Philosopher David Wallace has previously written many things about the foundations of quantum mechanics that – I believe – no competent quantum physicist may subscribe to. However, if he carefully avoids this particular foundational topic, he may look very intelligent to me. In February, he wrote
Naturalness and Emergence (PDF, HTML).The main conclusion is radical. He calls for a paradigm shift because the LHC null results and some facts about cosmology "undermine the entire structure of our understanding of inter-theoretic reduction, and so risks a much larger crisis in physics than is sometimes suggested". That's exciting!
OK, it is both exciting and ludicrous. But aside from these ambitious conclusions, he has written many things that seem correct to me – and that could earn an A grade if he were graded by someone like me.
In particular, I was impressed by his understanding of:
- the separation of physics' predictions to the laws of physics and initial conditions
- the fact that the naturalness is generally the "absence of extreme fine-tuning", not everyone gets it
- the existence of many types of "naturalness" – he talks about his own "Naturalness" that he capitalizes
- the sane definition of naturalness of probability distributions on the phase space – their \(\rho(q,p)\) is just given by some simple enough condition
- reasonable principles about the arrow of time (although he largely avoids the phrase); in particular, the precise distribution that evolves from a natural distribution (such as a "ball in a phase space") by a past-to-future evolution is unnatural
- basics and importance of the renormalization group and the basic patterns by which the parameters are reshuffled when the RG scale is changed (I think that most people who aren't real physicists but try to talk about naturalness don't even try to understand these issues although they're important)
- the fact that the naturalness on the parameter spaces of theories has something to do with the relationship between the more fundamental description and the less fundamental one
- the related relationship between naturalness and emergence
- the possible Bayesian (or other) interpretations of naturalness as a strategy, as a guess, as a theorem, and more
In the early 20th century, philosophers were a cherished caste between physicists – and physicists could really learn something from the philosophers. Positivism has played an important role for the rise of both relativity and quanta. Wallace's writing could perhaps be an echo of that era a century ago. His is an essay focusing on the truly conceptual things and he is much more talkative about them than a technical physicist who would think that "there's no reason to talk for hours or 27 pages" here.
On the other hand, he sort of sketches the basic conceptual ideas pretty much correctly (but my reading was excessively fast, not the super-penetrating deep analysis of every word that I have applied to a large number of texts – and books by others – in my life; numerous assorted authors have told me that no one has ever read their texts as carefully as I did). The text is a different genre than an introduction to a discipline of physics – textbooks of physics surely want to discuss some particular examples etc. that may be calculated, so that the textbooks are not just "philosophy talk". On the other hand, I've always had a lot of understanding for a "philosophy talk".
However, at some point, a miracle happens and Wallace concludes that we're approaching a truly deep paradigm shift in which naturalness in all of physics will be abandoned, and something dramatically changes in all of physics, some super paradigm shift takes place. OK, Dr Wallace, please stop taking the drugs if any and slow down a little bit.
First, extraordinary statements require extraordinary evidence, as Carl Sagan liked to say. The statement that "our recent observations require us to throw away all of naturalness and similar reasoning" is an extraordinary statement and the extraordinary evidence seems non-existent. A little bit more quantitatively: by some reasonable measures, at most 95% (and probably much less than that) of some parameter spaces of good enough supersymmetric models etc. may have been eliminated by the LHC experiment so far.
In the usual translation of the \(p\)-value to standard deviations, 95% is just two-sigma. So the LHC has experienced a two-sigma deficit of discoveries of new physics, if you wish. And two sigma is extremely weak evidence. It's surely too weak to change fundamental things about the physics reasoning. It's surely totally insufficient to throw away supersymmetry as a tool for phenomenology into a trash bin (supersymmetry is already "impossible to kill" as a broader theoretical tool in formal theoretical physics). Although a fraction of the people may prefer to work at other things, supersymmetry is alive and kicking. It's nowhere near to being ruled out or falsified in the usual scientific sense.
The idea that the "Bayesian-like reasoning such as the naturalness arguments should be thrown away" is even more ambitious than the hypothesized "death of supersymmetry". You would need even stronger evidence for that. We just don't have it. A two-sigma deficit of discoveries is surely insufficient.
But there's something even more implausible about Wallace's proposed conclusions. Because of the absence of new physics at the LHC etc., he wants to propose a paradigm shift in all subdisciplines of physics simultaneously. That's ludicrous for various reasons I have mentioned but another reason why it's ludicrous is that the various subdisciplines of physics are pretty much separated from each other.
By some discovery-less era in one field of physics, you just cannot directly change the thinking in other fields of physics (especially because "the absence of discoveries" really means "nothing happens, so nothing should change about the scientists' beliefs").
