From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wupost!darwin.sura.net!jvnc.net!netnews.upenn.edu!libra.wistar.upenn.edu Mon Dec  9 10:47:19 EST 1991
Article 1788 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wupost!darwin.sura.net!jvnc.net!netnews.upenn.edu!libra.wistar.upenn.edu
>From: weemba@libra.wistar.upenn.edu (Matthew P Wiener)
Newsgroups: comp.ai.philosophy
Subject: Re: Dennett on Edelman--what a total loss
Message-ID: <57864@netnews.upenn.edu>
Date: 1 Dec 91 20:07:28 GMT
References: <57730@netnews.upenn.edu> <1991Nov29.050859.21552@bronze.ucs.indiana.edu>
Sender: news@netnews.upenn.edu
Reply-To: weemba@libra.wistar.upenn.edu (Matthew P Wiener)
Organization: The Wistar Institute of Anatomy and Biology
Lines: 229
Nntp-Posting-Host: libra.wistar.upenn.edu
In-reply-to: chalmers@bronze.ucs.indiana.edu (David Chalmers)

In article <1991Nov29.050859.21552@bronze.ucs.indiana.edu>, chalmers@bronze (David Chalmers) writes:
>In article <57730@netnews.upenn.edu> weemba@libra.wistar.upenn.edu (Matthew P Wiener) writes:
>>Amazing.  You call me incorrect, and then start citing exactly what it
>>was that led to me write >> above.
>
>That's right.  To spell things out: Dennett calls Edelman's work
>"an instructive failure", and makes some remarks about how Edelman
>has ignored other relevant work, but this is far from "completely
>dismissing" Edelman.

Not really.  Who is Dennett to say that this other "relevant work" is
indeed relevant?  Nobody has an inside track here.  To imply that he
does, and that Edelman doesn't, and to throw in some personal remarks
about Edelman not knowing enough, is a complete dismissal in my book.

>>If Dennett thinks that Edelman is salvageable,
>>then why does he think it's ultimately unsuccessful?  I'm baffled.

>Because even wrong theories can have a lot that's right in them.
>Dennett presumably thinks that Edelman makes a number of insightful
>points out of which a good theory could have been built.

But he presumably doesn't want to encourage people to look at them?
The ad hominem, the deprecating "wait-and-see" until this theory has
been run through the paces, etc, are unnecessary.  I remain baffled.

>>Huh???  Edelman does not treat connectionism.  There is one reference
>>to it in his trilogy, and he says the models lack the precise neuro-
>>anatomical detail that he wants in a brain/mind model.  No more, no
>>less.

>See "Real Brains and Artificial Intelligence", in the 1988 special
>issue of Daedalus on AI.  This paper identifies among the core tenets
>of connectionism: [...] In other words, he has identified
>connectionism entirely with the use of Hopfield nets and Boltzmann
>machines, which in fact form a small and non-central subset of the
>field.

Considering that he avoids connectionism in his trilogy, but to
point out it's non-anatomical detail, perhaps we can blame Reeke
for the poor show here?  At the most, it's a side track.

>						    He appears not
>to have heard of backpropagation, for instance, which has been
>much more central to connectionist practice, and which fits none
>of the descriptions above.

This is my least favorite pro-AI argument.  You get all petulant,
pointing to yet another tree when the other fellow is trying to
study the forest.  As if one more programming technique is all
that's needed.  From the air, one more technique has always been
one more kludge.  The frame problem still can't find the frame.

>>Digitial computation is not as universal as you are blanketly asserting,
>>and Penrose knows this.  I mentioned this in a previous article: [...]

>Of course it is not a necessary truth that digital computation can
>simulate any physical process.  But the demonstrated power of
>digital computation so far makes it a sufficiently plausible claim
>that the burden of proof falls squarely on those who wish to argue
>the opposite.

"Falls squarely"?  Grow up.  And when you do, come back with a more
sophisticated argument than I-think-I-can-I-think-I-can.  AI has been
touting this argument for decades, and it as worthless as it is thin.
  
Five decades of digital computation have shown that there is a lot it
can do, and that there is a lot that we have no idea about what it can
do.  The burden of proof falls on no one.

And meanwhile, your original blanket assertion:
>>>Digital computation has this nice property of being universal, so that
>>>unless you have strong arguments about why very low-level processes
>>>might be uncomputable -- arguments of the kind Penrose would like to
>>>have but doesn't -- then there is little reason to reject the
>>>assumption that the behaviour of biological systems is computable.
is so much vaporware.  I have no burden or proof here.  I need no
argument stronger than "we don't know, but we have seen ugly tigers".

Digital computation does NOT have this nice property of being universal.
Within its own digital domain it does, no more, no less.  You were,
ironically, feeding us the very argument that Lucas gave that Goedel's
incompleteness theorem implies that our minds are not Turing.

>>The work of Deutsch, Landauer, Feynman, Margolus, etc has--well `shown'
>>is a bit strong, but `suggested' is a bit weak--anyway, they have
>>indicated that quantum mechanical computation is, in principal, a far
>>superior beastie, when it comes to speed, than classical Turing compu-
>>tation.

>This is irrelevant to the universality claim.  Of course Deutsch
>et al wanted to demonstrate the possibility of non-Turing-computable
>mechanisms, but as is well-known, they came up empty-handed.

I don't think they are irrelevant yet--this is a line of research that
has not finished.  Just the discovery of speedup was shocking.  They may
learn a non-computable trick or two from Pour-El and Richards.  I don't
think anyone in physics has come to grips with Geroch and Hartle, on
their possibility that quantum gravity is noncomputable.

