Why is there so much crackpot physics?

[dafydd]

For my claim to be invalidated however, it would seem necessary to demonstrate how the much stronger and well-established claims of Godel, Kitcher, and standard undergraduate philosophy of science pedagogy are wrong.

Go ahead. We await with bated breath.

 

[ben m]

For my claim to be invalidated however, it would seem necessary to demonstrate how the much stronger and well-established claims of Godel, Kitcher, and standard undergraduate philosophy of science pedagogy are wrong.

Wow. You think that your new idea of physics-project management is obtained by rigorous logical deduction from Godel, Kitcher, and "standard undergraduate philosophy of science"? Because that's what it means to say that you can't be proven wrong without proving your premises wrong.

Two points. First, neither Kitcher nor "standard undergraduate philosophy" are axiomatic systems from which you can prove things. Second, even in axiomatic logic, you can be proven wrong if your deductions are wrong. For example, if someone says "Ryan Gosling is a man. All men are mortal. Mortals die. Therefore Ryan Gosling is dead.", the problem is not with the truth of the premises, but with the arguer's understanding of them and ability to draw connections between them. Similarly, your conclusions may turn out to be wrong without a disproof of Godel's Incompleteness Theorem.

This is highly relevant to the thread. Why is there so much crackpot physics? In part, because it's extremely common among crackpots to say things like this. "I read some reliable sources. I used logical reasoning to arrive at conclusions. Therefore my conclusions are irrefutable." You get, for example, people proving that Special Relativity is inconsistent after applying "logic" to the Twin Paradox. I argued with a guy who had concluded that the Luminiferous Aether was a gas of spinning baton-shaped corpuscles, in which toroidal vortices (which were stable, he said) served as elementary particles. I spent a while walking through the actual predictions of his model, and how they differed from his mental picture of the model, how both my predictions and his mental-picture disagreed with experiment, etc.. After a while, exasperated, he said: "Well, I can't see where, but your predictions must be wrong somehow, because the ether HAS TO be particulate, and the particles HAVE TO be baton-shaped---I can prove it LOGICALLY."

These hold not only that such lack of documentation occurs in specific, easily documented cases (any math by any physicist, for example), but the infinite regress of complete documentation (for assump/aux hyp) makes such omission unavoidable and universal.

a) If this omission is "unavoidable and universal", why can't you provide an example of its occurrence?

b) If this omission is "unavoidable and universal"---not just in physics, as you've described it, but in all human endeavor whatsoever---why is it a good target for management?

c) I'm picturing the urgent management memo to, e.g., a database coding team:

From: Buck Field Project Management Consultants, LLC, CPA, WTF, BBQ
To: database-admin-dev-team-leaders-all
Cc: board-of-directors

Your PM team recently became aware of UNDOCUMENTED ASSUMPTIONS inherent in out database coding practices. From now on we encourage an effort to complete documentation of assumptions, according to standard PMBOK practice 13c4.6--10.

a) Coders may not declare an "integer" variable without documenting "integer" as a concept.

b) We observed multiple instances of the symbol "0" in the trunk release. This symbol was invented in India around 500AD and we are aware of no documentation of its properties. For example, will divide-by-zeros destroy the world? With what probability? Team leaders must file a risk-analysis document if your code contains both zeros and division operations.

c) We assume all of our coders are people. The board requests proof that none of them are p-zombies, Chinese rooms, or other non-person entities. To avoid errors, proofs must be provided in symbolic logic. The board requests documentation of the problem of personhood. What other assumptions are we making about our coders, the arrow of time, simulated reality, fiat currency, quaternions, etc. etc.? Highly respected philosophers tell me these problems are serious.

d) The Halting Problem is unsolved. This seems to be an unacceptable risk to an IT system. The PM objects to a documentation-practice that allows the possibility of unsolved problems going unmentioned in documentation. URGENTLY document the halting-problem impact on ALL systems. URGENTLY write a system that verifies the presence of halting-problem documentation. URGENTLY document the impact of the halting problem on the halting-problem-documentation-verification-system. Do the same for all other problems, known and unknown.

e) I see mention of "remote servers". Prove that these servers really exist. Document your proof. Proofs must halt in a finite time.

