Thursday, January 05, 2017 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Weinberg's trouble with quantum mechanics is not even wrong

Bill Zajc and Luke Lea have simultaneously sent me a link to a new (or future) article by Steven Weinberg in The New York Review of Books,

The Trouble with Quantum Mechanics
Weinberg's subjective dissatisfaction with quantum mechanics was growing in recent years, along with our fears about Weinberg's exceptional intellect. Many things in this article are similar distortions that have been discussed repeatedly but we may look at his thoughts again and emphasize those that look slightly new.

Well, the first new thing is the title, apparently a variation of the title of a notorious book by Lee Smolin, The Trouble With Physics. Before his 84th birthday, prominent Nobel laureate Steven Weinberg may have found a new goal – to become an assistant to the crackpot Lee Smolin. Or maybe he was just testing whether I would point out this similarity. Well, Prof Weinberg, be sure that I would.




Weinberg's troubled article starts with a picture from the book by an ex-colleague of mine Eric Heller showing two kinds of chaos, classical and quantum. I think that this picture is orthogonal to the main complaints that Weinberg's article is all about. The first real paragraph says:
The development of quantum mechanics in the first decades of the twentieth century came as a shock to many physicists. Today, despite the great successes of quantum mechanics, arguments continue about its meaning, and its future.
Weinberg is worried about the future of quantum mechanics. Well, quantum mechanics is pretty much guaranteed to remain the foundation of science. I would be much more worried not about quantum mechanics but about the people who gradually cease to understand science.

Incidentally, I think that the comment about the shock in the first decades of the twentieth century is misleading historically. It was actually straightforward for most physicists to switch to the new way of thinking that the discoveries of the quantum revolution clearly demanded. Sociologically, quantum mechanics became the mainstream almost immediately and for more than 30 years, critical comments about quantum mechanics were viewed as incoherent rants of irrelevant confused outsiders – despite the fact that Einstein was one of them. It was only in the 1960s when these confused outsiders began to hijack chairs in the Academia and the situation was getting worse in the following decades.




Weinberg continues with some history of the early 20th century physics. When talking about the wave-particle duality, we're told:
The world’s categories had become all muddled.
This is not how a top physicist talks about these matters. It looks like a sentence by a grumpy philosopher – exactly the same kind of philosophers' junk that Weinberg had been against so many times in the past, e.g. in "Against Philosophers" in his Dreams on a Final Theory. Why would a physicist talk about "categories" (unless they are categories from category theory in mathematics, e.g. derived categories to describe properties of D-branes)? Nature isn't obliged to be organized to "categories". Physical theories aren't about "categories". It's simply not a word that belongs to the scientist's vocabulary.

Also, the role and co-existence of (and "apparent" contradiction between) wave-like and particle-like properties could have been confusing for a while but a theory was found that is fully consistent and may be used to prove that all the feelings about inconsistencies are just illusions. So categories haven't "become" muddled. They only "looked" (and still "look") muddled to people who didn't look at the things carefully.
Worse yet, the electron waves are not waves of electronic matter, in the way that ocean waves are waves of water. Rather, as Max Born came to realize, the electron waves are waves of probability.
There is nothing "worse" about the laws of physics that use the waves of probability. The laws using the waves of probability amplitudes are demonstrably better: just compare classical physics vs quantum mechanics with the observations. Weinberg's claim about their being "worse" is demonstrably untrue.

After Weinberg mostly correctly says that the electron doesn't spread like butter, we hear:
Probability was not unfamiliar to the physicists of the 1920s, but it had generally been thought to reflect an imperfect knowledge of whatever was under study, not an indeterminism in the underlying physical laws.
This is exactly how the laymen misunderstand the role of probabilities in quantum mechanics. The trouble with Weinberg's sentence above is that the two "options" don't contradict each other at all (even though they could be shown to contradict one another if we assumed classical physics – but that assumption is invalid). The underlying physical laws (of quantum mechanics) are indeterministic, as Weinberg correctly says. But the first part of that sentence is wrong. The probabilities still do reflect an imperfect knowledge of whatever was under study.

