Thursday, December 17, 2015

Trouble with workshops on "testability"

Richard Dawid has organized a three-day-long workshop in Munich dedicated to claims that there's something wrong with contemporary theoretical physics because it's untestable. Two critics of science Silk and Ellis have previously called for such a meeting. So some of these critics and some of the string theorists gathered in Bavaria in order to agree about the big picture in which the string theory research takes place.

No agreement could have emerged from the meeting – which is good news because the world doesn't have too good experience with treaties signed in Munich. ;-)

Dawid and some participants are great but I can't imagine what the point of such a meeting could possibly be. String theorists meet with people who have no clue about contemporary physics but who believe that their ignorance is at least as good as the physicists' technical expertise. The latter pretend that they're (at least!) on par with the string theorists – like Gross and Polchinski – and they work hard to transform the workshop into a trial that could result in a ban of string theory, inflationary cosmology, supersymmetry, and other things. That's why I would find such events repulsive.

Accept my apologies but string theorists (and other groups) primarily do string theory research because they are free to investigate whatever they want – this freedom is a pillar of the Western society. Even if it made sense to describe string theory as a "religion" (and let's ignore the fact that this label is totally silly), string theorists would still have the right to believe in this "religion" and teach young people interested in this "religion" about the theory's explanation of the Universe. They would be free to do research and pay bucks to the researchers (and big bucks to the best researchers).

Everyone has this freedom. Because the string theorists are intellectually superior, they're capable of learning the theory that incorporates all the crucial valid older insights about Nature plus some more and seeing that it's probably an unavoidable paradigm shift. Others are not as intelligent so they don't understand it. The basic explanation is as simple as that. 99.999% of the people who criticize string theory (and a similar percentage of those who have never heard about it) just can't learn it. That is their primary characteristic that restricts and shapes their opinions. And you simply can't neutralize your ignorance by an ideology. If you don't understand string theory, you just don't understand it and every ideology trying to sell this ignorance as a virtue is just an idiotic piece of populism for the trash bin of the mankind.

Natalie Wolchover is a good writer but when it comes to her story about the workshop,
A Fight for the Soul of Science
I have lots of critical problems with that text.

At the beginning, she quotes David Gross. He said that physicists don't need philosophers – just like birds don't need ornithologists (this analogy goes back to Feynman). The following sentence is
But desperate times call for desperate measures.
It is positioned so that it sounds as a de facto quote of Gross. I don't believe that Gross has said something about "desperate times" and "desperate measures". He is much more likely to talk about wonderfully exciting times. Wolchover must have adopted the propaganda of the crackpots.

I am pretty sure that what Gross meant is that philosophers like Dawid are good for explaining to others what is the gnoseological status of the research they are doing. But philosophers are not needed for the scientific research per se – and this is what primarily matters. I think it's obvious that Gross' very point was that science is superior or at least primary and philosophical polemics are inferior or at least derived but Wolchover has obfuscated this key point.

Instead, she offered a long paragraph with bitterly anti-science remarks by Silk and Ellis. This is an intrinsic problem of the workshop. It's meant to resolve a problem but the masterminds who wanted the workshop to take place are extreme proponents of one of the two viewpoints, the totally deluded one. So their spirit would always fly above Munich.

The following paragraph is dedicated to Dawid but she doesn't quote a corresponding "preview" of his philosophy. It's only individuals such as Silk and Ellis who could enjoy this luxury. A seemingly innocent beginning of another paragraph is telling:
Gross, a supporter of string theory who won the 2004 Nobel Prize in physics...
Jesus Christ. Gross may be a supporter of string theory but I think it's much more important that he has been a great string theorist, and it's not just thanks to his co-discovery of the heterotic string. Wolchover is clearly placing string theory as such – the research and its findings – at least on the same level as the "string wars". And the sentence actually suggests that she is actually placing the "wars" above the "science" because Gross' being a "supporter" of something seems more important to her than his being an "expert".

This attitude by Wolchover is somewhat analogous to Lila Shapiro's description of Lisa Randall as a mere tool of feminism – instead of a top particle physicist.

