Monday, April 01, 2019

Skepticism about Standard Models in F-theory makes no sense

Four weeks ago, I discussed a quadrillion Standard Model compactifications that were constructed within F-theory by Cvetič et al. For some happy reasons, Anil at Scientific American wrote his own version of that story four days ago:
Found: A Quadrillion Ways for String Theory to Make Our Universe
I think that Scientific American hasn't been publishing this kind of articles about some proper scientific research – and Anil hasn't been writing those – for years. Some adult who works behind the scenes must have ordered this one exception, I guess. So I am pretty sure that the readers of SciAm must have experienced a cultural shock because the article is about a very different "genre" than the kind of pseudoscientific stuff that has dominated SciAm for years.

Well, there are some differences between my and Anil's comments about that article. But there exists "a very different" way of talking about these matters – a rant titled This Week's Hype (this title has been recycled about thousands of times) written by Mr Peter Woit.

OK, so he seems dissatisfied that SciAm writes about this fancy, rigorous research at all. If some people still read those vacuous anti-science tirades, Mr Peter Woit serves them the usual emotional gibberish. First, the paper and the SciAm summary are said to be "hype". Well, there is absolutely no hype in those articles. It's just a very technical research of 8-dimensional geometries that are relevant for particle physics thanks to the F-theory constructions; and a semi-popular summary of that research.

The first full sentence with a complaint says:
As usual in these things, the only physicists quoted are the authors of the article, as well as some others (Cumrun Vafa and Washington Taylor) who are enthusiastic about the prospects for getting the Standard Model out of “F-theory”.
People who have been doing research on F-theory – and especially phenomenology of F-theory – were chosen to provide their opinion on the paper by Cvetič et al. for a simple reason: They are the experts and the "only" experts. The opinions of non-string theorists would clearly be nothing else than random incoherent noise that would brutally lower the quality of the story in SciAm.

Even most string theorists could be expected to say misleading things about F-theory and its realizations of the Standard Model. It's just wrong to fill articles about a technical research where many details matter a great deal with some non-experts' emotions. These emotions wouldn't add any positive value. And you know, there is a very good reason why Cumrun Vafa was asked about his opinion. He is the father of F-theory. A detail, right? Maybe some people think that shouting "F-theory is evil" is as good an expertise as being the father of F-theory but I don't.

Wati Taylor is also highly qualified to comment; among other things, he co-authored a truly gigantic class of flux vacua (not resembling the Standard Model) in F-theory in 2015.

If someone doesn't know e.g. how to write a torus by a twelfth degree complex polynomial equation in \(x,y,z\), then he or she has almost certainly nothing useful to say about F-theory, period. And be sure that Peter Woit as well as over 99.99999% of the mankind belongs to this "not really promising" set. Science builds on evidence and calculations, not on "opinions" of the people who don't understand anything about the issue. People who can't reasonably say things like "oops, Mirjam, you forgot a term contributing to the first Chern class from a brane" should exploit the opportunity to shut their mouth because they have clearly nothing to contribute and it's terrible if some mass culture is trying to pretend otherwise. The scientific value of some knee-jerk "critics of F-theory" is exactly the same as the musical value of a drunk guy who penetrates into a concert hall and throws up on the orchestra. They should be a task for bodyguards, not researchers or musicians.

But Woit's whining gets more intense:
No one skeptical of the idea of F-theory compactifications of string theory...
A person who is "skeptical of the idea of F-theory compactifications of string theory" is exactly analogous to a person who is skeptical about the other planets in the Solar System or skeptical about the primes greater than 100. He or she is clearly a person who doesn't have the slightest idea about the topic that was discussed in the SciAm article.

The existence of F-theory compactifications of string theory is pretty much a rigorous mathematical fact. There is no "hype" or "commercial" or "exaggeration" hiding in this statement. It is really literally true. Even some 30 years after the First Superstring Revolution, the only known consistent theories of quantum gravity coupled to particle physics are the constructions linked to string theory. And they may be divided to five or so classes of compactifications – here I clump all of F-theory as one class. That's why actual physicists working on the top-down particle physics take string compactifications – and even F-theory compactifications, let's say a 20% market share of the stringy model building industry – very, very seriously.

One simply can't be both intelligent and "skeptical about them".
If such a person had been consulted, he or she might have pointed out: Models like this have been around for over two decades, see for instance this from 23 years ago.
It's nice if someone were capable of noticing that F-theory has been investigated since February 1996 (OK, is there something shocking about that timing information?) but this knowledge of a historical factoid that is extremely far from turning someone to an F-theory expert who could be reasonably "consulted" in articles about F-theory.
They have always come with claims that some sort of connection to experiment was right around the corner.
There is no comment about any "experiments around the corner" in the paper by Cvetič et al. and there is absolutely no reason why such remarks should be "mandatory" in papers that map the landscape of possibilities to get a realistic theory of particle physics from a consistent framework of quantum gravity.
This new work doesn’t even bother trying to make “predictions”. It just works backwards, trying to match the crudest aspects of Standard Model, ones determined by a small set of small integers.
There is absolutely nothing wrong about "working backwards". Indeed, the search for the right theoretical explanation of Nature is an inverse problem of a sort. To one extent or another, everyone who has ever searched for a better theory of Nature was "working backwards": Kepler, Newton, Maxwell, Einstein, Feynman, Glashow, and everyone else. It's just extremely embarrassing if someone misunderstands even such totally rudimentary facts about science.

