Sign, system and function: Papers of the First and Second Polish-American Semiotics Colloquia Ed. by Jerzy Pelc et al. (review)

234 LANGUAGE, VOLUME 62, NUMBER 1 (1986) as a 'retrogression into the morass of prestructuralism ' (109), constraints on the human articulatory and perceptual mechanisms seem to be most relevant in explaining phonological markedness—which, as S overlooks, is itselfan explanandum. Even S's notion of M-assimilation , which supposedly elevates assimilation from the realm of substance to Unguistic form, doesn't explain anything by itself: why, afteraU, should Japanese dental stops be [+ strid] before [ + diff] vowels, or EngUsh vowels [ + short] before [+tense] consonants? Assimüation of mvalues alone can't explain why preciselythese features are involved: only phonetic factors can account for this. Probably the most serious problem with S's theory is that it is too powerful. Often S assigns M-values (at times contradicting others) with Uttle or no justification except that they work. Surely these values must be determined independently ; otherwise they are arbitrary and circular . Surprisingly, however, S disregards this question; he cavalierly claims that circularity is inherent, and rejects demands for independent motivation as denying the hermeneutic nature of the task at hand. This is nothing more than an immunization strategy: without such evidence , how can we tell which claims are vaUd, and which are simply tricks of the analyst to make his theory work? Given seemingly arbitrary M-values, M-assimilation, reversal, and complementarity, there seems little that one cannot 'explain'. Despite S's disclaimers, some constraints are needed. In S's discussion ofRussian morphophonemics, we find such a startling series of arbitrary reversals that the reader begins to wonder if such alternations are not, after aU, mere historical residue; certainly not all MP alternations are semeioticaUy motivated, although this book would make it seem so. What is required is differentiation and refinement of 'markedness'. Instead of a monolithic concept, we need to distinguish different kinds of markedness—or of its opposite, naturalness —and to recognize them as scalar properties . Phonological naturalness cannot be equated a-priori with morphological naturalness , since they are two fundamentaUy different systems (cf. the 'double articulation' of language ). The phonological systems wiU include, inter alia, definitions of natural segments, processes , and syUable structures (these last often appear to be involved in the M-complementarity that S observes). The constraints found here wiU no doubt result mainly from human production and perception mechanisms. The morphological system, following W. Mayerthaler (Morphologische Natürlichkeit, 1981), wiU define markedness ofsymbolization, which is largely based on perception, and markedness of the encoded semantic categories, which indeed corresponds to conceptual complexity—as weU as optimal Unkages between the two, e.g. the natural pairings of semantic categories and parts of speech alluded to above. In this way, we could demystify the heretofore vague and undifferentiated concept of markedness by breaking it down into manageable subfields, hopefully leading to its ultimate foundations in human perception and cognition. Perhaps then one could return to the grand task of a unified semeiotic theory of markedness and Unguistic structure which S envisions . In the meantime, this book remains a valuable study for those interested in markedness , semeiotics, or Russian Unguistics. [Thomas F. Shannon, Berkeley.] Sign, system and function: Papers of the First and Second Polish-American Semiotics Colloquia. Ed. by Jerzy Pelc et al. (Approaches to semiotics , 67.) Berlin: Mouton, 1984. Pp. xiii, 503. Two coUoquia (held in Radziejowice, Poland, May 22-28, 1978, and in Bloomington, Indiana, October 1-3, 1979) produced this large, weUprinted volume. It contains 36 papers, a summary , and a preface (?-xiï) and appendix (493503 ) by Pelc. The summary is of T. A. Sebeok's 'Prefigurements of art', which has appeared in at least three different guises elsewhere. In a brief note it is impossible to even Ust the names of aU the authors (several of whom are represented twice), but a few papers deserve listing: most interesting to me were H. Buczynska -Garewicz & U. Niklas (2 papers) on C. S. Peirce, L. Koj on 'The empirical basis of Unguistics' and on 'Noncompositional semantics ', D. Patte on narrative semiotics, E. Stankiewicz on modern poetry, and L. Waugh on the Unguistic sign. Also valuable and thoughtprovoking were R. Kevelson on Peirce, I. P. Winner on theories ofnarration, K. Pisarkowa on "The use ofwords', and Pelc's...