Friday, January 23, 2015 ... Deutsch/Español/Related posts from blogosphere

Tom Siegfried's delusions about the reality of the wave function

In recent days, I've received links to many texts that looked insultingly and fundamentally wrong. One of these hyperlinks was contributed by Bill Zajc. Tom Siegfried, a journalist who can be good at times, wrote

Physicists debate whether quantum math is as real as atoms
in ScienceNews.ORG. It's the second part of a series. The first part said some basic things about Bell's theorem and except that the importance of that paper of Bell's was overstated by an order of magnitude (and except for the popular derogatory "weird" adjective used against quantum mechanics), the content was marginally OK. That can't be said about the second part. I was seeing red for an hour after I read that thing. This text is just terrible and serves as a testimony of the catastrophic degradation of the intelligence of science journalists and some loud people calling themselves scientists, too.

First of all, the title is obviously wrong. In 2015, no genuine physicists are debating whether "quantum math is as real as atoms". All genuine physicists know what quantum mechanics actually says about this question. What it says can't be exactly reproduced by the human language or the language of philosophers, other laymen, or stupid people in general who like to debate meaningless questions with ill-defined words rooted in tons of invalid prejudices such as "is it real?". Only armchair physicists, third-class physicists, and "philosophers" may consider these foundational debates about quantum mechanics to be a part of the actual research in 2015. Physicists know that all these questions have been settled since the middle or late 1920s.

Siegfried begins with Ernst Mach's opinion that "atoms may be useful for visualization or calculation but I don't believe that they are real". Not only because Ernst Mach was born in the Czech lands, I feel the urge to defend him. It is not quite clear whether he really disbelieved the derivations – such as those in statistical physics or the Brownian motion (he was already rather sick at that time) – based on the notion of the atom.

As far as I can say, his point could have been completely correct, namely that the derivations based on the atomic theory were useful and often insightful methods telling us something about our actual observations. But at that time, atoms weren't really directly observed which is why it was legitimate to consider them virtual, auxiliary theoretical constructs.

My broader point was that Ernst Mach was an important representative of positivism in philosophy and while no philosophy may be treated as a permanently solid basis for science, positivism was positively correlated with the way of thinking that was needed for quantum mechanics (and in fact, even relativity) to emerge and replace Newtonian physics.

Positivism states that all of our trustworthy knowledge about Nature comes exclusively from sensory experience – from observations, if we use a little bit more physical jargon. In the context of physical theories, it means that a physical theory a priori declares "all objects in the world around us" and words describing them as meaningless. Every physical theory has the right to do so. Every physical theory may work with its own fundamental and auxiliary concepts to describe our experience. The experience is the only "provably real" thing. But the point is that every question about Nature that can't be reduced to our experiences is considered physically meaningless or unphysical.

(The Vienna Circle elaborated upon these important positivist observations and probably was the culmination of philosophers' understanding of the essence of modern science. When other philosophers such as Thomas Kuhn or Karl Popper began to be popular, the degeneration and putrification of philosophy of science was already underway. To return to Mach and his positivist predecessors, they generally erred by forecasting that "something would remain unobservable forever". For example, the founder of positivism Auguste Comte would predict that the chemical composition of stars would remain a mystery forever – just years before spectroscopy answered all those questions. But this error must be seen as a side effect of their important and right understanding that certain questions and claims are physically meaningless.)

A reason why I consider Ernst Mach (and perhaps a few others before him) rather smart was that he was apparently able to switch to the new way of thinking that is so important in quantum mechanics – and, to a lesser extent, in relativity – decades before quantum mechanics actually emerged in physics. It's remarkable because almost all of today's philosophers of science are unable to think properly even 90 years after quantum mechanics was discovered.

With a grain of salt, we could argue that about 100 years ago, the philosophers were actually useful for the progress in physics. That's also why most of the greatest theoretical physicists of that time considered themselves part-time philosophers. After the Second World War, the philosophers had largely lost it. The greatest physicists of the post-war generation, the likes of Feynman and Schwinger, to mention two, were no longer admiring philosophy in the usual sense. They diverged. The propagation of the anti-quantum delusions by the contemporary philosophers may be viewed as a revenge against physics – revenge for the fact that science had to completely escape away from philosophy and the best physicists no longer have any intellectual respect towards the contemporary philosophers.

Let me return to Ernst Mach.

If you think about his philosophy, it makes quite some sense and in variations, it may be said to capture the spirit of relativity as well as quantum mechanics. Mach's positivist spirit allows us to say that the "(objective) simultaneity of events" in special relativity may be said to be meaningless because it cannot be operationally decided – it cannot be converted to an experiment producing some perceptions. If we try to prepare the experiment, we may find out that the results depend on the observer and his inertial frame. The simultaneity may very well be relative – and in relativity, it certainly is. There doesn't exist any "objective" or "absolute" way to experimentally decide whether two events occurred simultaneously, and that's why it's right to say that the question whether two events are (objectively) simultaneous doesn't belong to physics. This is really the key to special relativity.

In the same sense, the precise information about the "point in the phase space" (which is just a slightly more mathematical synonym for the "objective state of the world") may be said to be physically meaningless because the position and the momentum can't be simultaneously measured, thanks to the uncertainty principle. Even by incomplete arguments, it may be seen that the observation necessarily modifies the system, and we need two observations that "interfere" with one another, so they can't be done simultaneously. The fact that these pieces of "classical information" – which were meaningful to ask about in classical physics – are no longer meaningful is of course the key (or at least a key) to the new quantum mechanical thinking and the founding fathers of quantum mechanics actually praised positivist philosophers for that helpful attitude.

Ernst Mach also promoted his version of "extreme general relativity", Mach's principle. It inspired Einstein when he was searching for the right new relativistic theory of gravity. According to Mach's principle, there can't be a difference between non-rotating and rotating (or accelerating and non-accelerating...) frames in the otherwise empty Universe, so whether a bucket is rotating in an empty Universe must be determined by its location and orientation relatively to other objects – stars that may exist.

