Thursday, June 11, 2015

String theory and fun: a response to Barry Kripke

The Internet's most notorious anti-string blog posted a link to a view from a young impressive ex-string theorist (if you believe what's written there) that was posted on Reddit one year ago. That comment (plus the long thread beneath it) contains numerous ideas, thoughts, and sentiments. I will respond to the following points:
  1. the ex-string theorist thinks that he and similar people are really, really smart
  2. he left string theory for robotics
  3. \(h=15\) is approximately required to become a professor and it's too much
  4. he found out that string theory wasn't really fun for him
  5. string theory is only the journey, not the destination, and you should like the journey or leave
  6. Nima estimates that one gets a great idea once in 3 years or so
  7. there are too many trained string theorists
  8. there's too much competition and it's harmful
  9. the term "string theory" should be abandoned because it's too broad a subject
  10. the term "string theory" should be abandoned because most people work on different things
  11. Lisi's theory seems to be incomplete or provably wrong according to string theorists
There are lots of reasons why I think that the author of the text is a genuine trained string theorist with the stellar university credentials he has described. It's very likely that I know this person but I just haven't been able to figure out the name. Well, I have one extreme guess (a person much closer to me than the average) but I am not bold enough to speculate here. OK, let me start to respond.

Is he smart?

First, I do think that he is very smart and there is a couple of young researchers who are similarly impressive and whose background sounds very similar. The number of young people in this category who show up in high energy physics every year is comparable to one or two. It depends where one places the cutoff, of course. I do find it likely that this person belongs among 20,000 smartest people on Earth.

The author tells us that he left string theory for robotics. So he is a string theorist who left for robotics? Clearly, it must be Barry Kripke. In February 2015, I was stunned when Barry Kripke wrote a paper on light-cone-quantized string theory. He was supposed to do some plasma physics – and was even great at as mundane things as the construction of a wobot that destroys Howard Wolowitz's wobot. So how could Bawwy have the knowledge to write papers about the light cone quantization of string theory?

The Reddit posting makes it so clear that even Penny would understand it – he calls her Woxanna because Penny isn't sexy enough. Barry was trained as a string theorist but at some point, he decided to leave for robotics. I will call him "Barry" and I wish him good luck.

By the way, I have left the Academia (and the U.S.) mostly for reasons that had very little to do with any specifics of string theory which is a reason why I don't want to combine my experience with this story at all. It would seem off-topic to me.

Is a super-bright guy guaranteed to succeed in robotics?

Here, I can't resist to make one point. While people in string theory – but even other fields such as experimental particle physics (statistics in it etc.) – are likely to be able to do many other things, I believe that in most fields of science and technology, the success depends on good luck much more strongly than in high energy physics.

In string theory or experimental HEP, one may feel that he or she has made it to a field that is very selective when it comes to the required intelligence (of the mathematical type). However, in other fields, the intelligence just isn't that paramount and many other characteristics matter a lot. And so does luck.

To be concrete, I do think that while most string theorists belong among the 50,000 smartest people on Earth, most people who made or make the most spectacular advances in robotics only belong among the 10 million smartest folks. It's a much less strict selection that matters there. Because of these numbers, I don't think that a success of a person who switches to a less selective field is guaranteed – although there are obviously examples of greatly successful people who made a switch like that.

Is \(h=15\) too much?

An author of science papers has \(h=15\) if he has authored 15 papers with at least 15 citations, but he doesn't have 16 papers with at least 16 citations. It's a measure that was introduced as a superior compromise between the number of papers and the total number of citations. I think that it is not true that this criterion is applied strictly but I do agree that it is a reasonable rule-of-thumb.

Getting to \(h=15\) in a "predictable" way, by hard work, if you wish, is a lot of work, indeed. But at the same time, I do think it's a reasonable amount of work if the path to join faculty is supposed to be "straightforward". One needs to write "numerous" papers and they can't be "completely invisible".

On the other hand, I would disagree if Barry claimed that that \(h=15\) is a strict condition. In particular, I am pretty sure that if someone found something as clear, as understandably correct, and as revolutionary as the special theory of relativity, he could easily get a faculty job with one famous publication.

It's not a scenario that "most smart students" should assume to be theirs, but neither it should be. But the point I want to make is that some people are naturally more inclined to do a \(h=1\) spectacular work and that's another way to get through the system, too. I will discuss this point in the section about the journey and the destination.

