A bibliography of Russian word-formation By Dean S. Worth (review)

768 LANGUAGE, VOLUME 54, NUMBER 3 (1978) for reported), it is an interesting and often successful attempt to deal with the syntax and semantics of closely related but not equivalent constructions in an inflected language. I do wish, however, that Rijksbaron were less inclined to believe so strongly in the existence of hypothetical constructions, even when no examples can be found in his material. His hunch that certain unobserved clauses may in fact have occurred could well be right, but the matter deserves fuller attention. That so great a corpus as Herodotus should lack certain constructions is likely to indicate something fundamental about the style of Herodotian syntax. [M. E. Huld, UCLA.] Trakite i texnijat ezik. [The Thracians and their language.] By Vladimir Georgiev. Sofia: Izdatelstvo na BuIgarskata Akademija na Naukite. Pp. 348. Lv. 5.00. Thracian left such scant evidence of itself that it has understandably received little more than footnote attention in standard reference works on Indo-European linguistics. Thus Pokorny's Indogermanisches etymologisches Wörterbuch lists about a hundred Thracian (including Thraco-Phrygian) words—less than half the number listed and explicated in Georgiev's book. Specialist literature on the subject is extensive (see G's bibliography of some 400 items from 1873 to 1976, pp. 30519 ), but it is widely scattered. Thus G has performed a remarkable service to the field by bringing together into one up-to-date volume a thorough chronological survey of the literature on Thracian; a complete list of the forty glosses; a carefully organized discussion ofthe 1500 proper names (toponyms, ethnonyms, personal names, and divine epithets) that provide the bulk of what is known about the Thracian vocabulary (the 200 Thracian words deduced from proper names are listed on pp. 98-103); and several conclusions, made on the basis of the resulting 240-word lexicon, regarding the prehistory of Thracian as an IE dialect (see below). The chapter on Thracian inscriptions (pp. 105-58) includes treatment of several doubtful inscriptions, in addition to the three inscriptions of definite Thracian character; the inclusion of photographs thus makes the entire corpus of Thracian inscriptions available to the public. Especially interesting are the archeological-anthropological -linguistic discussions of the external relationships of Thracian to Daco-Mysian (and its descendant, Albanian), Illyrian, Phrygian, Armenian, Greek, and Macedonian. G documents the following principal sound changes from Proto-Indo-European to Thracian: (1) *o>a, as in Germanic, e.g. *gholto- 'gold' > Saldo-; (2) preservation and subsequent monophthongization of vocalic diphthongs, i.e. *oy>ai>e, *ey>ei>i (*aywes-dtti-ä-s 'eternally shining' > Ebisthitia , divine epithet); (3) development of a svarabhakti /u/ or /o/ before the vocalic liquids and nasals, e.g. *dhfs- > Durze, Dorses (personal names); (4) a Grimm's-Law-like Lautverschiebung, with the voiceless stops *ptk becoming voiceless aspirates ph th kh, the voiced plain stops *b dg becoming voiceless ? tk, and the voiced aspirates *bh dh gh becoming plain b dg; and (5) fricativization or assibilation of the palato-velars Kg to j/0 z/d. The last feature clearly places Thracian among the satem languages ; this and other isoglosses lead G into logical arguments on the ethnogenesis of not only the Thracians but of the other ancient Balkan tribes. The linkage of the Albanians to the Daco-Mysians is an especially intriguing display of the power of linguistic paleontology. A 26-page French-language résumé of the book's contents, after the last chapter, will make the data and conclusions in this book accessible even to those who cannot manage Georgiev's clear and direct Bulgarian. [Philip J. Regier, USC] A bibliography of Russian word-formation . By Dean S. Worth. Columbus, Ohio: Slavica, 1977. Pp. xliv, 317. One of the most striking features of modern Russian as a conservatively synthetic IE dialect is its rich system of word-formation, both grammatical and affective. The complexity of the system is shown by two facts pertaining to the linguistic description of Russian: (1) lexicons must specify the 'paradigmatic ' derived forms (nomina agentis, nomina actionis, relative adjectives etc.) for each base form, since grammars cannot predict these with any great reliability; and (2) any one facet of Russian word-formation (use of a...