From “Silent heads”

As manual and facial sign systems are highly effective expression modes, it seems reasonable to question the assumption that there is something about the PF component of oral languages that is ‘more linguistic’ than any other motoric mode of expression … Indeed, since there are various well-known ‘non-oral’ ways of signing at least parts of language, with smoke, flags, hieroglyphics, etc. (in the sense that hieroglyphics were deciphered without any access to their ancient oral ‘base’), we can see no reason why it should not be possible to, for instance, dance language to a considerable extent, given a persistent social or biological (hereditary) need to do so over a sufficiently long period of time. Such a language would presumably give rise to ‘dance linguistics’, with ‘dance morphemes’, etc.