Volume 26, Issue 4

The purpose of this article is to review recent proposals in phase theory. The notion of phase has acquired a central position in recent minimalist thinking, with several core concepts of minimalism intimately linked or even derived from phases. Notions such as cyclicity, movement and Transfer, the way the syntactic system interacts with the sensory-motor and conceptual-intentional systems of the brain, all are directly linked to the phase. We show how the notion of phase has been the driving force behind a lot of syntactic theorizing in the past few years, and what new ideas it has inspired. With new ideas also come new problems and questions. The second part of this article discusses several issues that the approach raises. Some of these find natural and straightforward answers, others seem more problematic. For example, parallel probing, one of the more conspicuous changes in recent models, raises practical problems for the No-Tampering Condition, and seems to require a theoretical move in the direction of a representational model of syntax. Furthermore, the empirical data it is based on appear to be questionable. We believe that it is not actually necessary to adopt parallel probing, and we therefore suggest an alternative that generalizes the notion of edge feature and makes use of Spec-Head instead of Agree as the main type of agreement operation.

It is argued here that an appealing approach to the distribution and behavior of rhotics in Catalan can be had by appealing to the maintenance, and the perceptual goodness, of contrast. Since the Catalan facts are very similar to those of Spanish, the relevance of these ideas to Spanish is also discussed. The analysis is cast within Dispersion Theory (Flemming, Auditory representations in phonology, Routledge, 1995, Flemming, Contrast and perceptual distinctiveness, Cambridge University Press, 2004), in a version following Padgett (Natural Language and Linguistic Theory 21: 39–87, 2003a, The emergence of contrastive palatalization in Russian, Kluwer Academic Press, 2003b). The analysis is interesting because it (i) provides novel insight into the rhotic facts, (ii) extends dispersion-theoretic thinking into the area of consonants, and (iii) contributes to a program of research showing how allophonic generalizations can be explained by appeal to contrast.

In light of the data regarding the focusing of the particle dake ‘only’, this article shows that in Japanese, wa -marked topic phrases are placed in the CP domain, and argues that this can take place either in overt constituent structure or in LF. Other ordinary phrases – including ga -marked major subjects that are sometimes assumed to reside in the same clause-peripheral position as topics – are shown to be located within TP. It is also argued that Japanese shows no evidence for “overt” wh -movement, despite Takahashi's claim (Natural Language and Linguistic Theory 11: 655–678, 1993, Journal of East Asian Linguistics 3: 265–300, 1994) to the contrary. The discussion suggests that the prominent use of the CP domain is subject to typological variation, and that Japanese, unlike English, is a topic-prominent language, in which a topic is placed in the CP domain – i.e., TopP located above TP – by the LF output in order to form a topic-comment structure.