Now, this was another possible excellent step to remove the shackles and possible wrong (because experimentally unsubstantiated) assumptions. I would say that Mach's principle, despite its role of having stimulated Einstein, is wrong in general relativity. Experiments in rotating and non-rotating buckets certainly differ and this difference may exist even without stars. But we may present the reason why Mach's conclusion was wrong in a pleasant way, too. Mach's conclusion (that the bucket's rotation only makes sense when there are stars) was wrong because he incorrectly assumed that the Universe without the stars is "empty" (carrying no information that could distinguish the rotation). But the Universe without the stars is still filled with the metric tensor field and that carries some information.

So we may say that the positivist philosophy was positively useful in 2 of the 3 revolutions, and it only seemingly worked in the third in which it was finally rejected. That's a pretty good, positive overall score for positivism, isn't it?

By the way, I mentioned both positivists and Popper. It may be useful to mention that the similarity between their principles is misleading. Both Popper and positivists say something like "if [something] can't be operationally measured, it's physically meaningless". But Popper was never a member of the Vienna Circle – he was named the "official opposition" – and it's important what is "[something]" in the previous sentence. Popper says (or at least his followers say) that a [theory] that doesn't connect to bread and butter is unscientific. But in some sense, positivists are saying the opposite thing because their "[something]" is often the "things that the critics of a theory demand to be observed". Positivists say that it is perfectly OK for a theory to say that many (a priori all) such things are unobservable! The winner of this disagreement is the theory, positivists say, and not those who demand some feature to be observable! The difference may be subtle but from some viewpoint, Popper tried to reverse all the rules of the game and these seemingly subtle changes are the reason why Popper was so much further away from the actual research scientists.

Reality of atoms vs wave functions

Now, Siegfried asks: Are wave functions as real as atoms? Well, that's a good or bad question, depending on what you mean by "something is real" and other things. There is a sense in which the answer to the question is obvious: The degree of reality of both atoms and wave functions is the same because atoms are nothing else than particular wave functions of the elementary particles. If wave functions are "not real", then of course we may say that "atoms are not real, either" because they're almost the same things!

People got used to saying that "atoms are real". But if we are careful about the meaning of this sentence in physics, it is only correct in the Machian sense. It means that the derivations working with the concept of an atom yield predictions for our perceptions that actually agree with the perceptions we have. There is no scientific justification for a much stronger interpretation of the sentence "atoms are real". If by "real", you mean an objective reality as imagined by classical physics, of course that atoms are not real. They may only be described by (complex) wave functions with a probabilistic interpretation.

Ontic vs epistemic

An armchair physicist and anti-quantum zealot Matthew Leifer is mentioned at the beginning of a would-be important discussion whether the wave function is "ontic" or "epistemic". The words are derived from Greek roots – because the philosophers incorrectly assume that they may hide their stupidity in front of everyone by using words that sound Greek to others – but they may be translated in a simple way. The wave function is "ontic" if it "objectively exists" (the root means "to be") while the wave function is "epistemic" if it describes what we "know" (the root "episteme" means "knowledge" or "understanding").

OK, so the would-be deep question is whether the wave function tells us something about "what exists" or "what we know". At this level, you see that it's a classic example of a meaningless philosophical question using ill-defined words. The right answer is to ignore the details of the question, to point out that it is sloppily formulated, and to explain that the correct theory – quantum mechanics – predicts the probability of perceptions from some knowledge about the initial state – obtained from other perceptions – using mathematical procedures that may be written in terms of mathematics. Is it "ontic" or "epistemic"? You may pick. But to demand anything else beyond the explanation how quantum mechanics works to predict observations is scientifically meaningless according to the positivist principles.

Well, if I had to pick, I would pick "epistemic". The wave function describes the state of knowledge. It's important that there is fundamentally no "objective reality" in quantum mechanics because quantum mechanics isn't equivalent to any classical theory. The previous sentence isn't any "interpretation" or controversial assumption. It is a basic fact about the status of quantum mechanics. Nothing about quantum mechanics makes sense if you are uncertain about the fact that it is a different kind of a theory than any classical theory.

So we may have picked "epistemic" because "ontic" was totally wrong. "Ontic" claimed that the wave function describes just another classical (probably deterministic) theory – that the wave function is a kind of a classical wave – which is surely not the case. Fine. The problem is that this way of choosing "epistemic" is negative. We refuse the first one so we may be forced to pick the second one. But the second option, "epistemic", is also wrong if you study what these philosophers actually imagine under "epistemic" in detail.

If you try hard to learn how they imagine "ontic" and "epistemic" wave function and translate these vague feelings of these would-be thinkers into some more accurate language of physics, you will easily find out that
by "ontic", they mean that the world is described by classical physics with some phase space coordinates and momenta \(x_i(t)\) and \(p_i(t)\), perhaps infinitely many, so something "objectively exists" and may be assumed to be exactly determined

by "epistemic", they mean that the wave function is in some way related to the probabilistic distributions \(\rho(x_i(t),p_i(t))\) in classical statistical physics.
I have read a dozen of papers by "this kind of writers" and everything I have seen is consistent with the claim that they assume that the world either works as a mechanical clockwork that is conceptually a classical theory with determined coordinates and momenta (that's the "ontic" option); or as a description of classical statistical physics with an evolving probability distribution (that's the "epistemic" option).

However, Nature around us doesn't work in either way. Just like the electron in Nature is neither a classical particle nor a classical wave, the wave function is neither "ontic" nor "epistemic". The world is simply described neither by classical physics evolving a point in the phase space; nor by classical statistical physics.

Let me mention that these two options would describe the same class of possible Universes, anyway. The description in terms of a single well-defined trajectory or history \(x_i(t)\) is a special example of classical statistical physics whose probabilistic distribution \(\rho\) is a \(\delta\)-function. On the other hand, the statistical description involving \(\rho(x_i)\) is relevant in a world with \(x_i(t)\) when we simply don't have the full information about the system, and we mostly don't have it.

