Introduction to quantitative analysis of linguistic survey data: An atlas by the numbers By William A. Kretzschmar, Jr. and Edgar W. Schneider (review)

BOOK NOTICES Introduction to quantitative analysis of linguistic survey data: An atlas by the numbers. By William A. Kretzschmar , Jr. and Edgar W. Schneider. (Empirical linguistics series, 1.) Thousand Oaks, CA: Sage, 1996. Pp. xii, 212. $49.95. This volume is the first in a series devoted to the study of empirical data. The series is intended to promote work from such areas of study as discourse analysis, sociolinguistics, dialectology, and corpus studies. This book focuses on the analysis ofdialectological data of the type collected for linguistic atlas projects, with particular application to the data from the Linguistic atlas of the middle and south Atlantic states (LAMSAS). The authors devote much of their book to describing the design process for the LAMSAS computerized database, and, thus, it reads as a kind of narrative history of this ambitious project. Along the way, some broader issues raised by the project are addressed , though the focus generally remains on explaining the decisions made by the database designers. The first chapter provides some background information about the database and the original LAMSAS project including a refreshingly balanced discussion comparing the methods and objectives of traditional dialectology with those of Labovian sociolinguistics. The current database project is viewed as bridging the gap between these fields as it makes possible quantitative analysis of linguistic atlas data. Some of the assumptions of such analysis are addressed in the discussion of fundamental statistical principles that closes the chapter. Ch. 2 considers the appropriateness of applying statistical testing to the LAMSAS data. Certain aspects of the data collection procedures (e.g. how informants were selected, how responses were recorded) impose limits on which statistical techniques can be employed and how the results are to be interpreted. Still, the authors make the case that the data meet the conditions for statistical analysis. Chs. 3 and 4 discuss how the raw field records are computerized and how statistical procedures can be applied to the database. This account provides readers with a sense of the types of retrievable information though the discussion is somewhat overwhelming in its technical detail. The database is put to the test in the final chapter, which offers a demonstration analysis of two lexical files. This presentation examines the regional and social distributions of the linguistic forms and serves to illustrate how these tools expand the possibilities for analyzing linguistic atlas data. However, the discussion also serves to remind readers of certain limitations related to this dataset (e.g. the rural bias). One wonders, for example, about the value of being able to detect statistically significant differences in the distribution of assorted terms for 'cow pen' (particularly when the data represent the speech situation from over 50 years ago). If there are broader lessons to be drawn from such analyses, the authors do not take adequate steps to indicate them. The authors are to be applauded for offering such a careful and detailed account of their experiences. The account will be most valuable to those who make use of the LAMSAS data and to those involved in designing similar databases. The narrow focus of the book as a kind ofcase study ofdatabase development may not appeal to other readers, even to those otherwise interested in linguistic variation. [Matthew J. Gordon, Purdue University Calumet.] The emergence of semantics in four linguistic traditions: Hebrew, Sanskrit, Greek, Arabic. By Wout Van Bekkum, Jan Houben, Ineke Sluiter, and Kees Versteegh. Amsterdam: John Benjamins , 1997. Pp. ix, 322. This book is the result of an innovative enterprise: Four Dutch historians of linguistics and experts in various languages and linguistic traditions from four different universities in Holland cooperated on this joint venture and compared four major linguistic traditions . As they explain in the preface, the common denominator of the Greek, Hebrew, Sanskrit, and Arabic traditions is religious texts and their preservation (v). This volume discusses the development of both linguistics in general and semantics in particular in each of these traditions. Van Bekkum's "The Hebrew tradition' (3-47) notes that semantics and exegesis always went together in the biblical, rabbinic, and medieval periods. Two grammarians stand out in his presentation: (1) the Egyptian Saadiah Gaon (882-924), who used the Arabic term taTwrl...