Grammar in the construction of texts Ed. by James Monaghan (review)

200 LANGUAGE, VOLUME 66, NUMBER 1 (1990) here to be novel. (There is a useful glossary of terms as used by the author at the end of the book.) I do find the author's discussion ofageneral notion of emphasis as a systemic construct in Indo-European (84ff.) and his conclusion about the IE grammatical system (1 16ff.) stimulating , but the specific rules (117-19) are too vaguely formulated and programmatic to be useful. A text or handbook of method this book is not. It is a rather vacuous and imprecise exercise in grand theorizing and a quite stimulating collection of speculations on specific methods with extensive illustration from IE and Italic. I leave it to the specialists to judge the analyses. [M. Lionel Bender, Southern Illinois University .] Grammar in the construction of texts. Ed. by James Monaghan. (Open Linguistics Series.) London: Frances Pinter, 1987. Pp. 155. $35.00. This slim volume contains 12 short articles based on papers presented at a 1983 Conference on Discourse Structure at The Hatfield Polytechnic Institute, in cooperation with the British Association for Applied Linguistics. All of the articles, representing a number of European linguistic traditions, deal with 'text', broadly defined , and/or with grammar; but, beyond this similarity, they form a rather diverse collection. The articles are grouped into three sections of four papers each: 'Discourse structures', 'Conversational strategies', and 'Comparative discourse studies'. The four papers under 'Discourse structures' address the issue of text 'meaning' and how readers or listeners understand text, but only three of them consider actual textual discourse. Paul Georg Meyer offers a theoretical and programmatic discussion of how hierarchical structure is signalled in technical discourse, focussing on the important role of conjunctions and lexical choice in signalling paradigmatic and syntagmatic relations between 'text elements'. Werner Hullen's 'On denoting time in discourse ' presents the suggestive outline of an analysis of the respective roles of tense markers and time adverbials in signalling time in the London -Lund corpus ofspoken English, suggesting, for example, that 'past-passages' are much more likely to be marked by concrete time references than are 'present-passages'. The third paper in this section, by Hans Arndt, is concerned with 'functional indicators' as aids in interpreting what an utterance does in social interaction. He concentrates on expressions such as / think in signalling 'expressive' (/ think he's a fool) vs. 'assumptive' (/ think he's a carpenter) functions , and shows how such distinctions are exploited in written fictional dialogue. The final paper in this section, Ivan Lowe's 'Two ways of looking at causes and reasons', discusses causal connectives in terms of several issues regarding causality—a scale of causality, the distinction between rationalizing events and rationalizing speech acts, and the distinction between necessary and sufficient conditions. The papers in the second section, 'Conversational strategies', are concerned with conversational structure. Bengt Altenberg provides a clear and convincingly argued discussion of the ordering of clauses expressing causal relations in conversational English in terms of four ordering principles, noting several recurrent interactive strategies employed by conversationalists . Anna-Brita Stenstrom offers some tentative answers to the question of her title, 'What does really really do?' by describing really in written and spoken English in terms ofits prosody and position in the clause. Margaret MacLure considers 'summons-answer' sequences in adult-child talk, persuasively showing how adults interact with 18-36-month-old children in such a way as to ensure topic collaboration and to repair trouble in getting a topic started. The last conversational paper, by Gillian Brown, offers a novel elicitation procedure for approaching the question of information packaging . In analyzing the interaction between pairs of subjects with two slightly different versions of the same map. Brown is able to draw interesting conclusions about the relationship between shared knowledge and the use of definite and indefinite noun phrases in English. The final section of the book is slightly misleadingly labelled; though each of the four papers in this section at least alludes to more than one language, only one of them is actually concerned with problems of text structure in two or more languages. This is the paper by Monika Krenn, who defines 'extended' reference as 'reference to ... a process or...