Constraints on variables in syntax. MIT dissertation. Van Valin, Robert D. , jr. 1985a. Case-marking and the structure ofthe Lakhota clause. Grammar inside and outside the clause, ed. by Johanna B. Nichols & Anthony C. Woodbury, 363-413. Cambridge: University Press. ------ . 1985b. Languages without government. Paper presented at the Symposium on Canadian Native Languages in Theoretical Perspective, SUNY Buffalo. Walton, Charles. 1983. Sama verbal semantics: Classification, derivation, and inflection . Temple University M.A. thesis. Williams, Edwin. 1984. Grammatical relations. LI 15.639-73. Woodbury, Anthony C. 1975. Greenlandic Eskimo, ergativity, and Relational Grammar . Grammatical relations, ed. by Peter Cole & Jerrold M. Sadock (Syntax and semantics, 8), 307-36. New York: Academic Press. [Received 12 June 1985.] The Sapir-Kroeber correspondence. Edited by Victor Golla. (Survey of California and Other Indian Languages, Report 6.) Berkeley: University of California , 1984. Pp. x, 494. Reviewed by Lyle Campbell, SUNY Albany Fascinating book! I couldn't put it down! Ostensibly, it contains 363 letters between Edward Sapir and Alfred Kroeber; but in reality it is much more—a treasure of information on such varied topics as the early history of linguistics and anthropology; the development of American Indian linguistics, particularly classification (with detailed arguments and data); the personalities of these two great men and of many of their contemporaries; phonetics and phonemics; psychology; politics; poetry; and Ishi. Golla's precise and copious notes abound in supplementary information, and appear unfailingly where additional context is helpful. Given their delightful style, I let the letters speak for themselves, ordering quotes around certain themes. (Note that these friends address each other as 'Dear Kroeber' and 'Dear Sapir'.) On classification. It has been claimed that the history of American Indian linguistics is characterized by enthusiasm for reduction ofthe number ofgenetic groupings, whereby originally preliminary proposals came to be widely accepted and expanded greatly in size as more and more previously unaffiliated groups were attributed to them (cf. Campbell & Mithun, The languages of Native America [Austin, 1979], pp. 3-69). The first-hand accounts in these letters give substance to this claim. Kroeber, 1913 (pp. 73-4): '...for two or three years past [R. B.] Dixon and I have been collecting lexical information on all the California dialects with a view to working out character and extent of borrowing between unrelated languages ... I could, however, get no intelligible result from our data until finally in desperation I dropped the assumption ... that all these resemblances were due to accident or borrowing and assumed genetic relationship between those languages that had the greatest number of similarities ... My wonder now is that we have overlooked the obvious so long ...' Kroeber, 1913 (p. 97): 'The Hokan group in particular lends itself to the supposition of being widespread ... I should be surprised if our Hokan group did not prove ultimately to be the nucleus of a very large stock.' Sapir, 1913 (p. 104): ? propos of larger linguistic units, which seem to be somewhat in favor 184LANGUAGE, VOLUME 62, NUMBER 1 (1986) just now, I may say that I have been occupying myselfoflate with Athabaskan, Tlingit, and Haida, and that I have collected enough evidence to convince myself at least of the genetic relationship of these three.' Sapir, 1913 (p. 105): 'Do you suppose Yurok might turn out to be related to Salish? Tillamook is not so awfully far away! Don't tell Boas—he'll think I've gone crazy!' Kroeber, 1913 (p. 106): 'We seem at last to have got [J. W.] Powell's old fifty-eight families on the run, and the farther we can drive them into a heap, the more fun and profit.' Sapir, 1913, to P. Radin (p. 113): 'The process of slaughter of linguistic families, upon which several of us seem to have embarked of late, is going on apace ... I now seriously believe that Wishosk [Wiyot] and Yurok are related to Algonki[a]n . . . The consequences of this latest theory are so great that I am hesitating very considerably, even in my own mind, about committing myself, and want to get more and more evidence before I confess to myself that I am convinced.' Kroeber, 1913 (p. 123): ? hope in any event that...