Probabilities strictly between 0% and 100% always reflect an imperfect knowledge about the proposition or the object that the proposition talks about! This is what the term probability f*cking means, Prof Weinberg.

Even though the laymen and Steven Weinberg believe otherwise, quantum mechanics doesn't invalidate that statement. Probabilities still mean that the knowledge about the objects and their properties is imperfect, just like they always meant the same thing. Instead, what is new is that quantum mechanics says that the imperfection of the knowledge is unavoidable. That's called the uncertainty principle. If you know something about (ideally the value of) the quantity associated with the operator \(G\), then you cannot know the value of another quantity associated with the operator \(H\) that obeys \(GH-HG\neq 0\). The previous sentence holds for any observer, not just you. It holds for any vantage point from which one may discuss the state of objects in Nature. Because the uncertainty is unavoidable, it's an "objective" law of physics. But the uncertainty in the knowledge about anything is still generally subjective – we must always ask "whose knowledge" we talk about. Probabilities refer to the knowledge and the knowledge is always subjective or observer-dependent.

If you prepare an electron with the spin directed along the \(x\)-axis with \(j_x=+1/2\), then the precise value of \(j_z\) cannot be known according to quantum mechanics. Both options, \(j_z=+1/2\) and \(j_z=-1/2\), are possible. In classical physics, an observer could have imperfect knowledge about classical bits analogous to \(j_z\), too. However, the difference is that in classical physics, one could assume that the imperfect knowledge about all these quantities could have gotten better – a more accurate observer could have known everything. He could have known \(j_x\) and \(j_z\) at the same moment. There existed an idealized observer – omniscient God, if you wish – and physics could have adopted His attitude and eliminate all the ignorance from physical discussions. According to quantum mechanics, it's not so because \(j_xj_z - j_z j_x\neq 0\) even for God.

In quantum mechanics, it's no longer the case that the perspective of an omniscient God may be adopted. Quantum mechanics postulates that no one – not even God – can have the perfect knowledge of all the properties of the physical systems. Pairs of generic operators don't commute and no observer can therefore know their values at the same moment. \(x\) and \(p\) represent the most well-known example. I think that Weinberg simply misunderstands what the uncertainty principle says if he knows the principle at all.

He writes lots of sentences that eliminate all my doubts that he misunderstands what quantum mechanics actually says about the uncertainty (and what the term "probability" actually means), for example:
Probability enters Newtonian physics only when our knowledge is imperfect, as for example when we do not have precise knowledge of how a pair of dice is thrown.
Again, probability different from 0% and 100% always means that someone's knowledge is imperfect. The real trouble with Weinberg's sentence is the word "only". Weinberg assumes that most of the time, our knowledge is perfect. That was really rubbish even in the epoch of classical physics. But one could postulate someone – God, if you wish – whose knowledge was perfect. That's no longer the case in quantum mechanics. The imperfection of the knowledge is a law of physics – the uncertainty principle. The word "only" before "our knowledge is imperfect" becomes ludicrous because "our knowledge of all the quantities is guaranteed to be imperfect" by the laws of physics. So it is "always" so. When some condition is "always" satisfied, it's silly to say that something else happens "only" when it's satisfied: It always happens, too.

But none of these things can change the fact that the probabilities mean that the speaker's knowledge is imperfect.