So this whole "choice" is spurious. You didn't get any choice about the "type of the Universe" at all. You were only offered two formalisms to describe the same kind of a Universe, a classical Universe. It's similar to being offered two political parties with different names and colors. But both parties promote the climate hysteria, bailouts, advantages for a specific "class", and similar toxic junk. Clearly, you are not getting much freedom to choose. The right option isn't there among the two at all. Similarly, the correct character of quantum mechanics is neither "ontic" nor "epistemic" in the sense understood by the "philosophers". Quantum mechanics is a conceptually new type of a theory that has never been seen in the classical world.

The wave function is "epistemic" in the sense that it forces us to talk about our knowledge and not about an objective reality – and to formulate our predictions in terms of probabilities. But it is "ontic" in the sense that unlike a smeared \(\rho(x_i,p_i)\) in classical statistical physics, it contains the maximum knowledge that anyone can ever have! The description in terms of a \(\ket\psi\) is "complete". That's simply how it works. There are no hidden variables. If you know as much as you can know about a physical system, then you describe it by a pure state. The mixed state means that you add some extra ignorance that could in principle be eliminated. But the pure state still means that most questions about the properties of your system remain "uncertain" – the answers are inevitably probabilistic. This is not some bug that may be removed. The very claim that this uncertainty can't be removed (e.g. by better or more extensive measurements of the initial state) is the content of the uncertainty principle!

So the wave function represents the maximum knowledge about the physical system but the maximum knowledge still forces us to admit that most questions have uncertain answers and we still need to talk about probabilities!

That's how it works and all of these basic things have been perfectly understood for 90 years. One can't say whether the wave function is "ontic" or "epistemic" because none of the options we are being offered is really right. Everyone who tries to say that something remains open about these basic matters – beyond the clear answers I have presented above so that intelligent schoolkids must be capable of understanding the point – misunderstands modern physics at a fundamental level and should shut up (and perhaps also calculate, a skill that these people are unlikely to possess).

Not surprisingly, Tom Siegfried also relies on some popular and really atrocious sociological argument, literally "consensus science":
Bohr’s attitude was incorporated into the view of quantum physics known as the Copenhagen interpretation, once the predominant approach but now widely challenged.
The reason why quantum mechanics (and yes, "Copenhagen interpretation" is being used just as an insulting synonym for quantum mechanics: there exists no "other" quantum mechanics than one described by the Copenhagen school) is being "widely challenged" these days is that the postmodern system has failed to eliminate incompetent pseudointellectuals and populist demagogues from the wider physical environment – science journals and even universities (especially their philosophy departments but not only philosophy departments). This fact changes nothing whatsoever about the point that physics has perfectly understood the status of the wave function for 90 years (without a few months: I guess that I will celebrate the 90th anniversary of the Heisenberg et al. results this year).

Of course that if someone will pay attention to what rather generic stupid – albeit sufficiently loud – people say about modern science, modern science will look like completely incoherent chaos to him. And even if there will be apparent consensus about something, the answers included in the consensus will probably be wrong.

Wojciech Zurek, the guy who coined the word "decoherence" and much of the modern description of this effect (that he sold under different brands but it's always the same thing) is the only living guy in Siegfried's article who is saying things that are mostly right. Everything else is a gallery of nuts. Sorry, in this case, I must include Mark Strednicki straight to this category, too.

Normally, I would be explaining how crazy the Pusey-Barrett-Rudolph paper (mentioned by Siegfried) was. It's a typical paper that is incapable of understanding that quantum mechanics is neither "ontic" nor "epistemic" in the sense above, another paper among (now) thousands of worthless texts by mediocre thinkers that are trying to force Nature to recant and accept classical physics once again.

And I won't dare to swim in the cesspool which would be needed for a really comprehensive reaction to the preposterous concept of the Boltzmann Brain – the idea that the Boltzmann Brains may be real. The mental midgets who promote this junk are not only failing to realize the important lessons of positivism – that questions that can't be reduced to operational tests are meaningless – but they are also hysterically demanding totally unjustified assumptions such as that "everyone is equally likely to be anyone", including "anyone" which are random pieces of matter at any time of the history of any Universe. This is clearly completely wrong – probabilities are almost never uniform – and the assumption is, in fact, mathematically inconsistent. I won't swim in this cesspool created by "scientists" such as Sean Carroll because I would have to wash myself for a week.

Much of this blog post was saying nice things about positivists and Mach. So let me end on the same semi-optimistic note. Siegfried's last paragraph says
“As the intellect … grows in discipline,” Mach predicted, “physical science will give up its mosaic play with stones and will seek out the boundaries and forms of the bed in which the living stream of phenomena flows.”
Times were promising 100 years ago. The intellect was growing and people were abandoning naive classical "stones". Quite symmetrically, we may say:
“As the intellect … decreases in discipline,” Motl observed, “physical science is returning to its mosaic play with stones and will cease to understand the boundaries and forms of the bed in which the living stream of phenomena flows.”
As you can see, because people still live 99 years after Mach's death, his prophesy that could be used in both ways wasn't too optimistic, after all. With Siegfrieds and Carrolls behind the steering wheel, the mankind is happily flowing towards complete idiocracy.

And that's the memo.

Add to Digg this Add to reddit

snail feedback (63) :

reader Guest said...

Would you agree that what's "real" in the wave function is the recipe that allows to prepare the system described by said wavefunction?
It is the label inside the ket, if you wish;
in a way, it contains "half" of the classical state, so this information has the same "reality status" granted to classical objects.
At least this is how I keep my sanity when thinking about QM :)

reader Luboš Motl said...

Apologies, I don't understand a word. Or more precisely, I understand every word but at the length scale of 3 words, the coherence is lost.

reader Avraham rosenblum said...

Here is a paper by a professor in Philosophy and I am curious as to your opinion.

reader Quantum said...

