I've been mulling over the idea that articulatory gestures should be looked as excursions from a home base that varies according to language and accent, and that defining sounds just in terms of the target of the gesture leaves out something important.
For example, the basic set up in French seems to have more tension somewhere than the basic set up in English. French speakers are not moving from the same place, or do not have the same perception of where neutral is.
If I'm thinking along the right lines here, how can home base / the neutral position be characterized? What variables would define it for a given pronunciation system, and is it possible to put values on them by analyzing recorded speech (or would you need to make direct physiological measurements)? In a language like English, which has weak forms of certain vowels, can you locate home base by drawing lines from the strong forms through the weak forms and seeing where they cross / come closest? Otherwise, could it be identified by looking at non-verbal utterances that might otherwise be language-independent (I mean the sounds people make when they are excited, hurt themselves are hesitating etc. etc.)