The volume under review is Alexander Vovin’s first full-scale monograph to touch directly upon the issue of the so-called “Altaic” theory (deep-level genetic relationship between several language families of Central Asia and the Far East). It represents a logical extension of a prolific series of papers on the subject—papers that, in the 1990s, were written from a stark “pro-Altaic” perspective, then gradually shifted toward a more skeptical and, finally, staunchly negative point of view. Having already provided a detailed summary of the main reasons for his disappointment in the Altaic [End Page 225] theory,1 Vovin now turns his attention to a more specific issue: the possibility of a genetic relationship between Japanese and Korean. This hypothesis may be pursued (and has been pursued) without the context of the larger Altaic theory but, as the author shows throughout the volume, shares similar theoretical and methodological problems—problems that eventually lead him toward decisively rejecting the evidence for “Japanese-Korean,” just as he did earlier with the evidence for “Altaic.”
Instead of presenting a detailed overview of the history of research on “Japanese-Korean” (only a few basic points of reference are given in the introduction), Vovin concentrates almost exclusively on the critical analysis of the most recent and, in his opinion, the most professional and potentially promising studies. These include multiple works by Samuel Martin (usually drawn upon in the section on comparative morphology) and especially John Whitman’s dissertation, in which a detailed system of phonetic correspondences between Japanese and Korean was set up on the basis of several hundred lexical and grammatical comparanda.2 The most recent systematic comparison of Japanese and Korean linguistic material, presented by Sergei Starostin, Anno Dybo, and Oleg Mudrak within a general “Altaic” framework,3 is consciously ignored, because, according to Vovin, this work is of a “pseudo-scholarly” nature (p. 92), a position consistent with his review of said work.4
Given the fact that the author is not only an acknowledged specialist in both Korean and Japanese diachrony but is also well known for his scrupulous analytical approach to linguistic data, Vovin’s monograph will undeniably be highly informative for scholars who have, for one reason or another, missed out on the latest developments in Korean and Japanese historical studies. It includes an entire chapter that briefly summarizes these developments, along with the author’s personal assessment of such controversial topics as the historical nature of lenition in Middle Korean or the reconstruction of initial *b- (accepted) and *d- (rejected) in Proto-Japonic (pp. 11–44).
However, the chief value of the monograph is summarized in its original conclusions. These occupy no more than four pages (237–40), with the bulk of the work given over to a detailed analysis of the lexical and morphological evidence on the basis of which they are drawn. The conclusions are exceptionally interesting in that the author does not merely sweep away all [End Page 226] evidence for “Japanese-Korean” as “non-evidence” but presents the reader with a constructive alternative to the “genetic” solution. Rejecting 261 out of Whitman’s 347 etymologies for reasons such as “irregularity in the phonetic correspondences, faulty morphemic analysis, philological problems,” and “vague or unreliable semantics,” he qualifies another 75 as “obvious loans” from Korean into the Central Japanese group of dialects. This claim is founded primarily on distributional analysis: most of the suggested undeniable parallels between Korean and Japanese, as Vovin shows throughout the monograph, are attested in Western Old Japanese (WOJ) but are generally missing in Eastern Old Japanese (EOJ) as well as the Ryukyuan branch of the Japonic family, that is, they are not properly reconstructible for “Proto-Japonic.” The conclusion, seemingly dictated by the standard practice of comparative linguistics, is inevitable: these parallels “represent comparatively late loans from Old Korean into Central Japanese and were probably borrowed between the late fourth and the late seventh centuries AD” (p. 239). The remaining eleven lexical parallels, whose Japanese counterparts are reconstructible...