Friday, April 05, 2019 ... Deutsch/Español/Related posts from blogosphere

"Search for holography in your kitchen" instead of the FCC is the return to alchemy

Anna has linked to a WUWT story about a $50 million fine that a fake journal has to pay. Much like Theranos, such fake open-access journals deceive their users about all the normal ingredients that are responsible for the quality control – about the identity or the very existence of referees, the existence of the review process, they heavily overstate the impact of the journal, and co-organize fake conferences (I really mean conferences whose scientific quality is non-existent but someone pretends it exists).

By the way, how many of you are getting daily "calls for abstracts" from some strange conferences that don't seem to be related to your interests?

Many armchair scientists who were ignored may suddenly find someone who wants to publish their texts, so they pay for the publication. Ambitious new "scientists" who can't publish, and therefore expect to perish, may suddenly survive. Some of them may even become "big leaders" after a few publications that appear in fishy outlets. At some level, people are happy – they get what they want. These "scientists" finally publish their stuff and the publishers get paid. The price is high – the whole ecosystem is being flooded with mostly wrong results and claims that pretend to be verified by someone who is careful but they are not. Readers get something else than they're told to get. Scientists waste time with bad papers – the wasted time is maximized in the ambitious yet truly marginal cases of papers that "almost" look like serious ones but ultimately turn out to be wrong for somewhat subtle reasons that would still be caught by a proper reviewer.

To some extent, this decrease of quality is an unavoidable consequence of the "open-access approach". While the "open-access" ideologues like to hide it, the "open access" – just like "open borders" – often reduces to nothing else than "the absence of a reliable enough quality (or security) control".



The courthouse problems for a fake journal may look like good news but at the end, this victory might be an exception. Solving intrinsically scientific problems through courts is an extremely bad idea. Ask Galileo Galilei who was right about virtually everything – but did it guarantee a smooth sailing in the court? Not much has improved about the reliability of the courts in such matters – an easy yet reliable fix doesn't really exist. In some cases, the controversy may be framed as one about fraud that is obvious even according to the judges (e.g. creation of a new identity of a reviewer – a trick that has nothing to do with the science itself).

But in most cases in science, it just isn't the case. In most disagreements between good science and fake science or bad science, some at least elementary scientific judgement plays a role. It's needed to verify the argumentation that someone is cheating etc. And the judges are rather likely not to have the proper scientific judgement. They haven't been trained for it. They can't even reliably distinguish a competent witness from an incompetent one – one needs quite some expertise for that, too. So it's just a bad direction for lawyers to become important players in influencing the direction of science.



Now I want to discuss something slightly different – and, from some viewpoint, a more particular technical topic in physics.

Recently, you could hear new things about the Craig Hogan's planned holographic experiments. If you click, you will see the most specialized blog post that I have dedicated to his holographic noise so far – and that blog post is already 7 years old now.

You know, holography "exists" in quantum gravity and it is exciting but this statement is a slogan. One must be extremely careful what the "holography in quantum gravity" actually means. Some detailed statements of this sort are right (some careful statements about two types of correlation functions in AdS/CFT, for example), others are wrong. Although laymen often "think" differently (or "fail to think"), it is not true that every sentence that superficially sounds similar to "holography exists" must be correct. You're always at a huge risk of overgeneralizing. "Holography" isn't some omnipotent divine power that guarantees the "truth" to everyone who comes close to It. Almost nothing in science is this "religiously omnipresent", a very general and important point that superficial non-experts often fail to get.

Hogan believes in some kind of "holographic noise" – he believes it is a consequence of the real holography in quantum gravity – and he wants to see it through some interference experiment. Well, that noise just doesn't exist. Much more generally, I believe – although I can't quite rigorously prove it – that all new observable effects implied by the gravitational holography

only exist if the local effective theory would want to squeeze too much actual information into a given region, the information exceeding \(S=A/4G\).
In other words, the "holographic entropy bound" – you can't compress the information more than what the black holes do – is the only constraint that may modify the results of the experiments. But the information that may be compressed to black holes – even tiny ones, comparable to the "size" of elementary particles as observed at the LHC – is just so huge that there's no practical way how it could become insufficient to store the actual data about the observed physical systems.

