This article attempts to provide a precise definition for topic and to derive most of the properties of topic from this definition. The main assumption is that the topic-comment construction is a syntactic device employed to fulfill certain discourse functions. Topic is always related to a position inside the comment. Since topic has no independent thematic role but always depends on an element inside the comment for its thematic role, it has no syntactic function of its own. This dependence relationship is subject to locality constraints.

This article has two main parts. In the first, the subject pronoun in a cleft sentence together with the cleft clause is shown to function pragmatically as a discontinuous definite description. Applying the givenness hierarchy (Gundel et al. 1993) makes it possible to explain the distribution of this-clefts and that-clefts in discourse, and predicts the more frequent occurrence of it-clefts. Clefts also semantically share existential and exhaustiveness conditions with definite descriptions. The second part presents a new syntactic analysis of clefts, which treats the cleft clause as an extraposed complement of the cleft subject pronoun, adjoined to the clefted constituent.

Some Korean-Japanese comparisons involving Japanese coronals complicate the internal reconstruction of pre-Old Japanese. Post-OJ verb forms that end uniformly in, for example, ki have distinct OJ final syllables (ki ≠ kwi) according to the form or paradigm of the verb. This is not true for OJ syllables like ti, but scholars have assumed that pre-OJ *ti ≠ *twi, etc., were distributed in corresponding verb forms in the same way as OJ ki ≠ kwi, etc. Whitman, however, has introduced K-J etymologies requiring that pre-OJ *ti > si, *ri and *ni > i, and hence that all OJ ti < *twi, etc. These conflicting results can be resolved if other pre-OJ sound changes supported by Korean etymologies are properly integrated into the internal reconstruction of Japanese verb paradigms.

In this descriptive report we outline the structural pattern of exclamative clauses in Paduan. Because of the close similarity between exclamative and interrogative clauses in this language, we begin by developing a number of tests which allow us to distinguish these two clause types. We then present the range of exclamative structures. A variety of factors interact to mark a clause as an exclamative, yielding a quite complex array of facts. We view this work as the basis for future study in the syntax and semantics of exclamatives.

In this article we argue that there is a large class of expressions, typified by plastic flower, stuffed animal and kosher bacon, that have a unique semantics combining compositional, idiomatic and decompositional interpretation. These expressions are compositional because their constituents contribute their meanings to the meanings of the wholes; they are idiomatic because their interpretation involves assigning dictionary entries to nonterminal elements in their syntactic structure; and they are decompositional because their meanings have proper parts that are not the meanings of any of their syntactic constituents. We argue that extensionalist semantics, on which the meaning of an expression is a function from domains to extensions in those domains, cannot provide an adequate account of the semantics of these expressions, and that supplementation with a theory of pragmatic interpretation does not improve the situation. We show how our account explains the intensionality and the productivity of these expressions.

The authors propose that the typologically uncommon combination of labial and palatal constriction in Twi has arisen from a convergence created by general patterns of coarticulation of consonants and vowels. This convergence has been systematized in a (consonantal) acoustic dimension partially independent from the original (vocalic) dimensions of contrast for which the rounding and palatal articulations were specified. These conclusions are based on an examination of distributional patterns, palatograms of the articulation of secondarily articulated consonants, and acoustic analyses. Contrastiveness and quantal considerations can be seen as contributing to the occurrence of typologically odd sounds, provided one keeps in mind how an articulatory gesture functions within a language's contrastive system.

This article is a reply to Kuno et al. 1999, which claims that a structural approach to scope should be replaced by an expert system. But the alleged theoretical and empirical problems faced by the structural accounts for scope are based on assumptions or interpretations that are not adopted in the structural accounts. Further, there are problems with the characterization and execution of the expert system, causing difficulty in the understanding and application of the system intra- and interlinguistically; the expert system is not empirically adequate and does not accommodate idiolectal variations. Finally, the expert system misses important correlations between scope and other properties in the grammar, such as binding, that follow straightforwardly from a structural approach. A structural approach to scope should not be abandoned in favor of an expert system.