Resemblance theories of musical expressiveness: the role of empirical evidence

Resemblance theories are a popular way of accounting for music’s capacity to be expressive of emotions. According to such theories, musical expressiveness depends on a perceived resemblance between music and human expressive behavior, both vocal (prosodic features, expressive vocalizations), and bodily (gestures, posture, deportment). This article asks whether the available evidence from the field of music psychology supports resemblance theories over their competitors, and answers the question in the negative. I divide the relevant scientific literature into two areas. The first consists in studies that show structural analogies between musical structures and expressive behavior. The second posits causal mechanism that link our experience of expressive music to the way we experience human expressive behavior. I show that neither kind of evidence should ground a preference for resemblance theories over competing views. In the final part of the article, I ask whether the same evidence could be used to settle an issue internal to the resemblance theory, that is, the relative importance to our experience of expressive music of vocal vs. bodily emotional expressive behavior. I conclude that an answer to this question depends on how exactly we interpret some of the resemblance theory’s commitments.