🚀 go-pugleaf

In article <7666@skye.ed.ac.uk> jeff@aiai.ed.ac.uk (Jeff Dalton) writes:

>So a lot turns on the definition of belief as disposition to 
>behave.  For otherwise we couldn't explain what we believe
>about consciousness merely by explaining behavior (though we
>might still explain what we _say_).

Well, the dispute about whether belief is a disposition to behave
is largely a semantic one.  Presumably we both agree that there are
two components to mentality: (1) the causation of behaviour, and
(2) consciousness (maybe these "turn out" to be the same thing,
but we can't beg the question).  So, maybe there are two
components of belief -- the role played in the causation of
behaviour, and its conscious feel.  It seems to me that the
conscious feel of a belief is at best a peripheral part of it,
and doesn't play a large role in determining what belief it is
(many of my beliefs don't seem to have a feel at all).  So
the components are fairly orthogonal, and might as well be
disentangled.  I choose to call the more central part belief.
If you insist on them both being present, then I'll say OK, maybe
so, but there is this thing which is just like a belief except
that we don't make consciousness constitutive of it -- call it
a schmelief.  And I'm happy to talk about those.  The point, then,
becomes: the functional story can explain everything about belief
except for its conscious feel.

>Behavior like mine might result in a different way.  Something
>that wasn't conscious, with no experience at all, might nonetheless
>have similar (maybe even identical) dispositions to behave.  _Those_
>dispositions would not be because of any experience of consciousness.
>They could be explained without involving consciousness, because
>consciousness is not in fact involved in their production.
>But that does nothing at all to show the same is true of _my_
>dispositions (or those of any other conscious entity).

Obviously, there's a story S we could tell to explain it's behaviour.
Presumably that story would also apply to us -- we're physically
identical, after all.  So it should presumably explain our behaviour.
The only difference is that we have something extra on top; but
I don't see how adding something could that destroy the validity
of explanation of what was there already.

>Suppose behavior is (entriely) physically caused.  Now, suppose
>consciousness is a pattern of physical causation.  How could you then
>say that the explanation of behavior needn't involve consciousness?

Obviously, in that case it would be.  This discussion is premised
on my arguments, which I'm sure you recall, that consciousness
isn't physical.

>So far as I can tell, it simply is the case that the behavior of
>conscious entities is caused in part by their consciousness.  This
>leaves us a choice between (roughly):
>
>  1. Behavior isn't (entirely) physically caused.
>
>  2. Consciousness is effectively physical.
>
>At present, I don't think I have a strong preference between those
>two, but I'd take either of them over an epiphenomenalism that makes
>consciousness irrelevant.

(1) implies that there are big gaps in physical causation, which isn't
impossible but is looking unlikely.  (2) is just wrong, because of
the logical possibility of zombies :-).  So, I'm afraid the
alternative is forced on us.  This epiphenomenalism is a bit
counterintuitive, but it's certainly consistent with the evidence.
What is "simply the case" is that there are regularities between
consciousness and behaviour; it's not *obvious* that these are
causal regularities, and it may turn out that they're not.
(Incidentically, I think there may be ways of reconciling this
property dualism with causal efficacy, but I won't go into them now.)

-- 
Dave Chalmers                            (dave@cogsci.indiana.edu)      
Center for Research on Concepts and Cognition, Indiana University.
"It is not the least charm of a theory that it is refutable."