Monday, August 31, 2015

Richard Muller's incredibly dumb comments about quantum gravity

I am often getting e-mail from, a server where people ask and answer questions. I am largely avoiding that server because while some texts over there may be insightful or interesting, there are tons of widely spread delusions written by the ordinary people for other ordinary people. They often drive me up the wall, I think that I've been exposed to that stuff many times, and I just don't want to add any exposure.

But today, I clicked at the Theoretical Physics category over there and was quickly led to a question about the unification of general relativity and quantum mechanics. It was the answer by Richard Muller of Berkeley that has simply stunned me.

Probably years ago, someone asked the following question:
Why do physicists believe that a mathematically consistent model that unites quantum mechanics and general relativity exists?

If mathematics breaks down when applied to black holes, why do scientists believe that mathematics can describe black holes? Perhaps the search for the mathematics that unites quantum mechanics and general relativity is pointless. Mathematics is a useful tool for modelling the universe in many ways, but maybe black holes are exceptional entities that can't be modeled by mathematics. Is it possible that the interior of black holes are so alien to our known universe and so different that current mathematical models/abstractions simply do not apply.
Well, the person who wrote this question is a layman who obviously doesn't understand what the words "mathematics" and "science" mean. Mathematics can't "fail" as a description of Nature because mathematics is, pretty much by definition, the language of the most accurate and reliable description of anything.

Mathematics may be connected with our observations differently than we thought and may imply different conclusions than what we thought but it can't "cease to apply". Only particular mathematically formulated statements and/or theories may "cease to apply". Even if the Universe were ultimately found to be governed by the mood swings of God, a malicious anthropomorphic dictator, the most accurate description of His mood swings would still be in terms of mathematics.

Most laymen simply do not have a clue but it was much more shocking to see Richard Muller, a well-known professor at Berkeley, who basically agreed with the hardcore layperson's misconceptions above (in his 2-day-old answer) and who actually added something even crazier ideas to the mix.

Richard A. Muller has worked on astro- and geo-related problems in physics (the Richard S. Muller in electronics is someone else!), became a very popular instructor at Berkeley, wrote some popular books about "physics for the U.S. presidents" and similar popular if not populist stuff, and was a major part of the team that created the Berkeley Earth Surface Temperature (BEST) dataset.

He must understand tons of ordinary things even though many of his claims may be controversial. But he decided that it was right for him to answer the aforementioned Quora question about the nature of quantum gravity. And his answer is juicy, indeed. It starts with an ambitious statement:
It is almost a physics religion to believe that relativity can ultimately be combined with quantum physics.
Wow. So it's a physics religion to "believe" that relativity can ultimately be combined with quantum physics. Dr Muller clearly stands next to the most insane pro-religion nuts who say that theoretical physics is just another religion – one that only tries to compete with The Church of Jesus Christ of Latter-day Saints.

This is a completely technical issue that can only be studied by the purely scientific method and that defines the "heart" of the domain where the scientific method is completely dominant. Why would a sane person start to talk about religion in this context? There is nothing to "believe". One may only collect the evidence, use the evidence to eliminate most of the candidate theories, interpret the evidence, and derive new conclusions from the theories that survive.

We know that the most accurate laws of Nature ultimately combine relativity with quantum physics for a simple reason – both relativity and quantum physics has been observed to hold in Nature. The theories that agree both with relativity and with quantum mechanics seem to be heavily constrained but that doesn't reduce the probability that both quantum mechanics and relativity is needed in the accurate laws. The probability is 100% because both parts of physics have been seen to control Nature. Relativity has been seen to hold. Quantum mechanics has seen to hold. Because each of the two – mostly independent – sentences hold, it's also true that they are "combined".

What Mr Muller (sorry, I just can't recognize the science degree of someone who makes moronic assertions of this caliber) is exactly as idiotic as the claim
It is almost a scientific religion to believe that the heliocentric theory of the Solar System can ultimately be combined with Darwin's evolution.
They have to be combined because both of them are true, aren't they? Even if heliocentrism makes Darwin's natural selection harder in any way, we indisputably know that the hurdles can't be insurmountable because both theories work, don't we? Does Mr Muller believe that science ends and religion takes over as soon as we discover at least two (independent) insights about Nature? This seems to be the only possible broader logic that may "justify" his claim about the "physics religion".

