Wednesday, June 15, 2016 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Research into totally settled questions is meaningless

Yesterday, I mentioned the obsession of Marxist "critics of science" with the politicization of everything in science and with assorted sociological conspiracy theories that have nothing to do with proper scientific research and that no serious scientists spend any substantial amount of time with.

But that doesn't mean that the likes of Ms Hossenfelder understand some other basic principles of the scientific method. Another totally elementary issue she clearly misunderstands is the following:

Scientific research may only be valuable if it investigates questions that are not completely settled. A complete certainty about the answers makes any research meaningless and worthless.
Climate skeptics often like to point out that some climate alarmists love to say that "everything is settled" in their field but they still want to be getting funds for the research. The criticism of this inconsistency isn't just some malicious harassment. It's absolutely fundamental that in science, you can't have both. Either you're certain about something, or it makes sense to do further research on it.

For example, people are reasonably certain that the Earth is round ;-) so not too many people are receiving scientific grants to cover their journey around the world (although, I am sure, some of them do have sponsors like that – but I would simply never call it science in 2016). If and when we know what the outcome, why would someone pay for such a journey as if it were science? The knowledge won't evolve at all but some evolution of knowledge is really the point of research.




I forgot whether it was Joe Polchinski or David Gross who has geometerized the knowledge or ignorance. There's some volume \(P\) of possible insights. We already know something so it's some initially small ball \(K\) (knowledge) inside \(P\) i.e. \(K\subset P\). But we want to know more so \(K\) is expanding. The real dynamics is happening on the boundary \(\partial K\). That's where the scientists are digging. It makes no sense to dig at the places where all the soil or rock has already been removed; and it is hopeless to try to dig very far from the boundary because we can't get there.

This metaphor ended with some interesting speculations what happens when \(K\) approaches \(P\), i.e. when we know a lot or almost everything. The complement \(\bar K\) may start to shrink. The area of \(\partial K\) may be called the "useful ignorance". At the beginning of science, the area of \(\partial K\) was getting bigger. The more we know, the more (interesting realistic questions) we don't know. ;-) There are always questions that are so far from our knowledge that we don't even realize that we don't know the answers. We don't even know the right questions. But at the end, the growth of \(S(\partial K)\) may get reversed. The number of interesting questions may really drop when \(K\to P\) and \(P\) is compact.




At any rate, what the likes of Ms Hossenfelder just don't get is that while the detailed character of the knowledge is always different, the basic dynamics of the growing knowledge is always the same. When it comes to the degree of a confirmation or certainty, string theory is totally analogous to any previous advance in science. In the discussion, a particular somewhat more modest one – the Higgs mechanism – was picked as the analogy.

You may want to selectively read the exchanges between Jonathan Tooker and Ms Hossenfelder.

First, Tooker was surprised that Hossenfelder was dissatisfied with the fact that the aesthetic value or positive ratings from other physicists matter in physics. Obviously, they always mattered and they need to matter for the fate of an idea. Tooker wrote:
SH: the first criterion they’ll take into account is aesthetic value, and the second is popularity with their colleagues.

JT: Why would you say Higgs' theory was found to be "successful" if not only because the math had quite a pleasing look to it when printed on the page and that the result was very popular with his colleagues?
Obviously, the theory involving the Higgs mechanism – the spontaneous symmetry breaking – was successful to the extent that people decided that Peter Higgs had contributed to physics. It was sensible to give him a job. And, as Tooker reminds us, Higgs has received a dozen of nontrivial awards before the Higgs boson was discovered.

Not only that. Theorists were vigorously building on the insights of Higgs (and other researchers of the Higgs mechanism in the 1960s). In particular, Steven Weinberg invented a more specific version of the Higgs mechanism and reconciled it with the \(SU(2)\times U(1)\) electroweak theory involving quarks and leptons, too. The "Higgs mechanism part" of the Standard Model was the main contribution of Weinberg's to the Standard Model – that was really the discovery for which he shared the 1979 Nobel prize in physics.

I want to emphasize that theorists have built upon (and had to build upon) the ideas of Higgs (and others) more than 40 years before the main new prediction of the Higgs mechanism, the Higgs boson, was confirmed at the LHC (in 2012). It would just be absolutely ludicrous to claim that Peter Higgs hadn't contributed anything to science up to 2012.

The Higgs mechanism is a pretty idea. One can formulate it in pretty ways. But one may also formulate more technical arguments why "some Higgs boson had to exist". You may basically add the particles and interactions one by one, using the experimental data and consistency (unitarity). You need to add the virtual W-bosons to guarantee the beta-decay (i.e. four-fermion interaction) from virtual W-bosons. At the end, you may study the longitudinal scattering of two W-bosons (WW-scattering) and realize that without some new virtual particle they may exchange, the unitarity is violated i.e. the predicted probability of the scattering exceeds 100% at energies below a TeV. What you need to add is a scalar boson with the interactions to the W-bosons that can also be derived from the spontaneously broken gauge theory.

