🚀 go-pugleaf

In article <1992Oct1.200238.23660@ttinews.tti.com> jackson@soldev.tti.com (Dick Jackson) writes:

>I would like to hear your opinion on how qualia, etc. came into being. I,
>and others here I think) find it satisfying to hold that our mental states
>have evolved to enhance our fitness in an exactly parallel way to our
>obviously physical parts. In other words, my introspection, my
>model of your mental states, my feeling of embarassment when you point
>out my error of logic, all make me function in a superior way to a being
>physically similar to myself, but having, perhaps, just a sort of
>"robotic" approach to survival, food gathering, etc.

I agree with this, but note that "introspection", "model of your mental
states", and even "feeling of embarassment" in some senses thereof
needn't require qualia, in principle.  All of those things are functional
concepts.  So evolution can explain how those processes might have
come along, but it doesn't explain why they're accompanied by qualia.
Incidentally, I certainly think that most of our mental states have
evolved to enhance our fitness -- e.g. belief, learning, emotion,
language capacity, and the like -- as all those mental notions can
be analyzed functionally.  It's only the raw experiential feel that
this doesn't seem to work for.

>If you can't go along with this position, then how would you speculate
>that human consciousness did arise?

Well, on the dualist position that I've been advocating (and that I find
the most plausible, although consciousness is such a mystery that I
can't claim to be certain of anything about it) there are laws of nature
that connect the facts about physical/functional systems to the facts
about qualia.  These laws of nature, just like the laws of physics,
predated the formation of life on earth, although it may have been the
case that before that there was no room for those laws to "come into
play" and produce qualia.  At some stage physical systems came along
with the right kind of configuration, and gave rise to qualia (by
virtue of their satifying the antecedents of those laws).

So, if you like, an evolutionary explanation of qualia will see
them as a systematic byproduct of the right kind of functional
states, where it is the functional states that get selected for.
Maybe you don't like the notion that qualia should be an
evolutionary "byproduct", but note: (1) any representational
content that those qualia seem to have is also carried by the
functional states within the system -- e.g. the perceptual
contents corresponding to perceptual experience -- so that content
will be selected for; it's just the raw experience of the content
that's a byproduct; (2) evolution doesn't add any new plausibility
or implausibility to this story about qualia -- if one accepts a
dualist position of the kind I've been putting forward, then one
must accept that there are laws by virtue of which qualia arise
from physical or functional states.  Given that one accepts that
those laws exists, then it's clear that the evolution of qualia
would work in just this way -- one wouldn't expect the laws themselves
to evolve, any more than one expects the laws of physics to evolve.

>Unless you take the extreme view
>that every collection of atoms has some level of consciousness,
>presumably it must spring forth at some level of system complexity. Do
>we allow that slugs are conscious? Rats? Dogs? Apes?

Well, that's not required for the position as I've been advocating.
It's perfectly consistent for the dualist to hold that consciousness
springs forth at some level of complexity -- i.e. that the laws
of nature concerning qualia are such that no simple system will
have them.  However, I personally find that implausible.  I don't
see why qualia should suddenly "blink in" when the systems reach a
certain degree of complexity -- there doesn't seem to be a place to
draw a principled dividing line.

Given the very close correspondence between perceptual qualia and
the functionally individuated contents of perception in the human
case, I don't see why this shouldn't hold down the line.  So it
seems to me that rats should have qualia: much simpler qualia than
ours of course, but at the least qualia corresponding to the contents
of their perceptual field.  Same for slugs -- although the qualia
will be getting quite simple by then -- and similarly, it seems to
me, all the way down to the simplest information-processing systems,
like thermostats, which will have utterly basic qualia corresponding
to the 3 states of information-processing that they can be in.

This also seems to make for the most parsimonious theory of qualia.
Given that we have to invoke new laws of nature to deal with them,
it seems reasonable to suppose that those laws should be as simple
as possible.  The notion that qualia are just another aspect of
information -- wherever one has information, one has corresponding
qualia -- seems to be about the most simple and reasonable theory
around, and doesn't require adding arbitrary-seeming new laws, such
as "systems with biochemistry X will have qualia".  It also maintains
a close coherence between the functional facts and qualia, which is
important for other reasons.

>Obviously we cannot (yet?) explain how the physical leads to the mental,
>but I claim that is analogous to asking someone who is using a computer
>with a multi-tasking real-time system with multiple I/O channels to
>explain the cause of a particular glitch in terms of the quantum mechanics
>of its silicon transistors.  It is, in a practical sense, impossible --
>despite the fact that at each level of the machine's construction we do
>have a complete physical/logical understanding of the processes involved.
>I find it quite plausible to suppose that computers in a few years time
>will be effectively non-deterministic as far as the user is concerned.
>To find out why it did certain things, we will have to ask it!  Running a
>diagnostic will then be very much like asking "How are you FEELING
>today?".

Note that I'm not asking for any kind of *causal* explanation -- I
agree that those can be very complex.  I'm asking for a *synchronic*
explanation -- corresponding to: given that the low-level physical
facts about the system *right now* are like so, why is it experiencing
a glitch?  The answer to that question is conceptual, not causal.
All it *means* to be experiencing a glitch is to be in a particular
functional state (e.g. one consisting in mess-ups of a certain kind).
It's abundantly clear how the low-level facts can imply that that
functional definition is satisfied, although of course one wouldn't
want the job of spelling it out.  It's just a matter of a role-filler
relation: the physical state does the job of fulfilling the functional
role which is definitive of "glitchness".  There's no corresponding
functional *definition* of consciousness (or, if you like, functional
analysis of the *notion* of consciousness), so no corresponding
explanation has a hope of working.  It would have to be a very
different kind of explanation indeed.

-- 
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."