So none of the "new holographic physics" is unavoidably observable in doable experiments. It may still be true that some new effects are observable "accidentally". But to know that these effects are predicted and what they are, you would need quite a detailed theory of quantum gravity. According to the best theories and detailed sketches we have, the holographic quantum gravity theory is emulating the local effective field theory approximation extremely accurately in practice. So it's almost guaranteed that you can't make any feasible experiments that could see the difference.

I think that the Papadodimas-Raju construction of the operators in the black hole interior shows this point most clearly. In some sense, their approach is "reverse" to the perspective of others, and they say:
We are given some limited space for information, \(S=A/4G\) nats, and we want to use the Hilbert space of this dimension and embed some approximate field operators in it so that the algebra and correlation functions between these field operators will emulate those from the unrestricted (non-holographic) Hilbert spaces as accurately as possible.
Papadodimas and Raju have made many steps to define these operators as engineers – they are solving an engineering problem of a sort. Or a homework exercise in mathematics. But the idea is that Nature is basically doing the same thing. Nature – a theory of quantum gravity that still allows quasi-local perceptions – is just maximally using the limited holographic space for information and tries to squeeze a good model of local physics into it.

And both Papadodimas+Raju and Nature can get very far to fulfill this task – much like modern PC games get very close to emulating real-world environments, despite the limited memory of the PCs. The result is that the expectation values of the operators and their products or commutators agree with the exact local, non-holographic results – up to the point when you include too many factors in the product. That's where, when, and how the holographic prediction starts to deviate from the local physics.

It means that if you measure some mean values of some simple local observables, you won't see a difference from locality ("locality" is meant to be the opposite of "holography" here – even the ordinary holography involving plastic films "delocalizes" an object to features of many strips in the interference pattern, right?). If you measure some correlation coefficient between two or three or four... field operators in some way, you still won't see a difference. Around the black hole, you will measure results that agree with the thermality of the Hawking radiation. You will need to study the correlation between a large number, \(N\), of field operators (or their integrals etc.), where \(N\) scales with the number of nats in the system, i.e. with \(S=A/4G\). Well, it scales with the black hole entropy. Again, because the black hole entropy is huge, you would need to study the correlation between millions or zillions of vigintillions or gadzillions of field operators, and no one can do it in practice.

Everyone who is at a similar frequency as Papadodimas+Raju, and this includes almost all serious quantum gravity theorists although the differences between them are still substantial, would agree with the expectation that a simple interferometry experiment can't see the gravitational holography. After all, if it were possible, the effect would have probably been seen before. By Michelson and Morley or anyone else.

At some level, these people assume that the measurable "deviations of a holographic theory from the local field theory's predictions" are as small as mathematically possible. There is no guarantee it's true, of course. But there are three big reasons to take this expectation seriously: One of them is that the local physics of quantum field theory is verified by an extremely huge number of experiments (everything we have seen so far, in fact), so there's quite reason to take the local QFTs seriously. Second, it seems hard to deform them while keeping their consistency and symmetry that look to be exact. Third, research in quantum gravity has repeatedly shown that it is often marginally possible to preserve the expected "maximal" accuracy of the field theories while obeying some quantum gravity principles and bounds, too. In many papers, you may see that "exactly when a contradiction could start to emerge, some effect kicks in and makes it impossible to detect the contradiction", and stuff like that. So the picture that the quantum gravity principles are right and the violations from locality are "minimal" seems to be a similarly robust theory as the uncertainty principle that is also robust against efforts to show that it's inconsistent.

The new holographic or nonlocal effects are only seen in unexpected correlations between huge numbers of degrees of freedom. This conclusion is morally equivalent to the explanations how the Hawking radiation carries the information away from the black hole. It is also stored in some correlations between the properties of a huge number of the emitted Hawking particles. If you could look at these incredibly complex correlations, you will see deviations from the thermality (predictions of a mixed ensemble). You will never be able to measure them in practice – for a sufficiently large black hole.

