Thursday, October 22, 2009

Roger Penrose is Much Smarter than I Am. But...

Roger Penrose is much smarter than I am. But I think he is completely wrong when he says

In my view the conscious brain does not act according to classical physics. It doesn’t even act according to conventional quantum mechanics. It acts according to a theory we don’t yet have. This is being a bit big-headed, but I think it’s a little bit like William Harvey’s discovery of the circulation of blood. He worked out that it had to circulate, but the veins and arteries just peter out, so how could the blood get through from one to the other? And he said, “Well, it must be tiny little tubes there, and we can’t see them, but they must be there.” Nobody believed it for some time. So I’m still hoping to find something like that—some structure that preserves coherence, because I believe it ought to be there.

We don't have any evidence at all that brains don't follow physical theories.


Joel said...

Well, to be nitpickety, as I read it, he's more suggesting that perhaps the brain cannot be fully explained on current physical theories, not that new physical theories cannot be found (and then, of course, tested on the rest of the universe) to fully explain the brain.

Myself, I don't see much terribly much reason to suppose that, but I must admit my inability to back that up very far.

Bayesian Bouffant, FCD said...

He can't even seem to figure out the limitations and non-reality of Schrodinger's cat. Why would the cat not be considered an "observer"?

Mitch said...

I'm not particularly enthralled by Penrose's theories outside of physics (I don't particularly know about those inside for that matter), but one could interpret his words to be not so contrarian (and self-aggrandizing).

Consider clouds (the puffy white things in the sky). Classical physics tells us lots about gasses and evaporation and suspensions and fluid flow etc. but one can't really take the classical physics -alone- and predict a classification of cumulus, numbus, cirrus, etc. Sort of in the same vein as there is a theory of chemistry underlying that of biology, but from the periodic table you can't really predict evolution.

I don't really think that that is Penrose's point (given other things he has said) but it makes sense out of context.

Larry Moran said...

Hmmm ... there seems to be a contradiction in your posting. You begin with, "Roger Penrose is much smarter than I am." but then you post a ridiculous statement by Penrose that reveals his ignorance.

What are you saying ... that you are even stupider than Roger Penrose? :-)

I don't believe that for a second.

Having read "The Emperor's New Mind" I conclude that Penrose's intellectual abilities are vastly overrated.

Eamon Knight said...

I learned about number theory and Godel from Penrose's book, then learned it all again from Hofstadter's -- interesting how they cover the same ground, but wind up in two different places. The Emporer's New Mind (as I recall, but this was over 15 years ago) covered all sorts of fascinating ideas; a great read -- then went a bit lame when he got to his pet theory about microtubules in neurons somehow channeling quantum effects up to the macroscopic level and accounting for our ability to go beyond formal logic.

Anonymous said...

We should never assume that we completely understand all physical theories, they thought this 100 years ago and now we have quantum mechanics. Penrose could be wrong but it doesn't make him less of a scientist to question whether current quantum mechanics can explain consciousness. Einstein questioning the universe's expansion with the cosmological constant, it moves in and out of fashion depending on your opinion of dark matter which could not exist? Questioning our knowledge with the goal of adding to our knowledge is always part of a good scientist's toolbox.

Eamon Knight said...

Well, to be nitpickety, as I read it, he's more suggesting that perhaps the brain cannot be fully explained on current physical theories, not that new physical theories cannot be found (and then, of course, tested on the rest of the universe) to fully explain the brain.

...which is either silly, or trivially true.

There are many phenomena for which we lack a complete, down to the last nut-and-bolts, description. Mental activity is one; evolution is another. That's why we have basic research science. The default assumption is that known fundamental laws provide an explanation; we just have to figure out how they apply to the details of the problem under investigation. You can always posit that there might be something "else" going on -- but absent a strong positive reason to think so (preferably accompanied by a coherent alternative hypothesis), you're really just blowing smoke. What Penrose is doing to brain function seems similar to what the IDists do to evolution -- offer at best a vaguely defined alternative for no very good reason.

Deane said...

Penrose is indeed a seriously smart guy and it is always risky and usually wrong to claim to be smarter than him.

I do find his more speculative writings hard to follow and, well, too speculative. However, on the few occasions that I've had the opportunity to discuss his ideas face-to-face, I find him to be quite sensible and insightful.

In particular, I don't think he is suggesting that consciousness requires a non-physical explanation. I believe that he is merely expressing his dissatisfaction with the current "understanding" of quantum physics. So for him (and me) it is not unreasonable to believe that there are new theories to be discovered or developed that explain better how quantum theory works. I believe Penrose is expressing the perhaps overly optimistic hope that these new insights will also lead to a better understanding of human consciousness.

Tom C said...

If you judge Penrose's intellectual abilities on one of his (highly speculative) books, then that's a little short-sighted. That's like judging the intellectual characteristics of a highly respected biochemist who speculates on the nature of spacetime, yet gets it completely wrong. Experts in one field can speculate all they want about ideas in other fields, but no one should judge their intellectual abilities based on these efforts alone.

