In the light of current morphological theory, this paper examines the analogical levelling of long/short vowel oppositions in certain inflectional and derivational alternations in a number of modern Swiss German dialects. The regular occurrence of levelling is shown to depend on the extent to which the alternation in question is ‘perceptually salient’ (Chapman 1994). That is, if the semantic relation between base and derivative is transparent and the derivative is uniformly marked, analogical levelling occurs regularly. On the basis of this evidence it is argued that all morphological alternations, both inflectional and derivational, are listed in the lexicon and that each one is assigned a different status according to its degree of perceptual salience.

In current linguistic theory, the theoretical status of adjunct extractions, as in for example How often do you think Robin sees Kim? is, somewhat surprisingly, an unresolved issue, with some investigators arguing that only arguments extract syntactically, entailing analyses of adverbial gaps via fundamentally different mechanisms from those posited for argument extraction. We adduce extensive evidence against such positions from a number of languages which exhibit morphological or syntactic phenomena which are sensitive to binding (extraction) domains and where this morphosyntactic flagging is present in instances of adjunct extraction as well as argument extraction. We also present language-internal arguments for the syntactic nature of adjunct extraction in English, including the coextensiveness of adjunct and argument extraction and their parallelism with respect to strong/weak crossover effects. Finally, we discuss the challenge which binding domain effects pose for accounts of adjunct extraction in various frameworks.

Relational Grammarians have proposed Union analyses for adversatives many languages. An odd asymmetry in base predicates contrasts Japanese (Dubinsky 1985), in which adversatives may not be formed on unaccusatives, and Indonesian (Kana 1986), in which adversatives may only be formed on unaccusatives. A close examination of adversatives in Javanese (a language closely related to Indonesian) resolves this asymmetry, revealing that Javanese adversatives (and by implication Indonesian adversatives) are best analyzed not as Unions but as passives. However, the passive analysis violates Perlmutter & Postal's (1984) I-Advancement Exclusiveness Law, which figures crucially in Dubinsky's elegant account of the distribution of Japanese adversatives. Gerdts' (1993a) Mapping Theory, rooted in an RG tradition, provides a solution, in which it is possible to capture the similarities of adversatives and other Javanese passives and at the same time preserve the insights of Dubinsky's analysis of Japanese.

A comprehensive view of German syllable structure is developed based on phonetic and psycholinguistic experimental results supplemented by phonological arguments. The account is based on the interaction between universal factors (sonority/strength constraints, effects of tempo) and language-specific factors (phonotactic constraints, effects of stress). Initial, phonological syllabification reveals the distributional constraints operating in the language. This syllabification is modified by stress and rate of speech in well defined ways. The effect of a faster tempo is to maximize and thus strengthen syllable onsets within the limits predicted by sonority/strength and to minimize and thus weaken codas. This is achieved through resyllabification of stronger consonants onto following weaker ones. The effect of stress is twofold. Stress blocks faster speech resyllabification. It also modifies initial syllabification by making stressed syllables heavier through the acquisition of a branching nucleus, and in faster tempos, a branching rhyme. The effects of syllabification adjustments due to tempo and stress are interpreted as an optimization of syllable structure. The study supports a theory of syllable structure based on the arrangement of segments according to the so-called Sonority Sequencing Principle augmented by language-specific constraints which, in the absence of counteracting factors such as stress, are relaxed in faster tempos.

In this paper I argue that the polysemy view is not capable of giving a unified account of the meanings of CAN, MAY, MUST and SHOULD, whereas the unitary meaning view does not encounter the problems facing the polysemy view. I propose unitary meanings which are rich enough to account for the range of interpretations these modals can have, but which are specific enough to account for why they get these interpretations and not others. Proposing unitary meanings implies that we have to look for a theory of pragmatics which can explain how we achieve the different interpretations of these modals in use. I will argue that adopting the Relevance theory view of what drives interpretation gives us the basis for such an explanation.

This paper argues that there are Modern Greek (MG) deverbal nominal predicates which take syntactic arguments. A mechanism for the derivation of these nominals requiring the existence of an ‘internal’ ([ — r]) argument is proposed which has broader coverage than the mechanism proposed in Grimshaw (1990). A small set of simple, unification-based operations is employed to model the relation between the argument structure of verb predicates and that of the corresponding deverbal nominals with an ‘ eventive’ reading. The Lexical Mapping Theory of LFG (LMT) is the theoretical framework of the present discussion.

Imperative Vs with distinctive morphology either have a distinctive syntax (Modern Greek, Spanish), or distribute like others Vs (Serbo-Croatian, Ancient Greek). The contrast follows from properties of the root C. The first type has a strong Imperative V-feature in C, and under Chomsky's Greed Principle, Imperative Vs raise overtly to check this feature. The second type is the Wackernagel language, whose C hosts no features, but V-features are in I. If no phrase fronts, Vs move to C to support second position items. V-to-C affects all Vs, is last resort, follows Lasnik's Enlightened Self-Interest, and escapes Greed.

