Volume 35, Issue 2 -

This article presents empirical evidence that disfavors using highly lexicalist minimalist models, such as the one presented in Chomsky (1995), when analyzing language mixing. The data analyzed consist of English – Spanish mixed noun phrases discussed in Moro (2014) as well as English – Norwegian mixed noun phrases and verbs taken from the Corpus of American Norwegian Speech. Whereas the lexicalist model in Chomsky (1995) only can explain a subset of the mixing patterns attested in both authentic English – Spanish mixed noun phrases and the American Norwegian corpus, we show that an alternative exoskeletal model can account for all of them. Such a model would entail that rather than assuming lexical items with inherent, functional features that determine the derivation, syntactic structures are generated independently from the lexical items that come to realize them.

In this paper we offer new arguments for bivalence of morphological features. In the domain of person, we argue in support of the system using the features [±speaker] and [±hearer], on the basis of plural marking in Cheyenne. In the domain of gender, we argue in support of the system using the features [±masculine] and [±feminine], on the basis of gender agreement in Serbian coordinate structures. The property of binary systems crucial for our proposal is that they allow combinations of disagreeing feature values, given that in such systems every morphological category is represented as a combination of two values. Our main empirical goal is to show that some languages treat such combinations of disagreeing feature values (as well as combinations of agreeing values) as natural classes (e.g., Noyer, Rolf. 1992. Features, positions and affixes in autonomous morphological structure. Ph.D. dissertation. Massachusetts Institute of Technology).

While VP-fronting in English has various properties that are generally taken to be hallmarks of A¯ $$\overline A $$-movement, other properties of the construction militate against an analysis in terms of displacement of VP to the clausal periphery. Such an analysis falls short of providing principled answers to the questions of why the trace of VP-fronting must be nominal, why certain kinds of morphological mismatches between the fronted VP and its trace are possible, why fronting of remnant VPs is impossible, and others. This article proposes an analysis of English VP-fronting – plausibly extending to English ’topicalization’ in general – based on the observation that the construction shares these non-movement properties with left-dislocation of VP in German. Connectivity effects, on such an approach, are the result of ellipsis. By assimilating VP-fronting to dislocation, the analysis furnishes principled explanations for the striking asymmetries between English-type VP-fronting (dislocation) and German-type VP-fronting (A¯ $$\overline A $$-movement).

The paper describes a puzzle in the semantics of a class of English adjectives which alternate systematically in meaning. We know from the literature on adjectival meanings that the STIMULUS vs. EXPERIENCER pair of roles forms the basis of two kinds of adjectives in English depending on whether the STIMULUS or the EXPERIENCER is the one who holds the property. Our puzzle adjectives alternate between these two readings in a way parallel to the morphologically marked adjectival participles (in - ed vs. - ing ) of Object Experiencer verbs. I argue further that there are reasons to think that the relationship between these two roles in this alternation is a stative version of causation, with the STIMULUS reading of the adjective being derived from the Psych EXPERIENCER reading. The existence of this alternation shows that within simple property ascriptions (of which adjectival predication is a prototypical expression) a distinction is made between Holders of properties simpliciter, and Holders of causational-properties. Thus, in analysing this adjectival puzzle as a parallel to verbal diathesis, the paper contributes to the debate on the similarities and differences between states and dynamic events.

This paper examines previously unnoticed facts about prosodic interactions between enclitics and their hosts in Bosnian/Croatian/Serbian (BCS). I show that there is a three-way split between BCS enclitics in this respect: (i) enclitics that always interact with the accent of their host; (ii) enclitics that interact with the accent of their host in some contexts, but not in others; (iii) and enclitics that never interact with the accent of their host. It is shown that this rather complex pattern can be accounted for in its entirety by a condition on prosodic incorporation of enclitics that holds at the point when syntactic structure is mapped to prosodic structure, which essentially requires the clitic and the host to be in the same spell-out domain in order for the clitic to incorporate into the prosodic word of its host, a prerequisite for the clitic to interact with the accent of its host. I also discuss certain idiosyncratic phonological properties of the auxiliary clitic je ‘be.3sg’ and the particle se ‘self’ that cause reordering of enclitics in PF. It is shown that this PF movement can have an effect on whether a clitic is spelled out in the same domain as its host, which in turn affects the prosodic interaction between the two. Based on the prosodic interaction of enclitics with the accent of their host, I argue for a phase-based approach to prosodic structure building (see also Dobashi 2003; Kahnemuyipour 2004, 2009;Kratzer and Selkirk 2007; Sato 2012; Sato and Dobashi 2016; among others).