For example, to better judge Penrose's abilities, please read "The Road to Reality" from which you can grasp the vast landscape of his understanding of physics and mathematics - fields that are his areas of expertise.

Of course, I'm not saying that Penrose's ideas about the universe are all correct; in this, no one has all the answers. At least Penrose is willing to wildly speculate (mostly in a non-conformist ways, e.g., twistor theory) in the hope of generating some glimpse of reality.

Nested said...

My impression is that Penrose believes the human brain is capable of solving tremendously complex problems well outside of even NP. Why he has this belief is unclear to me. The evidence seems to point in the direction that the human brain, like many other natural phenomena, has very fancy heuristics for problems of a limited size. Also he seems to think that fundamental new physics would have to be discovered to explain how the human brain works. Again, it seems much more likely to me that even the full power of the standard model would be over-kill. I don't pretend to be as smart as Penrose, but he would not be the first intelligent person to evolve into a crank, which I think in the case of his speculations about intelligence, he has indeed become.

Bayesian Bouffant, FCD said...

Tom C.: Experts in one field can speculate all they want about ideas in other fields, but no one should judge their intellectual abilities based on these efforts alone.

If someone persists in writing books and making public comments about fields in which they clearly hold no expertise then, yes, I feel it is fair to draw a conclusion about their intellectual abilities from that fact.

Is the Brain a Quantum Computer?
Abninder Litt, Chris Eliasmith, Frederick W. Kroon, Steven Weinstein, Paul Thagard
Cognitive Science XX (2006) 1-11.

Takis Konstantopoulos said...

I don't think that Roger Penrose said that brains don't follow physical theories. He said :

"[The brain] acts according to a theory we don’t yet have."

This is OK, because it's perfectly plausible that we need further physical theories to understand and describe, not just the brain, but the whole universe. He is only making a hypothesis.

But what I find strange is not Penrose's responses but the title of the article:

"Roger Penrose Says Physics Is Wrong, From String Theory to Quantum Mechanics."

Did he say so? I don't think so. That's why I am very suspicious of popular science magazines, especially the ones that were owned by non-scientific organizations (Walt Disney company in the case of Discovery). One exception used to be Scientific American but even this magazine has now become very lukewarm.

Blake Stacey said...

I rather liked Patricia Churchland's comment to the effect that quantum coherence in the microtubules has as much explanatory power as pixie dust in the synapses. Catchy, that.

(Philip Pullman's hypothesis is more parsimonious than Penrose's: the Dust in question also accounts for our cosmological measurements, because it is also dark matter.)

Field theorist said...

@Takis Konstantopoulos

Maybe Penrose means it indeed. In a Wilsonian vein, like, all quantum field theories are only "effective" theories, i.e. approximations, and each generation of approximation is accurate only within certain window of scale, and is completely wrong outside.

Nothing in the world behaves according to physical laws at all, because they are all only approximations, and all break down at a certain scale.

We don't have an exactly constructive theory of fundamental, basic building blocks of physics itself, witness fiasco of string theory and her contenders alike.

Penrose saying brain does not obey physics is just a catchy way of phrasing his bigger belief that the present physical theories are ugly, unconstructive, and ought to be wrong despite their apparent but miring correctness and efficiency.

Anonymous said...

I have a problem with the first three words, and as Penrose gets that wrong it is downhill all the way thence.

The conscious brain... You can't talk physics and then say something like conscious brain. That's bollocks. Physics works very well in explaining brain activity. The problem is people imagining things like consciousness when there is no such thing in physical terms.


Blake Stacey said...

"The conscious brain does not act according to classical physics" . . . ZOMG! Alcohol makes hyperquantum supercoherence leak out of the microtubules!

Jeff said...

On Friday, Larry Abbott of Columbia University gave a talk at the Quantum to Cosmos festival. He was asked what he thought about Penrose's theories on consciousness, and I seem to recall him using the word "crazy".

Argon said...

Everyone has a few wild-assed, largely unsupported or irrational ideas about how we think the world should work. We can only hope we don't display these beliefs in public or that they don't directly interfere with our professional work. For example, I know a yeast geneticist who firmly believes in reincarnation. Compare the impact of such a belief on this geneticist's professional standing (essentially none) vs. Jonathan Wells.

In Penrose's case, it's the microtubule stuff. He's a great physicist otherwise. Linus Pauling was a brilliant chemist but a nutter about Vitamin C.

Takis Konstantopoulos said...

Argon: I suppose when you say "everyone has a few wild-assed, largely unsupported or irrational ideas about how we think the world should work" you really mean "some have ..." It's too much to claim that everyone has.

Also, displaying beliefs in public or not is not the issue. For beliefs of this nature will influence anyone's professional work. Take, for instance, the case of Josephson, who received Nobel prize in Physics for the Josephson effect. Alas, his work, for years now, has been in the pseudo-scientific fields of the paranormal, parapsychology, etc. You know, ghosts and the like. They should be told that what they are doing, i.e. their using of irrational beliefs in their work, is not admirable. In fact, they are dangerous for the society. In any case, I don't think Penrose is a nutter of this sort.