Maša Bešlin / Lexical categories, (re)categorization, and locality in morphosyntax
Maša Bešlin / Lexical categories, (re)categorization, and locality in morphosyntax
Thursday 12/12/24, Maša will defend her dissertation, "Lexical categories, (re)categorization, and locality in morphosyntax," supervised by Masha Polinsky as Chair, to a committee of Norbert, Alexander and Dave Embick, with Michael Israel as Dean's Representative. Maša's abstract is below.
This dissertation is about the nature of syntactic primitives and principles, their status in the grammar, and their interaction with extra-linguistic cognition. The dissertation has two parts, brought together by two common threads, top-down (conceptual) and bottom up (data-driven). Both directions are taken with the goal of streamlining the syntax by asking whether some of its proposed primitives and principles are dispensable, as well as how those primitives and principles interact with other areas of the grammar and with other, non-linguistic domains of cognition. Core empirical data in both parts come from adjectival derivation in a number of languages, with a special focus on Bosnian/Croatian/Serbian (BCS). In the first part of the dissertation, I consider the status of lexical categories (LCs) in grammar. I begin with an overview of the received wisdom on LCs, critiquing in particular claims that LCs referenced by the syntax (noun, verb, adjective) (i) have universal syntactic properties, and (ii) map straightforwardly to some interpretive property. I then present two detailed case studies of active and passive participles, arguing that participles in the languages I consider are a derived category, namely deverbal adjectives. The distribution of a participle is fully explained by understanding the behavior of its subparts (verbs and adjectives), making it unnecessary (for the learner or linguist) to posit it as an independent category. I furthermore show that the interpretation a participle receives–eventive or stative–does not arise from having more or less verbal or adjectival structure. I draw the following conclusions: (i) positing innate LCs is unnecessary and unhelpful, and (ii) LC-features are purely formal features, ones that determine distribution but are independent of form and meaning. This claim is fully in line with the Autonomy of Syntax and the Abstract Morpheme theses, but in stark contrast with many generative and non-generative approaches which maintain varying degrees (and directions) of syntax-meaning isomorphism.
In the second part of the dissertation, I turn my attention to the formal principles that operate on grammatical primitives, asking specifically what kinds of locality constraints are employed by the grammar. Locality is a well-researched topic in generative linguistics, but the current offering of locality theories is stipulative, redundant, baroque, and/or empirically inadequate. After providing a state-of-the-art overview of the proposals to handle locality phenomena in syntax and in morphophonology, I investigate the compatibility of classical Phase Theory with Phase Theory in Distributed Morphology (DM). Looking at (de)adjectival derivation in BCS, I show that adjectivization imposes a DM-locality boundary (for allomorphy and morphological tone assignment), but not a ‘big syntax’-locality boundary (for punctuated movement paths). Partly in order to reconcile this clash, I offer a reinterpretation of successive-cyclicity effects from the literature in terms of Featural Relativized Minimality (“attract closest”). This reinterpretation of locality effects gets rid of the apparent redundancy between absolute locality restrictions (exemplified by Phase Theory) and relative locality restrictions (exemplified by Minimality) in syntax more generally. There is only one syntactic locality principle–Minimality–which regulates probe-goal relations. On the other hand, a modified version of the Phase Impenetrability Condition, namely Transfer, regulates the transfer of syntactic structure to the interfaces, but has no effects syntaxinternally. Elements that have been transferred to the interfaces can still be accessed for the purpose of syntax (displacement, agreement), but can no longer be internally manipulated for the purposes of the morphophonology or semantic interpretation.