Bleep Science

[Apathia]

I start this thread with some trepidation. I've seen some other threads here, as the "Materiallist" one, open a can of ugly worms.
To minimize the mess, I'll start with some disclaimers:

Since this is a Philosophy of Science topic gets tainted with psudoscience, I'm puting it in the Religion and Philosophy Section rather than in the Science or Paranormal and Scepticism section. Moderator may move it, if they see it as Science or Psudoscience. I call this stuff Philosophical speculation.

The words "Consciouness" and "Self-Consciouness" along with "Awareness" will be said here. These words have high woo conotation with some posters. Please don't point your guns at me in the assumption that I am talking woo or trying to let it in the back door. If, however, your position is that Consciouness is a load of dingos's kidneys, then please express that.

I'm not a mind-body dualist. In my opinion Self-Consciouness arises through natural processes. I'd say "humble opinion" but I doubt my humility.

I sympathize with people who are seeking an overarching Philosophy of Science that explains everything from how stars burn to how creative imaginations burn, but I've reached that age in my life where I don't really have to have one. If such elludes human grasp for eternity, I'm not losing any sleep over it.

One of my characters in a series of stories I'm writing has an obsessive enthusiasm for this subject. So this thread is part a search for material: not only what he would say, but what his girlfrined skeptic would.




If you haven't seen "What The Bleep Do We Know," see it just to get yourself riled up, and so you can tell your kids it was the worst movie you ever saw.
Staring in the Movie, right up there with Ramtha, is Fred Allen Wolf, aka "Capatin Quantum," whose passion is to find and explain Consciousness in Quantum Physics.

I can't post liknks yet, but you can google his website and read some of his stuff. od couse it dovetails nicely with the premise of the Movie that QM somehow proves that we can create our on Reality: six of them before breakfast if we choose.

Then there's Roger Penrose. I have his books but i haven't begun reading them yet. it's siad he find the basis of Self-consciousness in gravitation in Microtubuals. The Geometry of Consciousness appraoch is popular in some other circles as well.

Weigh in on these explorers, visionaries, crackpots, Wooists. or wahtever they are.

I sypathize with Wolf, but find him a bit silly.

[drkitten]

Then there's Roger Penrose. I have his books but i haven't begun reading them yet. it's siad he find the basis of Self-consciousness in gravitation in Microtubuals, tiny rolled up dimensions. The Geometry of Consciousness appraoch is popular in some other circles as well.


Roger Penrose is an extended example of an argument by ignorant authority. He knows neither physics nor neuroanatomy, but claims to find a cause for consciousness in their interaction. His claim to fame is that he is a first-class mathematician (possibly the best living mathematician today), and he holds a prestigious and therefore intellectually daunting position.

In general, the physicists find his neuroanatomy interesting and plausible, while they consider his physics to be appallingly ill-informed. The neuroanatomists find his physics interesting and plausible, while they consider his neuroanatomy to be apallingly ill-informed. In neither area does his theory have enough command of the facts to be taken seriously by the specialists.

Which is a pity, because I absolutely love his work in geometry....

[Mercutio]

Penrose : Microtubuals :: Descartes : Pineal Gland

[empeake]

There's a very good article about the movie at Skeptico (http://skeptico.blogs.com/skeptico/2005/04/what_the_bleep_.html). Should be a required reading for anyone has watched the movie or plans to do so.

[Apathia]

Roger Penrose is an extended example of an argument by ignorant authority. He knows neither physics nor neuroanatomy, but claims to find a cause for consciousness in their interaction. His claim to fame is that he is a first-class mathematician (possibly the best living mathematician today), and he holds a prestigious and therefore intellectually daunting position.

In general, the physicists find his neuroanatomy interesting and plausible, while they consider his physics to be appallingly ill-informed. The neuroanatomists find his physics interesting and plausible, while they consider his neuroanatomy to be apallingly ill-informed. In neither area does his theory have enough command of the facts to be taken seriously by the specialists.

Which is a pity, because I absolutely love his work in geometry....

Herein lies the difficulty for the layman, such as myself. I didn't have a poor Science education, but it was overall a Humanities one. So I can't spot his ignorance in those fields. Also, among Science writers, he isn't one of the clearist in his prose. So the layperson can only assume he knows what he's talking about. To really tackle with him, I'd have to go back to school and load myself up with graduate level physics. But even a presenattion for the layman is above the heads of most ordinary readers, so we fall back on whether the big idea makes sense to our own personal worldviews.

