Form with no substance: Substance-Free Phonology and amodal grammars
Year: 2026
Type: article
Abstract: Substance-free approaches to phonology hold that phonological representations contain no information about their physical realization. If this is right, phonology should require no modality-specific machinery, and what spoken and signed languages share should abstract away from the channel through which either is externalized. This article asks what, exactly, is shared, and what neuroscientific evidence could bear on the answer. The answer is not a set of features. If features are individuated by their relations within a system, and if the inventories emerging in the two modalities differ in size, dimensionality, and organization, then there is no way to ask if a signer and a speaker possess the same feature. What is shared is the combinatorial architecture that generates phonological structure, the range of oppositions a contrastive system makes available, and the principles by which inventories are acquired. The inventories are modality- and language-specific, though substance-free; the mapping to phonetic exponence is acquired, modality-specific, and lies outside the grammar. It is argued that a feature specified in terms of one channel is uninterpretable to a learner acquiring a language in another. Neural evidence cannot settle whether two modalities share features, since no linking hypothesis connects features to neural objects. It can address whether the relations among representations correspond where their contents do not. That comparison is available through representational similarity analysis in bimodal bilinguals, and has not yet been attempted.
Author: Alex Chabot
Institution: University of Maryland, College Park
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $551
OpenAlex ID: W7221892409