Doing things that are fun: most good HEP scientists are excessively rigorous

Barry told us that he thought that he should have enjoyed some activities and calculations and if he didn't, the following ones would have to be better. But they weren't. So at some moment, he reevaluated whether those activities were fun for him at all or not. And the answer was No.

Do most graduate students and postdocs in string theory enjoy making long calculations or formatting very technical papers? I strongly doubt it. This kind of difficult and "dry" technical work is generally assumed to be a necessary feature of high quality work.

String theorists are among those who have the highest standards. And indeed, the percentage of fundamentally flawed papers – with errors caused by mistakes that would be avoided with more rigor – are rare in the field. Maybe people think it's necessary because of the high competition etc.

Well, I think that string theorists actually heavily overshoot when it comes to the rigor and amount of "hard work" that should be visible in papers. The sometimes "excessive" degree of rigor and hard work in string papers – when combined with the realization that many authors of such papers may have a very hard time to find a job with an average salary – is making me upset, especially if I see how many people are getting lots of easy money for producing (or just shouting) pure junk.

String theorists, even the brightest ones, should realize that at the end, the criteria by which they are hired etc. are more diverse. They should do things that are fun. If they are almost sure about some result – if they have reasons to think that some calculation has to work – they should take the risk and submit the picture even with the less rigorous justification.

The reason why I say so is that I believe that a typical string paper is fully read by a truly tiny number of readers, often comparable to a dozen. One should think a bit rationally. The time you invest into something may turn out to be a wasted time under certain circumstances. Your technical paper is likely to have fewer readers than J.K. Rowling's fantasies but even if the number is small, you should still care about it and it should affect how much time you're investing into it.

All the detailed evidence may be needed if there's actually any uncertainty about the claims. But sometimes there's none. There's no controversy. So spending too much time with too boring calculations and writeups may be unnecessary. Moreover, papers may become less readable when the new key stuff is hidden in the pile of old and not so important material. And another point to make is that the readers who are qualified enough to read your paper at all are usually able to complete the calculations themselves – and they actually need to rediscover many things to become true converts. When the amount of rigor and boring hard work may be lowered so that it makes string theorists more relaxed and the quality of the work doesn't suffer much, it should be lowered and the community should collectively adapt to such a new format.

I believe that even with such a change, it would still be able to figure out what's right and what's wrong and who's a really good researcher and who's less so. Also, I believe that string theorists should realize that to some extent, even things like their charisma play some role in their jobs etc.

More generally, when it comes to their own well-being, people should think about their happiness, not about hiring and jobs. Those things may sound similar but are you really happy when you turn yourself into a follower of a career goal? Happiness often results from simpler things. I feel sort of happy. And I even believe that Stephen Hawking is a happy man!

The string theory community has to do lots of hard work for it to remain the subcommunity of the scientific community that has the best reason to consider itself #1. On the other hand, it's likely that it may do things less rigorously than now (although one could argue that the standards have already decreased in the recent decade or so). And maybe, it should because the happiness of the people is important, too, and too much boring work may make people less happy than certain more relaxed activities.

Bullšiting about very big questions has mostly been a greater fun than writing detailed pages of papers whose main message was known not to be a game-changer. ;-)

The destinations, not just the journey, is what makes string theory amazing

Barry claims that "string theory is all journey" and he realized that he didn't like the journey. Previously, he thought that the he would find some satisfaction in the "destinations" but there weren't any.

If you ignore the cynical message, his poetic wording sounds impressive – these are quotes that you could carve in stone. But when I think about the content, I don't really agree with it at all.

When you drive your car, you may spend lots of time on the road. But the road – and the asphalt on it – isn't really the purpose of the driving. You have some destinations. Sometimes you visit places, friends, or businesses that are not the most important or likable ones on Earth. But they are more important then the road itself!

It is perfectly fine to be bored in the middle of some boring calculation (on the road) as long as there are points – victories – that bring you satisfaction.

The most important destination is the holy grail, the "completion of a theory of everything". This destination sounds OK as a slogan but this goal isn't really well-defined. In practice, it can mean several great goals. The phenomenological portion of the holy grail is to find the right compactification of string theory that allows one to calculate all the observed features and parameters of observed particle physics. The formal portion of the holy grail is to find the complete definition of string theory and understand why it encompasses all the aspects and features of string theory that have been uncovered so far; perhaps, this achievement would unify and hack all the beautiful mathematics that can exist. One could add the "mini holy grail" to crack all the black hole information mysteries and similar things, too.