The difference between the "science" and the "wars" is that the science of string theory (much like the science about QCD and other things) that the mankind has learned has the value of uncountable billions if not trillions of dollars; I see no other good way to convey the point that it's basically priceless. Just like Gross has gotten a Nobel prize, several colleagues of Gross' have gotten $3 million awards for their discoveries in string theory, and so on. On the other hand, the jihad against string theory is a worthless enterprise pursued by worthless people. You can easily find billions of "soldiers against string theory" because well over 7 billion people on Earth have no clue about string theory (or QCD, for that matter). But such "soldiers" would still contribute zero information about the scientific question whether string theory (or QCD) is right, let alone more refined questions. Also, someone who would be just a "supporter" would only be helpful to the extent to which he or she helped other people do the actual research.

Several paragraphs in Wolchover's text redundantly convey the point that the scales of quantum gravity are far from the everyday life experience. At least, Wolchover quotes a pragmatic assertion by Gross:
“The issue in confronting the next step,” said Gross, “is not one of ideology but strategy: What is the most useful way of doing science?”
The point is that there doesn't exist and there can't exist any permanent, universal, everlasting algorithm that would tell a scientist what he should study and how he should approach problems. Cutting-edge science is always a creative business, not mechanical work that a worker may learn. Scientists have to choose or find a strategy and the research by various phenomenologists or formal theorists reflects their particular chosen strategies. Brilliant and educated people are "somewhat" more likely to pick a more promising strategy.

Afterwards, she discusses the history of the "rules of the game" (of the scientific method). She mixes Newton, Popper, and Shmoits as if they were parts of the same coin, as if they were peers. At least, she admits that falsifiability is in no way a sufficient condition for genuine science. Astrology is perfectly falsifiable – and falsified every day – but it is still around but it is not science. Massimo Pigliucci, a philosopher, announces that Popper is no longer so hot among philosophers. Whether this is true or not, the impact on the research in theoretical physics is clearly non-existent because these philosophers aren't familiar with the actual work, methods, and evidence that have emerged in physics. So their opinion just doesn't matter.

100 years ago, philosophers (like Mach and positivists in particular) were inspiring theoretical physicists (including Einstein, Heisenberg, and others) because they were offering some nontrivial ideas and were pointing out some problematic assumptions of the people's thinking. At least since the 1950s, philosophers largely stopped being inspiring for cutting-edge physicists – perhaps because most of the philosophers got reduced to ideological defenders of some old ideas (which are mostly known to the physicists, and many of them are also known to be debunked by the progress in physics). It's not a physicists' fault and the physicists have no "duty" to return to the early 20th century and build on philosophers' ideas again. 100 years ago, Einstein and pals were only doing that because they saw something valuable and new in the philosophers' thought. The physicists don't see things like that today. Just get used to it. Even to learn the conceptual "big picture" in physics has turned into a technically demanding field and someone who is just a philosopher is very unlikely to understand even the "broad skeleton" of the portrait of the Universe drawn by the state-of-the-art physics. You simply need some "technical" knowledge and experience for that.

The absurdity of the Popperian cult and discussions about it is illuminated by the following sentence:
Nowadays, as several philosophers at the workshop said, Popperian falsificationism has been supplanted by Bayesian confirmation theory, or Bayesianism, a modern framework based on the 18th-century probability theory of the English statistician and minister Thomas Bayes.
Wow, so the state-of-the-art modern contemporary realization by the philosophers are ideas by Thomas Bayes. How stupid do you have to be to believe that Bayes' ideas became new in the actual scientific reasoning of the recent decades? He may have been a Presbyterian apparatchik in the 18th century but Bayes has quantified some methods that people were and still are actually using to reason. No sensible person has ever disputed the validity of Bayes' theorem and other things.

If you believe (or have believed) some version of the "Popperian" or another philosophy that would sharply contradict Bayesian reasoning, then you are clearly a totally irrational individual: you contradict reason itself. Bayes has realized that confidence is a continuous quantity. What a shocking discovery: probabilities are continuous. But it really does seem that this discovery is new or controversial for the Popperian jihadists. Instead of the "dirty" science that works with different levels of confidence (and with arguments that may be both positive or negative), they are dreaming about a dogmatic framework that immediately attaches labels "good" and "bad" to everything according to some extremely oversimplified and misleading ideological criteria.

