reasons. First, because they undervalue the cognitive powers of pre-lin-
guistic children, animals, and earlier forms of hominid. ThusHomo erec-
tus, for example, was able to survive in extremely harsh tundra environ-
ments (presumably without language: see below). It is hard to see how this
could have been possible without a capacity for quite sophisticated plan-
ning and a good deal of complex social interaction (as argued by Mithen,
1996). And second, the views of Dennett and Bickerton are inconsistent
with the sort of central-process modularism defended in chapters 3, 4 and 5
above. Our view is that the mind contains a variety of conceptual modules
- for mind-reading, for cheater-detection, and so on – which are probably
of considerable ancestry, pre-dating the appearance of a modular lan-
guage-faculty. So hominids were already capable of conceptual thought,
and of reasoning in a complex, and presumably ‘oV-line’, fashionbefore
the arrival of language (we return to this point in chapter 9).
Why do we think that language evolvedafterthe central-process con-
ceptual modules? Why could not Dennett and Bickerton claim that lan-
guage evolvedWrst, with the various central modules making their ap-
pearance thereafter? Well, there does seem to be an emerging consensus,
grounded in a variety of kinds of evidence (and accepted by Bickerton),
that the evolution of language coincided with the appearance ofHomo
sapiens sapiensabout 100,000 years ago in Southern Africa (see Mithen,
1996, for reviews). In which case it is very unlikely that there would have
been time for a number of complex modular systems to emerge between
then and the human dispersal around the globe just a few tens of millennia
later. And it is, moreover, hard to understand anyway how language could
have evolved in the absence of quite highly developed mind-reading abil-
ities (Gomez, 1998), which in any case we have reason to believe are
present, in primitive form, in the common ancestor of ourselves and the
other primates (Byrne, 1995).
3.4 Conscious thinking 1: learning and inferringEven if language is not what underpinsconceptualthinking, as such, it may
be that it is the vehicle forconsciousconceptual thinking. It may be that
imaged natural language sentences, in ‘inner speech’, are the primary
vehicles for our conscious propositional (as opposed to visuo-spatial)
thoughts (Carruthers, 1996c; Mithen, 1996). On this view, a variety of
modular central-systems – crucially including a mind-reading module –
would have been in place prior to the evolution of language; as would have
been a capacity for various forms of sensory imagination (Wynn, 1993). At
this stage, hominids would have been capable of attributing thoughts to
themselves more-or-less reliably on the basis of self-interpretation, rather
The place of natural language in thought 217