Proto-Romance phonology By Robert A. Hall Jr (review)

REVIEWS689 rooted in theory and tradition, but not restricted by either. Fortunately, it is also good-natured. H's wit and occasionally light-hearted style provide frequent welcome relief from the massive amount of material and information. After an exhaustive analysis ofthe theories concerning the origin ofthe runes, he concludes : 'According to the later Icelandic literary sources, the runes were invented by Odin and given to mankind by him; perhaps this is as good a theory as any' (123). During the Viking period, the Christian nations Offered the Scandinavian kings membership in the club by adopting them as fellow Christians' (134); the Scandinavian princes and noblemen ofthe 12th and 13th centuries were all 'looking for a piece of the power' (182); the rune carvers couldn't resist 'the tempting wooden walls ofthe Norwegian stave churches' (192); and a sound change 'crept up the Oslo fjord' (207— undoubtedly transmitted by water-tight arguments). Icelandic and Faroese are 'fiercely' Scandinavian (23), and conservative dialects in general can be 'defiantly self-contained ' (254). Sometimes such stylistic flourishes can lead to vagueness, as when Danish is described as 'a supple and civilized language' (40); but usually they are themselves welcome examples of 'vigorous bursts of simple style' (379). Haugen's style in general shows the same 'admirable consistency' that he attributes to the orthography of Christian Vs Danish Law of 1683 (355); and just as 'realism dispelled the fogs of romanticism ' in Scandinavia by 1 864 (41 1), so has his own work on the Scandinavian languages dispelled many of the fogs of previous scholarship on the subject. [Received 24 June 1977.] Proto-Romance phonology. By Robert A. Hall jr. (Comparative Romance grammar, 2.) New York: Elsevier, 1976. Pp. x, 297. $22.50. Reviewed by Jerry R. Craddock, University of California, Berkeley* 1. The present work is the second volume in a projected set of six (for a review of the first, see Craddock 1976). In this installment, Hall's purpose is to provide a coherent account of the Proto-Romance sound system through the systematic comparison of cognate forms found in the medieval and modern Romance vernaculars . Proto-Romance can be defined as the set of linguistic traits that this comparison reveals. By definition, these traits must lie at the origin of all the Romance languages; any development affecting only a sector of the Romance domain, e.g. diphthongization or lenition, cannot be attributed to Proto-Romance, but rather must pertain to later stages. Since H quite properly includes Sardinian and Rumanian in his purview, his Proto-Romance bears a much greater superficial resemblance to Latin than do most standard versions of 'Vulgar' Latin—so much that skeptics may question the wisdom of documenting so elaborately multitudinous facts about Proto-Romance which are readily available to even the most modestly endowed Latinist. An initial chapter defines the comparative method and Proto-Romance. The core * I am grateful to my UC colleagues Suzanne Fleischman, Ruggero Stefanini, and Máximo Torreblanca for numerous valuable suggestions regarding the form and content of this review; and to Steven Dworkin, Arizona State University, for some leads to recent literature on the points discussed. 690LANGUAGE, VOLUME 54, NUMBER 3 (1978) of the work is Chapter 2, which presents the cognate sets used to determine the phonemic inventory of Proto-Romance. Chapters 3 and 4 describe the relationship of Latin to Proto-Romance and the major developments in early Romance. Finally, two appendices offer further cognate sets and an index of PRom. words. Not content with simply establishing the existence of a given sound, H has attempted to collect cognate sets that fully illustrate its phonotactic distribution. Thus the vowel /e/ is reconstructed in stressed, pretonic, intertonic, and posttonic syllables—the last category embracing both final-syllable and word-final position. Consonants are reconstructed as word-initial, medial, and final : at first singly, then as geminates, and finally in clusters. The reader is consequently able to form a rather complete picture of PRom. syllable structure in addition to the phonemic inventory itself. 2. H has obviously taken enormous pains to develop and present the data on which his reconstructions are founded. It therefore grieves me all the more to have...