676 LANGUAGE, VOLUME 74, NUMBER 3 (1998) absolute duration of the plosive and the duration of the vowel relative to the VC dyad—correlate with perception; and (2) that when absolute vowel duration is taken into account, stimuli are reliably differentiated according to plosive duration or F-value (188). Yet since the relative duration of V to C suffices for identification/perception of the quality of C for the words in Ws sample, measures such as absolute duration of C and relative duration of V and C in a dyad are superfluous (195). In the final chapter W relates his findings to Kohler's universalist model of a lenis/fortis distinction based on a concept of consonantal power and on constant duration of VC dyads based on V complementarity with (longer) fortis or (shorter) lenis C. W all too briefly concludes that his findings do not support Köhler's model, in fact render reliance on the power feature for the lenis/ fortis distinction superfluous. Substantial appendices (including questionnaires and histograms for perceptual test items) allow the reader to reexamine much of the data and methodology. The work ends with a short bibliography. Limited sample size prevents this study from being definitive, which the work does not claim to be. However , it may, as it intends, inspire more thorough examinations of the lenis/fortis distinction by indicating fruitful directions for future research efforts. [Désirée Baron, University ofMichigan.] Natural phonology: The state of the art. Ed. by Bernhard Hurch and Richard A. Rhodes. Berlin: Mouton de Gruyter, 1996. Pp. xii, 348. Cloth DM 168.00. This collection is divided into theoretical and descriptive sections plus an introduction by the editors (vi-xii) and a commentary, 'Natural phono(morpho )logy: A view from the outside' (1-38), by Rajendra Singh. Newcomers to natural phonology (NP) should start with the introduction and Wolfgang U. Dressler's 'Principles of naturalness in phonology and across components' (41-51). However , all the papers are accessible to anyone with average phonological training since NP does not make use of arcane notation or theoretical concepts. Nearly all the data cited are from European languages . Originated by David Stampe in the 1970s, the central idea of NP is that languages differ in the extent to which they suppress 'universal' phonological processes. This has obvious affinities with the currently popular view that languages differ in their ranking of 'universal' constraints. However, this collection dates from a conference which took place back in 1990, so comparisons with optimality theory must be supplied by the reader. Several of the papers deal with the syllable. Katarzyna Dztubalska-Kolaczyk ('Natural phonology without the syllable', 53-72) wants to replace syllables with beats. Marianne Kilani-Schoch ('Syllable and foot in French clipping', 135-52) and Elke Ronneberger-Sibold ('Preferred sound shapes of new roots: On some phonotactic and prosodie properties of shortenings in German and French', 261-92) illustrate and attempt to explain the distribution of open and closed syllables in clippings such as manif 'demonstration' and Fundi 'fundamentalist Green'. Michèle Loporcaro ('On the analysis of geminates in Standard Italian and Italian dialects', 153-87) argues persuasively that even 'true' geminates must be divided between codas and onsets rather than assigned entirely to onsets. Especially interesting is Eva Mayerthaler's exploration of 'syllable-timing' versus 'stress-timing' ('Stress, syllables and segments: Their interplay in an Italian dialect continuum', 201-21); she argues that this distinction is really a matter of whether the contrast between successive segments or that between successive syllables is optimized. Stress and its segmental effects are considered by Liliana Madelska and Wolfgang U. Dressier ('Postlexical stress processes and their segmental consequences illustrated with Polish and Czech', 189-200), Richard A. Rhodes ('English reduced vowels and the nature of phonological processes', 239-59), and Bernhard Hurch ('Accentuations', 73-96), who argues that primary word accent has priority over foot structure, so that it is a mistake to see primary stress merely as a matter of which foot heads the word. Other articles are by Carmen Pensado ('Portuguese secondary nasal vowels and phonological representations', 223-37), Julián Méndez DosuNA ('Can weakening processes start in initial position? The case of...