Large Language Models as Modal Models in Linguistics

2026-06-09

AI Summary

This study addresses the ongoing debate concerning the explanatory status of large language models (LLMs) in linguistic theory by reconciling three competing positions: insulationism, eliminativism, and conciliatory realism. Drawing on a modal modeling framework from philosophy of science and integrating mechanistic explanation theory with philosophical analysis of language, the paper argues that LLMs function as “minimal models” that offer “how-possibly” explanations of language acquisition and competence. It further delineates the conditions under which such models could advance toward “how-actually” explanations. The analysis demonstrates that current LLMs fall short of supporting “how-actually” accounts, yet they possess genuine cognitive value at the “how-possibly” level. Consequently, the explanatory power of LLMs is situated along a continuum between these two poles, thereby avoiding both overestimation and underestimation of their theoretical role in linguistics.

Abstract

The rapid advancement of large language models (LLMs) has intensified debates about their significance for linguistic theory. These debates are commonly divided into three positions: insulationism, which regards LLMs as irrelevant to human language; eliminativism, which claims that LLMs can replace traditional linguistic theories; and conciliationism, which views them as useful tools for linguistic research. To clarify these positions, this paper applies the framework of modal modeling from the philosophy of science. We argue that LLMs possess genuine epistemic value as minimal models, even without structural correspondence to human cognition. In particular, they can provide how-possibly explanations (HPEs) by testing modal claims about language acquisition and linguistic competence. We then examine the conditions under which LLMs could qualify as how-actually explanations (HAEs) of human language, drawing on the mechanistic account of scientific explanation. We argue that current LLMs do not yet satisfy these requirements. On the basis of this analysis, we propose understanding the explanatory power of LLMs as lying on a continuum between HPEs and HAEs. This framework avoids both overstating and understating their explanatory significance and offers a more precise basis for evaluating the role of LLMs in the scientific study of language.

Problem

• large language models
• linguistic theory
• modal modeling
• scientific explanation
• how-possibly explanations

Innovation

• modal modeling
• large language models
• how-possibly explanation
• mechanistic explanation
• linguistic theory