I did read his older book, The Emperor's new Mind
and kept protesting his conclusion that an A.I. with Self-Consciousness was an impossibility because we couldn't cough up an agorithym that would consistantly and completely describe the steps of the process.
It kept seeming a non siqitor to me, like th old piece of garbage that it's impossible for bees to fly, yet they do. A God of the Gaps sort of thing.
Smart systems that have reached a certain degree of complexity will naturally exhibit self-reflection. Heck, my Betta has already figured out it's just a mirror!

Just as in the case of Wolf, when I read about Penrose's conjectures of how consciouness works, I'm struck with this impression something fast and losse is going. the puzzle pieces aren't really fitting together.

In Wolf's case. it just seems he's using science terms metaphorically. Fine, but 100 years from now, it will all sound like what it is: medieval alchemy.

[drkitten]

Herein lies the difficulty for the layman, such as myself.

Yup, and it's how he's been able to get away with such drivel for such a long time. He knows enough about the fields to snow the layperson --- and even scientific non-specialists.

[Apathia]

Penrose : Microtubuals :: Descartes : Pineal Gland

Yep, still looking for the "Seat of Consciouness."
Nostrilldumass predicts that one of the next great places to search will be the Cerebrospinal Fluid.

[Mojo]

Also, among Science writers, he isn't one of the clearist in his prose. So the layperson can only assume he knows what he's talking about. have you read Intellectual Impostures by Sokal and Bricmont?

[Apathia]

have you read Intellectual Impostures by Sokal and Bricmont?

No I haven't. Thanks for the lead. I just read Dawkins's review.
Good criticism. The prose writing heros of my youth were Bertrand Russell and Issac Asimov. I haven't kept up with the intellectual scene, and the authors Socal criticises are part of the reason. I've alsway called that kind of prose "Obscurantist," and just haven't had time to bother with it. Also when it's as bad as the examples Dawkins gives from his review of Intellectual Imposters, it's clear enough even to someone like my self outside those areas of speciality that someone is talking rank nonsense of the sort you find scribbled in journals kept by patients in psychiatric wards.

Penrose and Wolf aren't quite in this league, but I still get that feeling of "fast and loose."

I said in a another thread we have hard QM Science. Then come the interpretations to relate it to science at large, such as the Copenhagen one, the most popular. The Philosophy of Science interprets the interpretation. I have no quarrel with that. It galls me when the New Age Carpetbaggers come round saying that Quantum Physics proves we have a soul and that we make our own reality and quote Wolf as their authority.

It's sad how by the time science trickles down to the masses it's supertstion.
take the word "tacyon." to some people I've met, this is the particle that carries qi. and is generated by the little motherboard componenets in a Q-link pendant. Theroretical Physics becomes an amulet!

[Beerina]

Which is a pity, because I absolutely love his work in geometry....

I believe he also claims there are tiling problems humans can solve that a Turing machine could not, which is a very interesting idea since a Turing machine is supposed to represent the abstract concept of "most powerful computational device that performs, at most, a finite number of computations in a finite amount of time." Which, if true, indicates that either the Turing machine is not that computational model, or that there's something infinite buried in the human mind, also of very high interest.

Of course, I doubt his conclusion, but does anyone have a link to this stuff?

[Apathia]

I believe he also claims there are tiling problems humans can solve that a Turing machine could not, which is a very interesting idea since a Turing machine is supposed to represent the abstract concept of "most powerful computational device that performs, at most, a finite number of computations in a finite amount of time." Which, if true, indicates that either the Turing machine is not that computational model, or that there's something infinite buried in the human mind, also of very high interest.

Of course, I doubt his conclusion, but does anyone have a link to this stuff?

I just have his books. Shadows of The Mind and The Road to Reality.
My thinking is that he's probably right in the old computation models. The brain has a whole different game for processing and storing information. With coming advances in quantum computing, A.I. may be able to duplicate what brains do.

My opinion is that consciousness is a thing that evolved naturally in interspecies communication. Get a system going with genetic algorthims in an environment of langauge communication and eventually it will reach a comparable complexity.
Penrose would probably shoot me down though with mathematics.

[Apathia]

Of course, I doubt his conclusion, but does anyone have a link to this stuff?

Here's one I intend to dig through.
http://psyche.csse.monash.edu.au/v2/psyche-2-23-penrose.html

Another:
http://www.dhushara.com/book/quantcos/penrose/penr.htm

[Apathia]

So, I've been doing a little homework, not enough to ruin my computer gaming life.
I'm not a mathematician, so penrose can be very persuasive to me. His arguments on noncomputability are impressive. Other but not all seem to think so as well.

