## Saturday, October 24, 2015

### Discoveries of monsters under the bed

This blog post is partially an advertisement of the blog written by a postdoc at the Perimeter Institute who studies fancy methods to calculate scattering amplitudes in gauge theory (very serious theoretical stuff!),
4gravitons.wordpress.com (by an anonymous serious researcher)
I am actually not sure about the name of the author. Because he or she has no nickname, either, I will call him or her Mr Tetragraviton. In the newest blog post
When to Look under the Bed,
Tetragraviton interestingly responds to the proposed semiclassical gravity experiments and the claim that the quantum character of gravity doesn't need to be tested. We agree that there's no way how gravity could be non-quantum which seems to make the experiment pointless.

But Tetragraviton also makes an important point that experimenters have a different motivation to do experiments. They're less attached to particular proposed theories, Tetragraviton says. They don't want to discover some NMSSM RPV gluino in a particular model but gluino – or supersymmetry – in general. I agree with that but I think that this generosity is inseparable from the experimenters' looser attachment to theories in general. If you talk about some particles but you don't exactly specify what is the theory that predicts its properties etc., it means that you don't care about the detailed properties much, either.

You shouldn't view these comments of mine as criticisms of experimenters. The different focus of theorists and experimenters is completely natural, understandable, and it doesn't conflict with the health of science in any way. However, I still believe that there have to be interactions between theorists and experimenters – interactions going in both directions:
1. Theorists have to care whether their theory or theories are compatible with the body of reliable experiments
2. Experimenters have to care about theorists' credible recommendations where to look
Experimenters may have their own opinions but they shouldn't run their "alternative theory departments" simply because they are not equally good at theory. They're free people, however, so they may sometimes "look for a monster under the bed", as Tetragraviton says. It funny because last night, I watched the 2013 comedy film "The Incredible Burt Wonderstone" about the magicians, their friendship, and their cruel competitive struggle. In the movie, an old magician in the hospital made himself disappear under the bed, too. The younger magicians – his soon-to-be partners – knew about the resolution of the trick but they pretended otherwise not to reduce the self-confidence of the old Gentleman.

Theorists obviously can't foresee everything – I mean every discovery. They are getting undoubtedly better at "foreseeing" things that were first seen at a fixed moment in the distant past. ;-) But when it comes to predicting things in the future, a given amount of time ahead of their current lives, they are probably equally bad. Many discoveries, including some very recent ones, were largely unexpected by the body of theorists – the positive cosmological constant is an example from fundamental physics. The positive cosmological constant is as close to a "monster under the bed that was actually discovered" as you can get.

So it's fine when some experimenters look for a monster under the bed even when theorists are saying that there can't be a monster in the bed. Sometimes, they may perhaps find a monster under the bed. But you know, such a discovery is unlikely – and may be either more or less unlikely. If the theorists say that they can show that a new effect is really impossible, they're probably right. They're increasingly right about fixed effects.

So I think that it's much more likely (and typical) for experimenters to find the monsters e.g. under the birds' excrements (where [almost] no one expected it, like the cosmic microwave background) than under the bed (a place where many people love to imagine a monster). The experimenters' opinion that a monster may or should be under the bed is actually narrowing the range of their search. It's better to search "truly randomly" than according to a totally wrong theory or prejudice.

If an experimenter didn't care about such voices ("you shall find nothing new in this way") coming from the scientific theories, his enterprise would be totally unscientific. He would be reduced to a shaman who looks for miraculous things obeying some vague unscientific rules – and his search for wonderful things wouldn't differ from the work of the shamans. It wouldn't be affected by the other people's experiments (and maybe even her own previous experiments) and the theoretical knowledge extracted from those previous experiments. So he would be repeating the search for monsters at random places picked by prejudices and irrational habits.

"Cargo cult" is an example of the experimental enterprise that doesn't take the (failed) results of the previous experiments and lessons (theories) that may be extracted from them into account. The dummy bamboo airplanes are shown from 6:00.

Of course, she wouldn't find anything just like the shamans haven't found (almost) anything. Doing experiments without any theoretical framework that suggests what outcome is "more likely" and "less likely" is simply not science and it leads to no scientific progress! The reason is that scientific theories – even relatively primitive ones – have a clear opinion on what actually matters in a shaman's experiment and they may confidently conclude that some wonderful new effect just can't materialize under given circumstances.