This paper deals with the clause structure of Old English. In the main body of the paper we adopt the ‘traditional’ analysis of the West Germanic languages in which it is proposed that VP is head-final. We will argue (contra Van Kemenade 1987, pace Cardinaletti & Roberts 1991, Pintzuk 1991, Tomaselli 1991) that the clause structure of Old English contains a head-initial functional projection whose head can be the landing site of verb movement in subordinate clauses. This claim is based on evidence related to the distribution and interpretation of negative elements in Old English and West Flemish. We will show that differences between these two languages with respect to Negative Concord phenomena can be accounted for straightforwardly in terms of an Old English clause structure which is different from the one traditionally proposed for the modern Germanic SOV/V2 languages.

In the appendix to the paper we briefly turn to the recent alternative approaches to the phrase structure of SOV languages in terms of a universal base hypothesis where all projections are head-initial (see Kayne (1993), Zwart (1993), Roberts (1995) for a discussion of Old English).

This study examines the referential properties of a class of complex pronouns labelled M-PRONOUNS, exemplified by Old English HIMSELF, French LUI-MéME and English HISOWN. It is shown that M-pronouns exhibit some properties commonly taken as characterizing reflexive anaphors, and that they also occur as ‘intensive’ pronouns. It is shown, however, that they are not anaphors, and that labelling them ‘intensives’ does not suffice to account for their distribution. It is argued that the semantic properties of M-pronouns may be derived from their morphological structure. Their pronoun component (Old English HIM, French LUI, English HIS) is not a pronominal, in the sense of the Binding Theory, but a bindable expression unspecified for disjoint reference and locality. In the complex form created by M-adjunction, the pronoun is crucially de-stressed and, correlatively, interpreted as endophoric.

In this paper we present an alternative approach to discourse structure according to which topicality is the general organizing principle in discourse. This approach accounts for the fact that the segmentation structure of discourse is in correspondence with the hierarchy of topics defined for the discourse units. Fundamental to the proposed analysis is the relation it assumes between the notion of topic and that of explicit and implicit questioning in discourse. This relation implies that (1) the topic associated with a discourse unit is provided by the explicit or implicit question it answers and (2) the relation between discourse units is determined by the relation between these topic-providing questions.

Does English have morphological case (as opposed to abstract Case)? Evidence is presented which suggests that it may be a completely case-less language like Chinese, contrary to the widely held view that the distinct pronoun forms and the ‘genitive’ 's involve morphological case. The existence of case in English has recently been accepted almost without question, but the question at least deserves serious discussion as it is easy to find alternative analyses. According to the analysis offered here, I and me are both personal pronouns whereas my, mine and 's are possessive pronouns; and the difference between I and me, like the one between my and mine, is handled by a very specific and local lexical rule which is sensitive to the syntactic structure but does not involve case.

Our purpose is simple: to apply a rudimentary kind of logical analysis to the problem, which sometimes arises in linguistic typology and in comparative linguistics, of whether there are any languages with fewer than two vowels. The principal reason why this question has occupied the attention of linguists is probably the fact that Proto-Indo-European is sometimes said to have had only one vowel phoneme (see especially Hjelmslev (1936–7), Borgstrøm (1949, 1954) and Lehmann (1952), following Saussure (1879). Yet this was precisely one of the features of PIE reconstruction that Jakobson selected for attack on typological grounds, reasoning that ‘a conflict between the reconstructed state of a language and the general laws which typology discovers makes the reconstruction questionable’ (1958: 23).

TWO VIEWS OF GRAMMAR

For many years, many people have used the term ‘grammar’ to indicate something which represents an individual's mature linguistic capacity and which arises in the mind/brain of that individual on exposure to some relevant childhood experience. The grammar interacts with other aspects of a person's mental make-up, in a modular conception of mind. Different experiences may give rise to different grammars in different individuals, but it is a plausible initial assumption that grammars arise in everybody in the same way, subject to the same principles, parameters and learning constraints, which are common to the species. This is a biological view of grammars. Countless questions arise about these grammars, about their internal properties, about how they are represented in brains, about how they emerge in young children. Under this view there is no grammar of English, rather various grammars which exist in the minds of English speakers; grammars hold of people and not of languages. Let us distinguish terminology from reality here: proponents of the biological view of grammars sometimes use ‘the grammar of French’ to refer to the grammars of French speakers in a kind of shorthand which abstracts away from individual variation. This usage may have been misleading but, as noted in Lightfoot (1991, henceforth HSP, 162), it is comparable to references to the French liver, the American brain, or the Irish wit; nobody believes that there is such an entity but sometimes it is a convenient abstraction.