I still have some speculation regarding nonalgorrithmic smart systems. You know, the kind of thing that goes on in your "smart" appliance and used to be called "Fuzzy" Math, but that was politically incorrect. I need to read some more in Penrose to see if he addresses that angle.

As for finding the basis of self-consciousness in fravitational effects upon microtubuals, that still feels funny.

It seems to me that Penrose, Wolf, and others have grabbed hold of Quantum Physics because the human thinking process and self-awareness don't readily lend themselves to a Classical Physics discription. So off they go looking for intelligence on the quantum level or some kind of relation between the quantum macro levels, and silliness follows fast. The larger hope is that someday we'll have a truely decent unification theory of quantum and relativistic effects and find within it the way we get self-consciousness. We can only hope.

[drkitten]

My thinking is that he's probably right in the old computation models.

The brain has a whole different game for processing and storing information.

The problem is that the computing theorists know this already --- and there's nothing new on the horizon in quantum computing that breaks the fundamental Turing machine limitations of the computable.

Again, Penrose is a layman with a gift for making plausible-sounding wrong statements. We've know about alternative models for computing -- basically, things that break the Turing limitations of computability -- for decades. You can read up on them under the heading of "computing with oracles." For a while in the 70s and 80s, it was an easy Ph.D. topic. Let's assume (counterfactually) that we have an "oracle" for the halting problem (or something like that -- so what else would we be able to do.

Unfortunately, neither brains nor quantum computers appear to be able to provide such "oracles." They can do interesting things regarding timing and decisions; we may even be able to build a classic non-deterministic Turing machine. But non-deterministic TMs are exactly as powerful as deterministic ones.....

[drkitten]

I still have some speculation regarding nonalgorrithmic smart systems. You know, the kind of thing that goes on in your "smart" appliance and used to be called "Fuzzy" Math, but that was politically incorrect.

Which? My "smart" appliances that use "Fuzzy logic" (it's not at all politically incorrect; it's a term of art) are still algorithmic.

That's part of the problem. Words have meanings; Penrose and his ilk seem not to know a lot of them, so he/they make stuff up on the fly that sounds plausible but is ludicrous if you actually know the underlying theory.

[Apathia]

Thank you for this very valuable input. it's why I strated this thread.
Yes, I have a lot of reading and thinking to do on this subject, so pardon if I start out with nonsense.
I was under the misguided impression that Fuzzy adpative systems could build instructions that didn't necessarily have corresponding mathematical equations (algorithims). But I see the error of my ways now.

"Oracles" seem to be a fudge. I started out with a Physics major in college with Astronomy in mind. I could understand the concepts of mathmatics but I was too slow at doing computations in examinations. calculators had just come out then and weren't allowed in math and physics exams, so I bowed out.
Anyway in my brush with Science, I found making fudge was essenstial to Theoretical Physics. I saw fudge again recently reading about the Higgs Mechanism.

While it's sound science to dismiss a pepetual motion machine immediately,
it's not wise to make pronouncements about what we will never be able to do regarding areas of Science that are on the theoretical frontier.
It appreas the consenses is that Penrose is right in speaking of the limitations of computability in a Turing machine. But it it doesn't follow that we won't someday understand how the brain works its wonder and be able to replicate it. But right now A.I. seems drifting in the Sargosso Sea.

As usual Creationists and Bleep people grab hold of anything in Science speculation that helps market their positions. So Penrose is famous among those who want to claim a supernatural component is necessary.

I just think nature is cleverer than we are. "Smart Evolution."
Our intelligence evolved in smart adaptations to the environment and in communication with each other. It's not a matter of just programming in some algorthims.

Anyway, it's more ignorant speculation on the part of a layman. So I appreciate it immensly when someone shoots any nonsense I say down.

[drkitten]

"Oracles" seem to be a fudge.

I'm not sure how you mean this. If you mean simply that they're counterfactual -- well, yes, that's the point. Rather than generating doublespeak about how something our current theories suggest is impossible, theoriests have taken the "honorable" way out and simply codified the magical part into a black box and listed its properties. (Physicists do the same thing, for example, when they postulate "exotic matter" with "negative energy density" or some such in a theoretical investigation of how black holes might lead to closed timelike curves.)



While it's sound science to dismiss a pepetual motion machine immediately,
it's not wise to make pronouncements about what we will never be able to do regarding areas of Science that are on the theoretical frontier.
It appreas the consenses is that Penrose is right in speaking of the limitations of computability in a Turing machine.

