Recent developments in syntactic theory suggest that phrase structure is cross-linguistically more uniform than assumed so far, and that the order spec-head-complement may be the only permissible one. The present article takes issue with this view, showing that the derivation of final complementizers from initial ones by means of IP-raising faces serious difficulties. The discussion focuses on Bengali and similar languages which may be called ‘hybrid’ because both orders, IP-C as well as C-IP, are attested. Five arguments are raised which indicate that these orders are not derivationally connected. The discussion bears results which may also be of interest for linguistic typology, language change and acquisition.

In this paper, we will argue that the LF Object Shift analysis of the infinite regress resolution of Antecedent Contained Deletion is untenable conceptually and empirically. Generalizing Baltin's (1987) CP Extraposition analysis, we will instead propose that any type of rightward movement, including NP and PP Rightward Shift, is available for the infinite regress resolution. Furthermore, we will show that our generalized rightward movement analysis explains a wider range of data than the LF Object Shift analysis. Finally, we will address a question as to why rightward movement, which is non-feature checking movement, is triggered/involved in Antecedent Contained Deletion. We will suggest a way to derive this property under the assumptions of Chomsky's (1993, 1995) Minimalist Program.

This is the second of a series of three papers that, taken together, will give an essentially complete account of inflection in standard German. In this paper we present that part of the account that covers nouns, one that captures all the regularities, subregularities and irregularities that are involved, but with a focus on the subregularities. Inflected forms are defined in terms of their syllable structure, as proposed in Cahill (1990a, b, 1993). The analysis is formulated as a DATR theory – a set of lexical axioms – from which all the relevant facts follow as theorems. DATR is a widely used formal lexical knowledge representation language developed for use in computational linguistics.

This article challenges the commonly-held assumption that words, feet and syllables fit into a single phonological hierarchy, and in particular that feet are always sub-units of words. A number of facts of English rhythm are cited which cannot be accounted for unless words and feet are taken to be units of different hierarchies: many of the differences of rhythm between phrases of otherwise similar phonological structure which are noted by Abercrombie (1964) are accounted for most naturally by differences in placement between, on the one hand, word- and stress-group-boundaries, and, on the other, foot-boundaries.

I investigate a diverse set of Czech words of unclear prosodic status. The analysis reveals that while it is possible to make a clear distinction between proclitics and nonclitics, certain nonclitics are disposed to occasional clitic-like behavior. I propose to capture this fact by distinguishing between inherent clisis vs. derived clisis, the latter being sensitive to certain semantic, pragmatic, and prosodic conditions. This approach allows us to address the striking similarities in certain behaviors between clitics and nonclitics without ignoring their equally important differences.

The paper explores the syllabic and segmental dimensions of phonological vowel disorder. The independence of the two dimensions is illustrated by the case study of an English-speaking child presenting with an impairment which can be shown to have a specifically syllabic basis. His production of adult long vowels displays three main patterns of deviance – shortening, bisyllabification and the hardening of a target off-glide to a stop. Viewed phonemically, these patterns appear as unconnected substitutions and distortions. Viewed syllabically, however, they can be traced to a single underlying deficit, namely a failure to secure the complex nuclear structure necessary for the coding of vowel length contrasts.

The hypothesis of an Altaic language family, comprising the Turkic, Mongolic, Tungusic, Korean and, in most recent versions, Japanese languages continues to be a viable linguistic proposal, despite various published claims that it is no longer accepted. A strong body of research continues to appear, developing and refining the hypothesis, along with publications that argue against a demonstrated relationship among these languages. This paper shows that many of the arguments against a genetic relationship fail to address the criteria demanded in modern historical linguistics, while many of the responses from proponents of the Altaic theory have failed to address the criticisms raised. We hope that arguments focusing on the real issues of phonological correspondences and morphological systems will shed greater light on the relationship among these languages.

Elements necessary for a proper interpretation of a sentence can sometimes be missing from the ‘surface’ representation involving coordination (ellipsis, gapping, etc.). Though different methods have been proposed to recover missing elements, it is not clear how such recovery can be accomplished when they correspond to bound morphemes. Assuming lexicalism, this paper shows how such missing elements are recovered semantically without employing empty place holders, abstract functional categories or invisible movement. Bound verbal morphemes in Japanese introduced here give rise to morpho-semantico bracketing paradoxes which prove to be problematic for a syntactic view of morphology.