 

[catsmate]

Wow. You think that your new idea of physics-project management is obtained by rigorous logical deduction from Godel, Kitcher, and "standard undergraduate philosophy of science"? Because that's what it means to say that you can't be proven wrong without proving your premises wrong.

Two points. First, neither Kitcher nor "standard undergraduate philosophy" are axiomatic systems from which you can prove things. Second, even in axiomatic logic, you can be proven wrong if your deductions are wrong. For example, if someone says "Ryan Gosling is a man. All men are mortal. Mortals die. Therefore Ryan Gosling is dead.", the problem is not with the truth of the premises, but with the arguer's understanding of them and ability to draw connections between them. Similarly, your conclusions may turn out to be wrong without a disproof of Godel's Incompleteness Theorem.

This is highly relevant to the thread. Why is there so much crackpot physics? In part, because it's extremely common among crackpots to say things like this. "I read some reliable sources. I used logical reasoning to arrive at conclusions. Therefore my conclusions are irrefutable." You get, for example, people proving that Special Relativity is inconsistent after applying "logic" to the Twin Paradox. I argued with a guy who had concluded that the Luminiferous Aether was a gas of spinning baton-shaped corpuscles, in which toroidal vortices (which were stable, he said) served as elementary particles. I spent a while walking through the actual predictions of his model, and how they differed from his mental picture of the model, how both my predictions and his mental-picture disagreed with experiment, etc.. After a while, exasperated, he said: "Well, I can't see where, but your predictions must be wrong somehow, because the ether HAS TO be particulate, and the particles HAVE TO be baton-shaped---I can prove it LOGICALLY."



a) If this omission is "unavoidable and universal", why can't you provide an example of its occurrence?

b) If this omission is "unavoidable and universal"---not just in physics, as you've described it, but in all human endeavor whatsoever---why is it a good target for management?

c) I'm picturing the urgent management memo to, e.g., a database coding team:

The sad thing is that's not too far removed from emails I've had from PMs.

 

[ben m]

I wonder if this isn't the key to the whole thread.

That's pretty close, however when the last policy studies at NSF were being conducted around TR, experts like Nersessian provided input, but it appears their recommendations may have been overly technical and poorly understood. This input seems to have been watered down over time, which would be expected.

BS has spend a month or two ignoring/deflecting/misreading questions which are variants of "How does it work? What are actual management recommendations? Is this or is it not a concrete proposal for managing scientific revolutions using cog-sci-history-of-science input?"

And only now he gets around to mentioning this? That the NSF conducted a transformative-research policy study, and actually got input from a cognition-of-science-concepts expert, whose recommendations were "poorly understood" and later "got watered down"?

Geez, that single factoid answers half of the questions BS has been dodging.

a) Yes, there are actual management recommendation in there somewhere. People other than BS are able to tell other people what it is. BS doesn't.

b) Yes, the idea is that these recommendations are supposed to be applied somehow, i.e. to affect something that happens at the agencies. BS has been very unclear at this point---whether, and when, and how, and who, someone on the management-end has to give cog-sci-informed instructions to a working scientist, or make cog-sci-informed funding decisions. But now we have (in spite of BS's steadfast refusal to explain) a hint of implementation---notice the past tense ("has been" watered down) suggesting that changes could have been implemented already.

c) BS has refused to discuss, in any useful way, the criticism that his ideas don't sound understandable. A long time ago, he floated specific claims about identifying "object vs. process concepts", with the latter being more likely to be revolutionary. I tried to criticize this (BS grew silent on the point) that this idea was so vague and subjective that you could never use it to, e.g., rate proposals---"proposal #13-12-5543 scores 95% on the broader-impacts scale but only 33% on the object-concepts scale, and is therefore denied". You'd think BS might have mentioned the time when the NSF "watered down" cog-sci-based ideas because they were "overly technical and poorly understood", but ... nope!

ETA: In a nutshell: do you *have* the recommendations to the NSF of "experts like Nersessian", BS? Would you like to post them here and discuss them? Better yet, would you please go two months back in time and post these recommendations instead of the content-free, goalpost-moving, gnomic non-discussion that you chose to conduct?

