On 10 July 2014 04:35, meekerdb <meeke...@verizon.net> wrote:
> I'm not sure about physics. I think the point of the MGA is that matter
> isn't primary? (As I've already mentioned, I'm not 100% au fait with the
> MGA.)
>
>
> It tries to show that by leading you to accept a scenario in which there
> is no physical action but which you believe is computing consciousness (of
> a dream).
That's Maudlin's argument, in which he uses a particular toy model to show
that the degree of physical action needed to implement any given
computation can be *trivialised* (though not in fact entirely eliminated).
However the MGA, in my understanding, exploits a different tack to reach a
different conclusion. It assumes a device, the systematic relation of whose
physical components is accepted as implementing a computation, which in
turn is assumed to correspond to some conscious state. The argument is then
that, even in the case that any or all of the original "computational
relations" (i.e. logic gates) is disrupted, an equivalent sequence of
physical states can still be made to go through. This can be by fortuitous
accident ("cosmic rays" or suchlike) or by the deliberate superimposition
of a recording of a prior iteration (for this reason the argument exploits
an optical computer).
So the conclusion is that the same sequence of physical states and the same
end product can persist even in the case that every *systematic relation*
between those states, originally accepted as 'implementing a computation',
has been disrupted. Hence, here it is the notion of *computation* itself,
not physical action, that has been trivialised. Essentially, the nub of the
argument is: "You show me a physical device that you claim produces a given
effect *in virtue of its systematically implementing a computation*, and
I'll show you a case in which every trace of said systematic relations can
be evacuated and yet the same sequence of physical states occurs. The real
point of the MGA is to make it obvious that, ex hypothesi physicalism,
derived notions such as computation lack any *effective* role in the
production of a given physical outcome.
An example closer to home would be that the PC on which I am currently
typing might have one or innumerable faults in its logic gates but those
faults are in fact being fortuitously compensated by a series of accidents.
In such a case I would be none the wiser because the same physical results
would be produced and as far as I am concerned those results *just are* the
computation. Or, even closer to home, I may unknowingly suffer disruption
to certain synaptic junctions in my brain, but if these deficiencies happen
fortuitously to be compensated in like manner, my consciousness would be
similarly undisrupted. This latter example is actually rather plausible in
that open brain experiments have shown that external stimulus of brain
cells can elicit memory recall, strongly implying that "fortuitous" events
do indeed elicit the same, or similar, conscious states as those produced
by "normal" brain function.
In short, under physicalism, a 'computation' *just is* a particular
sequence of physical states. Indeed what else could it be? The states, so
to speak, come first and hence the notion that those states 'implement a
computation' is always an a posteriori attribution that neither need, nor
can, bring anything further to the party. In this light the particularity
of physical structures such as my PC, for example, is that they happen to
be arrangements in which certain preferred physical outcomes normally have
a greater probability of occurring relatively reliably rather than
fortuitously. In terms of such outcomes the notion of "physical
computation" can only be a convenient fiction which, in the final analysis
can always be shown to be *effectively* redundant. And this is indeed the
conclusion of my own more general "reductio of reductionism".
David
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