Whose experience? By experience, do you mean qualia, or something else? Does Wigner's friend have experiences? Am I having an experience right now? Are you having an experience right now?

reader guest said...

Terry Rudolph call people like sean carroll "monkey realist" :)
very entertaining lecture.

reader RAF III said...

Lubos - What a fine post! It should be required reading for students of physics *and* students of philosophy.
Yeah, I know - but I can hope can't I?


reader Luboš Motl said...

Yes, I mean "qualia" although "qualia" is just another redundant incomprehensible word used by certain people to pretend that they are vastly smarter than they are.

By experience, we mean the experience of whoever or whatever who is using or checking the laws of physics.

The point of science, as clearly explained in this very blog post, isn't to say what physical effects are experiences and which are not. The point of physics is to predict or explain which of the experiences have occurred or will occur.

Wigner's friend has perceptions from his own viewpoint. From Wigner's viewpoint, the friend is just another collection of elementary particles that evolves according to laws of physics as described by a wave function measured by Wigner. Whether something is a perception is a subjective question. After all, perception itself is subjective.

reader Luboš Motl said...

Sorry, it's not a paper but a USENET posting by someone who starts by saying that "the theory of relativity is false". Several sentences are enough to see that the author of the rant is a compete idiot and I won't waste time reading the whole thing whether this idiot is employed as a professor of philosophy or the president of the United States of America.

reader Avraham rosenblum said...

Thank you for your time. I also thought that paper looked a little funny, but I wanted your opinion also.

reader Eclectikus said...

Brilliant once again. Lubos I'm not a theoretical physicist, but I loved it from my undergrad time, just I hadn't the courage (well, the inteligence) to go on. But every thing what you say fits perfectly with my background in the matter (the only A+ I got was in quantum mechanics and still keep the two volumes book of Cohen-Tannauddji).

Due that I never had real/research contact with this field (I devoted myself to mundane stuff as applied geophysics), I think if it were not for your writings (for its clarity, its conceptual depth and hilarity), I would be now hesitating on the basis of quantum mechanics. And I have no doubt because I am a regular reader of this blog.

So all I can say is many thanks for your work, and I join the club of readers of The Reference Frame who asks a book of yours (or a collection of your writings on this subject, Strings are way too much for me), because many of these writings, like this very one, are real jewels. Congrats.

reader physicsnut said...

In addition to the PingPong between Math and Physics, there is the question of how Stern-Gerlach could be done if Thompson hadn't discovered the electron. And how can one do experiments without vacuum pumps and all that. A lot of shoulders have people standing on them. Not to mention that Jordan was working with Courant - so one must not leave out where Courant was doing his work writing Mathematical Methods of Physics. What a time that must have been. And they did not even have TRS-80s, much less decent computers.

reader Gene Day said...

I am really impartial in the Apple/Microsoft wars, Lubos, but I may not be a typical user.
I spent almost $2k for the new Dell with Windows 8 and gave it to my wife, who soon gave it back to me. I am not going to sell it at a 50% loss; I assure you. It will get a fair trial with my MacBook Pro as the competition.
There certainly are things that I don’t like about MacOS; I am not viewing it through rose-colored glasses. My admiration for Apple is a business judgement only. If Microsoft takes Apple to the cleaners I will not cry.

reader Gene Day said...

I am not agains Windows 8; I’ve never used it but I think Windows 10 will be friendlier. I look forward to the upgrade.

reader Gene Day said...

I don’t think HoloLens will live up to the hype; they never do.

reader Guest said...

What I meant is that e.g. \ket{x=0 at t=0} is a shortcut for \ket{i've done this-and-that before t=0 and the value of something turned out to be y(0, this-and-that, something)}.
Not sure if it is more clear...

reader mesocyclone said...

Yes, Microsoft had some good ideas. C# is supposed to be good, and as you say Visual Studio has a good rep.

Apple had the advantage of starting on a BSD Unix platform. OS X and iOS are BSD Unix derivatives. Where Apple tends to screw up is in distributed systems. They have had some really bad iCloud disasters, and I just don't trust them in that realm.

I am also not a big fan of Linux, although I'm very glad that it is out there. It is probably the most widely used server OS today, and it is also in every Android device ever shipped. I just don't like the hassle of Linux systems admin - everything you want to do is a sys admin hassle, unless you are a long time, deep Linux geek, which I don't want to be.

BUT... I am a big fan of Unix and have used it, one way or another, for over 30 years. Don't bash bash! :-) Unix command line stuff is actually very elegant - a much better way of connecting programs together than the various kluges that Microsoft dreamed up, and it's too bad that Microsoft didn't use some of those ideas - it's not like they weren't floating around. If you want to script things, you can't beat Unix/Linux. But if you just want to just use a system, I think OS X wins.

reader Red Undancy said...

How about just call it the onstemic? We are part of this damn universe too.

reader Tony said...

Lubos, I'm a bit confused reading Pusey-Barrett-Rudolph paper. I mean, take two particles described by a Gaussian in x coordinate. One particle can have median in the nearby star, while another has it inside our Sun. Now, even then, Gaussians are non-zero and overlap in certain region. Thus particles don't have a X property, according to the article. Thus probability distribution is epistemic but, according to the article, that contradicts QM.

reader Tony said...

You are having an experience when reading this and that experience is about experiencing reading this, while I am having experience about writing this.

reader Quantum said...

OK, so by experience, you mean qualia. But you know, philosophers can't even agree on what qualia "are", or even if qualia "exist". So, if you're basing your philosophy upon qualia, you're on very shaky ground.

Whose experience? By experience, we mean the experience of whoever or whatever who is using or checking the laws of physics.

So, does a philosophical zombie using or checking the laws of physics have experiences, or are zombies not possible?

The point of science, as clearly explained in this very blog post, isn't
to say what physical effects are experiences and which are not. The
point of physics is to predict or explain which of the experiences have
occurred or will occur.

But how can one predict which experiences have occurred or will occur if one isn't even allowed to say whether something "is" or "isn't" an experience?