 

[Perpetual Student]

Quote:
Originally Posted by BurntSynapse
Finding out where I'm wrong is an important priority for me and I think: for all good researchers. A single omission from a single paper or calculation proves the claim that such omissions occur. Whereas the assertion no such omissions exist presents significant problems to defend.
For my claim to be invalidated however, it would seem necessary to demonstrate how the much stronger and well-established claims of Godel, Kitcher, and standard undergraduate philosophy of science pedagogy are wrong.

Can you provide any example of where the "claims" of Gödel have resulted in a single omission from any paper involving physics research? Providing one example might help towards reviving your hopelessly lost credibility.

Since it is not likely that any such paper exists, could you (at least) provide some hypothetical scenario in any area of current or future research in physics where Gödel's incompleteness theorems might have any effect on results? Could you also discuss why you believe that, even if Gödel's theorems were relevant in some hypothetical area of physics, those physicists would be more likely to avoid pitfalls due to Gödel through project management? Are project managers more cognizant of Gödel's theorems than physicists?

 

[BurntSynapse]

If you believe "a system cannot demonstrate its own consistency" makes no claim regarding unavoidable assumptions in relation to the execution of math & science processes, I've no problem agreeing to disagree.

 

[W.D.Clinger]

Before BurntSynapse posted the above, confirming that he intended his reference to Gödel as a reference to the incompleteness theorems, I had composed the following:

Addressing BurntSynapse, Perpetual Student wrote:

Since it is not likely that any such paper exists, could you (at least) provide some hypothetical scenario in any area of current or future research in physics where Gödel's incompleteness theorems might have any effect on results? Could you also discuss why you believe that, even if Gödel's theorems were relevant in some hypothetical area of physics, those physicists would be more likely to avoid pitfalls due to Gödel through project management? Are project managers more cognizant of Gödel's theorems than physicists?

Actually, we're just guessing that BurntSynapse's reference to Gödel and subsequent bafflegab involved some kind of crackpot misinterpretation of the incompleteness theorems. Because BurntSynapse wrote with his characteristic lack of precision and clarity, it's conceivable he was trying to refer to Gödel's completeness theorem, or to Gödel's proof of relative consistency for both the axiom of choice and the continuum hypothesis, or even to Gödel's bizarre solution of Einstein's field equations in which time travel would be possible.

Conceivable, I say, but unlikely. In my experience, people who try to hitch their empty ideas to Gödel and philosophy of science (while demonstrating near-total ignorance of Gödel's actual results or mainstream philosophies of mathematics and science) are seldom aware of Gödel's other contributions.
​

I will now respond to BurntSynapse's recent jibe:

If you believe "a system cannot demonstrate its own consistency" makes no claim regarding unavoidable assumptions in relation to the execution of math & science processes, I've no problem agreeing to disagree.

Note well that BurntSynapse had been claiming something rather different: That assumptions were not merely unavoidable, but could not be documented.

Gödel's second incompleteness theorem says absolutely nothing that precludes documentation of assumptions. In fact, if you've read and understood the proof of that theorem, you know full well that the proof proceeds by constructing a specific statement of the system's consistency that cannot be proved within the system. In other words, Gödel's proof already includes the documented assumption that BurntSynapse would have us believe cannot be documented.

 

[ben m]

If you believe "a system cannot demonstrate its own consistency" makes no claim regarding unavoidable assumptions in relation to the execution of math & science processes, I've no problem agreeing to disagree.

BurntSynapse, I presume that it is for your own good that you avoid specific examples. Your examples so far:

a) You don't understand quaternions.
b) You don't understand fractals.
c) You don't understand dimensional analysis.
d) You don't understand Godel.

Seriously, Godel's Incompleteness Theorem is a statement about the power (and limitations) of formal axiomatic mathematical-proof-generating algorithms. The mathematics actually used in physics does not utilize any of the fanciest set-theory trickery that qualifies it as "sufficiently powerful" under Gödel's work. Remember, the sort of problem that fires up the Gödel machinery are paradoxes like "how many members are there of the set of all sets that are not members of themselves?"; if you find a physics problem that requires that machinery, then maybe we'll need to worry about whether that machinery is complete and consistent. Until then, you have no grounds for generically citing Gödel as evidence of some sort of problem. None.

