3.1 The optionsWhat place can be found for sentences of natural language in central
cognitive processes? The orthodox cognitive science answer is: ‘None’. One
part of the reason for this has just been given. It is that researchers have
assumed that any sort of positive answer would commit them to views
which they regard as absurd, such as that it is conceptually impossible for
animals and infants to entertain genuine thoughts. But another point is
that language is, almost universally, believed to be a distinct input and
output module of the mind; which might seem to make it diYcult to see
how it could, at the same time, be crucially involved in central cognition.
And yet another reason is that, since non-linguistic creatures are believed
by most to share at least some of their cognitive functions with us (prac-
tical reasoning, say), natural language sentences cannot be crucially re-
quired to execute such functions (even if the modality of ‘required’ here is
just that of natural, rather than conceptual, necessity).
TheWrst of these reasons collapses as soon as the requisite distinctions
are drawn. To claim that some thought implicates natural language, either
as a matter of fact or out of natural necessity, is perfectly consistent with
denying that it isconceptuallynecessary thatallthought should involve
language. So we can reject the strong claims of some philosophers, while
continuing to maintain that natural language is crucially implicated in at
least some forms of central thought-process.
The second reason is also weak, as the analogy with the visual system
makes clear. This is the very paradigm of a modular input-system. Yet, as
noted in chapter 3, it is known to share mechanisms with imagination,
which is surely a central process, implicated both in reasoning and in the
Wxation of belief (Kosslyn, 1994). Just as central cognition can recruit the
resources of the visual module in visual imagination for purposes of
visuo-spatial reasoning, so it may also be able to recruit the resources of
the language module, in ‘inner speech’, for purposes of conceptual or
propositional reasoning. This point is worth elaborating on further.
According to Kosslyn (1994), visual imagination exploits the top-down
neural pathways (deployed in normal vision to direct visual search and to
enhance object recognition) in order to generate visual stimuli in the
occipital cortex, which are then processed by the visual system in the
normal way, just as if they were visual percepts. Normal visual analysis
proceeds in a number of stages, on this account. First, information from
the retina is mapped into a visual buVer in the occipital lobes. From here,
two separate streams of analysis then take place – encoding of spatial
properties (position, movement, and so on) in the parietal lobes, and
encoding of object properties (such as shape, colour, and texture) in the
The place of natural language in thought 209