Wolchover mindlessly repeats the critics' slogans about the "untestability" of string theory many times. But string theory is exactly as testable or untestable as quantum field theory. They're two general frameworks to describe Nature; and additional data (field content, interactions etc. in quantum field theory; particular compactification in string theory) have to be specified to make the theory relevant for particular observations around us. When they're specified, both frameworks are capable of producing detailed theories (not quite unique ones, in both cases) that agree with everything we have observed so far. Even string theorists often prefer quantum field theory over string theory to deal with real-world experiments because QFT is "closer" to the everyday life; but this is totally analogous to the fact that we often use Newton's physics for mundane enough problems even though we realize that relativity or quantum mechanics is more accurate. What is perhaps untestable (by low-energy experiments) is the difference between QFT and ST. We can't discriminate them by doing some straightforward experiments. But to selectively and asymmetrically use this fact against string theory – which is pretty obviously a more complete, more consistent, more explanatory among the two frameworks – means to misunderstand some complete basics about the relationships between various theories and ideas in modern physics. It's just stupid. Anyone and everyone who pretends that string theory has a completely different (qualitatively worse) epistemological status than quantum field theory is just a 100% crackpot. This is not a question you may organize meaningful conferences about. It's basically a question for organizers of the education system: How to make sure that fewer children are left behind?

In another section, Dawid is said to have become "fascinated by the confidence of string theorists in string theory" while he was a student. So he began to collect philosophical insights. This is a possible career move but the normal solution is to learn string theory at a technically deeper level. That's how students who stay in this part of physics do things. When someone is a young student (or postdoc) in physics, she may misunderstand many things – including the confidence of other researchers in certain claims because the confidence depends on some knowledge or experience that is not immediately available to everyone. But when they go through similar "lessons" (including learning of other people's work and calculations of some new things) that Nature and string theory teach us, they reach similar (but, due to progress, not the same) conclusions about the big picture as their older stringy colleagues. The philosophical summaries and the big picture follow from the detailed content of the theory. You can't start with philosophical dogmas and then demand that the technical science is "obliged" to conform. At least since Galileo's arguments with the Catholic heavyweights, we have known that this "philosophy first" method isn't how science may work. Too bad that the Popperian jihadists are incapable of understanding this basic characteristic of science.

So this whole way of thinking that "big questions about string theory should be reduced to philosophy" is just totally wrong. Big questions about string theory are still questions about string theory and one needs to learn physics, and not philosophy, to converge to the truth (or reasonable opinions) about those questions.

Fine. So Dawid has "discovered" those non-empirical reasons to trust string theory – it's the only game in town etc. Maybe it would have been enough to simply ask! He may have formulated the reasons why string theorists believe string theory in terms of bullets but one could have gotten similar answers from actual string theorists, too. And if the number of reasons to "trust" string theory and their detailed composition were different from Dawid's according to a string theorist, the string theorist could still be more right than Dawid. The question how many bullets you should draw when you discuss a theory is not a meaningful physics question.

I've mentioned how Gross' relation to string theory was described. Here's how Wolchover treated Polchinski:
The staunch string theorist Joe Polchinski of the University of California, Santa Barbara, presented...
What the purpose of the word "staunch" is supposed to be? What's the difference between a string theorist and a "staunch string theorist"? Is it some "badge of rank" used in the string wars? Just like a fundamentalist Muslim and a Muslim are the same thing, there is no difference between a string theorist and a staunch string theorist (but there is a difference between a string theorist and a Muslim, despite the fact that radical string theorists have just bombed the LHC LOL!). One may be a partial or applied string theorist in which case it becomes questionable whether it's right to call the person a "string theorist" (similarly, it's disputable whether a person who isn't sure about the desirability of the Sharia law should be called a Muslim). But as long as one knows that it's the right label, he or she is simply a string theorist. What annoys me is that for Wolchover (and others), whether someone fights in a war is more important than his contributions to science: that's how I read the word "staunch".