 

[W.D.Clinger]

BurntSynapse, I presume that it is for your own good that you avoid specific examples. Your examples so far:

a) You don't understand quaternions.
b) You don't understand fractals.
c) You don't understand dimensional analysis.
d) You don't understand Godel.

So far, so good, but ben m goes astray when he continues:

Seriously, Godel's Incompleteness Theorem is a statement about the power (and limitations) of formal axiomatic mathematical-proof-generating algorithms. The mathematics actually used in physics does not utilize any of the fanciest set-theory trickery that qualifies it as "sufficiently powerful" under Gödel's work.

Gödel's incompleteness theorems apply to any system that's capable of expressing a sufficiently large subset of arithmetic on natural numbers. The mathematics used in physics is certainly powerful enough to fall within the purview of Gödel's incompleteness theorems.

ETA: Gödel's incompleteness theorems also depend on a couple of other properties (mainly that valid inferences can be distinguished from invalid), but the math used in physics has those other properties as well.​

 

[ben m]

Gödel's incompleteness theorems apply to any system that's capable of expressing a sufficiently large subset of arithmetic on natural numbers. The mathematics used in physics is certainly powerful enough to fall within the purview of Gödel's incompleteness theorems.

What do you make of the statement below? From http://en.wikipedia.org/wiki/Hilbert's_program#Hilbert.27s_program_after_G.C3.B6del

Although it is not possible to formalize all mathematics, it is possible to formalize essentially all the mathematics that anyone uses. In particular Zermelo–Fraenkel set theory, combined with first-order logic, gives a satisfactory and generally accepted formalism for essentially all current mathematics.

Although it is not possible to prove completeness for systems at least as powerful as Peano arithmetic (at least if they have a computable set of axioms), it is possible to prove forms of completeness for many interesting systems. The first big success was by Gödel himself (before he proved the incompleteness theorems) who proved the completeness theorem for first-order logic, showing that any logical consequence of a series of axioms is provable. An example of a non-trivial theory for which completeness has been proved is the theory of algebraically closed fields of given characteristic.

 

[W.D.Clinger]

Gödel's incompleteness theorems apply to any system that's capable of expressing a sufficiently large subset of arithmetic on natural numbers. The mathematics used in physics is certainly powerful enough to fall within the purview of Gödel's incompleteness theorems.

What do you make of the statement below? From http://en.wikipedia.org/wiki/Hilbert's_program#Hilbert.27s_program_after_G.C3.B6del

Although it is not possible to formalize all mathematics, it is possible to formalize essentially all the mathematics that anyone uses. In particular Zermelo–Fraenkel set theory, combined with first-order logic, gives a satisfactory and generally accepted formalism for essentially all current mathematics.

Although it is not possible to prove completeness for systems at least as powerful as Peano arithmetic (at least if they have a computable set of axioms), it is possible to prove forms of completeness for many interesting systems. The first big success was by Gödel himself (before he proved the incompleteness theorems) who proved the completeness theorem for first-order logic, showing that any logical consequence of a series of axioms is provable. An example of a non-trivial theory for which completeness has been proved is the theory of algebraically closed fields of given characteristic.

I don't see anything wrong with what you quoted.

The first paragraph speaks of formalization, not consistency. Gödel's incompleteness theorems do apply to ZF (hence to ZFC), so ZF and ZFC are unable to prove their own consistency.

The axioms and inference rules of standard first order logic do not, by themselves, give you arithmetic. That's why you need additional axioms, such as (first order) Peano axioms, if you want to do first order arithmetic. In particular, Gödel didn't limit himself to the bare bones of first order logic when he proved his completeness theorem for first order logic; if I recall correctly, his proof used arithmetic, finite trees, and possibly some other stuff that goes beyond pure logic.

