New at Reason

Tim Cavanaugh | April 21, 2005

All the rest of you may be individuals, but I'm not. Kenneth Silber excogitates on free will and the physical brain while reviewing Jeff Hawkins and Sandra Blakeslee's On Intelligence and John R. Searle's Mind: A Brief Introdution.

Comments

|4.21.05 @ 4:12PM|#

"Mechanistic theories have enjoyed an authoritarian cachet in the past. Stalin?s regime embraced the work of Ivan Pavlov, famous for conditioning dogs to salivate at the ringing of a bell."

Associating Palov's theories with Stalin is somewhat unfair because as Leszek Kolakowski pointed out in *Main Currents of Marxism*, vol. 3, pp. 139-40: "It is fair to say that if Pavlov had been British or American his ideas would have been sternly condemned by Soviet philosophers on the ground that they explained mental functions by conditioned reflexes: he would have been accused of 'reducing' the human mind to the lowest forms of nervous activity, ignoring the 'qualitative difference' between men and animals, and so forth." But of course because Pavlov was a Russian, "the regime went to the opposite extreme and erected his theories into a dogma from which physiologists and psychologists were forbidden to deviate."

|4.21.05 @ 7:38PM|#

Pinker brings up the Chinese Room analogy in How The Mind Works, which is my only exposure to it. It's a fascinating problem, but I still can't shake the feeling that it's really just a semantic thing, like Zeno's Paradox.

Rikurzhen|4.21.05 @ 9:16PM|#

fyi: Stanford Encyclopedia of Philosophy - Chinese Room Argument - http://plato.stanford.edu/entries/chinese-room/

|4.22.05 @ 1:24AM|#

What a good review! I've been waiting for it to be blogged since I read it in the print issue.

Searle clears the way for more insightful thinking on these matters by laying waste to Descartes' mind/brain dualism. He looks to physical processes, but not the reduction of them to their smallest components, to explain and mental functions and mental functioning itself-Minds are what brains do. He further uses physical processes (Quantum Mechanics) to make the case for free will. I guess that really is the ultimate reductionism, huh? But he sees the mind as an emergent (non-reductionist) phenomena of the brain.

Full Disclosure: Searle is my fave philosopher. (non-political category) His books are really strong and he communicates complicated ideas with clarity. I'm just starting the reviewed volume.

The The Mystery of Consciousness is enthralling:

http://www.amazon.com/exec/obidos/ASIN/0940322064/reasonmagazineA/

BTW, Reason did an interview with Searle:

http://reason.com/0002/fe.ef.reality.shtml

It's groovy that the same book review also does On Intelligence, since Jeff Hawkins was inspired and guided by Searle's formulations, and says so in his book. I'm thinking that the folks at institute that Hawkins has founded and funded will push back the darkness on the nature of thought as they probe the neo-cortex. Will Hawkins' theory of intelligence as rooted in memories be corroborated by evidence?

BTW, Robert B Laughlin, co-winner of the 1998 Nobel Prize in physics makes the case that the whole universe can be better explained if we consider emergent phenomena. His book is A Different Universe - Reinventing Physics from the Bottom Down:

http://physics.about.com/od/philosophy/a/ADifferentUnive.htm

|4.22.05 @ 2:02AM|#

Todd,

I was going to say that; I'm thinking that Searle's Chinese Room is, at least, a lot less of a semantic thing than Zeno's Paradox since the Chinese Room is a definite proposition but thinking about it, I'm really not so sure how semantic Zeno's Paradox is. Never mind. ;) BTW, Zeno had more than one paradox.

|4.22.05 @ 3:17AM|#

I love this stuff! Good article by Silber.

(I hope it is realized that it does not reflect poorly on the article if this thread should get fewer comments than an article on Nazi popes, breast implants, Stone Age porn or What Women Want.)

|4.22.05 @ 4:52AM|#

From what I understand the relevant components of the brain are much too large to be noticibly altered by quantum mechanical effects. Searle's a smart guy, but I still think that the best way to find about the brain/mind is to just study the brain, period. I'm afraid (well not really, personally I think its great) that as neuroscience advances philosophy of mind (and traditional psych) will start to look more and more like astrology - just as chemistry (which was based more on observation and prediction) replaced alchemy. My main (but of course futile) hope is that people will leave neuroscience out of politics and vice versa.

|4.22.05 @ 10:01AM|#

I've never felt that the Chinese Room is a problem. Give the Chinese room guy enough symbols to accurately communicate his needs from inside the box, and you can't really say that your brain is doing anything different.

|4.22.05 @ 10:09AM|#

My main (but of course futile) hope is that people will leave neuroscience out of politics and vice versa

Yeah, no shit. Political discussions are stupid enough these days without people injecting their half-understood regurgitations of a science that's still in its infancy.

|4.22.05 @ 10:35AM|#

I've tried. I just don't get what that extra something is that Searle keeps postulating. It is biologically based but not even theoretically describable in terms of structure and program?

