Hebrew Studies 33 (1992) 119 Reviews Though nagging. these problems do not undermine the values and virtues of the study. What a pleasure to read! Darr writes with clarity, economy. and substance. She knows scholarship and how to teach it. Synagogues. churches. and introductory college courses can benefit from her work. Not least. the graciousness of its demeanor and the generosity of its vision are a welcome gift in an age of verbal assault. Phyllis Trible Union Theological Seminary New York. NY 10027 HEBREW LINGUISTICS: A JOURNAL FOR HEBREW DESCRIPTIVE, COMPUTATION AL, AND APPLIED LINGuIsTIcs. No. 31-32. Maya Fruchtman, ed. Pp ix + 114. Ramat-Gan. Israel: Bar-Han University, 1991. Paper. This double issue contains six articles (and a response to one of them) in Hebrew. with English abstracts and one short correction-note in English regarding the inappropriateness of characterizing Jewish languages as pidgins/creoles. The emphasis is on computational linguistics and discourse analysis. The issue opens with an article by Michal Ephratt, which proposes an algorithm for recognizing linguistic jokes, such as puns, which rely on multiple interpretation of sentences, where the "punch" is attained by the gap between the normal, "least costly" reading and the least expected. "most costly" one. Hanna David, and Hillel Weiss in his comments on her article, discuss a more general computational issue involving the application of complex mathematical models to analyze and characterize literary texts versus the use of the computer as a tool for (a) storing extensive. complex bodies of literary corpora. and (b) subsequent pulling out data that are relevant to precise determination of literary hypotheses. Weiss points out that use of mathematical algorithms in computational literary analysis has become marginal and that most computational work today centers on the building up of extensive literary data bases. At the same time. he outlines his own computational algorithm for distinguishing between poetry and prose. Hebrew Studies 33 (1992) 120 Reviews Zahava Goldstein and Michael Moore test a mathematical model for predicting active vocabulary on Hebrew-speaking children. In Israel, evaluation of vocabulary has always been performed by sampling words out of a dictionary and asking what they mean-an unsophisticated, often misleading procedure. The model applied here provides for reliable prediction of individuals' active vocabulary based on frequency of distribution in written or spoken samples. Yitzhak Zadka's note is a comment on an earlier article in Hebrew Linguistics, which discusses the modal meaning of ~eyn ~el mi lifnot ("there is nobody to tum to"). Zadka points out that the modality of this structure is not restricted to existential sentences of this type; rather, it covers a variety of patterns involving an attributive infinitive. Yitzhak Roeh and Raphael Nir demonstrate how Israeli news discourse tends to employ indirect speech to assure "objectivity," while keeping direct speech transmission to a minimum. It is therefore of particular interest to study partial deviations from the indirect speech standard in the news, as manifest in the use of direct speech elements in indirect speech or in mimetic direct speech. Such departures are permitted only to the extent that their effects (whether empathy, respect, etc., or suspicion, irony, etc.) conform with the "national consensus" and mainstream ideology, reflecting notions of "appropriateness," norms and values-and their hierarchical ranking. Lea Sarig shows how discourse analysis can account for linguistic dissimilarities emerging from comparison of a translation with its source text. "Discrepancies" in employing means of cohesion in translation from Arabic to Hebrew can be attributed to the translator's attempt to improve cohesion of discourse in the target language. Thus, grammatical anaphora may be replaced by lexical repetition/variation when the distance between the antecedent and the anaphor is substantial; the opposite conversion may occur when repetition is felt to be redundant. Connectives may be deleted, replaced by other grammatical items, or added, depending on the translator 's sense of optimal cohesion in the target language. The strength of Hebrew Linguistics continues to be in its interdisciplinary nature and in its offering Hebraists and scholars from other disciplines who wish to contribute to Hebrew language study a forum for discussion and exchange. With the general increase in interdisciplinary research and the "coming of age" of the...