>							       The
>computational complexity results are interesting, but a difference
>in speed falls far short of the radical difference in power that they
>hoped for.

But it also makes it much more plausible to say that QM is essential to
how our minds work.  Fragments of the intelligence question might best be
answered by saying we follow dumb algorithms with QM speedup.  And with
one QM mechanism in place, who knows how much more might be there?  Some
of which are not computations?

Deutsch, in fact, has a hard time thinking of quantum parallelism as a
computation.  To him, the only way to call it computation is to believe
in the many worlds interpretation!

>>In short, there is good physical speculation behind the notion that
>>Church's thesis is on its way out.

>I don't know what "good speculation" comes to, but there's certainly
>no good evidence. 

We do certain computations very fast.  How?

We observe a classical world.  Why?

These are two questions that must be answered, and a theory that does
not include proposals for dealing with them is incomplete.  (Although
the second one may have a purely QM answer that has no bearing on our
brains.  It may not.)

>		    Nevertheless, I wouldn't completely dismiss the
>possibility that the quantum level is noncomputable.  Even if it were,
>however, it would remain to be seen whether these quantum
>noncomputabilities amplified into relevantly noncomputable phenomena at
>macroscopic levels.

That's my point.  We don't know, but some of the most plausible answers
to the above two questions are firmly quantum mechanical.

>>This is pretty funny.  When Dennett dismisses Edelman, it's "correct",
>>but when Edelman does the same, it's "waffly" and "irrelevant".  And
>>this has nothing to do with whom you're rooting for?  Remember, none
>>of us know who is really right in the end.

>What makes you think you know who I'm "rooting for"?

My own bafflement at the lightness with which you view Dennett on Edelman.
It seems fairly ugly to me.  Do you think I was overstating my annoyance,
repeated below, regarding Dennett on Edelman re continuity?

>>>>At another point, Dennett mocks Edelman's discussion of the continuity
>>>>of consciousness.  What would it be like, if each morning, we woke up
>>>>with third person memories of our previous existence?  Who knows?  And
>>>>why don't we normally perceive space and time as fragmented?  Again, who
>>>>knows?  But we don't, and as Edelman puts it, and Dennett quotes him,
>>>>"one of the most striking features of consciousness is its continuity."
>>>>Dennett's response is to say, no, this is a "crashing mistake" and is
>>>>"utterly wrong", and that what's really striking is discontinuity, and
>>>>then enumerates an example or two, like blindsight.

>>>>I suppose it would be hopeless to explain to a Dennett, even in words
>>>>of few syllables, the meaning of the six blind men and the elephant.

As I explained to someone in e-mail, one can view this as a figure/ground
dispute.  To Edelman, decades of neurological immersion have made the raw
discontinous data of our nervous systems the ground, and the continuity
of consciousness the figure that stands out, in contrast to the intro-
spective modeller of the mind, who takes the ground as his own continuous
consciousness, and the figure that stands out are the experiments that
reveal the experinental underpinnings are discontinuous.

If Dennett wants to be on one side of this perspective dispute, that's
perfectly OK.  If he thinks Edelman is obligated to have the same point
of view as himself, he's crossed the line into ignoramosity.

>>>All AI needs to hold
>>>is that brain processes can be simulated by digital computation -- a
>>>claim that Edelman doesn't come close to refuting.

>>Why does he need to?

>I don't think he needs to, but he certainly tries.  E.g. on page 29
>of _The Remembered Present_, he seems concerned to refute the claim
>that "what the brain does may be described by algorithms".  [...]

Like I said, it's only a half-hearted attempt on his part.  Do you
agree or disagree with my more complete answer here:

>>Why does he need to?  Would Isaac Newton have been improved with an
>>explanation of why orrery construction is not the same as an under-
>>standing of gravitation?  Edelman is interested in the biological
>>basis of our minds.  It is merely the vagaries of modern intellectual
>>history that force him to explain why he is in the "let's understand
>>gravity" camp as opposed to the "let's construct orreries" school.

You agreed that he doesn't need to.  If you go further in your agreement
with me here, you'll end up concluding that this part of his "Proposals
and Disclaimers" chapter is of course a quickly fabricated hedge, and
not worth the time.

>>And grounded in the structure of actual brains, and loaded with evolu-
>>tionary plausibility.  That's not "just another guy".  These aren't just
>>"some ways" of being sophisticated.

>Edelman has a reasonably interesting theory that probably has a number
>of things right.  You'll note however that I haven't said anything
>substantive about his theory, partly because I don't consider myself
>qualified to comment on the neuroscience.  I've only been discussing
>his arguments against other approaches, all of which seem to me to be
>pretty feeble.  The point is that Edelman has been far more concerned
>to distance himself from others than is necessary.

Maybe this is the strong point?  A fresh start, away from preconceived
introspective notions or nonbiologically specified connectionisms?  What
appealed to me more than anything was his insistence on evolutionary
and morphogenetic plausibility throughout.  Actual biology.  And if the
only way to get there is to tell the non-biologists along the way that
they are running in the wrong race, then I say it's another plus for
Edelman.  (And if Prof McCarthy wants to add Edelman to the Dreyfus et
al crowd who haven't considered the significance of nonmonotonic logic,
that's McCarthy's problem, not Edelman's.)

I'll second the not-qualified-to-comment position.  As I said before,
I consider myself "shell shocked" by his trilogy.  And you'll note that
Edelman firmly rejects any QM considerations, leaving my two questions
above open.  So Weemba on Edelman is but a reserved thumbs up.  And
Dennett on Edelman remains a total loss.
-- 
-Matthew P Wiener (weemba@libra.wistar.upenn.edu)


