Brainstorming: Possible Seminar Topics (Spring 2026)

As usual, I post a few of the ideas I have for a seminar in order to get feedback. That is, each of the following topics is a possible topic for my Spring 2026 seminar, and I need to choose one of them. The default is ‘Inversion’, but I could be persuaded to do one of the others, if there is enough interest.

1. Inversion

This course will look at a wide variety of inversion constructions cross-linguistically, including: quotative inversion, locative inversion, inverse copular constructions, subject-object inversion in Bantu, as well as other related constructions. The goal will be to show how these constructions fit into the theory of argument structure and voice of Collins (2024) (‘Principles of Argument Structure’, MIT Press) (see also Storment 2025 ‘Projection (your) Voice: A Theory of Inversion and Defective Circumvention’, Doctoral dissertation, Stony Brook). Students will be given the opportunity to do fieldwork with a consultant during the course of the semester.

2. Foundations of Minimalism

This course will discuss foundational issues in syntactic theory, including the notion of Merge, copies versus repetitions and workspaces. Discussion will focus on Marcolli (2025) (‘Mathematical Structure of Syntactic Merge’, MIT Press) and related papers (e.g., Chomsky et. al. ‘Merge and the Strong Minimalist Thesis’ CUP). By the end of the semester, students will write papers in the framework of Marcolli (2025), or write papers critiquing that framework.

3. Large Language Models and Generative Syntax

This course will review literature concerning the relationship between Large Language Models and linguistic theory. Some of the topics of interest include: (a) What is the basic architecture of a LLM? (e.g., How are vectors used for lexical representation?) (b) How are LLMs able to produce fluent and natural English text? (c) Can LLMs be taken as a possible theory of the human language faculty? (d) If not, can LLMs contribute anything useful to the study of human language? (d) Can generative syntax contribute anything useful to the creation of LLMs? Students will be expected to do projects directly investigating these issues.