By the way, if there were some way to quantify "allegiance", I probably wouldn't call Polchinski "staunch" these days, mostly due to his ideas that the AdS/CFT can't describe the black hole interior (and perhaps other things) which, as far as I can say, are rather extraordinary speculations about unexpected limitations of string theory that aren't backed by any persuasive evidence. But his "allegiance" isn't too important. What's more important is the value of insights that Polchinski has brought and keeps on bringing to physics.

Polchinski has also offered his order-of-magnitude Bayesian calculation of the probability that the multiverse exists. He got 94 percent. Wolchover wrote:
(Polchinski also used Dawid’s non-empirical arguments to calculate the Bayesian odds that the multiverse exists as 94 percent — a value that has been ridiculed by the Internet’s vocal multiverse critics.)
Why is it important to mention that a bunch of crackpots has "ridiculed" Polchinski's argument? This was a nice example of a carefully crafted rational argument on an issue that can't be treated by more reliable or more precise tools now. Science should use such arguments if they're the best arguments for a given question. Science cannot be done by ridiculing. Whether the multiverse is a valid concept in science is unknown so one can at most present circumstantial evidence and calculate the odds. Polchinski is trying to do so (which doesn't mean that my calculation is the same: my selection of the arguments and the resulting figure differ); his critics aren't. To "ridicule" something is only legitimate if there exists evidence that the ridiculed claim is almost certainly untrue or absurd. There exists no robust proof or near-proof that the multiverse doesn't exist which is why "ridiculing" is nothing else than a sign of one's stupidity.

After some paragraphs on the risk that science disappears if people do science, instead of just ridiculing, Wolchover returns to Gross' pragmatism:
“Let’s give an operational definition of confidence: I will continue to work on it,” Gross said.
“That’s pretty low,” Achinstein said.
“Not for science,” Gross said. “That’s the question that matters.”
Exactly. Gross' operational definition may look narcissist but it's much better than anything that the Popperian jihadists etc. could offer as an alternative. The point is that the confidence is ultimately a subjective state of mind and whether someone has it may be deduced from his or her behavior. The confidence may depend on lots of things that sometimes require very complicated arguments or experience. And frankly speaking, only top minds – who have some knowledge, intuition, creativity – may have a reasonable chance to end up with true opinions about cutting-edge questions in science.

If you don't have the knowledge of the relevant concepts or if you have no experience that would allow you to guess which "encouraging evidence" or "discouraging evidence" is enough to largely embrace or disfavor a hypothesis, your conclusions are bound to be inaccurate or wrong. It may be "high" (or whatever is the opposite of "low") for Gross to present a simple rule that allows Achinstein and everyone else to decide about the probability that a statement on physics is right. But you would have to be high to believe that such a rule exists.

So even more simply: You simply can't have any method that would allow a layman or an outsider to correctly evaluate the probability that a very complicated proposition about the cutting-edge science is correct, especially not if it is conceptual in character (and if it is not directly connected to some experience). There are no easy solutions like that. Shmoits implicitly assume that such a simple recipe for total idiots like the Not Even Wrong readers to decide about the validity of difficult questions exists. But it doesn't exist. Simple answers to very delicate questions that they want to pick are guaranteed to be wrong.

The remainder of the text talks about Rovelli's and his "alternative" theory; and another person's sociological noise and calls to ban supersymmetry and other things. I've had enough of this garbage. At the end, I don't believe that workshops at which both experts and self-confident laymen and crackpots gather are not constructive for anything. Such workshops primarily help to amplify the meme that the crackpots criticizing modern physics are (at least!) peers of the top physicists. But because these "anti-string soldiers" are much closer to being peers of boars in the forest, I would never accept the basic justification of similar workshops.


