It is difficult to imagine the process of migration of the auditory discriminations from natural production to artifact. In the first instance, sounds are produced that generate synaptic effects in members of the species. Sounds and concurrent events are registered as auditory and visual cues, and are discriminated in some way. The precise character of this discrimination hardly matters for our purposes. What is significant is that auditory cues are eventually selected and reproduced so that indexical visual effects are recreated, as it were, on command.
The point is easily illustrated. Imagine two people engaged in an activity. Their engagement is primarily kinetic, but since their contributions are complementary, their awareness, though partly kinesthetic, is also partly visual, since each sees the other's contribution, and secondarily auditory, since the activity also produces (perhaps characteristic) sounds. In consequence, their activity produces concerted visual and auditory synaptic effects. The coordination of the activity yields a coordinated composition of neural effects. To borrow a human term of folk-psychology, we could say that these effects are associated in virtue of being produced in this way. The neural details of the association remain somewhat obscure. However, the significance of this association, perhaps the substance of it, is that one kind of stimulation is capable, in some measure of functionality, of doing the work of two.
Thus, even in the absence of auditory stimulation, just the visual cues generated by the activity might be capable of stimulating a sufficient range of auditory synaptic effects for some kinds of control; similarly, so might just the auditory cues be sufficient.8.1
The suggestion that neuronal structures can be arranged such that partial excitation is enough to generate complete stimuli is convincingly explored in Calvin's theory. The details given about structural features are somewhat speculative but the account is sufficiently consistent with experimental data to make a good argument. Calvin relates the experience of one humming a few bars of Beethoven's Fifth and suddenly the entire piece is playing in someone else's head. He goes on to explain how his model can explain this phenomenon. Most of the explanation has already been explored in the previous chapter.
In daily life the phenomenology of this neural set-up is taken for granted as a feature of our common experience. However, that there is evidence for it in the most fundamental facts of language suggests that one of the earliest stages of its pre-development was the exploitation of this kind of migration. The exploitation features the artificial reproduction of such effects; the artificial recreation, for example, of sounds producing effects functionally, like the multimodal effects of a natural activity.
In any adequate account of the evolution of language, it may be presumed, some such development would underlie the earliest steps. These would describe the shift from which the sounds of human, or rather pre-human, activities are eventually conventionalized and emancipated from their natural occurrences.