If you claim the "existence" of qualia is a pre-existing fact, what objective evidence do you have of this "fact"?

reader NikFromNYC said...

There seems to be a subtle reductionism here in both size scale and time scale. Atoms form molecules and extended solids with extended wave functions called molecular orbitals, that though strangely quantum in very rapid extreme detail, on the whole molecule scale over a timescale of seconds, a given drug or biomolecule indeed acts as a simple force affected and affecting real object. Additionally, there seems to be hierarchical entrainment of cause and effect in biology that can render strict reductionism impotent.

When I alter a drug molecule by a single atom addition, it can turn a nutrient into deadly poison, reproducibly, often predictably based on basic geometric and charge force reason. If something so crucial to health is something I can design rationally, based on consideration of a molecule as a real and rather classical or at least classically predictable behavior, then philosophically, at the level that it matters to drug design, molecules and thus their wave functions are perfectly real. I can now even feel molecular orbitals using virtual reality haptic feedback with a scanning probe microscope.

The caveat is that biologists are periodically invoking quantum effects to get around both astonishingly rapid protein folding and the curiously rapid associations between such macromolecules, things that no supercomputer can even begin to address since classical trial and error searching for low energy folded structures would take billions of years instead of nanoseconds, in the real world included. Another caveat is how a biological theory of mind hadn't been even remotely well formulated and there too quantum effects are often suggested.

It's one thing to ridicule charlatans and another to help both intuitive and logical chemists and biologists design better experiments with quantum effects in mind. In what ways and on what time scales and at what hierarchical level of order are molecules not real is the concern tinkerers have, not whether somebody is attacking the philosophy of physics.

reader Tony said...

I see you are having experiences too!

reader james said...

I have to admit that I don't quite understand what exactly he is asking about the reality of the wave function vs the reality of the particle. The wave function is a model of the behavior of the particle and, unless something spectacular has happened to the theory recently, it is an incomplete model. (But fixing up that incompleteness is what you're working on.) Incomplete, but valid wherever we've tested it so far, while the model of a particle as a little point has a more restricted domain of validity.

The particle is real "ontic"ally, the model is real "epistimic"ally. If tomorrow we had in hand a perfectly accurate model of the behavior of the particle, it would be real both "ontic"ally and "epistemic"ally, but not real in quite in the same way as the particle is. The physical universe would be like a perfect model plus the matter, the way chess is the rules of the game plus the pieces, if you will forgive a sloppy analogy. Unfortunately I don't know enough greek to come up with an impressive word to distinguish the two concepts. I could probably find something if I dug around in Aquinas for a while--the scholastics tried to be careful about distinctions.

reader NikFromNYC said...

An overly sober one, today.

reader Quantum said...

Modern neuroscience tells us almost all of what we claim to be experiences are in fact elaborate confabulations. So, a study of what we claim to be experiences would be heavy on the cognitive science of human brains.

If experiences are to be the bedrock of all of science, the act of "filtering out" cognitive perceptual "distortions" to describe what is "really" out there would be dubious because that presumes an objective reality out there undistorted by cognitive perceptions, which is precisely what positivism forbids us to claim.

reader John Archer said...

I've been working with computers since my late teens but I've been somewhat remiss about keeping up lately — but only for the last few decades.

Anyway, I was just wondering if anyone here would be kind enough to help me out by answering a few questions thereby saving me the bother of having to look anything up.

Do you know if Windows 10 HoloLens will run on my HP-41CX?

Or do you think I should just bite the bullet on the expense, lash out on an upgrade and get the full 32K BBC Micro?

Oh, I nearly forgot — does Windows come carbon credits? What are they anyway?


reader Luboš Motl said...

I've already said that I don't use - and competent physicists don't use - the word "qualia" because it's a redundant word used by intellectually inadequate individuals who just want to look smarter.

If they don't know what "qualia" - a word they use - means, it means that they don't know what they're talking about, right? The previous sentence is *literally* true. A person shouldn't use words he doesn't understand.

I don't suffer from this problem - and quantum mechanics doesn't suffer from this problem.

reader Luboš Motl said...

Experiences that are linked to various observables in complicated ways will be harder to be explained. The science explaining these experiences may be so hard that we may give up.

That's why science typically tries to discuss our experiences that are as clearcut as they get, that are as directly connected to the interesting properties of the studied "systems" as possible.

For those, the experiences may be almost directly connected with properties of the studied system. But one should never forget that it's just the experiences we have, and not the "direct" information about the external systems.

The "confabulations" in the brain that you mention play absolutely the same role as complicated - and perhaps dirty or broken? - lens in a telescope that we use to observe the sky.

If we look at the sky through pieces of broken glass, it will be harder to organize the patterns in what we see and related them e.g. to nuclear physics of stars. The same is true for "confabulated" experiences in the brain.

That's why we use as clean and unbroken telescopes and lens as we can, and the experiences that are the clearest possible among those we may have. Those are likely to have a more streamlined explanation in terms of science. Still,. experiences with confabulated experiences and broken telescopes exist, they are just not as useful in finding and checking the laws of Nature because they unnecessarily increase the complexity, and quantum mechanics can't be blamed for the fact that some glass is broken or some experiences are confabulated!

But it's always risky to assume that something specific exists (and some specific data or answers exist) "out of our mind". All such assumptions are just working hypotheses that may always be, and in most cases are, wrong.

The fact that quantum mechanics reduces all the tests and knowledge and predictions to sensory receptions in no way implies objective reality, it must be obvious that it doesn't, and your last paragraph claiming otherwise only proves that your brain is hopelessly confabulated and unusable.

reader Luboš Motl said...

The pieces are clear! ;-) The whole isn't.

So the question is whether "you've done this-and-that before t=0..." is real? It is a proposition. What does it mean for a proposition to be "real"? One claims that the proposition is true if he has knowledge about it. He has knowledge about it because of sensory experience (and/or plus memory).

