Linguistic Variation Yearbook - Volume 8, Issue 1, 2008

The aim of this paper is to argue that temporal marking in the nominal domain should not be taken as evidence for the existence of a syntactic category TP within the extended projection of the noun phrase. On the basis of two unrelated languages, Somali and Halkomelem, the paper re-interprets the role of temporal morphology as being linked to specificity in the former case and as being adverbial in the latter. Building on an analysis of Halkomelem, according to which the language lacks a syntactic category TP (Wiltschko 2003), the paper derives from this property the lack of a categorial distinction between nouns and verbs in Halkomelem. This is based on the proposal that becoming verbal necessarily requires a combination between some primitive predicative head and an Agreeing T in the syntax.

This paper discusses reflexive and control constructions in San Lucas Quiaviní Zapotec (Lee 2003) and Hmong (Mortensen 2003), where instead of a reflexive in the former and a null category in the latter, one may find a copy of the antecedent. The paper argues that these constructions provide compelling evidence for a movement analysis of control (Hornstein 2001, 2003; Boeckx & Hornstein 2003, 2004) and reflexivization (Hornstein 2001), as well as the proposal that the phonetic realization of copies generated by movement is regulated by linearization and morphological requirements (Nunes 1999, 2004).

We discuss the results of an L1 French comprehension study of the construal of present and imperfective past in (non) subordinate contexts. Our findings reveal that children accept (sometimes enforce) non-indexical simultaneous construals of both present and past under a matrix past — though present is utterance-indexical in adult French. Extending Kratzer’s (1998) zero-tense analysis of English past under past simultaneous construals to Japanese present under past simultaneous construals, we argue that zero-tenses in L1 French surface either as past (adult French) or as present (adult Japanese). That children allow multi-valued parameter settings is expected on the Multiple Grammars hypothesis where language acquisition involves grammar competition. We extend our zero-tense analysis of non-adult tense construals in subordinate contexts to non-subordinate contexts, by arguing that the binder of a zero-tense in child grammar can be a temporal adverb denoting the ‘now’ of the speaker or some salient time implicit in the context.

The crosslinguistic distribution of preposition stranding by A movement in pseudopassive constructions matches that of a marked A' phenomenon, namely extraction from Bare Present Participial Adjuncts. Moreover, both constructions show sensitivity to external factors of a sort that reanalysis-based theories of P-stranding are designed to capture, but which is not immediately predicted by P-stranding theories based on parametrisation of PP’s status as a bounding node or phase. However, an expanded version of Abels’ (2003) phase-based account of P-stranding, according to which the sensitivity of P-stranding under A- and A'-movement to PP-external factors is due to general constraints on movement and passivisation, captures the relevant data without resorting to a reanalysis operation.

I argue that the interpretation of expressions consisting of disjunction marker and wh-element (wh-disj expressions), which varies across languages, constitutes a case of semantic variation. In Hausa, these expressions denote universal generalized quantifiers, which give rise to free choice effects in intensional contexts (Giannakidou 2001). The universal meaning is derived in compositional fashion, where the disjunction marker expresses set union over the wh-domain. The free choice effects follow from the scopal interaction of universal quantifier and intensional operator. The account relates to Giannakidou & Cheng’s (2006) analysis of (quasi)universal FCIs, but it does not extend to Japanese and Malayalam wh-disj expressions, which are interpreted with existential force and should be analyzed as indeterminate pronouns (Jayaseelan 2001; Kratzer & Shimoyama 2002). Motivated by the analysis of FCIs in Menendéz-Benito (2005), we finally consider an alternative analysis of koo-wh expressions as selective indeterminate pronouns, which is rejected on conceptual and empirical grounds.

In this paper, we discuss some collective numeral constructions from Dutch, which display a remarkable plural morpheme -en: wij vieren “the four of us”, met ons vieren “with four” and met z’n vieren “with four”. We argue that these at first sight idiosyncratic constructs are formed in syntax by application of the regular syntactic rules: Merge and Move. A central claim we make is that these collective numeral constructions contain a silent noun PERSOON/PERSON (cf. Kayne 2002, 2003, 2007). Although the noun itself is silent, its plurality is realized as -en. The silent noun is licensed by a local (i.e., DP-internal) antecedent. Dialectal variation in the realization of these constructions is analyzed in terms of “externalization”. It is argued that the basic syntactic make-up is similar in all variants. It is the (morpho-) phonological realization of heads, which causes the differences at the surface.