This is really obvious: different fields of physics study different phases of matter, different corners of the Hilbert space, different types of effective field theories (and their real-world realizations), and/or different energy or distance scales. And "the knowledge of physics" is quasi-local in all these respects. So if something happens in one field, such as particle physics near \(1\TeV\), it just cannot possibly influence other disciplines of physics, such as nuclear physics of stable heavy nuclei, high-temperature superconductors, or the black hole information paradox! ;-)
You know, there exists some kind of locality or "sovereignty" of the different subfields or regimes of physics. Why is it so? Because the knowledge of physics is composed of theories that have limited domains of validity. They're limited by various parameters or qualitative constraints but they're always limited. And that's why events in "another discipline" are "internal affairs of another country", if you can kindly survive this geopolitical analogy.
And the different subfields of physics have their own calibration of standards, much like individual countries have different tax rates etc. in general. So different subfields of physics rely on "naturalness" to different extents, they use it a bit differently, and if some other subfield of physics becomes more enthusiastic or more skeptical about naturalness as a concept, it's their issue, not ours. If another country increases their tax rate, we don't have do do the same – unless the first country is Germany and we're in the integrated European Union, of course, but that's a kind of trouble in which the sovereignty really starts to fail.
In particular, if Finland hikes taxes after today's elections, it's not because Austria did the same. It will be because the victorious social democrats think that their own, Finnish, eponymous social system collapses without the tax hike. (Why would I make up metaphors if I can use literally accurate and relevant ones?) Analogously, if condensed matter physicists will use less "naturalness" next year than now, it's not because naturalness was weakened by non-discoveries at the LHC after the Higgs. Instead, the reason will be that they will have found a less fuzzy method to keep on doing research than naturalness – a strategy that is scientifically justified but still fuzzier than others.
So Wallace's "naturalness is bad in all subfields of physics" resembles our religious friends who wait for us to abandon Newton's laws, mathematics, and all this stuff, and accept Jesus Christ as our Savior and as the monopoly owner of all correct answers. What experimental results would need to emerge for a scientist to undergo this religious "paradigm shift"? Well, something rather brutal would have to happen to him, some terrible experience, or he or she would have to see some significant damage to his brain, either abrupt or gradual. It would be a personal change.
There cannot be an objective transformation of science that would legitimately affect all physicists or scientists that would make them e.g. abandon mathematics and use the Bible instead. If they did it, their activity would cease to be science – it would become religion. So while the paradigm shifts may be very deep, they cannot be "too deep", like "throw away all of mathematics and declare your previous usage of mathematics as a dead end in which the people are getting lost or a mortal sin". The relevance of mathematics in physics is pretty much a defining trait of physics, something that has worked and built the glory of physics as a field, and every subfield has its own reasons – independent of other subfields – why it uses mathematics. Subfields of physics can't be expected to decide about the "ban on mathematics" by copying some rituals from other fields because they have much better reasons to use mathematics than "being obedient parrots". Mathematics has really worked long before people accumulated the results of any nontrivial experiments.
And the abolition of all the naturalness-like arguments across the fields of physical sciences would be a revolution that would be almost as deep as the abolition of mathematics in general. After all, naturalness may be justified as an argument based on the probabilistic reasoning involving probability distributions on the parameter spaces. You just can't expect an earthquake like that. If you have arguments that this is what all physicists or scientists should do, they're simply invalid arguments. They're some irrational overgeneralization of some emotional experience of yours. Scientists must still stick to certain basic patterns of thinking, especially to the rationality of the thinking, and nothing can be changed about it by 7 years (or even 70 years!) without discoveries of new particles. Yes, the Higgs boson was discovered 7 years ago.
Physicists want to ignite as deep transformations of their field as possible. But other physicists, as long as they're still physicists, simply cannot accept some "super deep paradigm shift" just because someone tells them. Maybe such a mass transformation of people's faith would be great among the religious people but scientists work differently. They must be skeptical – which is really a special kind of "stubborn". They don't change their previous opinions without a good rational reason. If someone wants to profoundly change all physicists' opinions but doesn't have any valid evidence, the physicists give him or her a hard time – and it is both their passion and their professional duty to do so. Science would be a religion or a mass hysteria if the scientists were more "cooperative" with someone's random efforts to "change the opinions of others just for fun". The transformation needs some evidence and the call "let's ban naturalness in the XY subfield because our friends in another field haven't seen a new particle for 7 years" is simply not valid evidence supporting the plan, so it will not be accepted by physicists as long as they are physicists!
Do you get it, Dr Wallace? If you do, I am happy because it was my main goal to clarify these issues and at this moment, I am abolishing the temporary ban on any substances that any readers may like or need.