I don't want to expand this blog post by kilobytes of examples of experiments that have validated quantum mechanics, special relativity, or general relativity. In the quantum case, the confirmation has been most impressive. Calculate all the (up to) five-loop diagrams to find the prediction of Quantum Electrodynamics for the anomalous magnetic moment of the electron. The prediction agrees with the measurements perfectly – at the experimental error margin that is roughly a part per trillion. The agreement surely can't be a coincidence, can it? Especially because lots of other, sometimes comparably accurate predictions have been verified.

All non-gravitational observations that have ever been made have been nontrivially compatible with special relativity as well and all the gravitational phenomena we know seem to agree with general relativity. So whatever the "best theory of the Universe" that people will learn about in 2100 AD or 2500 AD is going to be, it will have to reproduce the predictions of relativity and quantum mechanics in the domains where those theories have already been confirmed.

In some extreme conditions where we don't have any experiments, the best theory in 2100 AD or 2500 AD may predict things that go beyond QM and GR separately – and maybe even beyond our current understanding of string theory. But the theory will still have to pass the tests that quantum mechanics and relativity have passed. So it will be a theory unifying quantum mechanics and general relativity. All other, "non-unifying" theories have been and will have been eliminated by the scientific method because they will have contradicted the empirical evidence!

Mr Muller elaborates upon the claim about "physics religion":
There is no evidence for this other than the fact that all the other forces of physics have been "unified".
Sorry, as I wrote above, there are billions of pieces of evidence supporting these two pillars of contemporary physics – all observations ever made by the humans agree with them. And because they have been done in one Universe, it is a "unification" of these two pillars that these observations make absolutely unavoidable. It's the separation of the knowledge to many theories that is an artificial sociological if not psychological sleight-of-hand. At the end, we have one Nature and we know many things about it – including the insights of relativity and quantum mechanics – that may be logically independent but that still "interfere" with each other because they fight to govern the same Universe. A viable description of the Universe is unavoidably a "unifying" theory. To suggest something else means to deny 100% of the evidence available in all of science. It means to be a complete lunatic.

It also means to deny tons of particular achievements of science such as the electromagnetic theory (the unification of electricity, magnetism, and light), the electroweak theory (the unification of the electromagnetism and beta-decay using the W-bosons and Z-bosons), and lots of other things that are really not questioned by serious physicists in 2015.
It is possible that relativity is different; that it is geometric and not quantum mechanical.
Mr Muller's sentence "relativity is different" is ill-defined, probably deliberately so. What does it mean for it to be "different"? Different than what? Different in what respect? Relativity is a theory "like" other theories in science in the sense that it is a scientific theory; it is different in the sense that its insights are independent from those of Darwin's theory or quantum mechanics, for example.

The second part of the sentence, "it is geometric and not quantum mechanical", only seems to reflect Mr Muller's ignorance of completely basic things in modern theoretical physics. The relativity is indeed geometric. But "geometric" does not mean "not quantum mechanical". The adjectives "geometric" and "quantum mechanical" self-evidently don't contradict each other. After all, Nature is known to be both.

The non-existence of the contradiction is, once again, totally analogous to the non-contradiction between any other two established facts of science. So Mr Muller's claim is as moronic as the claim:
It is possible that Darwin's theory of evolution is different; that it agrees with the natural selection and not with heliocentrism.
Well, a correct theory about the motion of planets and the life on one of them has to agree both with heliocentrism and with Darwin's theory of evolution. In the very same way, a correct theory of Nature has to agree both with quantum mechanics and relativity. The claim that these two parts of the contemporary physics cannon contradict one another is clearly wrong. They can't contradict one another because both of them are known to be true.