There's no good theoretical model or a sketch of a model/scenario that would make predictions just like Hogan. Hogan's derivations are heuristic and, according to "real" quantum gravity theorists (and/or Matrix theorists, for example, because he's also used Matrix theory to argue), these heuristic derivations are just not right. None of the real quantum gravity theorists has really adapted their theories significantly along the Hogan lines because his picture, while using some of the familiar building blocks, doesn't make sense as a whole to them.

But even if all these theorists were wrong and Hogan were right, and he would actually measure some "new interference caused by holography", it would still be unclear how to describe it because Hogan's description, while optimized to rationalize an experimental result he wants to get, just doesn't seem to work as a description of many other, more ordinary things in gravity – like the quantum realization of Newton's law or the gravitational waves.

In principle, it may always happen that some random heuristic theory wanting to get some particular new effect will be proven right. Theorists will be forced to accept it and construct the missing pieces (and a lot will be missing). But I think that sensible people understand that science just cannot go in this direction "just try any minor specialized weird theory that someone made up". If science went in this "adventurous" direction, it would cease to be science. It would become a revival of alchemy.

Hogan's heuristic explanations why he "should" see some holographic noise in his experiments are analogous to alchemy. The people who were creating gold from copper or the elixir of youth have also had some "rationalizations", some stories they were ready to present to Rudolph II the Habsburg who generously funded them in their Prague research centers. Our emperor was impressed enough. Given his excitement for mysterious arts and pseudosciences, it was probably easy enough for these people to impress their generous donor. And despite some "imperial" failures of Rudolph II, I imagine this era as a very happy and promising one, perhaps analogous to the 1990s. ;-) But was there a real reason to think that by adding some ingredients from mice and snakes into copper, the manipulation with the hot copper will turn it to gold?

And you know, Hogan's theory absolutely is analogous to these alchemists' "theories". It recycles some buzzwords – many of which are actually important in science that was recently found and that does work – but it adds lots of logic and ambitious assumptions and plans that are arbitrary and don't agree with the real science that is being done by real scientists. So the gold doesn't emerge from copper+mice+snake, and the holographic noise won't emerge from Hogan's experiments.

Finally, I want to combine the topics about the fake journals and Hogan's experiments.

The reason why I wrote about both is that Hogan's experiments are the kind of stuff that you get if suppress controllable huge experiments, like the FCC/CEPC, and replace them by lots of "sub-billion" experiments. These sub-billion experiments will unavoidably resemble Hogan's experiment – and they will be the experimental analogies of the fake journals. What they share is the insufficient quality control, the overlooking of serious counter-arguments, and the excessive importance played by the self-promotion, especially in front of the listeners who don't really understand almost anything about the topic.

There is no reason why the allocation of funds in experimental physics should be "either this big thing OR small things". Healthy physics has always built experiments of many types and sizes simultaneously and it will continue to do so, unless we're really entering into new dark ages in which the scientific research is being intentionally stopped and phased out.

But if you chose the answer that "it's OK for science, we just stop the large ones and replace them with small ones", you will end up with a landscape of Hogan-like experiments that really don't make any sense to the best experts in the most relevant subfields. For this reason, the actual knowledge accumulated in the scientific research so far – which is imprinted as the best experts' expertise – will play no role and it will be totally accurate to describe almost all the experimental industry as Alchemy v2.0 – an unrestricted search for elixirs and gold according to totally random or arbitrary theories based on some wishful thinking – a collection of efforts that always ignore everything that other people have found before.

You wouldn't get far with this Alchemy v2.0. The only truly lasting value of alchemy (the first edition) was that it was a human activity that had the potential to gradually turn into proper chemistry. It's very sad how many people seem to accept the philosophical explanations whose real purpose is to return science (not just chemistry) back to the era of alchemy – a method that isn't really scientific because it ignores the conclusions extracted from the previous experiments. Some people apparently love this "search for great discoveries" that always ignores everything that science has previously learned and that starts from scratch. But it's alchemy, not science.

So sorry. The fact that the FCC may see new things according to experts, while Hogan's experiments probably can't, is a difference that definitely should matter when some agencies are allocating funds for experiments. To ignore these expectations of experts means to give up the scientific method. The mankind may afford to fund a bunch of alchemists but to promote them into the new majority culture – that replaces the science where the accumulated knowledge matters – would be incredibly destructive.

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

snail feedback (0) :