Needless to say, the situation of string theory is absolutely analogous. The graviton scattering also leads to an inconsistency due to a fast growth at high energies, new virtual particles have to be added, and collections of excitations that follow from one vacuum of string/M-theory or another seem to be the only ways to cure the problem.

It would have been ludicrous to say that "Peter Higgs had achieved nothing in science" before 2012. It would have been absolutely insane not to take his paper into account when it came to the decision whether he should be hired – and especially whether his followers should be hired. It just doesn't matter that the new particle hadn't been observed. This uncertainty was a reason why the research of those issues was meaningful. People weren't absolutely certain about the theoretical arguments that imply that the Higgs boson had to exist. But if they had been certain, the research would have been meaningless, as I reminded you, so this uncertainty is a supportive argument for the research. What's important is that the theory seems consistent with the available evidence and more convincing than proposed alternatives.

If you decided to celebrate Peter Higgs only in 2012, it would be very stupid, indeed. In fact, as a textbook example of a one-hit wonder, Peter Higgs only wrote some 13 papers or semi-papers after 1966 and 12 of them were completely unimportant (a paper from 1979 was the only exception). So if you waited for the experimental discovery of the Higgs, you would literally be delayed by 48 years and you would be hiring hasbeens.

Also, you would probably not hire people like Weinberg who were working on followups of the work by Peter Higgs – an experimentally unverified paper.

I think that only a complete imbecile like the Marxist "critics of science" could have doubts about the fact that a scientist (and also his boss or potential sponsor) must make many of these decisions well before the final experimental proof becomes available. When it's available, it's just too late. You surely don't contribute anything to science if you observe that the Higgs boson exists today. This simple sentence was a vanishing contribution to science already in the 1960s.

OK, let's return to Ms Hossenfelder's troubled brain. What's the answer to Tooker's question why Higgs was successful before 2012?
SH: Jonathan, Higgs' idea was found to be successful because... they found the Higgs-boson. I don't think that the idea at the time was thought of as particularly beautiful.
Wow, so this Marxist bitch basically tells us that she would have made sure that Higgs and his papers weren't successful before 2012. Can't she realize how absolutely hopelessly idiotic assertion hers is? To make sure that she would have killed and banned the Higgs research already in the 1960s, she informs us that she doesn't find the mechanism beautiful, anyway. Haven't you considered the possibility that it's because your aesthetic sense is completely f*cked-up, Ms Hossenfelder?

The Higgs mechanism – the spontaneous symmetry breaking – is beautiful because it shows that some seemingly arbitrary and asymmetric physical phenomena are actually resulting from symmetric, more beautiful, equations because of the general and largely unavoidable tendency of unstable systems to "randomly pick directions".

At any rate, isn't this example enough to show that the approach she promotes would have been absolutely devastating in the context of the most recent big discovery in experimental particle physics? And I can give you dozens of other examples. Something like roughly 50 percent of scientific discoveries were first made by theorists. Sometimes they were ahead by many years or decades (or perhaps millenniums if we generously add Democritus' predictions of atoms) – and this fact didn't make them less scientific in general. It made them deep and visionaries instead! It just isn't a heresy and can't be a heresy for a scientist to be ahead of his time. Ambitious scientists naturally want to be as much ahead of their time as possible or speed up the progress by a maximum amount. That's what makes others wow.

Tooker realizes very well that Hossenfelder's answer was insane so he refined his question to make it a bit more suggestive. Maybe Ms Hossenfelder will be able to fix her "minor mistake":
JT: I guess another way to ask my question is why did Higgs win all these prizes if not only because his theory was nice to look at and his colleagues liked it?

• Hughes Medal, Royal Society (with T W B Kibble) 1981
(and other 9 prizes from 1984 to 2010 are listed)
Great. Tooker succeeded and Hossenfelder was capable of adjusting her previous answer:
Because it solved a problem.

Anyway, you're not making sense focusing on a theory that was popular, was tested, and turned out to be correct. You should be asking how many theories are there which were (are) popular and turned out to be not correct (think: geocentrism, mechanism, steady state, vortex theory, etc), and how many theories are there which are not popular and their lack of popularity might have the consequence we'll never find out whether they are correct.
Right. The Higgs mechanism was an important insight because it solved a problem. That's exactly why string theory is so critically important for all the competent people who study quantum gravity.

And yes. Tooker mentioned a historical example where the answer is already known but when we have to make decisions, e.g. whether to work on string theory or its part, we are never certain. But as I said, this uncertainty is absolutely unavoidable whenever the research makes any sense.

Her comparison of string theory to geocentrism and other examples is absolutely backwards. Like the Higgs mechanism, string theory is a new set of equations and rules that weren't known a long time ago. String theory, like the Higgs mechanism, is some nontrivial addition that some people added or a modification that some physicists did. Geocentrism is nothing of the sort. Geocentrism is the default opinion that was held by the humans since the first moment when they thought about such questions – so geocentrism is obviously analogous to the opinion that strings are not needed.