One quadrillion Standard Models in the paper refer to one quadrillion of 8-dimensional topologies that, if used as the hidden dimensions of F-theory, produce a particle physics spectrum whose low energy part agrees with the Minimal Supersymmetric Standard Model. It is nontrivial that all the quantum numbers work – and the authors were capable of translating these conditions into geometric constraints on the 8-dimensional topology. It is an impressive piece of work whether or not an anti-physics heckler prefers to spit on it.

Also, it's laughable to describe the reproduction of the Standard Model spectrum just as "crudest aspects" of the model. All physical predictions are totally determined by the theory given the knowledge of the spectrum and the values of some continuous parameters. In this sense, the correct quantum numbers describing the spectrum are about "one-half" of all of physics (and virtually all of the "qualitative aspects" of physics), not just the "crudest aspects".
Given the huge complexity and number of choices of these F-theory constructions, that some number of them would match this set of small integers is not even slightly surprising.
One quadrillion is much more than any "package of explicitly constructed Standard Models" that was found ever before. So by the sheer size, it is surprising. Cvetič et al. deliberately tried to look for such a class and they found a quadrillion solutions. Some people could have expected more, some people could have expected less. Surprises are a subjective matter. It is meaningless to talk about "surprises" in an objective way.

What is surprising to me is the very concise way how the geometric conditions equivalent to the "Standard Model spectrum" may be written down. I find the topological condition with some "three terms" to be much more economical than the usual QFT ways to describe the Standard Model spectrum.
The authors seem to argue that it’s a wonderful thing that they have found quadrillions of complicated constructions with this kind of crude match to the SM. The problem is that you don’t want quadrillions of these things: the more you find, the less predictive the setup becomes.
These assertions are absolutely irrational. Every consistent theory of quantum gravity that also includes the Standard Model spectrum and perhaps a few more things that are needed is a viable candidate to describe Nature in detail. So until they're ruled out by a wrong detailed prediction, these quadrillion F-theory vacua are viable candidates, too. In science, one simply can't refute or eliminate possible theories by incoherent emotional rants. Only the falsification by conflicting empirical evidence may eliminate models – that's true for every element of this "set of one quadrillion vacua", too!

At this level of fineness, it is simply another mathematical fact that the number of viable candidates is at least one quadrillion. Someone could find a number smaller than one quadrillion "more philosophically pleasing" but in that case, he would simply be discarding most of the real possibilities and therefore heavily reducing the probability of finding the right theory.

Realistic models of quantum gravity coupled to the Standard Model are rather rare (string theory is the unique solution, it seems), but because of the multiplicity of the stringy vacua, they may also be considered numerous. Is the number of possibilities large or small? It depends on what you mean by "large" or "small" e.g. what you expected. At any rate, this is the relevant class one has to work with and to say otherwise means to be detached from the basic facts.

So of course it is a wonderful thing that these one quadrillion F-theory Standard Models were found and explicitly constructed.
What’s being promoted here is a calculation that not only predicts nothing, but provides evidence that this kind of thing can’t ever predict anything. A peculiar sort of progress…
These compactifications are (supersymmetric) Standard Models. So they make the same qualitative predictions – of the spectrum and the particles' interactions etc. – as the Standard Model as a QFT, or any other realization of the Standard Model within a complete theory. F-theory isn't just some numerology producing quantum numbers; it actually does include all of the QFT dynamics as a limit. So to say that the F-theory vacua make "no predictions" is as silly as saying that the Standard Model makes no predictions. But an intelligent person understands it. He doesn't have the need to talk about "predictions" all the time.

Woit's pathologically obsessive usage of the word "predict" is a sign for every intelligent reader indicating that he's just doing a propaganda for the least demanding readers, not anything that is related to science. While his short emotional rant uses the word "predict" a whopping six times, this verb doesn't appear on the 6 dense pages of the Cvetič et al. preprint at all.

Science isn't being done and cannot be done by obsessively screaming buzzwords. It is obvious to everybody with a brain why the mapping of Standard-Model-like compactifications in F-theory, 1/5 of string theory, is an important enough research. SciAm hasn't consulted "F-theory skeptics" because it realized that consulting people who are totally and completely unfamiliar with the topic would be heavily counterproductive for the quality of the resulting article.

And that's the memo.

P.S.: Some commenters realize that Woit's negative remarks are just "mean" and uninformed. But there's one other commenter who cannot understand anything about the derivation but who still feels entitled to demand a "worsening" of the title. "A quadrillion standard models in F-theory" is no good for that commenter, you know, because they're supersymmetric models and they may have various proton decay operators. So the title will hopefully be bastardized by a reviewer, the commenter hopes.

Holy cow. Every reader who has at least 1% chance to get anything useful out of the paper knows that all realistic, detailed models of particle physics incorporated in a theory of quantum gravity must be supersymmetric models. In fact, all the promising potential readers almost certainly know that all models ever described by Cvetič were always supersymmetric. So all these people simply know that the title refers to a verbally nice shortcut of the "MSSM". The MSSM really is the "standard" model in the string model building community.

Also, it's terminologically correct to use the term "Standard Model" even if e.g. the proton is much less stable there. The "Standard Model" is a standard phrase defining models that have the same qualitative low-energy spectrum as the theory we need to explain the LHC data. I sincerely hope it's still impossible for these individuals from comment sections on the Internet to corrupt the peer review process but sometimes I am no longer sure.

No comments:

Post a Comment