Massam (1992), challenging the view that the presence of an implicit argument is essential to middles, argues that middles should be defined in terms of genericity and modality. This paper shows that Massam's hypothesis is factually untenable. Genericity and modality are only typical, not essential, properties of middles. Furthermore, middles in Massam's conception do not form a natural class with middles as normally understood. Rather, the presence of an implicit argument is a defining characteristic, as is widely accepted.

It is further shown that in order to handle an implicit argument properly, middles need to be represented by conceptual structure in the sense of Jackendoff (1987a, 1987b, 1990). The proposed analysis of middle by means of conceptual structure accommodates even the implicit argument of middles that express specific events.

Recent work in Relevance Theory (henceforth RT) illustrates the coming of age of modern pragmatic scholarship in creating an environment in which a particular theory of pragmatics can be taken for granted, without explanation or justification, and an analysis of a phenomenon previously unaccounted for within that theory can be advanced. One is reminded of much of the recent syntactic work within GB/Principles & Parameters/Minimalist Theory: the dominance of the Chomskyan approach – particularly in certain geographic regions – allows researchers, for better or worse, to simply assume the correctness of the theory in their work and proceed to illustrate how that theory might (or must) be extended or modified to accommodate a new class of data. In this volume, Žegarac & Clark (1999) provide the latest illustration of a similar strategy in pragmatics: the correctness of RT is assumed and an analysis of ‘phatic communication’ proposed within that framework. On the one hand, this constitutes an advance for pragmatic theory, since until recently there was no comprehensive all-inclusive framework within which certain pragmatic generalizations could be stated. If nothing else, RT has served to raise a number of important issues surrounding the semantics-pragmatics interface, helping to crystallize the debate and make explicit many assumptions that had been either implicit or non-existent in other frameworks. In particular, RT work on scalar implicature/explicature and on echoic mention and metalinguistic negation (e.g. Carston 1988, 1995; Récanati 1989) has represented major advances in our understanding of these phenomena and their theoretical implications. Thus, whatever one may think of RT, it is a theory that must be taken seriously by anyone working in this area. On the other hand, we find that RT suffers from one of the principal afflictions of the aforementioned work in the GB/P&P/Minimalism mainstream: a remarkable failure to address, come to terms with, and incorporate the extensive previous literature on the topic under current consideration.

This paper considers how the notion of phatic communication can best be understood within the framework of Relevance Theory. To a large extent, we are exploring a terminological question: which things which occur during acts of verbal communication should the term ‘phatic’ apply to? The term is perhaps most frequently used in the phrase ‘phatic communication’, which has been thought of as an essentially social phenomenon and therefore beyond the scope of cognitive pragmatic theories. We suggest, instead, that the term should be applied to interpretations and that an adequate account of phatic interpretations requires an account of the cognitive processes involved in deriving them. Relevance Theory provides the basis for such an account. In section 1, we indicate the range of phenomena to be explored. In section 2, we outline the parts of Relevance Theory which are used in our account. In section 3, we argue that the term ‘phatic’ should be applied to interpretations, and we explore predictions about phatic interpretations which follow from the framework of Relevance Theory, including the claim that phatic interpretations should be derived only when non-phatic interpretations are not consistent with the Principle of Relevance. In section 4 we consider cases where cognitive effects similar to those caused by phatic interpretations are conveyed but not ostensively communicated.

The Dutch dialect of Venlo has a lexical tone opposition comparable to the distinction between Accent I and Accent II in Scandinavian. The two word tone patterns are realised in a variety of different ways, depending on the intonation contour, on whether the word has a focus tone, and on whether it occurs finally or nonfinally in the intonational phrase (IP). Twelve such contexts are identified, and an autosegmental-metrical analysis is presented of the contours for the word tones in each of these. The analysis is instructive because of its clear illustration of the distinction between the phonological underlying representation and the phonological surface representation, as well as of the distinction between the latter representation and the phonetic realisation. In addition, because of the complexity of its tonal phonology, the dialect is of considerable typological interest for the study of word prosody and intonation.