OK, I will skip a few paragraphs painting quantum mechanics as "very strange" and "weird" because they're annoying propaganda that is not new and I have wasted enough time with this irrational garbage.
Even the most adventurous modern speculations, such as string theory, are based on the principles of quantum mechanics.
String theory is neither a speculation nor an adventurous theory. It is the state-of-the-art, conservative, and only consistent theory of quantum gravity that we know of. But yes, it obeys and has to obey the universal postulates of quantum mechanics.
Many physicists came to think that the reaction of Einstein and Feynman and others to the unfamiliar aspects of quantum mechanics had been overblown.
Feynman was in no way an anti-quantum babbler like Einstein. He understood the right way to use quantum mechanics very well and his teaching of quantum mechanics was flawless. His quote about "no one understanding" meant that no picture that people want to imagine may be the right picture – that quantum mechanics isn't a classical theory. Feynman has never claimed that quantum mechanics had a problem (in the sense of an inconsistency or incompleteness). On the contrary, he repeatedly stressed that he believed that all the efforts to find something else underneath quantum mechanics were doomed.

After some correct comments that it was bad for Schrödinger and Einstein to isolate themselves from progress in physics, we read:
Even so, I’m not as sure as I once was about the future of quantum mechanics. It is a bad sign that those physicists today who are most comfortable with quantum mechanics do not agree with one another about what it all means.
In some cases, it just means that some of these people are wrong. In others, it means that they like to use different human words but the human words are really inconsequential for the physics. None of these observations is a valid scientific argument in favor of a "trouble with quantum mechanics".

OK, quantum superpositions are compared to musical chords – I sometimes do the same when I explain that C+E is something else than D (musical notes) and similarly, the superposition of "dead" and "alive" is something else than a tired/injured cat or "something in between" dead and alive. But one must be careful about the differences as well. The wave underlying a musical chord is still a classical wave – while the wave function is not.

After Weinberg sketches what he imagines Born's rule to be, we read:
The introduction of probability into the principles of physics was disturbing to past physicists, but the trouble with quantum mechanics is not that it involves probabilities. We can live with that. The trouble is that in quantum mechanics the way that wave functions change with time is governed by an equation, the Schrödinger equation, that does not involve probabilities.
That's simply wrong, too. There is nothing "troubling" about a well-defined "deterministic" equation governing the evolution of probabilities or probabilistic distributions. After all, there is nothing intrinsically quantum mechanical about those, either. Even in classical physics, the probability distribution on the phase space was evolving in time according to a "deterministic" equation, the equations from Liouville's theorem. As I have shown in some detail in several previous blog posts, Schrödinger's equation for the density matrix is just a straightforward rewriting of those Liouville's equations.

It's absolutely standard for probabilities to evolve according to some laws. That's what probabilities did in classical physics, too. Bayes' theorem is just another example of a law that fully determines the change of probabilities, in this case in a discrete step. Bayes' "deterministic" law for the change of probabilities after evidence is collected has its counterpart in quantum mechanics as well – it's the reduction of the wave function after a measurement. The very outcome of the measurement is not fully determined when probabilities are between 0% and 100% – but that statement was true in classical physics, too. A probability different from 0% and 100% means and has always meant that the outcome is unknown i.e. not fully determined.

When it comes to well-defined laws governing the evolution of probabilities in time, it's just a plain stupidity to suggest that these laws are "troubling" in any sense. Quantum mechanics doesn't change anything about the meaning of probabilities, their relationship to imperfect knowledge, the existence of well-defined equations by which these probabilities evolve as well as discontinuous Bayesian/collapse changes by which the probabilities jump after an observation. The only thing that is new in quantum mechanics is the uncertainty principle (due to the nonzero commutators) which, among other things, forbids any perspective in which two generic observables are perfectly known at the same moment. The nonzero commutators make it unavoidable to talk about probabilities between 0% and 100% i.e. about imperfect knowledge. But what the "knowledge", "imperfect", "probability" etc. mean is exactly the same as it always was.