One may see that the actual "drift" of Mr Muller is nothing else than the drift of a rank-and-file anti-quantum zealot. He dreams about a future event in which quantum mechanics will be "undone" once again. But that can't happen because quantum mechanics may be pretty much defined as a consistent theory of physics that is not classical, and we need this theory because classical physics has been falsified. It's dead so it can never return again.

Muller is vague about "which relativity" he means. If the discussion is about special relativity, well, special relativity and quantum mechanics have been perfectly unified from 1930 or so – by quantum field theory (and perhaps from the 1927 Dirac's equation, we could say). These days, we know two related frameworks that unify them: quantum field theory and string theory. There is nothing "religious" about quantum field theory or Dirac's equation. They're vital parts of modern physics whose validity should be understood by every professional physicist (among theoretical high-energy physicists, the same thing is true for string theory as well) and I really feel highly uncomfortable if electronics allows people to become professors even if they are missing this Modern Physics 101.
But most physicists think that will not be the case, in large part because quantum physics has been so successful in the past. That's why they are looking to unify them. But it is worthwhile to recognize that this is based on hope, not on any firm physics or mathematics reason.
There is nothing based on "hope" here. Everyone is free to propose and develop a theory that ultimately "isn't" quantum mechanical. The only problem is that "hope" isn't enough for these alternatives and the fact is that no such alternative that would make any sense is known. On the contrary, the proofs that the principles of quantum mechanics have to apply universally, including on gravity, seem to be pretty much rigorous and to deny them means to be a scientifically illiterate idiot.

What Mr Muller says not to be the case – that there is a firm physics and mathematics reason – is the actual key to meaningfully analyze all these questions. And Mr Muller simply has nothing whatever to do with science. He acts as a shaman. Sichu Lu responded appropriately to Mr Muller's musings yesterday:
With all due respect. the reason that physicists believe that QG exists and it exists in such a form that would incorporate both GR and QM is because both of these theories work to extremely high precision and are extremely well tested, yet we don't know how they would work in much higher energy scales then we can currently test. So they are effective field theories. When we have a better theory in physics, we don't throw away what we previously know. Indeed the new theory would have to able to reproduce the old theory at some limit. We expect them to unify because we expect to keep being able to understand nature through physics. Sure, there is no "evidence" that this will continue to be the case. But it's like saying to a biologist, tomorrow you might discover an organism on the planet that isn't subject to evolution, and it might have genetic information coded in some way that isn't a double helix. It just wouldn't make any sense in the light what we know about nature already.
Amen to that. What is known, is known. And this knowledge may be classified as a collection of effective theories. Any viable theories in the future will have to agree with the effective theories in the realm where the effective theories have been successfully validated. The more complete theories will continue to be "quantum and relativistic" for a simple reason: these adjectives mean some "upgrade" of the physics toolkit onto a new level, and the un-upgraded theories – classical and non-relativistic theories – have simply been irreversibly ruled out.

We don't know what the future in science is exactly going to look like. But we know that it won't build on theories that have been proven wrong. A miraculous twist could hypothetically make the adjectives "quantum" and "relativistic" doubtful. But to hype this possibility is exactly on par with hyping any other hypothetical "revolution" that will e.g. imply that the DNA doesn't play any role in biology, or something equally radical. It's just totally stupid for someone to turn these insane speculations into pillars of his thinking about a scientific discipline – and to dismiss the actual science as "religion" or "hope".

As Sichu Lu correctly hinted, Muller's comment basically suggests that physics will cease to hold as a science. Muller's comments can't conceivably mean anything less than that. It's remarkable that Mr Muller is wrong about all the assertions he makes, including those that are not needed for his broader claim:
Mathematics doesn't break down for black holes. True, there is a singularity, but this singularity is invisible from the rest of the universe; it is cloaked behind an event horizon.
First, the claim that every singularity is cloaked behind an event horizon is just a hypothesis, the Cosmic Censorship Conjecture. It seems to be "morally" hold in the real, 3+1-dimensional world. But almost every rigorously formulated version of this conjecture seems to be provably wrong and in higher-dimensional spacetimes, the conjecture turned out to be wrong even "morally".