I don't want to discuss her Marxist and completely misguided obsession with the sociology all the time because this staggering garbage appears virtually in every sentence. But let me assure you once again that not a single principle I mentioned has anything to do with any communities. Even every individual scientist knows that it's meaningless to investigate questions whose answers have been absolutely settled (according to this scientist and the evidence available to him). Even individual scientists know that theories have to build on each other. Even individual scientists know that the construction of a theory requires many steps and it's sometimes absolutely needed to make them simultaneously instead of being only allowed one step at a time. And so on.

Ms Hossenfelder has absolutely no idea what the scientific method is.

Tooker kept on realizing that what she was writing made no sense. So he added the following comment:
JT: What is the difference between the case where one solves a physics problem only using math tools and the case where on develops a theory that can be considered successful even before it is tested? I don't get it. It seems like if your theory was good enough that someone might want to test it, that means it was a successful theory.
Right. The Higgs theory solved a problem and one understands why there is a problem and why there is a solution only with some mathematical arguments. The same is true for string theory. But it can't be forbidden to use mathematical arguments. So even when a mathematical argument or a calculation is needed, it's still true that a scientist may identify successful theories and solutions that have good reasons to be correct which is why it's a good idea to study them! It's really common sense and this common sense was important in most advances in physics of the recent century or a few centuries.

Needless to say, when a comment on her blog makes sense, Ms Hossenfelder goes ballistic:
SH: Jonathan, you write

JT: "It seems like if your theory was good enough that someone might want to test it, that means it was a successful theory."

SH: I sincerely hope you are not a scientist. Because this sentence of yours sums up pretty much all that's going wrong in science right now. The purpose of science is to describe nature.
Wow, so this piece of Marxist pseudoscience "hopes" that Mr Tooker – whose comments are absolutely sensible and would be endorsed by every genuine scientist – isn't a "scientist" because Ms Hossenfelder believes that the scientific method "sums up pretty much all that's going wrong in science right now". But the scientific method – which Mr Tooker described – summarized what was going right and wrong with science at every moment in its history, too.

It is incredibly offensive that such a nasty idiotic bitch is allowed to scream at the most sensible readers of her blog in this way. Mr Tooker has about 50 times greater credentials to be a scientist than she has.

She says that instead of the sensible reasoning that Mr Tooker, only theories that "correctly describe Nature" are OK. Except that the words "correctly" and "describe" are vague and the evaluation whether the proposition holds for a particular theory requires one to do some nontrivial work – work of the kind understood by Mr Tooker but not by Ms Hossenfelder – to figure out whether it's true or not.

Even before the 2012 discovery of the Higgs boson, physicists were convinced that the Higgs mechanism correctly describes Nature. And analogously, quantum gravity experts now are convinced that string/M-theory correctly describes the physics of quantum gravity and the unification of all forces. So what's your problem, bitch?

At the end, her focus on the word "correctly" suggests another fundamental issue in science that she totally misunderstands, and it's the following:
Fully defined physical theories have never been proven to be "completely correct". Instead, they are temporarily compatible with the experiment up to the moment when they're falsified and superseded.
A theory of everything – probably string/M-theory – may be the only exception to this principle. You know, the important point is that in science, theories are not earning their right to live by some "perfect recommendation letters" making them infallible, as the Pope. Theories in physics have never been "completely correct". They have the right to be considered from the very moment when someone proposes them and what may decide about their fate is, on the contrary, when they get falsified.

So a theory saying and allowing us to calculate something is doing very well if it is not falsified, or up to the moment when it's falsified. You just can't change this elementary rule of science. You can't kill or get rid of a scientific theory without falsifying it. Get used to this fact, crackpots. If you would demand that theories may only be investigated when someone has a proof that they're infallible, you could never do anything in science, especially because theories (at least theories as understood and formulated by humans) are generally never infallible.

Also, you may believe e.g. that the correct theory of all forces should have a small number of solutions (like the vacua of string theory). But there are still two reasons why this belief can have no impact on the physicists' rational selection of the focus of their research. First, the belief that the "number of solutions should be small" is absolutely unjustified by any empirical or logical or mathematical evidence – it's purely your prejudice. Second, all known theories with a smaller number of solutions are known to be wrong so even if the "small number of solution" were an important virtue, string theory could still be winner among the survivors.

Obnoxious "critics of modern science" such as Ms Hossenfelder are hysterically trying to find some qualitative difference that would show that all the modern science is defective in some way, fundamentally different from all the "good science" that they're willing to approve of. But no such difference exists. Fundamental physics has never abandoned the scientific method and the physicists are using basically the same strategies and considerations that have been useful and successful in many cases in the past.

They're adjusting what's needed but the basic principles of science just cannot be changed. Because of their screams that only fully verified theories should be investigated; or that the falsification in science should be replaced by the "absence of a complete proof", it is these "critics of science" who are absolutely abandoning even the most basic and universal principles of the scientific method. Nothing like science can ever work according to the principles they are describing.

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

snail feedback (0) :