But the positivist picture is that the only truly "irreducible" or "unquestionable" source of knowledge here is the sensory experience itself. The assumption that the experience reflects particular things in the external world - or that these things are "objective" - is unjustified in general.

I don't think it's possible to discuss these would-be philosophical matters rationally. One is constantly being asked about "is wave function real"? "Is some memory about the past real"? "Is a pure state real"? Etc. With some definition of "real", the each answer may be Yes, with another definition, it is No. At any rate, one is learning exactly zero about Nature by these meaningless exchanges. Science isn't meant to tell us about everything and every sequence of symbols or screams whether they are "real". Science is meant to give us explanations and predictions of observations - that we ultimately complete by sensory experience.

reader Luboš Motl said...

Dear Tony, I don't understand the overall point of these comments of yours.

They do seem to be obsessed by "eliminating some options completely" which is not the way how quantum mechanics describes (almost all) things: in QM, the probabilities of virtually every option are always nonzero, although they are often very small.

Probability distributions are by definition said to be "epistemic". But "epistemic" is also being used in a narrower sense, as probability distributions that are used in a model of classical statistical physics, and QM obviously doesn't have anything that is "epistemic" in this narrower sense because it's not classical.

I wrote most of these points above and they're linked to parts of your comment but I probably don't understand how you wanted to link all these pieces and facts in your comment - in some new way that is esoteric to me.

reader cynholt said...

"Brilliance is always about the ability to do something with the highest-hanging fruit."

Great line, Lubos! A most brilliant statement on the subject of brilliance. I'll have to file it away in my memory banks and pray that I've got enough brilliance to reach for it for future use.

BTW, what are your thoughts on Elon Musk's Hyperloop project?

Is this just more overblown virtual reality hype, or does it possess enough serious staying power to survive as a viable and safe mode of transportation?

reader Guest said...

I think we are talking at different levels of abstraction (and I agree that these kind of discussions are ultimately futile, but just for the sake of debate i'll try to clarify further :D).
In an hbar=0 world, a point of a phase space can be called "real" as it can consistently be put in 1-1 correspondence (by means of perceptions, of course) with something "out-there".
The question I was answering is "What's left of this concept in our hbar>0 world?"

reader Quantum said...

I think you're still missing my point. If science is based entirely upon experiences, but the concept of experience remains very fuzzy, then science has a rotten foundation indeed. And you just claimed what you call experiences is the same thing as what philosophers call qualia, even if you don't like that word. So, if philosophers don't know what they mean by qualia, then you don't know what you mean by experiences either.

Let me give you a specific example of my experiences. By your philosophy, I ought to claim that subjectively for me, you wrote a superposition of comments and posts on your blog, and your writings only "collapsed' to definite texts whenever I visit your blog. So, should I fail to read your blog for a week, you and your recent blog activities would remain in an indefinite superposition for a week. Is this a fair characterization of your philosophy?

reader Luboš Motl said...

Sorry, James, none of the things you write - except for the proposition that these things have been settled for decades - is right.

Quantum mechanics is a *complete* theory, in the sense that there exist no additional degrees of freedom (hidden variables) that are relevant for a physical system but overlooked by the quantum mechanical description; and there are also no additional patterns or relationships that are not included in the probabilistic predictions of quantum mechanics. What quantum mechanics says about a particular system is the most complete description that may in principle exist. This most complete description still implies that most questions have uncertain - probabilistic - answers, due to the uncertainty principle.

The universal postulates of quantum mechanics are not enough to describe a particular system. We also need a particular Hamiltonian (or other observables on the Hilbert space) and perhaps some information about the relevant initial conditions etc. But these additions cannot be considered signs of "incompleteness" of quantum mechanics because quantum mechanics, by the definition, is the collection of the general postulates, or the "template" for a physical theory. One wants to have more than a template but each particluar quantum mechanical model is precisely following the postulates of quantum mechanics, and the adequate model is complete.

Whether quantum mechanics was complete has been a topic of the debates starting from the Einstein-Bohr debates and Einstein, who claimed that QM was an incomplete theory, was simply wrong and the number of proofs that he was wrong only grew since the 1920s. There are no hidden variables in Nature and the probabilistic character of the predictions doesn't mean that quantum mechanics is a provisional theory that may be replaced by an inequivalent one in the future. It cannot.

Quantum mechanics also teaches us that a particle, or anything else, is not real ontically in the usual i.e. classical sense. On the contrary, the model *is* real, perhaps ontically if you wish, because the laws of physics work, are complete, and are the same for everyone. It's only the subjective knowledge that is inserted to the laws of physics - or that results from them - that is subjective.

reader Luboš Motl said...

Or epintic. ;-) It's great but I can't imagine what such a new word would be good for.

reader scooby said...

I know a few white hat security experts who are excellent "professionals". But beyond that I thought your comment about mesocyclone was gratuitous and insulting.

reader scooby said...

Well Powershell on Windows was certainly inspired by the Unix shell scripting environments. It is very powerful when combined with .NET or objects exposing a COM interface.

reader Luboš Motl said...

What's left of the phase space at nonzero hbar is the quantized phase space which may be effectively divided to cells of area/volume (2.pi.hbar)^N - of a fuzzy but arbitrary shape.

What's left from "x=0, p=0" is that this can't be true simultaneously. Instead, one may be "pretty certain" that a certain inequality holds both for "x" and "p" - as long as these rangles don't conflict with the uncertainty principle, the probability that we are in the "cell" may be very close to 100%.

In QM, one may also measure (and later "know" i.e. consider as well-known properties of the initial state) quantities separately, like J_z (t) = 0. But the uncertainty can only be zero for observables with discrete spectra, and this information can never pinpoint a point in the phase space because that would conflict the uncertainty principle.

At the end, I think that you are really asking what is left from the strict deterministic accurate description for all observables at nonzero hbar, and the answer is that nothing is left from that.

reader Claes Johnson said...

Here is a discussion of a new approach to give the wave function a physical meaning going back to Schrödinger's article IV from 1926.

reader Peter Risdon said...

