ticipatory perspective – as if we were alien scientists – we would see that
all the basic cognitive capacities are shared by members of this species
wherever they have spread over the planet, resulting from a common
cognitive endowment (see, for example, Brown, 1991).
Whatever the merits of his arguments, Locke’sEssaywas very successful
in establishing developmental empiricism as a dominant paradigm,Wrst in
philosophy and later in psychology. For this reason Chomsky’s ‘Cartesian
linguistics’ (1965, 1975, 1988) – the thesis that innate cognitive structures
are required for the acquisition of grammatical competence in one’s native
language – has had a revolutionary impact throughout cognitive studies.
Before reviewing the central Chomskian argument, however, we ought
to pause over the extent to which the developmental empiricist paradigm is
rendered implausible by a more general theoretical perspective, namely
that of evolution through natural selection. The essential features of the
paradigm are these:
1 Human cognition is moulded in individuals through the experiential
environment to which they are exposed.
2 There are only a small number of in-built mental capacities (such as
selective attention, abstraction, copying, storage, retrieval, and com-
parison) for processing input from the environment.
3 These capacities aregeneralones, in that thesamecapacity can be
applied to representations of many diVerent kinds.
We will argue that cognitive science and developmental psychology have
progressed far enough for us to judge the empiricist paradigm empirically
inadequate to the facts within its own domain – mainly, facts about the
development and functional organisation of cognitive processing. But
there is also a serious question whether a system of such a kind – an
empiricist mind – could ever have evolved in a species whose ancestors had
more limited and inXexible cognitive systems.
Marvellous and intricate as evolutionary adaptation is, we need to
remember that evolutionary development is constrained in terms of both
its resources and its goals – it works on what it has already got (give or
take the odd mutation) and what it ‘designs’ need not be a theoretically
optimal solution. Now, it is clear that modern humans are descended
from creatures with special systems for controlling behavioural responses
to various kinds of environmental information. So we should expect to
Wnd that selective pressure has operated on a range of special systems in
shaping human cognition, and hence that evolved special systems will
structure human cognitive capacities. It is very hard indeed to understand
how a big ‘general-purpose computer’ (which is the way in which develop-
mental empiricists conceive of the human brain) could have developed
Some background on empiricism and nativism 51