Grammaticality Judgments in Humans and Language Models: Revisiting Generative Grammar with LLMs
By: Lars G. B. Johnsen
Potential Business Impact:
Computers learn grammar rules from reading text.
What counts as evidence for syntactic structure? In traditional generative grammar, systematic contrasts in grammaticality such as subject-auxiliary inversion and the licensing of parasitic gaps are taken as evidence for an internal, hierarchical grammar. In this paper, we test whether large language models (LLMs), trained only on surface forms, reproduce these contrasts in ways that imply an underlying structural representation. We focus on two classic constructions: subject-auxiliary inversion (testing recognition of the subject boundary) and parasitic gap licensing (testing abstract dependency structure). We evaluate models including GPT-4 and LLaMA-3 using prompts eliciting acceptability ratings. Results show that LLMs reliably distinguish between grammatical and ungrammatical variants in both constructions, and as such support that they are sensitive to structure and not just linear order. Structural generalizations, distinct from cognitive knowledge, emerge from predictive training on surface forms, suggesting functional sensitivity to syntax without explicit encoding.
Similar Papers
Do Large Language Models Grasp The Grammar? Evidence from Grammar-Book-Guided Probing in Luxembourgish
Computation and Language
Tests if computers truly understand language rules.
Comparing human and language models sentence processing difficulties on complex structures
Computation and Language
Computers understand sentences like people do.
Syntactic Blind Spots: How Misalignment Leads to LLMs Mathematical Errors
Computation and Language
Fixes math problems by changing how they're asked.