Now that Microsoft has almost finished cloning UNIX, with the recommended 'Core' install for servers being without a pointless GUI, and Powershell bringing reasonable scripting, and UNIX-style pipes, to the party, they're shaping up to have a reasonable operating system. Their last important innovation as XMLHttpRequest, though, back in 1999. Hopefully we'll have another contribution before the 20 year mark is reached.

reader Luboš Motl said...

I understand your point perfectly but what you fail or refuse to understand is that your point is totally and fundamentally wrong.

Precisely separating "what is perception" and "what is not perception" doesn't belong to science, it is in no way needed for science, and obviously has no scientifically accurate answers, anyway.

Science is a tool of insight, rules, and equations that describe all possible (since 1925 probabilistic) patterns relating different perceptions.

The statement that experiences - observations that are communicated all the way to observer's awareness or knowledge - are irreducible concepts in quantum mechanics exactly *means* that your requests to "decompose" them to something more elementary - to define them in terms of something else - are just totally illegitimate. There is nothing more "elementary" than perceptions conveying the observations to the observer, so these perceptions can't be reduced to something else.

If one describes a different person totally using exact accurate laws of quantum mechanics, of course that the different person - just like any other physical system - evolves into a general superposition of states that includes macroscopically very different states (like "something has been written" or "not").

This is not my "philosophy". This is how quantum mechanics has always worked since the 1920s. It has always applied to all physical systems and physical systems' wave function evolves according to Schrödinger's equation - it spreads - up to the moment when the physical system is observed. Whether the observed physical system is called a "human" is totally irrelevant. Physics doesn't give any exceptions to physical systems who claim to be "human". Indeed, it would be very bad if such exceptions existed because the difference between humans and non-humans is a vague and undefinable pseudoscience.

reader cynholt said...

What is wrong with the no.9, and what's so great about the no.10! You'd think that 13th-century numerologists were in charge of Microsoft.

reader Luboš Motl said...

As Heisenberg wrote to Pauli in 1926,

The more I think about the physical portion of Schrödinger's theory, the more repulsive I find it...What Schrödinger writes about the visualizability of his theory 'is probably not quite right,' in other words it's crap.

What has changed since 1926 is that the crap had 89 extra years for its smell to flourish. I wouldn't like to smell you on the same street.

reader Luboš Motl said...

Hi Cynthia, thanks!

I love Hyperloop but I can't guarantee that it will be used and commercially successful.

There's small but nonzero chance that it will start to make airplanes obsolete in a few decades.

reader Luboš Motl said...

Dear Cynthia, I actually have compassion for that omission.

The digit "9" is typical in version numbers that are "beta" or preliminary or provisional.

It's like having VLC 0.9, or something like that, which indicates "before it's really completed, before the bugs were fixed and the interface polished" and so on. Microsoft probably decided that Windows 8 was already good for that description and the next one could already be a serious final release, like Windows 10 (a new 1.0 of a sort).

reader Claes Johnson said...

What says the Heisenberg is more clever than Schrödinger? After all, it was Schrödinger who invented quantum mechanics, not Dirac, nor Lubos.

reader Guest said...

That's why these philosophical discussions are futile: one always ends up into an infinite regression of definitions.
By "this concept" I was referring to the 1-1 correspondence between the formalism and something "out-there", not to the phase space...
I am fully aware of the implications of QM on realism; my point is that IF one whishes to attach the term "real" to something in the QM formalism (but that's merely a definition of the term), then that something is the ket "label" in the sense specified above: it is operationally well defined, and it represents the maximal information that may be possesed on the system.

reader Tony said...

Their definition is that, if pair of distributions are disjoint, then label L is uniquely fixed by lambda, distribution parameter, and we call it a physical property. This having a property would, I assume, characterize an ontic distribution.

But if distributions overlap, then corresponding quantum states can justifiably be regarded as "mere" information, (this is literal quote) thus this would be a case of an epistemic distribution.

In this latter case, they claim that then there is a contradiction with the predictions of QM.

The distributions they are talking about are not Schrodinger wave functions, but some functions of state u(x,p).

I then wonder why not consider QM wave packets like Gaussians, where the property in question is the position. It would then follow that QM would lead to contradictions just by using the description of its own invention.

That would be ridiculous, of course, so I guess that the secret is in how these hidden distributions get associated with the quantum state.

reader james said...

Don't exaggerate: I also correctly said that I didn't quite understand what he was asking. :-)

I am not a "hidden variables" partisan; what I meant was simpler. QM is, last time I looked, linear, while GR doesn't seem to be. Granted, it has been a long time since I studied GR, so perhaps my memory is faulty, but if not then I think it safe to say that QM isn't a complete theory yet. I seriously doubt that a complete theory would be less "weird" than the current theory. (I take "weird" to mean "non-intuitive" where the layman's intuition is based on human-scale sizes and masses and speeds.)

reader Peter Risdon said...

I think it's because some software looks at the Windows version, and it's possible that some legacy stuff didn't look beyond the first 9 of 95 and 98. So it's to prevent some issues stemming from that.

reader Quantum said...

Why the chip has external wires (why consciousness exists) and what
exactly qualifies for a wire doesn't belong to science anymore and it's
meaningless from a scientific - any rational - viewpoint. If
consciousness doesn't exist, the job of science to explain things is
limited to zero. It only makes sense to look for laws of Nature if some
sensory experience and consciousness with properties waiting to be
explained exists.

Let me get this straight. You believe that consciousness exists, but to discuss about it, is meaningless. However, the "reality" of the "external world" is a meaningless language game. Science is based upon qualia, but to ask who has qualia or what the qualia are about is meaningless, which seems to suggest science is about meaningless entities. I take it you're an idealist who doesn't believe in meaning.

reader W.A. Zajc said...

Lubos, I am sorry if the article made you see red, but the good outcomes from that unpleasant process are your very clear replies to both Mr. Siegfried and to the many comments you received.