In practice, almost all the experimenters know the theory up to some moment. Some experiments have only mastered classical physics up to a point in the 19th century. Some experimenters, including those in particle physics, have their "threshold" some 30-70 years ago. They haven't understood anything that theorists have figured out in the recent 3-7 decades. But even this "partial ignorance" of theoretical physics is a problem that can make the experimental activity largely unscientific. When you want to make discoveries in the quantum aspects of gravity, you simply have to know something both about gravity and the quantum mechanics.

Tetragraviton also suggests that it's OK to do one experiment to see whether the cell phones cause cancer. He links to a page that looks ambiguous but I think that Tetragraviton would ultimately agree with me that the answer is almost certainly a simple "No".

I've spent quite some time by explanations to the laymen of the difference between the "microwaves" (and electromagnetic waves with frequencies below those of the visible light) and "UV, X, and gamma rays" (higher frequencies than the visible light). Only the latter contain photons with sufficient energy to kick electrons (or even nuclei) out of their places and cause ionization, DNA dislocations, or cancer. The former are "slow, nice, and long waves" that at most manifest themselves by "heating things" (like in the microwave oven). Funnily enough, a friend was alarmed by that description, too. Cellphones could heat up your brain, terrible, she shouted! ;-) Well, yes, they may warm your head but it's a small amount that you feel and when you feel it, the effect is exactly the same as if someone touches your head with his warm hand! This is surely not lethal, is it? Well, sometimes when you talk to the laymen and you feel that you have reduced the scientific problem to the "physics available at the basic schools", it's still not enough to settle the problem.

(Just to be sure. I can imagine that some mechanisms in our bodies are monitoring low-frequency electromagnetic radiation – in some semi-intelligent way – and get disrupted by the cell phone signals. Such a mechanism just doesn't seem to be helpful for the explanation of anything known that actually does work inside our bodies, so I don't think it's too sensible to actively search for such a "problem". If this influence existed, I would probably wait for the "problem" to become obvious. If too many people at Cupertino were dying of a mysterious illness or from flu, Apple would probably have something to do with it. But would I actively investigate whether iPhones make flu more lethal? No. And I also find it important that while the experiments can't "completely" settle the question whether the cancer-causing effect of cell phones is nonzero, they have already made it obvious that the effect, if nonzero, is so weak that you can ignore it unless you belong to top 0.01% of the chronic users.)

Let me return to the main problem. Experimenters may sometimes look for things that, according to the (best) theorists, have basically zero potential for a new discovery that would contradict the predictions that the theorists can make, at least in principle. But I think it's obvious that most of such experimenters fail indeed. And experimenters who are doing nothing else than these failed experiments have to be eliminated, otherwise the experimental physics doesn't work as an industry. When meritocracy works among experimenters, experimenters naturally stop doing silly things like that because they have led to no successes before.

Another fact is that theorists like Tetragraviton shouldn't be expected to talk as experimenters. They're still theorists and from their (or our) viewpoint, there is virtually no possibility that an experiment will discover something that would imply that "gravity isn't quantum" when the results are correctly interpreted. So the theorists are not only free to express this opinion. They are pretty much obliged to do so! And that's what I have done in this case, too.

You may try to perform certain experiments that you claim to be potentially able to prove that "gravity isn't quantum". But my duty is to tell you that this is as "sensible" as e.g. an experiment designed to produce lots of gold by mixing milk with crude oil. It just can't work! For an experimenter, it may be easier to see that the milk experiment won't work; but for me, both cases are pretty much equally easy. From my viewpoint, both proposed experiments are approximately equally silly (I may be hypothetically wrong about "gravity has to be quantum like everything else"; but I may also be wrong when I say "you can't get gold by mixing milk and crude oil") and it's wise for an experiment to "empathize" with the people who clearly understand some relevant theories better than she does.

An ideal versatile scientist has to verify all her assumptions herself. But an experimenter who doesn't know the state-of-the-art theories at the technical level is not an ideal versatile scientist and she must acknowledge this fact and take it into account. For her search to be sensible and promising, she simply has to rely on some feedback from the people who are better theorists.