Friday, May 29, 2009

String theory: a middle way?

Peter F. has kindly brought my attention to an article that Jessica Griggs wrote for Nude Socialist,
What string theory is really good for,
that primarily focuses on the application of AdS/CFT methods to condensed matter physics.

She did a fair job in describing the history of string theory, its role in unification, holography, AdS/CFT, its application to condensed matter phenomena, especially high-temperature and "holographic" superconductors, and related issues. And it's pretty clear that she had to pay a lot of attention to the testimonies of her informers, real physicists like Sean Hartnoll or Clifford Johnson.

However, the article also repeats many widespread laywomen's rudimentary misunderstandings about theoretical physics.

Let me start with the issues that are simple to understand. The first one appears in between the lines of the first paragraph. String theory - you love it or loathe it. Well, indeed. What the lady failed to report is that the two groups are not quite isomorphic: the people who "love" it are usually those who have studied it, did (or are doing) research on it, and whose average IQ may be around 145.

Those who self-confidently scream how they "loathe" the theory are uniformly pompous fools who have no idea what they're talking about and whose average IQ may be around 95. When it comes to their sociology, be sure that your humble correspondent is the world's #1 expert in these people - a social scientist who has researched them, one by one, in thousands of interactions that took thousands of hours. And you can be sure that every single one of them is a crackpot or a whackadoodle. I could spit hundreds of their names here but let's not transform this article into white pages with whackadoodles. :-)

This general principle about the origin of the difference between the two groups had to be clear even to Jessica Griggs because it directly follows from the stories she reported but she wasn't able to deduce the proper conclusion. Jan Zaanen, a Dutch condensed matter physicist, was annoyed by the fact that the string theory talks were the only ones he couldn't understand. Unlike many others, he decided to study the theory and fell in love with it.

This story is, of course, completely general. Those who haven't appreciated the breathtaking intellectual power of string theory either haven't studied it at all, or are not capable enough to collect the "critical mass" of its inner workings and squeeze it into their brains. Although many of them seem to be proud about their ignorance and the limitations of their memory, creativity, and imagination, there is really nothing to be proud about.

At any rate, Ms Griggs counts herself as a member of another group that believes a "middle way", a universal method used by many people to feel wise. Well, being "in the middle" is a good recipe to be a mediocre person but it is surely not a good strategy to converge to the right answers in science where the truth is rarely in the middle of the people's opinions. In fact, the truth often exceeds the expectations of the most extreme group of scientists - but it is not a priori clear in which direction. ;-)

The previous paragraphs talked about simple sociological issues. But there are many basic technical misconceptions that Ms Griggs promotes.

The first one is her suggestion that the tiny magnitude of the length scale where quantum gravity becomes relevant is some kind of a recent disappointing discovery. Well, it's surely not. Natural units began to be understood in 1881 when George Stoney put universal constants G (Newton's constant), c (the speed of light), and the electric charge equal to one. They were extended to Planck units by Max Planck in May 1899 when he added Planck's constant.

From that time, it was clear to every competent physicist that the natural scale where the speed of light, the novel quantum phenomena, and gravity simultaneously matter is close to 10^{-35} meters, a ludicrously short distance. In other words, every adult physicist born in the 20th century who was "surprised" by the observation that the fundamental length scales are tiny and hard to be directly probed - after he got his physics degrees - has badly messed up with the gears and wheels in his skull.

Another bizarre concept in the article is that Maldacena may not fully deserve the credit because another disappointing discovery showed that the spacetime in his correspondence is not the Universe around us. Well, such an idea is crazy, too. The AdS/CFT correspondence has never been studied as a part of phenomenology. The conjecture originated from a detailed analysis of coincidences involving branes and black holes in highly unrealistic backgrounds of string theory, usually with 8, 16, or 32 supercharges (at most 4 supercharges are acceptable as a starting point in phenomenology).

No one has ever thought that this new kind duality should be relevant for model building tomorrow or on the day after tomorrow (and various unparticles and conformal windows are the closest projects to become counterexamples). It was always a part of the conceptual research of one of the paramount questions, "What is string theory and quantum gravity". Sean Hartnoll has even made it clear to Ms Griggs that string phenomenology is just a relatively small part of the research of string theory (one that he was never too thrilled by) but she's still not getting it.

It turned out that the AdS/CFT is helpful for heavy ion physics, superconductors, atomic physics, and many other disciplines.

But these applications have emerged as a "bonus" and they are in no way essential to justify the research of fundamental conceptual issues about quantum gravity. The latter are being studied because they are arguably the deepest part of science of the early 21st century, a part that is likely to keep (or expand) its fundamental importance for decades or centuries, not in order to construct one particular dirty experiment in condensed matter physics that will be fashionable for two weeks.

