[Wittrs] Re: Is Computation too Static to Sustain a Mind?
- From: "gabuddabout" <gabuddabout@xxxxxxxxx>
- To: wittrsamr@xxxxxxxxxxxxx
- Date: Sat, 07 Aug 2010 20:13:23 -0000
> Charlie said:
>
> > I observe:
> >
> > 1. Computing is the method of applying or performing algorithms on static
> > frames of data.
> >
> > 2. Consciousness and thinking are dynamic processes, of which computing has
> > none.
> >
> > Given these self-evident truths, how do you propose to reconcile
> > "consciousness in a computer?"
> >
> > Best regards,
> >
> > Charlie
> >
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> My response below:
>
> Well we don't know what consciousness (or the array of features we group as
> "consciousness") is so, of course, there isn't a genuine logical problem
> posed by your two seemingly incompatible statements below [above--Budd].
We have a starting definition that awaits an analytic definition after enough
empirical inquiry: We start with the idea that consciousness is a qualitative
state of ontological subjectivity--ontological because one is really conscious
sometimes (except during some philosophical debates where consciousness is
defined in behavioral or functional terms..).
>
> You speak of consciousness as consisting of "dynamic processes", just as you
> speak of computing as "static". But the issue is what underlies that which
> you call "dynamic"?
BP.
After all, why shouldn't it be the sort of thing you call "static" in
computers?
Because he's distinguishing the formality (PP and serial processing) of
computation with the dynamics of BP (even though any BP system can be given,
willy-nilly, a computational description.
>
> These terms seem to have a meaning to us and the meanings seem to be clear.
Just as the preliminary idea of cons. as ontological subjectivity except for
arbitrary definitions which exclude our considering ontological subjectivity
from a scientific point of view..
> What is dynamic is not static and vice versa. And I would say that's true.
> But I would add that there are ranges of meaning and that what we mean in any
> situation always depends on context, on the use.
Right, so one can give a computational explanation for any dynamic process--we
just describe the most dynamic BP as the most complex type of computation;
computations can be described in terms of easy computations along a range to
the most complex (that we can think of). In the limit, BP = PP, even though
Searle would still insist that the meaning of PP (even in the mouths of strong
AIers) involves PP as a type of BP but with necessary computational
explanation. For Searle, _all_ computational explanation can be dropped if we
truly understand the system. That's why some insist that the hard problem is
to be dropped in favor of explanations that really don't explain ontological
subjectivity.
>
> For instance, while I think we would both agree that "static" means without
> movement, without change, etc., I'm not sure that THAT meaning is applicable
> to computers or computation. Yes, in a computational machine the processes
> can be described as a series of fixed states. But once you introduce the idea
> of "a series" into this, we are back to being "dynamic" aren't we?
Not really. That the zeros and ones happen serially or in parallel doesn't net
you a dynamic system at all if the explanation is a computational one. That is
why some insist on randomness of PP--but then the randomness doesn't have a
fixed computational explanation. It is BP without that type of explanation.
So then, what is the point of the strong AIer? Just BP after all? Or BP plus
PP?
>
> Well you may say that's not what you mean by "dynamic". You have in mind the
> flow of things we encounter in our own experience, our own consciousness.
Or BP regardless of thinking about consciousness.
> Surely THAT is dynamic!
Don't know if anyone ever argued the contrary!
> But note that that says nothing about whether THAT sense of changing states
> is or isn't grounded in a series of fixed states in brains, states that are
> fixed in roughly the same way computational states in a computer operation
> are fixed.
Right, BP may be another way of explaining PP and vice versa. For Searle, we
can drop the PP if we understand the system. More, we can simulate any system
we understand serially (all PP can be simulated serially).
>
> If by "dynamic" on the other hand you just mean the ability to respond on the
> fly, so to speak, to incoming stimuli, i.e., the capacity of the platform to
> operate with flexibility and even creative autonomy in the way most (perhaps
> all) programmed computers cannot operate today, I would just note that we
> aren't speaking of those computers but of a complex system designed and built
> to more robust specifications which is yet to be developed and implemented.
Right, maybe something like BP after all!
> The question of the viability of computationalism as a theory of
> consciousness is not whether any given machine available today is conscious
> but whether a machine that is built of nothing more than such components ever
> could be.
Otiose, given that we're after BP-type explanations which don't need
computational explanations.
>
> If being "dynamic" and being "static" are not strictly contradictory as you
> assume (above), then there is no reason why a machine that is described as
> consisting of static states in the way you mean (by your reference to
> "computers") could not, at least in principle, be constructed to operate in a
> dynamic way vis a vis the world or to have the dynamism of experience that we
> find in ourselves.
>
> <snip>
>
> SWM
That sounds right, too. The point about distinguishing the static and dynamic
is just the familiar Searlean reason for harping on the fact that programs have
a formal description and we're after scientific explanations which don't
involve programs.
If, on the other hand, you want to talk of dynamic PP, you may just be
considering the PP system you think may be equivalent to the causal power of
brains which actually do cause consciousness, as if BP is equivalent.
But then no computational explanation per se would be the type where we
understood the system in terms of BP without PP.
If you want to insist on PP as some different way of causing cons. (as opposed
to the way brains do), then one might offer PP as a computationalism that is
dynamic enough to emulate the causal powers of actual brains. If so, is the
different way in terms of computational complexity as if BP complexity? Yes.
There are BP systems of various types, some of which (all of which?) may be
interpreted as composed(?) of computations.
But at the end of the day, notice that even you will insist that the
explanation of a computational system is a physicalist one. How? By noting
that all PP = BP-type explanation.
Or not.
If not, then one has to show how the notion of "information processing" is
carved off the physics, rather than as Searle describes the situation: read
into it.. :-)
Cheers,
Budd
=========================================
Need Something? Check here: http://ludwig.squarespace.com/wittrslinks/
Other related posts:
