From: Brian Holtz [brian@holtz.org]
Sent: Sunday, February 23, 2003 8:23 AM
To: Daniel Pech
Subject: RE: philosophy of physics (AI/intuitionism2)
P: I meant that, for us, true philosophical problems are not (meant to be) resolved by a direct attack upon the problem, because for it to be a true problem for us means that it is greater than what we at first see of it. The problem is irreducibly complex (IC). If the calculation lacks some of the fundamental facts, then the commitment to establishing a conclusion will result in an unsound conclusion.
?
P: the solutions to all problems are, without exception, irreducibly complex
H: "All"?  I strongly disagree.
P: I'm not sure you understood what I meant by IC in relation to a problem. In light of my first new reply, above, do you wish to retract your disagreement?
No.
H: That a conclusion is reachable by a mind does not imply that the mind always had "implicit" knowledge of it.

P: I'm not sure you understand what I mean. As a very simple example of what I mean, if, regarding math, you have attained only to the conscious knowledge that 1+1=2 (I'm talking about the math, not the symbols), then you have implicit (subconscious) knowledge that 2-1=1.

Not necessarily. Also, implicit does not imply subconscious.
P: The problem with this "Chinese room" experiment is that the rules must have been originated by a person who could see.

H: The problem with your analysis is that you assume that a system (e.g. earth's ecosystem before hominids) can never produce knowledge or intelligence (e.g. H. sapiens') that was not already built into the system. This assumption is now known to be mistaken.

P: Known by whom? Not by me.
It's known by anyone who understands modern biology.
But, what you have said is an analogue of implicit knowledge: if the pre-H.sapiens eco-system of the earth had a future-H.sapiens built in, this is to say that H.sapiens was implicit in that system.
H. sapiens was of course not "built in" (in the sense of being nomologically necessary).
[..] fuel-injection system (FIS) is implicit in a car [..]
This analogy was unhelpful, but mercifully brief. :-)
P: Qualia (by which I mean to include not simply things like 'red', but the sense of beauty, disgust, pleasure, pain, etc., and every kind of each of these) cannot be simulated. You either have it or you don't.
That is merely your intuition, based on a naive conclusion from the simple simulations you've experienced. My analysis is that qualia are an almost inevitable epiphenomenon created by the functional and dispositional properties of perception.
The simple and random forms of the non-living world do not apparently have the ability to intelligently select patterns above a certain level of complexity (such as the first 100 primes,
Simulations aren't "simple and random".
Simulated gravity [..] simulation of an avalanche or an aerodynamic test is not the real thing.
I didn't say they were; I said music and math proofs were. Are you saying intelligence is more like gravity than like music/math?  Your examples are almost as question-beggingly lame as Searle's lactation example.
Deeper Blue is a simulation of chess-playing intelligence,
You thus concede my point.
and it is not apparent that such a machine could evolve from the non-living world,
Who said it could?
while it is self-evident, to say the least, that such a machine is the product of a qualia-based (i.e., living) intelligence.
You baldly assume that intelligence is somehow "based" or reliant on qualia.  It's not.
It has not been proved that functional intelligence can be produced without qualia.
Indeed, as I say in my book: The functional role of certain sorts of perceptions in a conscious system necessarily and understandably entails that the system will report qualia.
P: First, I reject that a zombie can be made (see my replies above).
Ah, but do you reject that a zombie is metaphysically possible?  If so, you agree with us who say qualia are mere epiphenomena.
P: Second, your assertion that "The experience of the redness of red consists in the operation of a complex set of functional components for processing information" is not self-demonstrative,
It's not supposed to be.
and I can distinguish between correlation and consists-of-ness. All one can observe in the brain in regard to the origin/nature of qualia is that there is a correlation between a specific mechanical function and a specific qualae. The seeming ability of the function to produce the qualae does not mean that the qualae consists of the function,
It strongly implies it.
and only a commitment to physicalism will force the conclusion that the qualae consists of the function.
I could as easily say: only a commitment to anti-physicalism allows the conclusion that qualia are more than epiphenomena.
 Function, per se, is already distinguishable from qualia
No, not in the sense that anyone can show that qualia are necessarily independent of mere functional properties and dispositions.
Steve Grand, of Cyberlife Research, said " any more than you know whether I really feel or simply behave as if I do."
Right; this is the standard observation that if zombies are possible, then everybody else could be zombies.
P: To insist that these devices do understand just as the technologically primitive man might think they do would be to implicitly [say] that there is no such thing as a mistake in thinking that something which you thought was a real man was just a manecquin.

H: Obviously false. It's not to say such mistakes can't happen; it's to say that the likelihood of this being such a mistake is low.

P: That is an inconsistent response on your part. See my last new reply, above.

I don't see how anything you've written above indicates any "inconsistency" in my response.
H: some of us are apparently unable to think our way past our primitive intuition that an entity can't think or feel unless it is wet or soft or warm.

P: Speak for yourself. Some of us are unable to see our way past the primitive physicalist idea that "subjective" intelligence (i.e., sentience) is, or is produced by, a mechanical function of a non-sentient substrate.

I say you fail to imagine how quantitative complexity can yield a qualitative phenomenon like intelligence. You say I'm failing to realize the existence of some kind of barrier or limit.  I can easily call your imagined barrier a naive and primitive idea (like the idea that a plane couldn't fly), but it's odd to call my appreciation of complexity "primitive".
Rhetorically, how short can an one-second duration of qualia be divided up before it is no longer there?
Don't ask me; you're the one saying qualia have independent existence.
P: What's the limit of human non-intuitiveness regarding the facts of reality? How far can reality part with human mechanical intuitions?
In principle, reality can be as non-intuitive as is logically possible.
The following story illustrates the problem of determining which of two competing paradigms is the primitive/false one. [..]  the intellectual and technological culture of the ten-thousand-plus people in the station had long ago declined to that of the proverbial Stone Age tribe
I see no problem here. The authentically primitive intuition is the one held by authentically primitive persons. Your space station inhabitants aren't authentically primitive.
 

This was your fifth reply to me, and you are still nowhere near substantiating your original contentions that

So as we stray ever farther from any substantive criticism of my original points about Resurrection probabilities, I'll be decreasing the frequency of my responses.