Linjedansere og pantomine på sirkhus: Folkeetymologi som morfologisk omtolkning. [Tight-rope artists and pantomime at the circus: Folk etymology as morphological reinterpretation.]By Helge Gundersen (review)

894 LANGUAGE, VOLUME 73, NUMBER 4 (1997) troductory Ch. 1 succinctly but adequately presents the vast literature on the subject and outlines the author 's own approach (1-33). Chs. 2-6 deal with direct discourse utterances that comprise one, two, three, more than three, and volitive clauses respectively (34-231). Ch. 7 discusses the verbal forms in narrative context (232-97), and Ch. 8 deals with 'Tense in the subordinate clause' (298-318). The main conclusion of this work, summarized m the ninth chapter (319-25) is that the qatal form is marked for past tense and complete aspect, whereas yiqtol is marked for nonpast tense and incomplete aspect (exception being made for Stative verbs, to which these distinctions do not seem to apply). As to the simple and compound forms, E claims that their use is conditioned by the functional opposition of sequentiality vs nonsequentiality. In a sequential context, biblical Hebrew uses wa-yiqtol instead of qatal, and we-qatal instead oí yiqtol. '[T]he non-sequential form stops the flow of the story (i.e. stand still), whereas the sequential form lets the story flow on' (321). Further, these functions are not affected by the position of the verb in the clause—initial (default ) or noninitial. 'Rather, the fronting ofa constituent of the clause seems to function as a topicalization' (320). This book largely confirms conventional wisdom on the subject, synthesizing several lines of thought. Still, the present reviewer finds it a bit disappointing. There is a certain imbalance between the treatment of direct discourse and narrative material, the argument sometimes lacks precision, and the material hardly allows for generalization. Too much attention is paid to translations. Finally, the book shows less advantageous evidence of camera-ready submission to the publisher. [Arian J. C. Verheii, Netherlands Organization for Scientific ResearchlVrije Universiteit, Amsterdam.] By word of mouth: Metaphor, metonymy and linguistic action in a cognitive perspective . By Louis Goossens, Paul Pauwels, Brygida Rudzka-Ostyn, Anne-Marie Simon-Vandenbergen, and Johan Vanparys. (Pragmatics and beyond, 33). Amsterdam & Philadelphia : John Benjamins, 1995. Pp. xii, 252. The publication of Lakoff and Johnson's Metaphors we live by (Chicago & London: University of Chicago Press, 1980) laid the foundation of what came to be known as the cognitive approach to metaphor . By word ofmouth, a collection of seven corpusbased studies, is firmly situated within this tradition. The authors' main concern is to elucidate the nature of metaphorical and metonymical expressions relating to linguistic action. The following question figures prominently throughout the volume: How is linguistic action expressed in English? Some papers have brought to the fore intriguing details which usually escape our notice. Vanparys undertakes an investigation into the source domains available for conceptualizing linguistic action. The author marshals ample counterevidence against Michael Reddy's claim that our language is dominated by the so-called conduit metaphor. In his lucid essay on 'metaphtonymy' , Goossens sharpens our sensibility for the interaction of metaphor and metonymy. Of particular interest is Rudzka-Ostyn's paper. She presents a stimulating analysis ofschematization processes underlying metaphorical transfer. This book has a number of shortcomings in presentation . Most papers are too long to hold the reader's attention. Frequently, the authors present painstaking accounts of individual metaphorical expressions in terms of donor domains or image-schemas. The generalizations offered are not always illuminating. In general, the reader will often encounter difficulties in discerning the theoretical import ofthe investigations at issue. Many papers dwell on irrelevant details—to the point of elevating mere footnotes to the status of major findings. For example, Pauwels and Simon-Vandenbergen expand on the 'dichotomy ' between 'context-dependent' and 'contextindependent ' value judgments as expressed by linguistic action metaphors. Fumble, which metaphorically refers to 'clumsiness in speech' (58), is cited as a context-independent value judgment since the expression is generally derogatory. The authors set great store by the observation that all value judgments may be reversed in certain contexts In a subsequent article by Simon-Vandenbergen, which is exclusively devoted to value judgments, this finding is supported by a dubious example. For Simon-Vandenbergen the prototypical positive value judgment associated withfluency is reversed in sentences like / hate his fluency (because I'm jealous of it...