Most science journalists still seem to be unable to distinguish pure research driven by concepts that are going to be taught for centuries to come from the applied research that is justified by more short-term, often practical needs.

The typical reason is that they're completely unfamiliar with the first, purely theoretical group of activities. They just can't distinguish quantum gravity from their desire to have a tastier cheeseburger on Friday night. The cheeseburgers keep on distracting them. They don't understand how a theoretical question unrelated to food or sex could keep scientists busy for decades or a century - or for millenia, if I include the Greek philosophers.

The article also includes a comment showing the anti-landscape bias. What do I mean? String theory has countably many stabilized solutions - and a (probably) finite but very large number of them (people like to say 10^{500} even though the calculation behind this number is not terribly canonical, well-defined, accurate, or important) are a priori semirealistic. Does it mean something for the validity of string theory?

Because we can't determine the "right" number of vacua in the multiverse - we can't measure it - it clearly means nothing. Even after all those years when this basic question was intensely discussed at all conceivable levels, many people still don't understand this basic point. The anthropic people think that a scenario realized by a (much) larger number of vacua is (much) more likely to be true. The anti-stringy crackpots scream that a metascenario (in this case, the whole string theory) that is realized by a (very) large number of vacua is, on the contrary, (very) bad.

Of course, as we have repeatedly explained, both of these groups are being completely irrational. The counting may only become relevant for our estimate of the probability of a theory (or a scenario) once we find out an independent way to derive the correct number of vacua admitted by the equations of the world we inhabit. Because the latter is not empirically known today (and it is not known from other, independent theoretical considerations either), the particular structure of the landscape (or its size) is just a prediction of a theory that has been neither verified nor refuted.

Ms Griggs also suggests that the condensed matter applications suggest that people study a different string theory or that the theory has different properties than what is expected from its being a unifying theory of all forces of matter. That's another nonsense, of course. As we have explained many times, it is easy to see that "the" string theory appearing in the AdS/CFT correspondence is the same theory that is studied in the context of the unification: these two subdisciplines of string theory just focus on different parts of the set of its solutions.

There are many other misunderstandings that are written either explicitly or implicitly, between the lines, in Ms Griggs' article, even though this article is incomparably better than the junk that we used to see two years ago when certain two crackpot books were so popular with the media. (The blue one was released in the Czech language three weeks ago but fortunately, almost no one has noticed its publication.) But let me conclude with another widespread misunderstanding about all of science. People don't distinguish active disciplines on one side and theories that are likely to be true on the other side.

These are totally different things. There are many theories that are no longer active because they have either been completely settled or they have been settled (or, on the contrary, not-so-settled) to the extent that seems realistic given the toolkit that is accessible today. People are working on other questions because they see a greater potential to make progress over there.

Mathematicians are no longer classifying finite groups because the task has been completed. It doesn't mean that there is something wrong with the classification of finite groups: quite on the contrary, it's totally and beautifully completed which makes it kind of dull, too.

In natural sciences, things are usually never "quite" completed. But nevertheless, good theories often reach the point where their agreement with the empirical world is satisfactory enough and attempts to make it substantially more satisfactory seem too difficult. For example, biologists "roughly" understand the evolution of birds and they are relatively satisfied - even though their picture of the evolutionary tree leading to birds is incomplete, somewhat inaccurate, and maybe even slightly uncertain. But most of them are simply looking to other questions because they seem to have some ideas how to make progress. But that doesn't mean that they immediately start to believe that birds haven't evolved from small theropod dinosaurs.

The same forces operate in theoretical physics, too. Some people study string phenomenology, others investigate conceptual issues of black hole physics or higher-dimensional geometry, while yet another group does research of the applications of holography in low-energy physics. Each of these groups has many subgroups and people are jumping in between them.

But the amount of activity is not a measure of the immediate value of the probabilities that various hypotheses in the given subdisciplines are true: the amount of activity may be comparable to the expected rate of changes - i.e. the time-derivative of the probabilities (which can come with both signs). But if there's no activity relevant for a particular question, it means that the probabilities of different answers to the question are not changing; it doesn't mean that they go to zero (or one - or another number that an external demagogue could find convenient as his interpretation of the "silence").

For example, as of 2009, the Standard Model is quite surely the correct effective theory below 100 GeV while string theory is very likely the correct unifying theory of all forces and particles. One needs progress to refute one of these claims or to increase the probabilities, but no progress means no change of the probabilities.

Some people would like to change scientists' beliefs about particular questions without making any research and without presenting any scientific results. That's simply not possible in science and every person who is not an idiot should be able to understand why.

And that's the memo.

No comments:

Post a Comment