Through a glass darkly: a critique of the influence of linguistics on theories of music

If music is treated as a kind of 'language', then it makes sense for musicology to borrow from linguistics in order to define exactly what sort of a 'language' music is. However, not only does this avoid the challenge of defining music on its own terms, it also brings across a whole lot of unnecessary historical baggage. Through a close analysis of selected texts, the current article critiques some recent work in musicology, showing how it is based on a far too narrow understanding of what constitutes linguistics, i.e. basically formal linguistics. It points out some of the dangers of interdisciplinary work such as Brown's (2001) 'musilanguage' model, and shows how a less than careful borrowing of linguistic concepts can vitiate the usefulness of such system-building. It then traces the source of much of this borrowing in Lerdahl and Jackendoff's (1983) highly influential work drawing on generative linguistics, and shows how a framework that privileges structure over meaning, system over text, and the cognitive over the social, is unable to provide a broader understanding of music beyond pattern recognition. It calls for a greater methodological scepticism among musicologists towards linguistics, and a greater self-consciousness about borrowings across disciplines.