It also stimulated me to wonder if part of this persistent confusion with QM is due to nomenclature, starting with “wave-function”, which carries with it the unfortunate analogy of a classical wave. It would be a challenge to teach an introductory QM course always saying “probability amplitude” instead of “wave-function", but that would be a start. Yes, there might still be an issue with “amplitude”, since it reminds students of wave amplitude like an electric field, where a stronger field has a stronger amplitude, while it makes no sense to speak of a “stronger" wave-function. But I would hope that “probability amplitude” would provide a strong (no pun intended) hint for those above some modest intelligence threshold that the complex square of this object must be normalized to one. Also, as you have noted many times in TRF, we should teach students that in QM “+” should be thought of as “OR”, not “AND”. Doing this, along with consistently reminding students that QM clearly is and must be non-classical, but it clearly is not and must never be non-local, would do a lot to immunize the next generation against the disease vector of QM revisionistas.

reader Tony said...

Let me try it in a different way. They assume some, yet unknown, distributions associated with discrete states in their argument.

I admit that I don't follow the argument and the resulting contradiction with QM yet.

However, following the same line of thought, eventually one must start considering continuous variables like position.

Thus two delta functions would be fine, both as QM wave function and their, yet unknown, distribution describing particle position. There is no overlapping support.

Two Gaussians would not be OK from their standpoint, because of the overlap.

Thus, to each QM Gaussian we should associate some other distribution, two of them with non-overlapping support, respectively, in order to get an objective description that wouldn't eventually lead to contradiction.

But that is only possible if two delta functions always remain delta functions and never overlap, otherwise even distributions with non-overlapping support may eventually end up overlapping, as the particles move closer, for example.

Thus the only way to satisfy their requirements is if the world is classical.

reader Arthur said...

As I am a new poster here I will test the waters with a brief comment and see what the response is.

The Copenhagen Interpretation of QM has two distinct processes. One is described by the wave equation and expresses in mathematical terms all that can be said about what happens when nature is not being directly observed, as in the double slit experiment where interference effects occur.

The second process is the process of measurement, also referred to as the collapse of the wave function to a single state.

The proponents of the Copenhagen Interpretation of QM were well aware of the logical and physical discontinuity of these two processes. Heisenberg and Bohr were the principal exponents of this viewpoint,as opposed to their contemporaries Einstein and Schroedinger.

As true positivists, they disliked any discussion of what Einstein might say is 'really happening' when we are not observing nature in the very small. Bohr went so far as to deny that an electron had an orbit within the atom. He would assert that the idea of a continuous path through time and space traced by an electron inside an atom was meaningless.

The succeeding generation of this school, purged of any metaphysical leanings, coined the expression 'Shut up and calculate'.

Bohm and Bell did calculate, but they did not shut up.

They asked for a physical explanation of such things as 'measurement' and 'collapse', and for an understanding of how the many different complex amplitudes of the wave function are related to the one final state of an event, or why that particular state resulted.

They have been answered by silence. For as the young Wittgenstein wrote, "Whereof one cannot speak, thereof one must be silent."

The danger to society of this extreme form of positivism was not foreseen. But within 25 years, the world learned the tragic consequences: Whereof one does not speak, thereof one will be silenced.

So let us not repeat the errors of the past. Let us speak and be spoken to; listen and listened to; whether we understand what Wittgenstein's Lion says or not.

reader Luboš Motl said...

Hi crackpot, I assure you that it was Heisenberg - and then Heisenberg, Born, and Jordan - who invented quantum mechanics.

Schrödinger wrote a paper about the wave equation one year later but he misunderstood the meaning of the wave function.

The founding fathers quickly completed the right understanding of the postulates and it got developed by the first-generation of founders of QM including Bohr, Heisenberg, Jordan, Born, Dirac, Pauli, and I assure you that Schrödinger wasn't among them.

reader Luboš Motl said...

Having seen an eigenstate of some observable allows one to say with 100% certainty that the observable has the corresponding value. So the value of this observable is "real".

But statements about all observables that don't commute with this observable - which is virtually all observable - are uncertain - or "unreal".

You may try to redefine the word "real" so that it allows things to be uncertain etc., but I don't know what this version of the word "real" has to do with the previous one and what it is good for to use the (already previously worthless) word real with an even more distorted and worthless meaning.

reader Pavel Bažant said...

Powershell pipes are object streams, not byte streams, making them less UNIXy and much less fragile. .NET was released in 2002 and it is a significant innovation. A GUI is no more pointless than a CLI. It is useful to have both.

reader Tony said...

Nah, Gene is right. Nothing in mesocyclone's statements, the topics that he picks, the way he talks about them, indicates a professional.

He is at best semi-technical user. I interviewed hundreds for the software developer job and many are much more eloquent and know even few technical details when they mention specific technology or language.

More than half of them end up refused or moved to test jobs.

Semi-technical people are the worst. With non tecnhical ones you at least know they don't really have a technical education, and they admit it.

Semi-technical, they know everything about the technology when it comes to talking and making sweeping statements about things that they barely understand, but again understand enough to be believable to those who know even less.

reader scooby said...

Sweeping statements.. you mean like those you made about Apple users and Linux geeks?

Myself I prefer to stay out of discussions on favorite platforms/OS etc... They always tend to sound like arguments between religious fanatics.

reader Tony said...

I made a sweeping statement that it was Microsoft behind IT revolution. Mesocyclone made a sweeping statement that Apple has superior tehnology.

One of these statements is correct, another is wrong.

While I agree with you on these discussions often turning quasi-religious, well, now and then one has to say what one has experienced while practicing his profession.

reader Tony said...

What about Linq, what about Rx, introduction of functional programming, WPF, task pools and asynchronous programming?

Where is Apple or even Google here?

reader DonGateley said...

The disk drive at IBM. Many things at IBM.

reader QsaTheory said...

I have reported on this company many months ago(TRF), they have an actual working product you can buy.

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');