Volume 30, Issue 1

The concept of convergence, from the study of language contact, provides a model for better understanding interactions between cognitive systems of the same type (for example, in bilingualism, subsystem instantiations of the same kind of knowledge representation and its associated processing mechanisms). For a number of reasons, musical ability is the domain that allows for the most interesting comparisons and contrasts with language in this area of research. Both cross-language and cross-musical idiom interactions show a vast array of different kinds of mutual influence, all of which are highly productive, ranging from so-called transfer effects to total replacement (attrition of the replaced subsystem). The study of music contact should also help investigators conceptualize potential structural parallels between separate mental faculties, most importantly, it would seem, between those that appear to share component competence and processing modules in common. The first part of the proposal is to determine if the comparison between the two kinds of convergence (in language and in music) is a useful way of thinking about how properties of each system are similar, analogous, different and so forth. This leads to a more general discussion about the design features of mental faculties, what might define them “narrowly,” for example.

In this paper, I propose that Croatian relative clauses (RCs) introduced by the complementizer što ‘what/that’ do not form a homogeneous class with respect to their derivation: some are derived by movement, and some are derived by a non-movement strategy. Unless the relativized element is the subject, što- RCs normally require a resumptive pronoun to appear in the site of relativization. However, this requirement is removed under morphological case matching between the head of the RC and the resumptive pronoun: the resumptive pronoun may be omitted if the pronounced head of the RC, case-marked in the matrix clause, appears in the morpho-phonological form that it would have if it were case-marked by the embedded predicate. I argue that the absence of a resumptive pronoun in a što- RC indicates that the RC is derived by movement. I propose that the matching analysis of RCs (Bhatt 2002; Citko 2001; Hulsey and Sauerland 2006; Sauerland 1998, 2002), coupled with a version of the Inverse Case Filter (Bošković 1997, 2002; Martin 1999) and a particular view of inverse (case) attraction offers a natural explanation for why such što- RCs may receive a movement analysis only if the matching requirement is satisfied.

In this paper, I explore the Defective Intervention effects in the past transitive structures in two dialects of Kurdish, where variation is found in how agreement shows up on the verb. In the Northern dialect, in the presence of a Dative subject, the verb displays full agreement with the Nominative object (in person and number), whereas, in the Central dialect, in the same context, the default third person singular agreement obtains. It will be shown that the Defective Intervention as proposed in Chomsky (2000) and assumed in much subsequent work in the Minimalist Program is not adequate to explain the varied pattern of agreement attested in Kurdish past transitive structures. Drawing on the structural differences in the past transitive constructions in the two dialects, I will introduce a modification into the current formulation of Defective Intervention to the effect that in Dative-Nominative structures, agreement with the Nominative DP is blocked by the intervention of a φ-complete head that licenses the Dative DP and not by the intervention of the Dative DP itself. This approach to Defective Intervention predicts that in structures where the Dative DP is licensed by a φ-incomplete head, the Defective Intervention effects will not arise.

The location or even the existence of stress in Korean has been controversial. At the same time, it has been widely assumed that Korean has phonemically long vowels, allowed in the phrase-initial position. This article considers these two key issues of Korean word-level prosody anew. I claim that the long vowel in Chonnam and Traditional Seoul dialect is an acoustic manifestation of the metrical head, and show that an analysis based on accent rather than phonological vowel length yields a simpler account of the vowel shortening in verbal suffixation. I propose that in Chonnam and Traditional Seoul, there is a two-syllable left-edge window, where stress falls on the initial syllable if accented and, otherwise, on the second syllable. Acoustic analyses show that duration, pitch, and intensity all have significant effects in predicting the location of metrical head in Chonnam. In Traditional Seoul, however, only duration and second syllable pitch have such effects. I argue that the weaker relics of metrical prominence in Traditional Seoul reflect the historical shift of its prosodic system from lexical to phrasal. Implications of the metrical head on the Accentual Phrase theory of Korean (Jun 1993) are discussed with focus on the interpretation of pitch in Korean phonology.

Contrastive topics and non-contrastive topics in Japanese generally receive separate treatments in the literature: although they are both marked by the particle wa , the former carry tone prominence and only optionally move to clause-initial position, while the latter (Kuno's [1973] ‘theme’) do not carry tone prominence and typically occupy clause-initial position. This paper presents arguments that contrary to this standard view, topics in Japanese, contrastive or non-contrastive, must occupy clause-initial position. Evidence comes from examining the syntactic and interpretive properties of both types of wa- marked phrases in various discourse contexts. First, a tonally prominent wa -phrase must move to clause-initial position in contexts that require a contrastive topic. Second, in contexts that allow a tonally prominent wa -phrase to remain in situ, the relevant phrase is not interpreted as a topic and cannot move to clause-initial position. Third, tonally prominent wa- phrases displaced to clause-initial position show the syntactic distribution of topics that are predicted by considerations at the syntax-information structure interface, but those in situ do not. Finally, despite the general consensus, there are limited circumstances in which tonally non-prominent wa -phrases can occupy a non-initial position. The article presents syntactic and interpretive evidence that such wa- phrases are also not topics.