This is a response to Carston's critique of my account of presupposition-cancellation. While accepting her demonstration that (contra ‘On Horn's dilemma’) presupposition-cancellation does not involve a linguistically encoded contradiction, I show that this is nevertheless consistent with the account of presupposition proposed in Burton-Roberts 1993/7. In fact, Carston's and my accounts of presupposition-cancellation both treat it as involving a pragmatically derived contradiction. I also reconsider the nature of so-called ‘metalinguistic negation’, arguing against Carston that there is a special use of negation (!MN) which involves a use-mention mix and a pragmatically derived contradiction, and is non-truth-functional. I show that, although !MN is echoic, not all echoic negations are examples of !MN.

Ray Jackendoff, The architecture of the language faculty. Cambridge, MA: MIT Press, 1997. Pp. xiv+262.

In The architecture of the language faculty Jackendoff responds to the Minimalist Program (MP) of Chomsky (1995). There are three major themes: the conceptual foundations of MP and its historical antecedents; the interfaces between syntactic structure, phonetic structure and conceptual structure, and the nature of the lexicon. The basic approach is that of Jackendoff (1983, 1990), with roots going back at least as far as Jackendoff 1975. In the first chapter Jackendoff sets out the more or less standard views about universal grammar and subjects them to a critical analysis. Chapters 2–4 focus on the various ‘interfaces’ between levels of representation, and on how the interface relations are reified in individual lexical entries. Chapters 5–7 explore the properties of various types of lexical entries and the relations between them. Chapter 8 concludes with a speculative essay on the relation between language and thought.

This book is a rich compilation of observations, analyses, suggestions, perspectives, speculations and proposals. One could easily write a review at least as long as the book, developing and responding to the many ideas that it contains. Since space here is limited, I will concentrate on the main conceptual and foundational issues that are addressed by the book. In many ways, Jackendoff's work points to significant departures from classical perspectives on the organization of grammar. I will focus some attention on those areas where even more radical steps might be worth pursuing.

In Žegarac & Clark (1999) we try to show how phatic communication can be explained within the framework of Relevance Theory. We suggest that phatic communication should be characterized as a particular type of interpretation, which we call ‘phatic interpretation’. On our account, an interpretation is phatic to the extent that its main relevance lies with implicated conclusions which do not depend on the explicit content of the utterance, but rather on the communicative intention (where ‘depends on X’ means: ‘results from an inferential process which takes X as a premise’).

Frederick J. Newmeyer, Generative linguistics: a historical perspective. London: Routledge, 1996. Pp. x+218.

In this book (henceforth, GL), Newmeyer has collected 11 of his articles (including two co-authored ones), two of them previously unpublished, on the history of generative grammar, supplemented by reprints of two sections of Newmeyer 1986.

Metalinguistic negation (MN) is interesting for at least the following two reasons: (a) it is one instance of the much broader, very widespread and various phenomenon of metarepresentational use in linguistic communication, whose semantic and pragmatic properties are currently being extensively explored by both linguists and philosophers of language; (b) it plays a central role in recent accounts of presupposition-denial cases, such as ‘The king of France is not bald; there is no king of France’. It is this latter employment that discussion of metalinguistic negation has focused on since Horn (1985)'s key article on the subject. While Burton-Roberts (1989a, 1989b) saw the MN account of presupposition-denials as providing strong support for his semantic theory of presupposition, I have offered a multi-layered pragmatic account of these cases, which also involves MN, but maintains the view that the phenomenon of presupposition is pragmatic (Carston 1994, 1996, 1998a).

David Lightfoot,The development of language: acquisition, change, and evolution. (Maryland Lectures in Language and Cognition 1.) Malden, MA & Oxford: Blackwell, 1999. Pp. xii+287.

The central thesis of The development of language is that there are no principles of grammatical change, so that ‘historicist’ or deterministic approaches to diachronic change are misguided. Instead, Lightfoot argues that language change can only be understood by taking the perspective of the ‘growth’ (i.e. acquisition) of an individual's biological grammar, which may end up with a different parameter setting from the parent's generation when the trigger experience changes. Such events of grammatical change are abrupt and unpredictable, and Lightfoot suggests that they can be understood better from the point of view of catastrophe theory and chaos theory than under a deterministic theory of history as was common in the nineteenth century.