In lieu of an abstract, here is a brief excerpt of the content:

REVIEW ARTICLE Histoire de l'accentuation slave. By Paul Garde. 2 volumes. (Collection de manuels, 7:2.) Paris: Institut d'Études Slaves, 1976. Pp. x, 525. Reviewed by Morris Halle and Paul Kiparsky, MIT* In his preface to this book, Garde tells us that his purpose was 'to show in broad outline how the accentual systems of the modern Slavic languages, compared with one another and with those of the Baltic languages, derive from a prior "Balto-Slavic" system, and how the latter derives from an anterior "Indo-European" system ...' (p. vii). He discovered that the present state of studies is such that writings on the topic are accessible only to readers who 'are already familiar with the discussions dealing with these subjects'; his original intention was simply to remedy this unsatisfactory state of affairs. But he found it was an impossible task: 'The things that we thought were the results of prior research, thoroughly tested by being subjected to a methodologically coherent exposition, turned out in a great many instances to be inconsistent and self-contradictory, and we were [thus] forced in spite of ourselves, in a host of cases, to take positions which do not correspond exactly to those espoused by any of our predecessors' (viii). This is the only originality to which G explicitly admits. Otherwise, his aims are much more modest—he wants to provide his reader with information about what generations of scholars have securely established (x): 'Such as it is, this book, based entirely on second-hand materials, does not pretend to any new discoveries; it will have attained its aim if, in developing a coherent view ofthe accentology of Slavic (and of Indo-European), it contributes to rendering this body ofknowledge somewhat less esoteric ... and if its readers are never again tempted to ask themselves the ritual question (cf. Lunt 1963): "What are they talking about?" ' It will, we hope, become clear below that G is too modest in his evaluation of his own work. While most of the pieces of the puzzle have been around in the literature for many years, they have never been put together in quite so coherent a fashion, with so few pieces left over and with so few places where the fit is poor. Moreover, on a number of issues G makes important new proposals that are so well-founded that they should become part ofthe standard textbook account of the topic, to be mastered by all students in the field. The clear superiority of G's account over that of his predecessors results from the fact that G's account is based on an explicit descriptive framework which determines in great detail the outlines of his exposition. Although G does not especially insist on the theoretical sophistication of his approach, his empirical results derive in considerable part from the fact that he takes his descriptive framework seriously, and views the facts consistently in terms of the categories and relationships implicit in that framework. In this review we * This work was supported in part by a grant from the National Institute of Mental Health #5 POl MH13390-13. 150 REVIEW ARTICLE151 have attempted to make clear the underlying assumptions of G's descriptive practice, and to show how further advances in understanding the empirical subject matter—the accentual systems of Baltic, Slavic, Sanskrit, Greek, and of the IE proto-language—can come from improvements and elaborations of G's theoretical framework. 1. Garde's framework. G assumes that, in the IE accentual system, three prosodie characteristics—quantity, tone, and accent—played a definite role. 'Quantity', according to G, 'is a non-accentual prosodie feature which is associated with each individual morpheme' (2). G accepts the traditional view that the tonal system had the following properties (5): 'short syllables (containing short vowels) are subject to no tonal opposition. Long syllables (containing a long vowel or a diphthong) can be either acute or circumflex. These interrelations (rapports) affect all syllables regardless of whether or not they are accented ...' Thus there is tonal contrast only among long syllables; short syllables exhibit no contrast in tone, and are treated on a par with toneless syllables. Finally, G's system includes the...

pdf

Share