Bill Zajc has picked this sentence from Weinberg's essay as the predicted locus that causes irritation:
So if we regard the whole process of measurement as being governed by the equations of quantum mechanics, and these equations are perfectly deterministic, how do probabilities get into quantum mechanics?
Probabilities have been a part of quantum mechanics from the beginning and they have to be a part of it by definition – or because of the universal postulates of quantum mechanics. Thanks for asking. The question how "probabilities got to quantum mechanics" is as silly as the question how the terrestrial life forms got to Earth. "Terrestrial" is an adjective meaning that they are on Earth so something is terrestrial exactly when it's on Earth. In the same way, quantum mechanics is a theory whose mathematical variables (I mean mainly the wave function now) that may be evolved according to well-defined ("deterministic" is misleading) equations have to be (physically interpreted as) probability amplitudes. If it didn't need probabilities, it wouldn't be quantum mechanics.

In following paragraphs, Weinberg tries to promote theories that say that the underlying theory doesn't have any probabilities – the underlying theory violates the uncertainty principle because everything fundamental is certain in it. Sorry, all such attempts are guaranteed to fail and only deluded folks study this kind of garbage in the first place. It's an established scientific fact that the uncertainties i.e. probabilities between 0% and 100% cannot be eliminated from the fundamental laws – this established scientific fact is known as the uncertainty principle. The question whether \(x\) and \(p\) – which can demonstrably be measured – are knowable at the same moment is mathematically equivalent to the question whether \(xp-px\) is zero. The difference \(xp-px\) may be measured and was measured experimentally and the result, \(i\hbar\), is not zero. In fact, it's known with the precision one part per billion. The chance that it's zero is excluded at least at the one-billion-sigma confidence level. It's just a full-blown denial of reality to suggest that the observables commute i.e. that they are simultaneously perfectly knowable.

Weinberg tries to present decoherence as something that provides us with the pseudorandom generator. That's completely wrong. Decoherence doesn't collapse the density matrix to a particular outcome – i.e. a diagonal matrix with zeroes and the number one once. Decoherence is an approximate description of the evolution of a density matrix whose final result is a diagonal matrix with diagonal entries (probabilities) which are numbers in between 0 and 1 that add up to one. But the number of diagonal entries strictly in between 0 and 1 is greater than one – all possible outcomes are confirmed as possible after decoherence takes place. Decoherence in no way changes the fact that the outcomes are random and only probabilities are calculable.

So almost everything that Weinberg writes about decoherence is just wrong. What he writes about the correct quantum mechanics is even stinkier cr*p, however:
One response to this puzzle was given in the 1920s by Niels Bohr, in what came to be called the Copenhagen interpretation of quantum mechanics. According to Bohr, in a measurement the state of a system such as a spin collapses to one result or another in a way that cannot itself be described by quantum mechanics, and is truly unpredictable. This answer is now widely felt to be unacceptable.
Sorry but the unpredictability of the outcomes is an established fact that every credible physicist agrees with and only cranks may find it "unacceptable". If Weinberg considers the intrinsically probabilistic character of quantum mechanics to be just an idiosyncrasy of Niels Bohr's, universities in Cambridge and Austin should probably demand him to return the salaries he has received as an instructor of quantum mechanical courses.
There seems no way to locate the boundary between the realms in which, according to Bohr, quantum mechanics does or does not apply.
There is an easy way to locate it. The boundary doesn't exist or is at infinity. Quantum mechanics applies everywhere. The only boundary that is found at a "finite place" and that may be discussed is the boundary bounding the regimes which are also described by classical physics or classical concepts reasonably well. The shape of this boundary may be calculated from quantum mechanics and has a complicated shape depending on the kind of problems we calculate. But quantum mechanics applies on both sides of this boundary! Is it already so bad with Weinberg that he misunderstand this trivial point?