Now, the primary holy grail hasn't been found and it's totally plausible if not likely that it won't be found in the next 10 or 20 years, either, if ever. But that doesn't mean that this dream (these dreams) isn't motivating young scientists. Even though I am very far from an unlimited optimist when it comes to this dream, I still count myself as the "dreamer". And I am still eagerly waiting for someone fresh (and maybe someone very experienced!) who brings us the holy grail in 2015. And if that won't work, in 2016. And then... you know how to add 1. I still want to do it myself although it's hard.

As far as I can say, it is perfectly fine for young folks to be motivated by this big dream. After some time, they may find out that the dream is very far and lose their hope completely. In that case, they may decide to do something else (slightly different or something completely different), too. But I think it's right when people keep on trying to find really important things. One of them – perhaps a very young guy (or babe, to make it even more spicy) – may very well succeed.

But as I mentioned, string theory isn't just about one big dream. It has many spectacular results that are just "one level less groundbreaking" than the holy grail. And I am extremely thrilled by those, too, even by those that were found or that I learned 20 years ago. Similar things are being found in recent years, too. Perhaps the frequency is lower than it was in the mid 1980s or mid 1990s but the field keeps on moving. And the rate may accelerate abruptly in the future is someone unlocks a new room of marvelous insights.

When Barry says that "it is all journey", it is hard not to imagine that he actually means that "he would never find some important results" or "he couldn't recognize important results found by others". If true, it's bad for him. But many important results – not extremely far from the holy grail – have been found and are being found. And those results bring people the satisfaction that may compensate the dissatisfaction from many and many hours of the boring ride on the superhighway.

As a driver, you shouldn't be devastated by the need to spend some time on the road. As a string theorist, you mustn't throw up when you need to do some complicated calculations at some point. But it's simply not true that this boring stuff is everything that you do or the motivation why you do it.

Nima's great discovery each three years

Nima Arkani-Hamed is quoted as saying that one makes a very cool discovery once per three years in average. Well, I would just say that Nima's own frequency has been substantially higher than that, according to my definition of a very cool discovery. He's pretty amazing.

By the way, if this estimate were right, it would imply that the usual job contracts are approximately appropriately long. If you're hired as a postdoc for three years and if you do a great discovery each 3 years, it's great because chances are high that you will make a great discovery during the job – and you may get another job later. In that case, you don't need to spend too much time with the boring work. ;-)

There is no guarantee that you will make such a discovery in those 3 years but you know, the world is a risky place, anyway.

I want to reiterate my point that the mechanical production of papers – with lots of nearly copied stuff and a little percentage of new original content, as Barry wrote elsewhere – isn't the unique strategy how to do physics. Many people might agree that it describes their attitude – this may include most of the people who would agree that they're doing the research because of the career – but other people have other attitudes and motives.

Now, someone who has spent time at Stanford, Princeton, and Harvard may have gotten all the great grades and other official stellar endorsements and safely belong among the 20,000 or 50,000 smartest people in the world. But I am afraid that if he thinks that the hard mechanical work is how one has to do things to succeed, it may be because he doesn't really have the extra X Factor, some extra mysterious talent (perhaps combined with some specific dose of independence) that some colleagues may have, however.

Even if most people need to work hard and organize their work and dreams as almost everyone else, geniuses sometimes emerge. Sorry that not everyone belongs among them.

Do too many people learn string theory?

Barry says that just the string theory PhDs from Princeton would easily fill all the string theory jobs in the U.S. That's why the competition is so extreme. And Barry's implicit (or explicit?) recommendation is that fewer people should learn string theory.

I disagree with these assertions, too. Relatively to the importance and clarity of string theory, the number of people in the world who work hard to learn string theory is painfully tiny. I think that less than 5,000 people in the history of the whole mankind could "mostly credibly" claim that they had learned string theory at the level of a two-semester graduate course or better. This is less than one person in one million!

And I think that it's right when lots of people learn string theory even if they are going to do something else. Many people learn so many other things that they don't do for living. Just count how many people do useless things such as sports even though they don't earn even a fraction of what Messi does. String theory may be more demanding (time, talents) but I think that the number of people who should try this path should be much higher than a few thousand. People should only be realistically told how many candidates per job are out there.

