Language form and language function By Frederick J. Newmeyer (review)

168LANGUAGE, VOLUME 76, NUMBER 1 (2000) cophonology. In effect, there are still levels, but the levels are not extrinsically ordered. If there is a usual order of levels (as there undoubtedly is in most languages with sufficiently complex morphologies), Inkelas and Orgun claim that it is because affixes with a particular cophonology are ordered by other principles of syntax, semantics, or morphology. The propensity for similar or neighboring affixes to undergo similar phonological processes (giving the appearance of adjacency among the affixes of a given stratum) is due, say Inkelas and Orgun, to the historical processes by which affixes were added to a language. Affixes which are added at a particular period in the history of a language take up and fossilize the common phonological processes of that time and also tend to appear adjacent or to be paradigmatic alternates. This is another case of a synchronic description but a diachronic explanation. Finally, there are two shorter papers. Mark Aronoff ('Isomorphism and monotonicity: Or the disease model of morphology', 411-18) suggests that while in syntax, 'juncture strength' increases monotonically with the level of the constituent structure, this is not always so in morphology. For example, since inflectional affixes typically attach outside ofderivational affixes (i.e. higher in the constituent structure), one would expect the juncture between an inflectional affix and the stem to be stronger than the juncture between a derivational affix and a stem, much as a word boundary juncture is stronger than a word-internal juncture. To the extent that this is not true—and Aronoff cites examples to show that it is not always true—morphology is to blame, implying a distinction between morphology and syntax. Finally, the conference organizer, Steven G. Lapointe ('Some remarks on the morphologysyntax interface', 419-25) offers several general comments on various issues brought up in other papers having to do with argument structure, preceding a floor discussion of the relative responsibility of the domains of syntax, semantics, and morphology (but not phonology). The editing has been very well done. Indeed, while I looked for discrepancies between the example numbers mentioned in the commentary papers and discussion sessions, and the numbers in the actual papers, I could find none, despite the fact that some authors apparently revised their papers substantially after the conference. The only significant typo I found is example (33) on page 280, where the association lines are missing; the example appears again in the commentary paper on page 286, this time with the lines drawn in. There is unfortunately no index. REFERENCES Baker, Mark. 1996. The polysynthesis parameter. New York: Oxford University Press. Dixon, R. M. W. 1979. Ergativity. Language 55.59-138. Pinker, Steven. 1989. Learnability and cognition. Cambridge, MA: MJT Press. Prince, Alan, and Paul Smolensky. 1993. Optimality theory: Constraint interaction in generative grammar. (RuCCS Technical Report 2.) New Brunswick, NJ: Rutgers University. Sherrard, Nicholas. 1997. Questions ofpriorities: An introductory overview ofoptimality theory in phonology . Derivations and constraints in phonology, ed. by Iggy Roca, 43-89. Oxford: Clarendon Press. Stump, Greg. 1993. On rules of referral. Language 69.449-79. 7809 Radin Road Waxhaw, NC 28173 [Mike_Maxwell@sil.org] Language form and language function. By Frederick J. Newmeyer. (Language, speech, and communication.) Cambridge, MA & London, England: The MIT Press, 1998. Pp. xii, 428. $40.00. Reviewed by Edith Moravcsik, University of Wisconsin-Milwaukee* The book starts out with a delightful make-believe dialogue between two recent PhDs in linguistics. True to their last names, blissfully genderless Sandy Forman and Chris Funk represent * I am grateful for comments from Jessica R. Wirth. REVIEWS169 opposite sides of the formalism-functionalism debate. Sandy Forman's basic claim is that language structure is best described as an autonomous system; Chris Funk's guiding principle is that language structure has been shaped by semantic and functional forces (7-11). The balance of the book is an argument to show that there is no real conflict between the two positions and that their amalgamation is highly desirable. The line of argument offered in support of this thesis is simple and powerful. Three basic points emerge. (1)The unification offormalism and functionalism is possible in principle because the respective claims are not contradictory...