In the comments under Wolchover's report, Matsuko says that string theory is an amazing project and one should prefer it over the seemingly simpler theories. At the same moment, Matsuko recognizes that string theorists are smart and want them to think about more down-to-Earth problems. Like quantum gravity, Matsuko says. Well, quantum gravity and string theory are two approaches to the very same thing, with different focuses. Also, it's a pity that Matsuko's basically pro-stringy comment looks a little bit like astroturf which is a pity.

Sadly, Matsuko's comment is the first one and the last reasonable one, too.

Peter Bolt demands to use the name "hypothesis" for string theory etc. Under normal conditions, this would be purely a terminological issue and the word "hypothesis" is simply not used in this way in modern physics. A "hypothesis" is meant to be something you may formulate in plain English on three lines, a (possibly vague) belief about a (usually particular and limited) issue. String theory isn't a hypothesis in the sense that it is given by some "completely specified" equations, not just words, and it describes a "whole world". That's when people talk about "theories". After all, physicists use the term "theory" even for systems of laws that certainly don't apply to the world around us, like Chern-Simons theory (or maximally supersymmetric gauge theory – or any other theory in physics). The "hypothesis" says that string theory applies to the Universe around us. But it's clear that for Bolt, the call isn't just a matter of terminology. He has a tangible goal as well – to delegitimize theoretical physics. He believes that the word "hypothesis" would make theoretical (or hypothetical) physics look less legitimate which is why he would prefer this wording. He's an activist crank just like Shmeeter Shmoit and several others.

Paul Mays parrots a slogan "if a theory is not falsifiable, it is not valid" blah blah blah. But string theory is exactly as falsifiable or unfalsifiable as quantum field theory or other frameworks that physics can't exist without.

Richard Gaylord attacks Pigliucci for daring to say that astrology is both non-science and falsifiable. The philosopher must misunderstand the difference between a necessary condition and a sufficient one. Except that falsifiability, as understood by folks like Gaylord, is neither sufficient nor necessary, and the kind of "more falsifiable" ideas they would prefer are analogous to astrology. They may be more falsifiable and they are indeed more falsified and obviously wrong, too. At the end, science looks for ideas that are more true, not ideas that are more falsifiable, and that's a basic fact that the likes of Gaylord are incapable of getting.

Ali Reza Assar uses about 10 exclamation or question marks in the first sentence where he expresses the shock that some people including Witten dare to say that the beauty of string theory is a reason to take it seriously. It looks ugly to him. Well, instead of learning to recognize the beauty of string theory, let's be more modest and hope that unlike his countrymate, Ali will be able to learn the concept of a toilet.

Howard Landman says that string theorists like Gross only work on that because there is a mysterious establishment that corrupts them to do so. Sure, Gross would die of hunger if he weren't getting money as a supporter of string theory. And ask the young people who are struggling to get jobs, even postdoc jobs. Tell them that they're drowning in easy money. Landman, you are a complete idiot and a nutty conspiracy theorist.

John Merrymen suggests that string theory and the multiverse are symptomatic of [some crackpot theory of his how the past becomes the future which mixes 10 different physics buzzwords in chaotic ways].

Roger proposes that aside from physics and metaphysics, there is a better third path to proceed which would be endorsed by Popper and which needs 3D modeling software. This almost sounds like a joke but he writes stuff like that.

Elliot McGucken lists numerous quotes of "authorities" that he believe are arguments against string theory. Sorry but science doesn't acknowledge these non-technical arguments from "authority". Moreover, to claim that Born and Einstein (who died in 1970 and 1955 respectively; and similarly Planck, Poincaré etc.) as authors of arguments against string theory (that slowly began to emerge in 1968) is just dumb. These top physicists were doing totally analogous things as string theory in their own times.

Dan Hess, an engineer, similarly believes that Bohr and Feynman would eviscerate string theorists. That's great and in the case of Feynman, it was true to some extent, but there were also people outside string theory proper – like Gell-Mann – who provided string theorists with tons of support (at the same Caltech). None of these things may be a valid argument for or against the validity of string theory.

Finally, Patrice Ayme talks about hyperlocal and nonlocal tunnel vision in the context of string theory. It's time to reduce your drug consumption, Patrice.

No comments:

Post a Comment