So the claim that every singularity is cloaked behind an event horizon is an example of a claim that should be classified as a "hope", unlike the indisputable facts that Mr Muller has called "hopes".

This hope, the Cosmic Censorship Conjecture, doesn't really have any good reason to hold. One "needs" to believe in this conjecture if he wants to believe that the classical general relativity is basically enough for all the macroscopic phenomena. But it doesn't have to be enough! A consistent and complete theory of quantum gravity may maintain its consistency and predictive power even in the presence of objects that look like singularities from the classical GR approximate viewpoint.

So there is no rational reason to believe that all singularities have to be cloaked. A more complete theory has to fix some (at least small) problems of the classical GR, anyway; so there's no reason to expect that it won't have any "larger purpose" at the same moment.

Now, even if they were cloaked, this cloaking doesn't imply what Mr Muller wrote that it implied. The cloaking doesn't imply that "mathematics doesn't break down"; he clearly meant "the mathematical description of general relativity doesn't break down". Pretty much by definition of a singularity in GR, the particular mathematical description does break down in the presence of a singularity. (String theory allows completely consistent physics of strings on geometries that look singular as GR geometries, but I don't want to go into this advanced stuff.) The breakdown may only be experimentally observable by an infalling observer who "remains around" until the very end. But this is just an excuse.

After all, the Hawking radiation is known to subtly depend on the detailed degrees of freedom that may also be described as those "inside" a black hole. So a singular value of any of these degrees of freedom may be a problem even outside, even if the event horizon basically cloaks the singularity.

To summarize, Mr Muller wants to make the classical GR look more healthy than it is. When one is rigorous enough, the appearance of singularities in classical GR actually does imply that a more complete theory – a quantum theory – is needed. This logic is fully analogous to the observation that the "states in the Hilbert space of a compact object" should better be discrete because classically, there are infinitely many points in the phase space which should formally imply that the entropy is infinite.

You may see that Mr Muller tries to make classical GR look healthier than it is. As a rank-and-file anti-quantum zealot, this is a part of the pattern. The following half-paragraph he wrote makes it totally clear that he wants to sling mud on quantum mechanics:
Mathematics does break down for the standard model of electromagnetism-weak-strong forces, which is based on quantum field theory. This theory has infinities that have to be "renormalized" in a way that isn't based on mathematics.
So the methods that people use to renormalize the Standard Model and QFTs are "not based on mathematics"? Wow. What are they based upon? Theology or psychology? Mr Muller is clearly batšit crazy.

One needs some advanced mathematical knowledge to renormalize quantum field theories; and in all these methods, one can reliably say which algorithms are right and which algorithms are wrong. And whether you aesthetically like the intermediate formulae or not, the calculations assume experimentally measured finite parameters and produce finite and experimentally verifiable (and successfully verified) predictions for further experiments. What can possibly be his basis for the claim that renormalization is "not based on mathematics"? Clearly, any "problem" that one encounters during the renormalization of a renormalizable quantum field theory is just an illusion. There is no genuine problem here.

It is complete nonsense that "mathematics breaks down" for quantum field theories in general.

In particular, QCD is a fully consistent theory making predictions at all length scale – including arbitrarily long and short ones. It needs renormalization but renormalization is a rock-solid well-defined collection of mathematical rules that lead to unambiguous and arbitrarily accurate results for the observables.

The Standard Model, which also includes the electroweak theory which also has the Higgs boson, is perturbatively renormalizable. So if we calculate anything as a power law expansion in the small coupling constants, this theory is as well-defined as QCD. Non-perturbatively, when "tinier than any power of the coupling constant" effects are taken into account, the Standard Model is an inconsistent theory due to the Landau poles for the \(U(1)\) gauge coupling and/or the Higgs self-interaction. But one needs to probe processes at energies much higher than the Planck energy to be sensitive to those. We know that at those high energies, the Standard Model is already inadequate because it ignores gravity – which becomes more important than those non-perturbative Standard Model effects at these trans-Planckian energies.