The problem as I see it is that he is leading us to an answer he likes by putting a person in the Chinese Box. The whole point of the Turing test is to determine if there is a person (or something with person like intelligence) in the box. The question to ask is, "Assume that there is a blob of Jello that can do everything the guy in the Chinese box can do. You interact with the Jello on the streets of Hong Kong. Demonstrate that the Jello can't speak Chinese." I don't think you can.

|4.22.05 @ 11:45AM|#

Leif Oines:

From what I understand the relevant components of the brain are much too large to be noticibly altered by quantum mechanical effects.

What Searle is saying is that QM effects are the part of the universe that is non-deterministic and they are a way to explain the free will in our brains. It's true that QM effects seem too small by them selves to effect nuerocellular processes in the brain. One theory is that they are ramped up via a chaotic butterfly effect so that they can break the deterministic lock. Any way, Searle's point is that QM is the place to look.

I still think that the best way to find about the brain/mind is to just study the brain, period.

Searle absolutely agrees. This is Hawkins' main inspiration for his endeavors.

I think that as neuroscience advances philosophy of mind (and traditional psych) will start to look more and more like astrology.

Yeah, I think that this is the fate of most philosophies as science advances. Searle is pointing the way in this direction. Social philosophies, such as political philosophy, are an exception to this.

My main (but of course futile) hope is that people will leave neuroscience out of politics and vice versa.

I agree and I'll up the ante and hope that we will leave most of existence out of politics and vice versa.

|4.22.05 @ 11:52AM|#

Jason:

You interact with the Jello on the streets of Hong Kong. Demonstrate that the Jello can't speak Chinese." I don't think you can.

But you can certainly note the lack of any evidence that the Jello is speaking Chinese, Right?

|4.22.05 @ 12:40PM|#

Rick, I was speaking of the 'arrow that never hits the tree'.

I consider Zeno's Paradox (the one referenced above) to be semantic because the 'halving of distance' only takes place in perception, not in external reality, therefore the arrow and the tree have no need of it.

The Chinese Room might also be semantic because it all hinges on the definition of 'understanding'. But it's not clear to me (or anyone else it seems) what the solution is. But I think we all have an apprehension that understanding isn't just mechanistic, else we could say that computers understand too, while they clearly don't. We may not be able to define what the 'something' is, but I think we all feel it. Until the sense of feeling can be translated into something concrete, and therefore testable, it remains a paradox.

|4.22.05 @ 1:08PM|#

Todd,

I see the point you're making about Zeno's Paradox. But did you mean to say: "Until the sense of feeling can be translated into something concrete, and therefore testable, it remains *semantic*."

|4.22.05 @ 2:21PM|#

I've never really grasped the Chinese Room thing, for two reasons:

1) If a man knows "the rules for using Chinese words correctly" well enough to apply them correctly, how is that distinct from knowing what the words "mean"? Where is the dividing line between "knowing how to use words" and "knowing what they mean"? Can you possibly have the ability to correctly apply all the "rules" that let you construct an expression such as "1 + 1 = 2" without understanding what 1, 2, "plus" and "equal" mean?

2) OK, assume it is possible to be able to correctly apply the rules for using words correctly without actually "understanding" what the words mean. If that's so, can't you say the same thing about a conversation between you and me? If there's an emotional, experiential, "feeling" element to "understanding," then it's impossible for the two of us to have precisely the same understanding of such words as "freedom," "big," "motherhood," etc. Who decides which of us correctly "understands" the meaning of those words? Regardless, even if our understanding is not exactly congruent, we can still understand the "rules" of language well enough to discuss them with each other, and exchange meaning. I think that still requires sapient awareness.

I might be missing the point.

