Causal Interventions Reveal Shared Structure Across English Filler-Gap Constructions
This work addresses linguists seeking to refine theories of syntax using mechanistic insights from language models, though it is incremental in applying existing causal methods to a specific linguistic domain.
The paper tackled the problem of understanding whether language models learn shared abstract structures for English filler-gap constructions, and found that they converge on similar analyses, revealing overlooked factors like frequency and context that could influence linguistic theory.
Language Models (LMs) have emerged as powerful sources of evidence for linguists seeking to develop theories of syntax. In this paper, we argue that causal interpretability methods, applied to LMs, can greatly enhance the value of such evidence by helping us characterize the abstract mechanisms that LMs learn to use. Our empirical focus is a set of English filler-gap dependency constructions (e.g., questions, relative clauses). Linguistic theories largely agree that these constructions share many properties. Using experiments based in Distributed Interchange Interventions, we show that LMs converge on similar abstract analyses of these constructions. These analyses also reveal previously overlooked factors -- relating to frequency, filler type, and surrounding context -- that could motivate changes to standard linguistic theory. Overall, these results suggest that mechanistic, internal analyses of LMs can push linguistic theory forward.