Large Language Models as Modal Models in Linguistics: A Framework for Understanding Their Explanatory Value
A new preprint proposes using the philosophy-of-science concept of 'modal modeling' to evaluate what large language models (LLMs) can and cannot explain about human language. The paper argues that LLMs occupy a middle ground between being merely speculative tools and being genuine explanations of how human language works. This framework aims to move debates in linguistics beyond polarized positions and toward a more precise assessment of LLMs' scientific value.
A preprint submitted to arXiv by Haruto Suzuki applies the philosophical framework of modal modeling to clarify the role of large language models in linguistic theory. The paper identifies three dominant positions in current debates—insulationism (LLMs are irrelevant to human language), eliminativism (LLMs can replace traditional theories), and conciliationism (LLMs are useful research tools)—and argues that none fully captures the epistemic situation. The authors contend that LLMs have genuine value as 'minimal models' capable of providing 'how-possibly explanations' (HPEs), meaning they can test whether certain outcomes related to language acquisition and competence are even theoretically possible. However, the paper concludes that current LLMs do not meet the stricter requirements for 'how-actually explanations' (HAEs), which would require structural correspondence to human cognitive mechanisms as understood through mechanistic accounts of scientific explanation. The authors propose treating LLM explanatory power as lying on a continuum between HPEs and HAEs, offering a more calibrated basis for integrating LLMs into the scientific study of language.
What's missing
As a preprint, this work has not yet undergone formal peer review, and its central philosophical claims—particularly the application of modal modeling to LLMs—have not been independently evaluated. The paper does not appear to include empirical tests of its proposed framework, leaving open the question of how researchers would operationally determine where a given LLM falls on the HPE-to-HAE continuum.
What different sources said
- arXiv cs.CLCenter
Large Language Models as Modal Models in Linguistics
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.