By the way, one of my favorite SF books (The Star Fraction) concludes with a chapter that's "speaking" in the voice of an Artificial Intelligence," as a denounement explaining its actions during the course of the novel, it's motivations for doing so, and what happened after the novel's end. The final line is an expression of warmth for the human race: "I hope I see you again." The chapter is titled "What I Do When They Shove Chinese Writing Under the Door."

|4.22.05 @ 3:52PM|#

I'm surprised that Julian hasn't commented on this thread. I'd be interested to read his views on John Searle and these matters.

|4.22.05 @ 4:21PM|#

Rick:

"But you can certainly note the lack of any evidence that the Jello is speaking Chinese, Right?"

I don't note anything of the sort. My experience tells me that if the Jello can respond in English just like you can, I have no reason to suppose you are distinguishable from Jello in terms of English speaking ability. Searle is postulating something extra, that even though Jello appears to be speaking English, it isn't 'understanding' it. Like Stevo, I don't know what that means. There is no way to make a distinction based on objective criteria between you and Jello given our starting point. I don't know that I am doing anything different from the Jello, and neither does Searle.

I keep wanting to apply the Razor to Searle's argument.

|4.22.05 @ 6:31PM|#

Jason,

Whoa! When you said, "interact with the Jell-O on the streets of Hong Kong." I didn't think that you meant speaking English. I thought you were watching it shake or something.

But back to Searle's Chinese Room. I think that the thought experiment is a strong argument against the claims of Strong AI(he makes no such arguments against the claims of Weak AI) because AI is indeed just manipulating symbols according to rules without ant contextual relationships.

The whole point of the Turing test is to determine if there is a person (or something with person like intelligence) in the box.

But that's just it. The guy inside the box can pass the Turing test and still not understand Chinese. Even after doing all the symbol manipulation, if some one asked him to order some garlic chicken in Chinese, the fellow would go hungry.

|4.22.05 @ 7:00PM|#

...without *any* contextual relationships.

Sorry

|4.22.05 @ 11:23PM|#

Rick,

Why would "someone asking him to order some garlic chicken in Chinese" demonstrate a lack of understanding? Couldn't you say the same about someone who is born completely deaf in China and then somehow given some sort of cochlear implant that allows him to hear? Would you say that the ability to read, but not understand the spoken word, means that someone isn't truly concious, or human? I wouldn't think so, as your posts in the past regarding the inherent value of individual people have been quite clear.

I think Stevo and Jason have brought up good, fairly convincing points. I could easily imagine being able to read and write Chinese (or Japanese, or Korean) but be completely in the dark as to how it sounds. I seem to remember that I've heard that Chinese especially makes differences in intonation an important part of determining the meaning of words. I could be wrong on that point, however.

|4.23.05 @ 1:24AM|#

Shawn,

The idea is that even though he could respond to the questions written in Chinese with answers written in Chinese since he has the symbol manipulation rules and would pass a Turing test, he still doesn't understand Chinese and so certainly couldn't order garlic chicken in Chinese.

It's not a question of being conscience or human or how Chinese sounds. It's a question of understanding, which our fellow in the room lacks vis a vis Chinese and computers lack in general.

|4.23.05 @ 1:40AM|#

...BTW, I don't know anything about Chinese, so I don't understand how just knowing the rules for character manipulation would allow one to formulate answers to questions that he/she had zero understanding of, but that procedural part of Searle's thought experiment is not in question, so I know that that's the nature of Chinese.

|4.23.05 @ 11:57AM|#

"Even after doing all the symbol manipulation, if some one asked him to order some garlic chicken in Chinese, the fellow would go hungry."

No. If he passes the Turing Test, he answers questions about food in a manner indistinguishable from the way a person would. You are asserting that 'all he is doing' is manipulating symbols. I am challenging you to demonstrate that you are doing anything else.

|4.23.05 @ 1:08PM|#

Jason,

What I meant was, that he would be unable to take the initiative and order food in Chinese because he really doesn't understand Chinese. He is just manipulating symbols according to the rules to answer the questions that they pass into him. We manipulate English symbols AND understand English.

Your challenge to me to demonstrate that I am doing anything else is an interesting one and one of the challenges to the Chinese Room. I've gotta split now to attend a talk (on QM and philosophy!!-I'm gonna OD) But I will respond back on this thread within a few days.

BTW, our exchange on Hume a while back led me to some groovy ponderings and investigations.