Monday, October 10, 2016 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

A cute but flawed PBS video on QM and realism

I just became aware of a YouTube channel named PBS Spacetime that is probably run by the public TV station. A guy named Matt is talking about various topics close to fundamental physics. Almost every video gets hundreds of thousands of views which is impressive.



When watching the first one in my life, a two-week-old video on the quantum entanglement, local realism, and the Einstein-Bohr debates, I was amazed by the visual quality of the computer animations and some creative memes.

For example, at the very beginning, we were told that babies are great quantum mechanics because they're excited that a person disappears when they cover they eyes and reappears when the cover is removed. They realize that by observing or not observing something, the object or its aspects may be made disappeared. That's a cute idea but babies seem to overstate how much this "quantum disappearance by covered eyes" applies in the ordinary classical limit.

And I want the babies to know the commutators of observables, anyway, before I admit that they're good at quantum mechanics. :-)




OK, up to the middle of the video or so, I also had the feeling that it has avoided unquestionably wrong propositions, especially those that are omnipresent in this discussion.




But in the final portions of his monologue, Matt has said almost all the wrong things that are being said by most popularizers of these issues, too. He endorsed non-locality as a possibility and even though he had previously said, correctly, that Bell has proved that locality and realism can't be simultaneously true, he claimed that locality and realism may be simultaneously true, after all – if you adopt the "many worlds".

I am sorry but if you are careful, define your notions carefully, and evaluate the truth values of propositions carefully, then the "impossibility of locality and realism at the same moment" is proven as rigorously and unambiguously as \(2+2=4\). So whatever name you choose for the denial of this proven fact – whether it's "wakalixes" or "many world interpretation", this thing is demonstrably full of šit. The demonstration is really at most a 2-page-long elementary proof.

Again, realism is wrong, locality is right in Nature. It's so simple yet almost every layman – and almost every popular writer addressing the laymen – gets it incorrectly.

The otherwise good video actually started to spit incorrect statements earlier than that. We have "learned" that relativity doesn't require locality, it only requires causality. Rubbish. Relativity definitely does require locality because the relativistic causality and locality are exactly the same thing. Causality is the principle or law demanding that causes precede their effects.\[

\forall {\rm cause}, {\rm effect}:\quad t_{\rm cause} \lt t_{\rm effect}

\]Within relativity, physical laws must apply in all inertial frames. Relativity is a meta-theory, as Einstein was describing it, that constrains the properties of all (not meta) theories in this way: they have to apply equally in all frames.

So the relativistic causality says that the causes precede their effects according to all inertial observers. But if they do, you can easily see that it means that the cause and the effect are (null or) timelike-separated (and correctly ordered in time). It has to be so because the ordering of two spacelike-separated events depends on the chosen frame of reference. This spacelike possibility is therefore forbidden by relativity: the cause and the effect can't be spacelike-separated. Equivalently, the influence propagates at most by the speed of light. In other words, locality holds.

The general public has been so infected with the wrong statements about nonlocality and the echo chamber in which the laymen parrot all these statements is so isolated from any sensible physics research that we're guaranteed that Feynman's rule that "nobody understands quantum mechanics" will remain true for a very long time.

This video has gotten over 250,000 views. That's modest relatively to a two-month-old video that claims that the delayed choice quantum eraser rewrites the past. Sorry, it doesn't rewrite the past and nothing in the world can rewrite the past. The delayed choice quantum eraser, like every sufficiently complex or generic quantum experiment, only makes some interpretations of the past easier to talk about.

In this experiment, the shape of an interference pattern \(I(x)\) has some properties that depend on a "discrete piece of information" \(B\) (the letter stands for a "bit") that may be measured after the particle contributes to the interference pattern \(I(x)\). But that doesn't mean that the later bit \(B\) has affected the earlier interference pattern \(I(x)\). These two were only correlated with each other. But correlation doesn't imply causation, especially not in the direction that someone could find cooler.

The most careful description says that \(B\) and \(I(x)\) are correlated because of some events (which are not mysterious at all in this experiment) that took place before the measurement of both \(B\) and \(I(x)\), at a moment when the two later entangled subsystems were in contact. But even if you need to imagine that the "measured correlation" has to arise from some "current influence", which it doesn't, the more correct description would still be that the earlier measurement \(I(x)\) affected the odds for the later measured bit \(B\).

This causal influence is the kind of an influence that many people dislike because they imagine that a discrete choice may affect a continuous curve but not the other way around except that this proposition written in italics is completely wrong once again. It's totally OK for continuous observations – e.g. of the position of a particle on the photographic plate where it landed – to affect the odds for different outcomes of a later measurement of a discrete observable. There is no metaphysical difference between discrete and continuous observables. In particular, both may be observed earlier or later, both may be causes or effects.

If you need to imagine a "mechanism of a simulation" (even though it's not right to imagine that this is how Nature works), the earlier measurement produces an outcome \(x\) according to the "integrated" probability distribution \(\rho(x)\) which is summed (well, traced) over all possible results of the other, independent observables.\[

\rho(x) = \sum_{i} \rho(x,B_i)

\]Once \(x\) is measured, i.e. the particle lands somewhere, the wave function for the two subsystems collapses to the portion that has the right \(x\). This also means that you will get \(x\)-dependent predictions (conditional probabilities) \(P(B_i) = \rho(x,B_i) / \rho(x)\) for the possible outcomes of the bit \(B\) that is measured later. A point is that this bit \(B\) is also random and the experimenter can't "dictate it by her free will" so the assumption that the value of this bit must be a cause, and cannot be an effect, of this correlation is just plain wrong.

But blog posts like mine are probably a pure waste of time. For every 2,000 people who read my blog post about any related quantum mechanical issue, there are 250,000 (the newer video) or 570,000 (the eraser video) viewers of some wrong take on these topics. Of course, the audience I have or had in mind was more special – it should be 2,000 people who should know better than an average layman – some popularizers close enough to physics. And those should ideally make their videos for PBS etc. that are correct.

The only problem with this optimistic plan of mine is that there are just no popularizers who would be at a higher level than the 570,000 laymen. You either understand quantum mechanics or you don't. There's almost nothing in between. Like virtually everyone writing popular texts and shooting popular videos on quantum mechanics these days, Matt of PBS Spacetime doesn't understand quantum mechanics. After some time, I decided that George Musser just doesn't understand it, either, and I could give you a list of 50 other popularizers who don't understand quantum mechanics although they write about it. So a TRF blog post about quantum mechanics is affecting 2,000 people of the "same kind" as the viewers of the PBS Spacetime videos (apologies to the several exceptions who know very well to be exceptions) and the result is guaranteed by a simple comparison of the numbers.

I lose because the human stupidity, delusions, and bombshell claims about impossible miracles are more aggressive and attractive for the laymen than the modest scientific truth and theories that actually work. The mankind's march towards idiocracy will be increasingly full of non-localities, retrocausality, and other idiocies before people stop talking about physics altogether.

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

snail feedback (0) :