Good question. The initial claim was that certain models of capacitors have an intrinsic set of properties, a characteristic impact on sound that is pretty much independent of their actual role in a circuit. Now that's just kooky bananas.what are we trying to prove here
There are more reasonable claims pertaining to the idiosyncrasies of some dielectrics (ceramics in particular) and/or methods of construction, relevant for a particular set of circumstances. I.e. a ceramic coupling cap at the input of a high-gain rig would be microphonic due to its piezoelectric properties. Proximity to a loud speaker might result in positive or negative feedback, which would modulate the sound in an audible fashion, that is to say it would sound different if the cap wasn't microphonic, ceteris paribus.
Things that fuel a 30-page debate and prevent us from reaching a consensus:
1. People think we don't understand capacitors well enough, that they have some metaphysical properties or that tiny variations in capacitance or parasitic/nonlinear properties have a significant impact on the signal. Learn about electronics.
2. People mistake their subjective aesthetic values with falsifiable, testable claims. Learn about basic aesthetics and epistemology.
3. People overestimate their powers of perception and ignore their plasticity. Learn about cognitive psychology.
4. Lots of people believe batshit crazy things and they get sore when we don't take them seriously. Stop saying stupid shit.