An observer also invents a boundary, the Heisenberg cut, between himself and the world he directly perceives – the observer side of the cut whose information may be treated as classical information; and the observed side that needs the treatment in terms of probability amplitudes. The side treated classically has to agree with the fact that classical physics is a good information. But otherwise the cut is arbitrary and objects on both sides of the cut may appear on the quantum side of the cut as defined by another observer.
As it happens, I was a graduate student at Bohr’s institute in Copenhagen, but he was very great and I was very young, and I never had a chance to ask him about this.
Dr Weinberg may easily ask me. I can give and I have given Bohr's answers – although probably much more clearly stated than Bohr could. ;-)

Otherwise, Bohr remains very great even now, when he's dead, at least in the extent to which he understood the fundamental properties of the laws of physics.
Today there are two widely followed approaches to quantum mechanics, the “realist” and “instrumentalist” approaches, which view the origin of probability in measurement in two very different ways. For reasons I will explain, neither approach seems to me quite satisfactory.
The separation to "realist" or "instrumentalist" may be interpreted either as a separation of actual physical theories that are used to make predictions or as a separation of people's preferred language and philosophy to think about partly unphysical, philosophical questions. When it comes to the theories, the situation is unambiguous. "Realist" theories are wrong, stupid, and excluded because they just can't produce correct predictions for all the phenomena while all the correct theories require the "instrumentalist" attitude.

When we talk about language, the word "instrumentalist" may sound misleading to many people as well because of two reasons. One is that it suggests (and Weinberg explicitly confirms the hidden meaning) that there is something incomplete about quantum mechanics and quantum mechanics is only "useful for a practical man". I don't think it's true. Quantum mechanics is almost certainly complete and no "deeper" description may exist. The second reason why "instrumentalism" may be misleading is that the fundamental quantum mechanical theory allows us to discuss any "instrument" – it doesn't break down whatever we choose so the suggested dependence on the "instruments" is almost as misleading as if we suggested that quantum mechanics is "anthropocentric". What the application of quantum mechanics needs isn't observers that look just like humans with instruments that look just like those produced by humans. What the application of quantum mechanics needs is just observers that can become aware of the values of some physical observables.

At any rate, all these questions were already completely settled in the mid to late 1920s and no evidence whatsoever has emerged in the following 90 years that would indicate some trouble with those conclusions.

OK, Weinberg interprets, reinteprets, and misinteprets the word "instrumentalism" in various ways. For example, he claims:
It seems to me that the trouble with this approach is not only that it gives up on an ancient aim of science: to say what is really going on out there.
If by "what is really going on out there", he means that there is a theory that violates the uncertainty principle and admits the description of all the fundamental variables that are observable in principle and completely certain at the same moment, and the rest of his essay indicates that that's exactly what he "wants", then the answer known from 1925 is that these theories – known as theories of classical physics – have been ruled out. They are f*cking wrong because they violate a fundamental axiom of physics, the uncertainty principle.

Values of (non-commuting) observables aren't perfectly knowable at the same moment and this is a fact that every viable law of physics has to agree with. This is an important fact describing what is really going on in Nature while Weinberg's fantasies about the world without the uncertainty principle are what is never going on Nature. They're old science that is exactly as debunked as creationism and if Weinberg believes that he is behaving more reasonably than the creationists towards the scientific evidence about these fundamental questions, then he is just fooling himself – and others – and abusing the fact that the ideologically biased university environment allows some types of stupidity more than others.

When Weinberg says that it's "troubling" when quantum mechanics doesn't allow the fundamental, in principle observable quantities to be perfectly known at the same moment, i.e. when he directly assaults the uncertainty principle, he displays exactly the same kind of metaphysical bigotry as a creationist who finds Darwin's theory "troubling" because it's not compatible with the constancy of species according to the Bible, God's one-week creation contract, or with some verses in the Bible that have described that contract. Science has simply proven these prejudices wrong. Species have been scientifically proven to evolve and observables have been scientifically proven to have nonzero commutators i.e. to be simultaneously unobservable with perfection.
Thus humans are brought into the laws of nature at the most fundamental level. According to Eugene Wigner, a pioneer of quantum mechanics, “it was not possible to formulate the laws of quantum mechanics in a fully consistent way without reference to the consciousness.”
This criticism is absolute rubbish, too. Wigner stated that quantum mechanics requires consciousness. Consciousness isn't the same thing as humans. Consciousness is just a "spiritually sounding" synonym of the ability of an observer to be aware of a measured value of an observable. But the observer capable of "feeling" this information doesn't have to resemble a human in any way. It may be a smart chimp or a silicon-based computer or anything else. The idea that members have some special features that separate them from the rest of the Universe is a silly medieval superstition. And it's Weinberg, not Wigner, who implicitly assumed this superstition.

