3.1 The case for teleo-semanticsOne source of attraction is to note that the mind is an evolved system as
well as the body. Since the mind, like the body, has been shaped and
selected by evolution, we should expect toWnd within it systems and
mechanisms withproper functions– that is, systems which aresupposedto
act in one way rather than another, in the sense that they only exist at all
because theyhaveacted in one way rather than another in the past, and
proved successful. And some of these systems will be those which process
information, set goals, and execute plans. Indeed, it seems natural to think
that propositional attitudes – beliefs and desires, in particular – will have
proper functions, being supposed to operate in one way rather than
another within our cognition. Desires aresupposedto get us to act, and
beliefs aresupposedto guide those actions towards success, in any given
environment, by providing correct representations of the state of reality.
And then it is but a small step from this to the thought that thecontentsof
propositional attitudes will have functions too.
(Note that this small step is resistible, however; as Fodor – 1990, ch.3 –
points out. From the fact that hair has a function – to protect the scalp
from sunburn, say – it certainly does not follow that any individual hair
has a unique function, distinct from that of the others; indeed, it seems
most unlikely that it has. So from the fact that beliefs in general have a
function, it does not follow that thebelief that Phas a function distinct
from that of thebelief that Q.)
This gives us the project of teleo-semantics. If we could say what it is for
a state to have the content that P in terms of what that state issupposedto
achieve in cognition, then we would have eVected a naturalistic reduction,
provided that the notion of proper function appealed to in the account is a
genuinely biological one. Roughly, the idea will be that the content (truth-
condition) of a belief is that state of the world which enables the belief to
achieve those eVects (namely, successful action) which it issupposedto
achieve (that is, which it is its function to achieve).
Another way of motivating this sort of teleological account, is to notice
that many naturally occurring signs employed by biological systems only
rarely co-occur with the phenomena which (one wants to say) they rep-
resent. For an evolved feature does not have to be always or often suc-
cessful in order to be selected for. It just has to confersomeadvantage on
organisms which possess it (without incurring any signiWcant disadvan-
tage). For example, suppose that a particular species of ground-squirrel
uses a series of alarm-calls to warn of potential predators (somewhat like
the vervet monkey, only rather less discriminating) – one call foreagle, one
forsnake, and one forbig cat. It is quite possible that the call which means
168 Content naturalised