Meaning and mental representations Ed. by Umberto Eco, Marco Santambrogio, and Patrizia Violi (review)

Abstract

224 LANGUAGE. VOLUME 68. NUMBER 1 (1992) Ch. 4 describes D's cross-sectional study, which aims to remedy these theoretical deficiencies . The study traces the emergence of speech acts and conversational meanings in the language of 19 children aged between 3;6 and 7;0 years, against the template of their social development. The linguistic task involves the children in conversation with their mothers as they negotiate the rules of a board game. The sociocognitive tasks, based on responses to a series of pictures, focus on the abilities to make inferences about others' psychological perspectives , to connect events, and to understand social contexts. Chs. 5 and 6 discuss the results of the linguistic and the sociocognitive tasks, respectively . Speech acts are found to differ significantly with age as children become more equal conversational partners, and conversational relevance and adaptability to the hearer improve dramatically. The sociocognitive tasks also show age-related improvements in the ability to take roles and to empathize. Significant correlations were found between all the variables for conversational competence and social knowledge, supporting D's thesis about the interdependence of these two developmental areas. Finally, in Ch. 7 D acknowledges some limitations of the present study, but goes on to define the large-scale implications of her findings . These include thoughtful insights about the facilitating role that caretakers and educators can play in social-interactional development by respecting children as creators of their own meaning. Theoretically, this book may not meet its own requirements of comprehensiveness, at least in the linguistic analysis. For example, only certain elements of Grice's four maxims are considered ; and politeness theory is limited to a summary mention (135-6) of universal cooperative strategies. Although D stresses the primacy of intersubjective negotiation in conversation, she explicitly sidesteps such features as turn-taking and repair in the empirical study. Since she herself acknowledges that speech-act theories alone cannot capture 'the properties of natural dialogue' (132). her decision to overlook ethnomethodological methods that can capture them is unfortunate. Nevertheless, this book will offer valuable insights for child-language specialists and developmental theorists. To pin down the relationship between linguistic and social development is a large and very important challenge , and D takes it on here with imagination and lucidity. [Elizabeth O'Dowd, University of Colorado.] Meaning and mental representations. Ed. by Umberto Eco, Marco Santambrogio, and Patrizia Violi. (Advances in Semiotics.) Bloomington: Indiana University Press, 1988. Pp. 237. Cloth $29.95, paper $12.95. In this volume ten original essays present new approaches to the definition of meaning and mental representation. An introduction (3-22) by Marco Santambrogio & Patrizia Violi provides background information on AngloAmerican views of meaning as well as on European semiotics. Although the essays seem to fall into three groups, they are arranged alphabetically according to the authors' names, with no clear progression from one article to the next. Three essays attempt to refine truth-conditional semantics: 'On the circumstantial relation between meaning and content', by Jon Barwise (23-39); 'How is meaning mentally represented ?', by Philip N. Johnson-Laird (99118 ); and 'Identity in intensional logic: Subjective semantics", by Bas van Fraassen (199218 ). A fourth essay, 'On truth, a fiction', by Umberto Eco (41-59). dismisses truth-conditional semantics. Three articles propose cognitive models of mental representation: Ray Jackendoff's 'Conceptual semantics' (81-97), George Lakoff's 'Cognitive semantics' (119-54), and Gilles Fauconnier's 'Quantification, roles and domains' (61-80). The essays by Yorick Wilks ('Reference and its role in computational models of mental representation ', 221-37), Wendy Lehnert (The analysis of nominal compounds', 155-79), and Roger Schank & Alex Kass ('Knowledge representation in people and machines', 181-200) are concerned with meaning as it relates to computer programs emulating human cognition. The essays, though accessible to the reader with a basic knowledge of semantics, are not overly simplistic. The quality of the articles is uneven. The chapters by Barwise, Eco, and Schank & Kass stand out as clearly written and thoughtful. Those of van Fraassen and Faucon- BOOK NOTICES 225 nier are convoluted and often difficult to follow, due to excessive formalism and an insufficient amount of clear explanation. Several of the articles fail to be convincing. Johnson-Laird argues...