I'm not sure what you mean here, either. If you mean that the consensus is that he's right that there are things that a Turing machine can't do -- well, yes. That was Turing's original proof in the 1930s, and he hardly gets much credit for being able to correctly read a paper that was written before he was born. Where Penrose falls down is in his assertion that this means anything about the human mind/brain system. Basically, he asserts -- how did your review put it -- "The bounds of computability, Penrose says, are related to Godel's theorem, which holds that any mathematical system always contains self-evident truths that cannot be formally proved by the system's initial [axioms]. The human mind can comprehend these truths, but a rule-bound computer cannot."

This is, not to put too fine a point on it, a misreading of Godel's theorem. I don't know where the phrase "self-evident" crept in from, but just as a simple example, it's been proven that the Generalized Continuum Hypothesis is independent from the standard axioms of ZFC set theory. It's not at all "self-evident" to me whether the GCH is true or not, and I would love to see any sort of rational argument about it.

But more to the point, Godel's theorem does not say that any mathematical system must contain unreachable truths. It says that any mathematical system must contain either unreachable truth or demonstrable contradictions. If Penrose's standard for truth is the self-evidence of human intuition -- well, the demonstrable contradictions in human intuition are too numerous to bother listing. So if the mind/brain is a formal mathematical system (a statement that Penrose has never been able to support), then it's demonstrably an inconsistent one -- the human mind is capable not only of intuitively comprehending truth that cannot be formally demonstrated, but also capable of intuitively comprehending and believing demonstrable falssehood.

And there's nothing magical about this that distinguishes us from computers in this regard. We could easily create a computer program that can't tell truth from falsehood, but most of the time, we don't want to.

Penrose's argument hinges on an old and unsupported assertion -- the idea that human reason is somehow special. And the arguments he uses in favor of his assertion have long been discredited.

[Apathia]

You understood me acurately on both counts about Oracles and Turing.

As I read more (bits and pieces over days because I'm busy with other things)
any motivation to read Shadows of The Mind is evaporating.
For one it's clearer to me now why this QM explaination for Consciousness stuff seems incongruent to me. It's simply looking in the wrong place. very simplistically speaking: Consciousness is mysterious. QM is mysterious, so it all must be the same mystery. No, Murder on The Orient Express isn't Hound of The Baskervilles. Where we should be looking for the workings of consciousness is in Biology and Psychology.

On Penrose and the whole Turing thing, he makes a big deal about problems of computability with Formal systems, as if human thinking and consciouness proceeded along the lines of Formal logic and proceedure. Heck, it ain't so.
We reach conclusions the damndest of ways, even when they are true. The human "computer" thrives on fragmentation.

Thanks again for you contributions.

[LW]

I'm not sure how you mean this. If you mean simply that they're counterfactual -- well, yes, that's the point. Rather than generating doublespeak about how something our current theories suggest is impossible, theoriests have taken the "honorable" way out and simply codified the magical part into a black box and listed its properties.

As surprising as it may be, there are also practical uses for oracles. Of course, we can't use oracles for unsolvable problems in real life but luckily nothing in the definition of an oracle demands us to use uncomputable oracles.

In particular, NP-oracles are useful in solving many "difficult" problems, where "difficult" means that their computational complexity falls into some complexity class where no one has been able to find an efficient algorithm to solve them (or when we go higher up in the complexity hierarchy, where people have proven that no efficient algorithm exists). n symbols long, then the algorithm might take n2 computation steps. An inefficient algorithm takes an exponential number 2n of steps in the worst case.]

The complexity class NP is the "easiest" of the "difficult" complexity classes. A problem is in NP if we can efficiently verify that an existing solution to it is correct. The most important NP problem is SAT (http://www.satlive.org/) --- the question whether a given formula of propositional logic is satisfiable. There is a great host of existing solvers for it and their quality keeps improving year to year. If a SAT problem doesn't belong into the really difficult part of NP and isn't excessively large (best solvers can handle cases with hundreds of thousands or even millions of logical variables if the instance has a convenient structure), the solvers can likely solve it in a rather short time.

We can use these existing SAT solvers as oracles for solving more difficult problems. For example, there's a theoretical result that states that the complexity class PSPACE (that is, problems that can be solved in space that is polynomial with respect to the instance size) is exactly the class that could be solved efficiently if we had an NP-oracle. So, we can get a provably-correct algorithm for solving some PSPACE problem by writing a program that calls some SAT solver to solve a series of NP-queries.