Furthermore, any first order formalization of arithmetic (such as first order Peano arithmetic) will be incomplete in the sense that it will contain unprovable statements that are true of the standard integers but not necessarily true of the theory's nonstandard models. That doesn't contradict the completeness theorem, because the completeness theorem asserts completeness with respect to validity (truth in all models), not completeness with respect to arithmetic truth (truth in the standard model of arithmetic).

Induction is the problem. The natural statement of the induction principle is second order, not first order. To obtain a first order approximation to that second order axiom, first order Peano arithmetic uses an axiom schema standing for an infinite set of first order axioms. Even with that infinite set of axioms, the theory you get isn't quite as powerful as what you get with the second order axioms, and that shows up in the nonstandard models and in Gödel's incompleteness theorems.

 

[The Man]

Appearance depends overwhelmingly on our focus and perspective.

If there's any objective metric to suggest that group represents a statistically significant percentage of the general population, or that they produce a level of harm for which it would be rational to invest ourselves in preventing, I'd be interested to know what it was, especially given the classic demarcation problem in philosophy of science.

So you’re just going to ignore the questions I did ask to instead try to address one I didn't ask?

 

[BurntSynapse]

If it is that simple, you could perhaps give one of those specific examples that are easily documented? Take any math of any physicist and let us see how it works!

Perhaps, but I'm not offering to guess what a particular discussion member might consider within that category. I've invited a counter example to what I and others perceive as the implications of Godel(sp), Quine, et al.

 

[The Man]

Perhaps, but I'm not offering to guess what a particular discussion member might consider within that category. I've invited a counter example to what I and others perceive as the implications of Godel(sp), Quine, et al.

I don't recall anyone asking you "to guess what a particular discussion member might consider within that category" but just what you "might consider within that category". Unfortunately, that as well as what you "and others perceive as the implications of Godel(sp)" suffer the same affliction of what you "and others perceive as the implications of Godel(sp)". In that the failure of that perception can be attributed to something outside that perception.

 

[dafydd]

Perhaps, but I'm not offering to guess what a particular discussion member might consider within that category. I've invited a counter example to what I and others perceive as the implications of Godel(sp), Quine, et al.

So you have no examples. No surprise there.

 

[ben m]

Perhaps, but I'm not offering to guess what a particular discussion member might consider within that category. I've invited a counter example to what I and others perceive as the implications of Godel(sp), Quine, et al.

a) Conversation skills 101: when someone asks "Can you provide an example", they are not looking for an answer like "yes", "no", or "perhaps". They are looking for the actual provision of an example. Please offer an example as Steenkh requested. If you're not sure whether your example fits what Steenkh had in mind, post multiple examples and ask for clarification.

b) It's ironic to see you, after spending months avoiding all requests for examples of what you're purportedly talking about, "inviting a counter example." No, you first. Really, go ahead.

c) You say Godel's Incompleteness Theorem is relevant somehow to physics, and causes (or is evidence of, or something) a problem. I (and Steenhk, I think) find that unlikely. I don't see how the idea of a "counterexample" even applies.

 

[Perpetual Student]

Don't hold your breath! People who harbor crackpot ideas can never provide real specifics. Hand-waving and non-specific references to esoteric concepts are their stock-in-trade.

 

[steenkh]

Perhaps, but I'm not offering to guess what a particular discussion member might consider within that category. I've invited a counter example to what I and others perceive as the implications of Godel(sp), Quine, et al.

It was you who said that it was easily documented. I just took you on your word and asked you to do it, particularly because it is so easy.

But now you retract, and would rather not give examples. What a pity, for a moment I thought that you believed in this yourself, but apparently you are aware that no such examples exist.

 

[BurntSynapse]

So you’re just going to ignore the questions I did ask to instead try to address one I didn't ask?

Absence of response is evidence of being ignored?

I'd disagree even if I didn't know it was materially inaccurate.

My reply here technically ignores the question being asked because of its fault premises. Since I'm interested in obtaining real criticism of my actual positions & claims, other discussions tend to be of insufficient priority, just as they are for all of us.

 

[BurntSynapse]

It was you who said that it was easily documented. I just took you on your word...

I claim it is easily documented and that I welcome examples that skeptics would consider valid test cases.

I cannot and did not claim to provide examples skeptics might like.