Telling the American story: A structural and cultural analysis of conversational story telling By Livia Polanyi (review)

734 LANGUAGE, VOLUME 62, NUMBER 3 (1986) knowledge in the internal lexicon, referred to as ILRRR ('the internal lexicon ... as the cognitively reflected representation of reality'; 29). Based on the ILRRR are 'semantic-logical' and 'action-oriented' inferences. It is not clear to me how a theory of discourse meaning will contain these three components in a systematic relationship ; V pays no attention to the matter, because he is describing 'discourse processing'. V emphasizes that inferences about the natures , significances, and relationships ofhuman actions are relevant to the establishment of textual coherence. But surely it is not just human actions that have this value; potentially, it is everything of which humans have knowledge. As empirical evidence for Vs theory of discourse processing, he surveys some experiments with disordered language. He reports that, for 85% of Broca's aphasies, syntactic markers are overridden by ILRRR inferences; however, schizophrenics, Wernicke's aphasies, and global aphasies all have disordered ILRRR's—so that when they talk to normals, cooperative coherence is impeded on both sides. V speaks ofWernicke's and global aphasies as having problems with action processing because of their disordered ILRRR's; he does not mention whether schizophrenics also have apraxic problems, though his theory predicts that they should. This is an interesting book, though it contains no surprises. It is practically oriented, but requires theoretical underpinning. Each chapter is summarized, which is good; but this obviates the need for a final chapter which summarizes the arguments yet again. [Keith Allan, Monash University.] Coherence and composition: A symposium . Ed. by Nils Erik Enkvist. (Publications of the Research Institute of the Âbo Akademi Foundation , 101.) Âbo, Finland: Âbo Akademi , 1985. Pp. 134. Here are five papers from a March 1984 symposium at Âbo, the purpose of which was to study the analytical tools for describing and teaching coherence in written composition, primarily in EFL. Enkvist's 'Coherence, comprehension , and text linguistics' very briefly surveys four models for textual cohesion: sentencebased (M. A. K. Halliday & R. Hasan), predication -based (G. Källgren), cognition-based (R. Schank), and interactional (H. P. Grice; P. Brown & S. Levinson). E concludes that 'Connexity is ... a matter of an interaction between the text and the receptor's knowledge base' (16), and that 'every text should be evaluated in relation to its purpose and situation' (20). L. Bjork, 'TUAP and the teaching of writing in Sweden', argues that students' inability to write coherent compositions results from a lack of instruction, and proposes a five-point program for instructors; he notes that coherence is improved by organizing the content before writing , and emphasizes that auto-/alter-reviewing ofpreliminary drafts improves the coherence of a final draft. L. Evensen, 'Discourse-level interlanguage studies', describes an on-going survey of Norwegian school children's EFL composition , including a pilot study on the use of connectives. The outstanding paper in the collection is A.-C. Lindeberg, 'Cohesion, coherence patterns, and EFL essay evaluation': in a nicely presented analysis of EFL essays, 10 judged good and 10 bad by native assessors, she shows that suchjudgments are not indicated by the number of cohesive ties à la Halliday & Hasan, but by the functional roles of utterances within a text—e.g. as asserting, specifying, giving a result, cause, contrast etc. Although L finds that different topics require different sets of functional strategies, poor essays have too many strings of assertions without further development of what is asserted through specification , explanation, contrast etc. Finally, examining a corpus ofstudent papers, E. Wikborg exemplifies 'Types of coherence breaks in university student writing' with unclear inference ties, misleading connective and paragraphing, unheralded topic shifts, and unspecified topics. The symposium was prompted by the inability of some students, even postgraduates, to write coherent texts—whether in their Scandinavian mother tongues or in EFL. There is a need to discover why we judge that a text lacks coherence , so as to establish guidelines for teaching composition. This book takes a small step toward achieving that aim. [Keith Allan, Monash University.] Telling the American story: A structural and cultural analysis of conversational story telling. By Livia Polanyi. Norwood, NJ: Ablex, 1985. Pp. xii, 158. $29.50. BOOK NOTICES 735 This book has two themes; the 'linguistic...