On categories, semi-lexicality, and quantifying expressions

Traditional linguistic theory assumes discrete categories like “noun”, “verb” and “adjective.” In my work (largely found in my dissertation), I focus on the nominal domain, in particular, what it means to be a noun; I explore cases of “semi-lexicality,” where the categorization of “noun” is simply not sufficient. For example, quantifying nouns like a lot, a tona number, and a bunch look very much like nouns (they are preceded by an indefinite article and followed by of, e.g. a bunch of books), but are also very different from nouns: they have the property of quantifying (bunch, lot, ton = ‘many’), and they cannot control verbal agreement (a lot of books are/*is). Such elements are “semi-lexical” and calling them nouns fails to capture their idiosyncrasies. Interestingly, the problem of semi-lexicality seems to recur with quantifying expressions (e.g. numerals like 100, a hundred), and part of my work is interested in understanding why that is. My work in this domain is primarily concerned with understanding/exploring (a) the basic building blocks of nominal structure, (b) the types of semi-lexicality (e.g. deviations from “noun”) which occur (are there limits to what can exist?), and (c) the syntactic status of semi-lexical quantifying expressions, and their recurring similarity to particular categories, such as nouns. In this line, I’ve explored the morphosyntax of Polish numerals, English quantifying nouns (lot, ton, bunch, number), and English kind-words (kind, type, sort).

Case and agreement

In my dissertation, I explore the mechanisms of case and agreement, and propose a system of case assignment and agreement (heavily based on existing analyses in the literature), which is suitable to model the case/agreement facts of Polish numerals, English quantifying nouns (bunch, lot, ton, number), and English kind-words (kind, type, sort, as in these kind(s) of cars). What is particular to these constructions is that case assignment seems to be DP-internal, as opposed to between DPs, suggesting that it must also be possible to assign case to something smaller than a DP. These constructions also (to a certain extent) argue against the common assumption that agreement results in case assignment, which leads me to argue for separate (but interacting) case and agreement processes. In other work, I also explore properties of oblique case, which shows different requirements from structural case when the element in question is semi-lexical.

Modified numerals

In joint work with Lisa Bylinina and Jakub Dotlacil, we considered the syntax and semantics of modified numeral constructions in English. The construction is characterized by an obligatory indefinite article when a numeral expression is modified by an adjective, regardless of the value of that numeral, e.g. a  whopping 23 communities, an additional 200 applications, a boring five songs, an estimated hundreds of thousands of dollars. In addition, only certain types of adjectives are permitted in the construction (e.g. *a tall five men), with some interpretational effects (e.g. a required consecutiveness with expressions like a boring five songs). Our work was concerned with (a) the syntax of the construction (what triggers the article?), (b) the semantics of the construction (which adjectives are permitted? why?), and (c) the historical development (when did the construction arise, and how did the use of adjectives in the construction change during its development?).

EPP and Dutch expletive er

In joint work with Franca Wesseling, we looked at the wh-extraction of subjects in Dutch, collecting data through a large scale grammaticality judgment task. Extraction of a wh-subject causes the expletive er to surface (which is further subject to a definiteness effect in relation to the object in transitives). We argue this to be an EPP effect, and propose that the wh-word is decomposable, but strands a DP layer (subextraction), which is realized as the expletive er.

In further joint work with Franca Wesseling and Coppe van Urk, we expanded our domain of inquiry to include impersonal passives, showing (via a large scale grammaticality judgement task) that intransitive impersonal passives require the expletive er unless a locative proform (hier ‘here’, daar ‘there’) is present. We further propose that the Dutch EPP involves a locative feature, such that locative elements can satisfy the EPP. In her dissertation, Franca Wesseling further pursues an analysis of er which captures these seemingly different types of er.

The morphosyntax of genericity

A topic which came out of my dissertation concerns the representation of genericity morphosyntactically. In my dissertation, I explored the morphosyntax of the semi-lexical elements kind, type, and sort. These have the peculiar property of showing apparent agreement with the noun they combine with (subject to certain systematic exceptions), e.g. these kinds of books, *this kind of books. In my dissertation, I argue that this follows from their semi-lexical status (in terms of function, they create reference to the kind-domain, but morphosyntactically, they seem to be nouns which combine with something smaller than a DP), and in further work, I argue them to be restructuring nouns, which trigger a number-sharing effect. I hope to further explore the functional structure of the DP in relation to genericity, and whether genericity has a syntactic correlate, beyond kind-words.