Linguistic Variation Yearbook - Volume 4, Issue 1, 2004

This paper adds to the ongoing debate about the quantificational status of English FC any. It argues that any is a universal, though its universality is such that it can be conjoined with an existential statement. This is established by showing, first of all, that various properties that are amenable to a universal account of FC any remain elusive in recent accounts that treat it as an indefinite. Secondly, it presents a detailed description of supplementary/numeral any, cases in which any occurs with an indefinite. Such cases have been assumed to provide evidence for the indefinite-any view but an explicit semantics for the constructions is given, showing that any invariably contributes universal quantificational force. Existential force always comes from other elements. The paper also discusses alternative formulations that preserve the key insights of the universalist position and the role of cross-linguistic variation in the study of FC items.

DPs and BNs are used parametrically in Arabic (as well as in other languages) to express generic/existential contrasts, and mass/count oppositions. Unlike English and Romance, Arabic BNs behave like overt indefinites, give rise to bare singulars, and to numeral BNs, obviating scope or opacity distinctions between bare singulars, duals, or plurals. These numerous BN varieties are accounted for via various N-to-F computational processes, involving namely Numeral and Generic positions. Second, indefinite and definite generic types also make use of covert/overt D oppositions, yet they are subjected to a unifying treatment. Third, the mass/count dichotomy cannot provide an adequate base for a classificatory system. ‘Atomicity’ and ‘singulativity’ values are needed to characterize more appropriately Classifier and Number interactions in so-called ‘classifier’ and ‘number’ language types. ‘Inner’ atomicity (associated with Classifier) and ‘outer’ atomicity (provided by Number) can be properly identified, making it possible to avoid confusion of Classifier and Number contributions to structure and interpretation (e.g. the kind/generic confusion). Finally, the referential/quantificational split, usually expressed through the BN/DP opposition, finds its counterpart in proper noun (PN) and common noun distributions.

This paper offers a new analysis of Japanese scrambling, under which some instances of scrambling phenomena are derived from the process of linearization. It is specifically proposed that the absence of formal agreement in Japanese enables Spell-Out to apply solely to an argument of the verb. The spelled out argument is “dislocated” at PF by the mechanisms of linearization of spelled out syntactic objects. Radical reconstruction effects, along with various other properties of Japanese scrambling such as the proper binding effects, are captured as a natural consequence of the proposed analysis, because the scrambled constituent actually does not undergo any syntactic movement but rather stays in the base-generated position. It is also argued that an analysis of scrambling ought to be eclectic in the sense that another strategy, which employs null operator movement to establish the relation between the dislocated element and the gap, is also available in Japanese, as originally proposed by Ueyama (2002). Thus, the optionality of Japanese scrambling is shifted under the proposed analysis to the optional application of Spell-Out (made possible by the absence of formal agreement) and the optional selection of a null operator in the numeration. The paper is concluded with the speculation that the availability of the latter strategy is due to the rich use of predication in Japanese.

The hypothesis of the copy theory of movement forces us to look at mismatches between syntax and LF on the one hand, and syntax and PF on the other in particular ways, often revealing new insights. Through such a lens, we examine the syntactic expression of pejorative mood through echo reduplication, focusing on shm-reduplication in (varieties of predominantly American) English. It is argued that the two elements in a reduplicated structure form a chain of two left-peripheral positions that, due to a distinctness requirement within a Spell-Out unit for the Transfer to PF, cannot be mapped onto linear order. A number of seemingly unrelated facts are derived rather naturally: (i) English shm-reduplication cannot appear in an argument position; (ii) the two copies involved in shm-reduplication are strictly adjacent; (iii) the phonological phrasing of the two copies is not the intonation of a compound; (iv) the discourse context felicitating shm-reduplication is not out-of-the-blue; (v) no echo reduplication process yields the reverse order (e.g., with the echo reduplicant preceding the base); and (vi) echo reduplication is never the exponent of a Case- or wh-feature. One of the more general issues advanced in this paper is an investigation of the clausal realization of pejorative mood expression, whether expressed through shm-reduplication or by other means.

The dominant position among generative grammarians with respect to typological variation is that it should be captured by parameters, which are either directly tied to principles of Universal Grammar (UG) or to functional projections provided by UG. Parameter-setting approaches, however, have failed to live up to their promise. They should be replaced by a model in which language-particular rules take over the work of parameter settings and in which most typological variation follows from independently-needed principles of performance. In such a model, UG specifies the class of possible languages, but not the set of probable languages.

This paper discusses the very general question of how syntactic features of individual languages relate to the universal set of syntactic features. It is pointed out that Chomsky’s approach (2001a) to this fundamental issue is paradoxical. On one hand he argues that language is uniform in the relevant sense (L-uniformity), but, on the other hand, he also assumes that languages make different selections of features from a universal feature set (L-selection). The paper argues strongly that L-uniformity is the only conceivable possibility. However, if that is correct, a great deal of what languages have is common is ‘silence’, that is, categories that are present in Narrow Syntax but silent in PF. In other words, language has innate elements and structures irrespective of whether or how they are overtly expressed. It follows that language variation is to a substantial extent ‘silence variation’, that is, much of it boils down to languages being explicit vs. silent about different (syntactically active) categories. This claim is coined as the Silence Principle, saying that any meaningful feature of language may be silent.

This paper examines variation in the interpretation of the Chinese verbal suffix ‑le and attributes it to a process of ongoing change in which an original aspectual system is gradually developing into a new tense system. The paper argues that the incremental pattern of re-grammaticalization can best be captured in a minimalist feature-based approach to the licensing of morphology, which factors clines of development into formally discrete, individual steps.