The debate on internalism and externalism is often constructed in the
form of an outright opposition between mental
content and mental causation. This reinforces a tendency to take
sides in the debate. Some may claim that because
there is no proper explanation of the role of externalist mental content
in the internal causation of behavior,
physicalism has failed. Others (most notably Jaegwon Kim) side with
physicalism and argue that this leaves no room
for a causal role of mental content (or of mind in general).
We submit, however, that the debate on internalism versus externalism does
not need a winner, but rather a
dissolution of the opposition between the two. In this paper we will
indicate a part of the way out of the 'winner takes
all' approach to the debate by focussing on the internalist physicalist
claim that externalist mental content cannot
play a genuine causal role in the production of behaviour. Physicalism,
according to Kim, provides a dilemma:
either the mental can be reduced to the physical or it cannot. In the
first case mental content is becoming a mere
epiphenomenon, and in the second case the irreducibility of the
mental leaves its causal powers unexplained and
hence mysterious.
Contrary to Kim, we will attempt to retain physicalism while escaping
the dilemma. We will indicate that this
'epiphenomalize or mysterialize' dilemma is false because it presupposes
a view of explanation and reduction that,
though it is standard in cognitive science, does not do justice to
the dynamical nature of cognition. On closer
examination, the standard cognitivist explanatory strategy (i.e.
reduction via functional analysis (decomposition and
localization)) turns out to be valid for aggregative systems in which
internal component interaction is minimal.
Obviously, this coincides well with the traditional cognitivist view of
the mind as a system of representation-passing
components.
However, we maintain that the mind is an embodied, embedded system
of a dynamical nature to which the standard
cognitivist explanatory strategy does not apply. In many cases these
systems display circular causality which means
that higher level variables constrain the behaviour of lower-level
components. Thus, the notion of circular causality
exemplifies the importance of higher level variables for the causal
processes operating at a lower level. We will
provide several examples to illustrate the circular causality involved
in cognition to illustrate the inadequacy of the
standard reductive explanatory strategy.
We conclude, then, that Kim's dilemma presupposes a model of reductive
explanation that is inappropriate for the
dynamical aspects of cognition. Specifically, physicalism does not
have to lead to epiphenomalism or mysterialism.
More generally we claim that a proper understanding of the dynamic nature
of cognition can provide a way to
escape from the perennial opposition between externalism and internalism.