Wigner only said that quantum mechanics needs an observer – and presented this fact in words that people who care about their spiritual life like to use. But the content is exactly the same as the content of some universal postulates of quantum mechanics.

After additional confusing paragraphs about humans, Weinberg correctly says that "objective probabilities" are untenable. Yes, probabilities at a particular situation must always refer to the state of mind of someone who knows something – and it depends whom we pick. Probabilities at a given moment are always observer-dependent or subjective. Quantum mechanics allows us to calculate values of probabilities according to objectively valid laws of physics. But the resulting probabilities have to be interpreted as subjective probabilities. In a particular situation, probabilities are always subjective. And that's what viable laws need according to quantum mechanics.

Some more paragraphs talk about the "trouble" with "realist" theories. I think that it's good that he sees the trouble but he doesn't see the main trouble of those theories because he shares the basic misunderstanding of the concepts with the champions of these pseudoscientific "realist" memes. I don't want to read these paragraphs carefully.
Strange as it is, the entanglement entailed by quantum mechanics is actually observed experimentally. But how can something so nonlocal represent reality?
Entanglement doesn't represent any nonlocality. There is zero nonlocality in quantum field theory on a Minkowski background. The absence of nonlocal effects is mathematically encapsulated in the vanishing (graded) commutators of spacelike-separated field operators. The entanglement is nothing else than the most general correlation between two subsystems that have previously interacted (or have common ancestry) – which explains why the correlation is possible in a local world – and expressed in the fully quantum mechanical description of the wave function (which allows some properties of various correlations that wouldn't be possible if the world were classical).
What then must be done about the shortcomings of quantum mechanics?
What must be done for the idiotic talk about shortcomings of quantum mechanics go away is for totally confused people like Weinberg to shut up – or, even better, shut up and calculate. Nothing else is needed, nothing else has been needed since 1925. Quantum mechanics is a perfectly consistent and complete framework for studying and applying theories that describe everything in Nature at the most fundamental level.

Thankfully, Weinberg's first answer is "shut up" but he utterly fails to apply it to himself. He must believe that he stands above the laws of quantum mechanics. He doesn't. No one does.

The remaining portions of his essay say a few words about the Ghirardi-Rimini-Weber "objective collapse" theories and Weinberg's own efforts to experimentally prove that quantum mechanics is wrong – and what he says is nothing less than that. Sorry, all these things are wrong for the reasons I have described many times and it is guaranteed that this kind of "research" won't lead to anything useful.

Years ago, I was only gradually noticing weird comments about quantum mechanics that Weinberg sometimes made. I cannot be 100% certain whether the transformation was mostly due to Weinberg's becoming a more full-fledged anti-quantum zealot, or due to my increasing allergy to these falsehoods that Weinberg was saying about quantum mechanics. I do believe it's mainly the former.

At any rate, I consider Weinberg to be a 100% anti-quantum zealot – and a full-blown peer of Gerard 't Hooft – at this point. It's sad. It was probably the last text by Weinberg about the foundations of physics that I wanted to (partly) read.



Off-topic: Willie Soon has sent me documents showing that the IPCC has adopted a new catastrophic scenario. Higher CO2 levels affect titanium dioxide nanoparticles and after several additional causal influences captured by the 55-million-view video above, this threatens rice in China. ;-) Good job, guys. Time to hear "You're fired".

Add to del.icio.us Digg this Add to reddit

snail feedback (0) :