Korean.

• Author: Miller, Roy Andrew

By SUK-JIN CHANG. London Oriental and African Language Library, vol. 4. Amsterdam and Philadelphia: JOHN BENJAMINS PUBLISHING COMPANY, 1996. Pp. xvii + 251.

The professed goal of this new S.O.A.S. series is admirable: "to make available . . . reliable and up-to-date descriptions of the grammatical structure of a wide range of Oriental and African languages, in a form readily accessible to the non-specialist" (p. [ii]). Unfortunately neither Professor Chang (of Seoul National University) nor his London editors have come even close to delivering what they promise. The "non-specialist" will be forced to decode and decipher much of the English in which the book is written, while knowledge of Korean will only entail spending even more time correcting many of the Korean exampies. (E.g., "The Silla vernacular handed down to us in the writing of Chinese scripts is called Hyangka 'Silla Song'" [p. 227] means: "a small corpus of poems in Silla Old Korean, the Hyangka, has been transmitted in texts recorded in Chinese characters"; "J. in short" [p. 235] means "Japanese is abbreviated 'J.'")

But the inadequacies of this book go far beyond such simple if annoying errors that careful writing, editing and proof-reading might easily have avoided. The author has been unable to settle upon a relevant grammatical model in terms of which to tackle his language. The question of what kind of grammar he should write, indeed the question of what possible kinds of grammars exist, clearly has never troubled him; but it will trouble his readers. Simply because he nowhere specifies his grammatical model does not mean that he has none. Lurking behind every page is the spectral shape of English grammars of the last century, as transmogrified into Japanese school grammars, and from there on into Korea. None of this fitted Japanese very well, and it fits Korean equally poorly. But that is the way those things are still done in Seoul, not to mention Tokyo; and now apparently in London, as well.

The spectre of Japanese is all the more spectral in these pages because Japanese itself is rarely mentioned, or when it is, mostly cited incorrectly (e.g., pace p. 235, the verb suru is not used of neckties). More often it is Banquo's ghost; alpaithu 'part-time student job' is neither, as Chang claims, from German arbeiten nor does it illustrate how "foreign words, including verbs, are borrowed as nouns" (pp. 24, 25). The form is ultimately from the noun Arbeit, but it entered...