There is definitive evidence the brain contains multiple mechanisms for pitch ratio detection.[1] Spectral theories of dissonance like Sethares', meanwhile, fai

There is definitive evidence the brain contains multiple mechanisms for pitch ratio detection.[1] Spectral theories of dissonance like Sethares', meanwhile, fail to explain some very basic phenomena.[2] It should also be noted that people have reported trouble reproducing Sethares' dissonance curves which Sethares has, to my knowledge (we've corresponded on the subject for almost 20 years), never addressed.[3] Lastly, there are serious problems with the ethnomusicology in Sethares' book. The scales derived from timbres, e.g. of Indonesian instruments, are designed to match the dissonance curve of the timbre only in a single mode. Intervals not measured from the chosen tonic will in general not be present as minima in the corresponding dissonance curve (except in the case of highly symmetrical scales like the diatonic scale). [1] e.g. http://dx.doi.org/10.1152/jn.00271.2001 http://dx.doi.org/10.1152/jn.00271.2001 [2] Spectral models are all based on pairwise interactions between partials and therefore predict roughly equal dissonance scores for harmonic and subharmonic chords (including the major and minor triads). But subharmonic chords are dramatically more dissonant, as you can hear for yourself http://lumma.org/music/theory/demo/majmin/ http://lumma.org/music/theory/demo/majmin/ (files beginning "o" are harmonic; "u" subharmonic) This is a consequence of chords larger than dyads being almost completely ignored for the entire history of psychoacoustics. A strangely analogous situation may exist in economics, where the theory of comparative advantage is thought to be settled by most economists despite that there is no consensus on how to extend it beyond Ricardo's 2-country, 2-goods model of 1817. [3] Briefly, the shape of the dissonance curves is unrealistically sensitive to the relative amplitudes of the input partials.