In the literature on Null Subject Languages (NSLs) since Rizzi (1982), the three properties that are commonly thought to be connected are (i) the richness of inflectional morphology, (ii) free subject inversion, and (iii) the COMP-trace effect. The connection between them is that if a language (e.g., Italian) has the option of having a null subject (NS) pronominal pro in subject position (i.e., SpecIP) as a result of having rich inflectional (i.e., verbal agreement) morphology, as represented in (1a), then the same element may appear in the same position in cases of subjects occurring postverbally as in (1b) and in cases of long-distance subject extraction where the postverbal trace is properly governed by the verb, as in (1c):

Locative inversion is a common linguistic phenomenon that has been studied by linguists of various theoretical backgrounds (see, among others, Bresnan 1990, Demuth 1990, Hoekstra and Mulder 1990, and Rochemont and Culicover 1990). This paper is an attempt to analyse such a phenomenon in Mandarin Chinese within the framework of Government and Binding (GB) Theory. Its purpose is twofold: to investigate the syntactic properties of this particular construction, and to illustrate the relevance of Lexical Conceptual Structure (LCS) to syntactic analyses within the GB paradigm.

Constraints, which go also by the names “conditions”, “filters”, and less transparently, “principles” and “parameters” are currently opposed to (contextual and non-contextual) arbitrary rules, and are argued to be more explanatory in the scientific sense that they can: 1) reduce the number of sources and/or causes of a given phenomenon; 2) link apparently unrelated facts; and 3) make more predictions, if formulated adequately and related to universal grammar (UG). Although the notion of phonological constraints was not born with generative phonology (see, e.g., Hockett 1958:282), it is within that framework that they have begun to receive more formal treatment (Paradis and Nikiema 1993:45).

In phonology, constraints are finally coming into their own. There is no real consensus, however, about how constraints behave, and specifically how they interact with other aspects of the phonology and with each other. Theories of Declarative Phonology (DP) approach this “Interaction Problem” (Scobbie 1991a) in a radical way. In DP everything is a constraint — generalizations and lexical entries alike constrain the form of the intended phonological representation (“surface structure”). Moreover, constraints are mutually compatible, indefeasible and are declared non-procedurally. Declarative Phonology is therefore an extremely simple and restrictive type of constraint-based phonology, eschewing constraint conflict, violation and destructive repair. For details of various declarative theories as well as phonological and computational motivation for the declarative paradigm, see Scobbie (1991a) and the other papers in that volume, Bird et al (1993), Bird (1990), Broe (1993), Coleman (1992), Russell (1992), Scobbie (1991b) and Wheeler (1988).

Few topics in the phonological literature have inspired the amount of attention that has been lavished on the notorious “mute-e” of contemporary French. Perhaps the most perplexing difficulty for phonologists is how (or even whether) to include schwa in phonological representations. Solution of the representation problem, moreover, is necessary for a proper identification of the conditions under which schwa is deleted, maintained or inserted, and for an analysis of alternations linking schwa and other vowels. Many different solutions to the representation problem have been suggested: schwa as a distinct vowel; schwa as absent from the phonemic inventory (hence from underlying representations) and inserted as a “lubrifiant”; or (in the current non-linear literature), schwa as an empty nucleus; schwa as an unlinked vowel.

On définit généralement les séries verbales (ou constructions verbales sérielles) comme des constructions dans lesquelles apparaît une séquence de deux ou plusieurs lexèmes verbaux pour former le prédicat de la proposition.

La problématique des séries verbales a reçu de la part des auteurs plusieurs interprétations parfois divergentes. Dans cet article nous allons proposer une analyse des constructions dites à verbes sériels en les reliant d’une part à la possibilité qu’ont les verbes pleins de fonctionner parfois comme auxiliaires et d’autre part à la possibilité pour un V′, c’est-à-dire une projection de niveau 1 d’un verbe en termes de X-barre, de recevoir en position adjointe des constituants syntaxiques de type CP, IP, VP, V′ et V.

Nous soutiendrons l’hypothèse que certaines constructions verbales sérielles qu’on observe dans les langues Gbe2 sont à relier à la possibilité pour ces langues de présenter des sujets nuls dans le cas d’adjonction d’un CP ou d’un IP à un V′.

Since the paper by Barss and Lasnik (1986), the fact that the double object construction exhibits Theme/Goal asymmetries has been the object of an important debate in the literature. The central question of this discussion is whether these asymmetries should be accounted for in terms of dominance (e.g., Larson 1988, 1990; Aoun and Li 1989; den Dikken 1991) or precedence (e.g., Barss and Lasnik 1986; Jackendoff 1990; Tremblay 1991). On the dominance account, the Goal must asymmetrically c-command the Theme (e.g., Larson 1988, 1990). On Barss and Lasnik’s account, the Goal and the Theme of the construction may symmetrically c-command each other but the Goal must precede the Theme. This proposal is captured in their definition of domain of (1986:352).

The idea that constraints on well-formedness play a role in determining phonological alternations, which dates back at least to Kisseberth’s (1970) pioneering work, has by now achieved almost universal acceptance. A tacit assumption of this program, largely unquestioned even in recent research, is the notion that valid constraints must state true generalizations about surface structure or some other level of phonological representation. Anything different would seem antithetical to the very idea of a well-formedness constraint.

In the past few years, a number of articles have argued that the default articulator for consonants is Coronal (see, for example, Kiparsky 1985; Iverson and Kim 1987:187; Shaw 1988; Avery and Rice 1989; and the contributors to Paradis and Prunet 1991). Paradis and Prunet (1993b), with a view to strengthening and formalizing this proposal, have expressed the Weak Coronal Hypothesis (WCH) in (1).

The following pages sketch a system of constraints intended to sharpen and organize a number of current assumptions surrounding the concept of “prosodic licensing”, as proposed in Itô (1988) and further developed in later work (Bagemihl 1991; Charette 1990; Goldsmith 1990; Inkelas 1991; Itô and Mester 1991; Kaye 1990; Lombardi 1991; Piggott 1991; Zec 1988; and others). The proposal to be developed below draws on Optimality Theory (Prince and Smolensky 1993); one of the goals is to explicate the notion of the coda as a “secondary licenser” (Goldsmith 1990) and to clarify the privileged status of geminate and partial geminate clusters (Prince 1984) with respect to syllabification. After laying out some background assumptions (Section 1), we present the definitions and constraints that form the core of our theory of licensing (Section 2). We illustrate the approach with an analysis of the syllable structure of Japanese (Section 3) and conclude the paper with a discussion of remaining issues and problems relating to the general typology of syllabification systems (Section 4).

It is increasingly accepted in generative phonology that phonological alternations can be caused, directly or indirectly, by phonological constraints. The Theory of Constraints and Repair Strategies (TCRS) proposed by Paradis (1988a, 1988b, 1990, 1992, 1993) claims that when a constraint is violated, a repair strategy must apply which, in repairing the violation, produces a phonological alternation.