hinduwoman

AI could perhaps evolve by themselves if we allow computers to go on redesigning themselves.

Also what does greater mentality mean? The ATM machines do their work as well as any human teller. So it should be the equal to a teller.

If the question is if it can know emotions like human beings, that is a different ball-game; 'greater' mentality does not come into it.

Plump-DJ

Hello Hello.

Ok, i'm trying to get a grip on the main criticism of Penrose's argument as presented in "Shadows of The Mind" so i found this surmation here : (http://www.artsci.wustl.edu/~philos/MindDict/quantum.html)

Penrose's main argument
-------------------------

Penrose's main argumentative line can be summed up as follows (this summary is taken from Grush and Churchland (1995)):

Part A: Nonalgorithmicity of human conscious thought.

A1) Human thought, at least in some instances, is sound , yet nonalgorithmic (i.e. noncomputational). (Hypothesis based on the Gödel result.)

A2) In these instances, the human thinker is aware of or conscious of the contents of these thoughts.

A3) The only recognized instances of nonalgorithmic processes in the universe are perhaps certain kinds of randomness; e.g. the reduction of the quantum mechanical state vector. (Based on accepted physical theories.)

A4) Randomness is not promising as the source of the nonalgorithmicity needed to account for (1). (Otherwise mathematical understanding would be magical.)

Therefore:
A5) Conscious human thought, at least in some cases, perhaps in all cases, relies on principles which are beyond current physical understanding, though not in principle beyond any (e.g. some future) scientific physical understanding. (Via A1 - A4)

I guess the first couple of premises are the big ones, and the rest just follow if they're true. So are they true? Tell me what you think.


Plump

tronvillain

There is no apparent reason to accept A1.

Bill

To elucidate a bit (since tronvillian gave an accurate but cryptic reply), A1 requires that human thought have some mechanism for achieving "soundness" in result while being "nonalgorithmic."

Stop and ask yourself: just how could that be true?

One possible answer is that human thought can draw upon some sort of resevour of "sound thoughts" that has some sort of independent existence outside of our physical reality. We metaphysical naturalists reject the idea of such a realm existing in fact without any evidence of the ability of a human mind to communicate outside of the scope of our own skulls. (In a theist sense, this would be the equivalent of claiming that God-like knowledge was given to the recipient of this sort of a universal truth in the form of a sound thought that was not produced by some algorithm within our brain).

The other possibility is that the human brain has some sort of unique innate ability to produce sound thought through nonalgorithmic means. This, too, suffers from a total lack of scientific support. Again, brain studies have gone to excruciating lengths to uncover such mechanisms, with literally a total absence of clues to even the possible existence of nonalgorithmic means of obtaining sound thoughts.

=====

In effect, Penrose is making yet-another "God of the gaps" sort of an argument: we don't know how it is that humans have this ability to frequently infer sound conclusions from only sparse and partial data inputs, but it is clear that we do have those sorts of abilities, so therefore (some equivalent of "Goddidit," which, in this case, amounts to the claim made in premise A1).

What Penrose is really claiming is that we still (as yet) have no clue as to how these wonderful inference engines inside of our all-too-human brains happen to work to achieve these clearly wonderful (and nearly "magical") results. And yet, we know that the results are achieved, but that the lowest level operators (the neurons) are very simple biochemical machines (apparently operating on a purely algorithmic basis).

=====

Yes, there remain gaps in our knowledge as to how this entire situation manages to pertain. But no, we have no reason whatsoever to accept the valitity of proposition A1, even provisionally.

== Bill

Nu

Anyone remember all the 80s scifi movies that were written in response to the public's concern that computers could be "smarter" than humans?

The counter-argument was typically that "humans are unpredictable and thus better than computers that can otherwise outperform them in every other way".

ROFLMAOPIMP

tronvillain

Well, just think about the old Star Trek episodes, where logical paradoxes were always making computers and robots explode while Bones rambled one about characteristics that would always make humans superior.

[ August 23, 2002: Message edited by: tronvillain ]</p>

Plump-DJ

Hello Bill. Since you gave the most productive response i shall reply to you.

Penrose's A1 really rests on the argument he makes *from* the implications of Godel's Theroms. So if i may say -- all the scientifc objections or the philosophical objections you raise (Like 'magical' resivours of thoughts and so forth ) are irelevant, at least as i see it.

Since Godel's Therom is true, the only place for objection is in the implications or argument he makes *from* the Theroms, and not the scientific or philosophical objections to that idea itself. (Since Godel's Therom is true and undeniable. ) Now correct me if i'm wrong but Penrose is arguing the following.

1) Taken from the review of his book by Stephen M Barr - "There are, he (Penrose) argues, mathematically demonstrable limitations to the kinds of things computers can do, and these limitations are not shared by the human mind."

So.. this is what he has to prove or at least argue for.

2) Again this is taken from the article -- "What Godel showed, however, and rocked the mathematical world by showing, was that mathematics could not be so mechanized. In particular, he demonstrated that if one is given any consistent formal mathematical system rich enough to include ordinary arithmetic, then there exist propositions (called "Godel propositions") that (a) can be properly stated or formulated in the symbolic language of that system, (b) cannot be proven using the mechanical symbolic manipulations of that system, and yet (c) can nevertheless be proven to be true-by going outside the system. Because the human mind can grasp the structure of the formal system and the meaning of its symbols, it is able to reason about them in ways that are not codified within that system's rules."

Now how does this affect us and how does this relate to Penrose's argument i ask?? (i'm just working this thru so be3re with me)

3) And more from the article.. "The relevance of all this to computers is that all computers involve- indeed are-systems for the mechanical manipulation of strings of symbols (or "bits") carried out according to mechanical recipes called "programs" or "algorithms." Now suppose that there could be a computer program that could perform all the mental feats of which a man is capable. (In fact, such a program must be possible if each of us is in fact a computer.) Given sufficient time to study the structure of that program, a human mathematician (or group of mathematicians) could construct a "Godel proposition" for it, namely a proposition that could not be proven by the program but that was nevertheless true, and-here is the crux of the matter-which could be seen to be true by the human mathematician using a form of reasoning not allowed for in the program. But this is a contradiction, since this hypothetical program was supposed to be able to do anything that the human mind can do."

So *this* is the meat and bones of the argument. He seems to be saying that if we created a computer that could do all the things we could do, it *couldn't * create a Godel proposition -- only we can create a Godel proposition. He's also assuming that we could in fact create a Godel proposition for this computer. (ie US) Hmm.. now i think i see the problem. How could Penrose and Lucas (A philosopher) overlook two obvious and foundational (to the argument) error of thought? I must be missing something.

Scorpion

You're not missing anything When theorizing about consciousness you need to stick to some intuitions about its nature and abandon others (since folk-psychology concept of consciousness is ambiguous at best and self-contradictory at worst), and Lucas & Penrose simply choose to take these assumptions for granted (instead of, say, deciding to rule consciousness out of existence as functionalists do). Of course this does nothing to enhance the credibility of the argument.

-S-

Ierrellus

IMO:

Neocortexial enlargement in the mammalian brain has placed in consciousness algorhythms necessary to provide us with rational comprehension of prior, experiential, organismic retention of dimensions that define possibility of survival in a given environment. The evolutionary change is from experience to calculation of experience.

Before the definitions of physics and mathematics was the experience of motion in space, change, distance, time, etc. It is unfortunate that Penrose, et. al.,with would find problems with algorhythms simply by claiming that their predecessors are an entirely different animal, not a part of any continuum in which there is proven advances in complexity.

Ierrellus
PAX

hinduwoman

But what about Commander Data?