That's why the Landau poles (tiny, faraway inconsistencies) of the quantum field theories are usually much less harmful than some general inconsistencies that sick theories could suffer from. But in principle, the correct theory must be free of the small inconsistencies, too.
Much of the enthusiasm for string theory is that it addresses this problem, while introducing many more (extra dimensions, huge numbers of parameters, etc). Personally, given the problems of string theory, I am not optimistic that it will be with us 20 years from now.
Mr Muller counts himself among the anti-string crackpots, too, and adds his own creative idiosyncrasy. As of 2015, string theory obviously doesn't have any known problems that would indicate that it is going to be eliminated. Everything we know makes it rather clear or at least suggestive that it is a framework capable of reproducing all the successes of QFTs and GR; and curing at least a big part of their diseases.

Extra dimensions are an example of a wonderful far-reaching prediction of string theory, not a "problem" as crackpots including Mr Muller sometimes love to suggest. And string theory may be seen to admit no continuously adjustable dimensionless non-dynamical parameters whatsoever. Mr Muller is clearly a complete idiot if he hasn't been capable of noticing this fact even though he has already spent more than 70 years on Earth.

The claim about "optimism" is clearly a piece of fraudulent demagogy. Folks like Mr Muller don't want progress in string theory. They are dreaming about some problem with string theory that someone will find. Such a problem could "undo" Mr Muller's and other men's striking intellectual inferiority in comparison e.g. with the Berkeley string theorists. He wants something to "kill" string theory because he doesn't want to learn it. He wants to remain not only an anti-string crackpot but even an anti-quantum zealot.

But such an insight doesn't seem to be likely in the next 20 or 50 years. Quantum mechanics has irreversibly become a pillar of modern physics and the same thing largely holds for string theory, too. They have been connected to the actual observations that modern physics builds upon so tightly that to dream about their re-segregation means to be utterly unrealistic about the future of physics.

Even if a problem were found with string theory, what would it mean for it "not to be with us in 20 years"? It's obvious that the theory has already led to some striking results of mathematical character that can't go away and won't go away. So at Berkeley in 2035, Mr Muller can teach the future U.S. presidents or hippies (or students who "unify" both groups) that "string theory is no longer with us" – and prevent the students from studying important questions about the Universe – but those people in 2035 who will be more intelligent physicists than Muller's obedient students will surely continue in the research of it.

The last paragraph about the black hole coordinate systems is stupid, too:
The math of black holes, when treated consistently by a reference frame that is not falling into it, is in good shape. Most of the unknowns are based on attempts to make it a quantum theory. Hawking radiation is a first step, but we don't even know for sure that Hawking radiation exists; it has never been seen experimentally, and maybe never will be seen.
The classical GR that Mr Muller has previously excessively defended works well, on the contrary, in the reference frames connected with observers who are falling into the black hole! The coordinate systems of not-infalling observers include the Schwarzschild coordinates and they have a problem – the coordinate singularity – already at the event horizon. This problem makes these systems unsuitable for the description of observations by all the observers who do fall in.

It may be true that most of the unknowns are based on the quantum aspects of general relativity, if I make his statement a little bit more accurate and sensible. But most of the knowns are about the quantum aspects, too. There's just a more diverse spectrum of physical phenomena to be studied in the quantum theory than in its classical counterpart.

Now, the Hawking radiation is the "first step" and everything that the future U.S. presidents and Mr Muller need to know about it is that "we're not even sure whether it exists". Sorry, you and your future U.S. presidents don't even know whether the Hawking's radiation exists because you are a bunch of arrogant morons who are as dumb as a doorknob and who would love to define the standard of "what people should know and what they shouldn't know". Other physicists not only know that this effect discovered in the early 1970s exists. They have also made hundreds or thousands of additional steps – steps described in lots of papers – that your brains lacking talent, patience, and curiosity aren't even dreaming about.

It's nasty for Mr Muller to try to sell his staggering intellectual limitations as a virtue. I find it amazing that Mr Muller found it a good idea to embarrass himself in this way. Isn't there a theoretical physicist at Berkeley who would find some time to tutor Mr Muller?

No comments:

Post a Comment