Competition is too strong, the number of jobs is low

The number of jobs is low. I agree with that. But first, I think that the number of jobs for theoretical physicists should probably be substantially higher than it is today. Second, I think that almost all people who are hired to do a similar kind of fundamental or formal high-energy theoretical physics simply should be expected to have mastered string theory. It doesn't mean that they will work on actual string theory throughout their lives but I think that the mastery of string theory is a healthy and sensible filter for similar jobs. Pretty much all string theorists must learn "some" basics of particle physics. I don't see a reason why the converse shouldn't be expected from (those considering themselves primarily) particle theorists. It's 2015 now!

In other words, I think that the jobs are being stolen from string theorists by occupations that are "totally outside theoretical physics" as well as "occupations inside the broader theoretical physics" where the demands are lower.

Try to estimate the number of people in the world who are collecting taxes, for example. Czechia is employing about 15,000 people who are collecting taxes. Do you really think that the number of paid string theorists below 5 is too high? Well, I probably won't create new jobs for Princeton string theory PhDs by revealing this comparison. But I am still writing these numbers because if the public were a bit more reasonable and cultural, I actually would create them by this paragraph! ;-) At any rate, the tax system could be simplified. Some streamlined computer-assisted universal value-added tax could be administered by 5 people and the remaining 15,000 could do string theory. That would surely be a better world.

Barry says that the competition is too intense and it's bad. Barry and other commenters write that it burns people out and it is responsible for making people writing papers which are about the quantity, not breakthroughs.

Well, I partly agree with the former. Too much competition makes most people exhausted. But this is true in any field and there are many fields with strong competition. On the other hand, if someone arrives and makes a huge single contribution, he will be able to avoid the stereotypes and find a different route through the system. And if he doesn't, he may always choose to do something else than a similar career even though he has learned string theory. For this reason, the excessive quantitative work that someone doesn't like is a self-inflicted injury.

Concerning the second point, I don't agree with it at all. I don't believe that higher competition may reduce the number of breakthroughs. It still makes people work harder, spend more time and energy with the research. This unavoidably increases the probability that a big discovery is made – even if some of the discoveries could be made by exhausted people.

There is this idea that Albert Einstein is relaxed, he has lots of time for things in the patent office, lots of time for women etc., and that's the best setup to do science. No real job competition etc. Well, I don't think that this reasoning is right. Albert Einstein was highly creative and extremely interested in physics. He would have almost certainly made the discovery if he were hired as a physicist before 1905, too. These days, he could see some young colleagues who write not so original papers. But that doesn't mean that he would be doing the same thing. He would surely insist on his relaxed approach to physics – and freedom to do the same things with the women. He always did. ;-)

String theory is a bad name: the subject is huge

String theory is a name of the theory that has undergone the second revolution and it turned out that in a more accurate and complete understanding of the theory, strings are just some quasi-fundamental objects that are really important in some limits of the theory. But the theory has other limits and in these other limits and in the bulk of the configuration space, there are many objects that are as important as strings.

The actual theory is "the theory formerly known as strings," as Michael Duff liked to say. It's a fun slogan but it is unusable as a name. "M-theory" was thought to replace the term "string theory" for a while but it was at the times when people thought that the understanding of the 11-dimensional (more dimensions than string theory) limit would solve all other mysteries about the whole theory.

This expectation was shown to be incorrect. M-theory is really "just another limit" that, unlike the five 10-dimensional supersymmetric string vacua, has 11 spacetime dimensions (and it contains no strings, just M2-branes and M5-branes). Since these realizations, string theorists use the term "M-theory" only for descriptions of/and situations in string theory that are UV-complete in the sense of string theory but where some 11-dimensional supergravity may be seen as a long-distance limit or an approximate description.

The full theory including the new M-theory limit(s) is still called "string theory" these days. However, the term "string theory" is singular. There is one string theory while before the mid 1990s when string dualities were discovered, people thought that they were studying many string theories.

To summarize, the term "string theory" isn't quite capturing what the theory is according to our today's understanding but it's just a technical term that means something, a very specific theory, it has a damn good reason – the strings are still very important – and I don't really see a reason to change it.

Barry also suggests that the term "string theory" is bad because the "actual topics that string theorists study are wider". Well, when they do something that isn't "quite" string theory or that isn't string theory at all, they shouldn't call it string theory! They may still call themselves string theorists either because they're simply proud about it ;-) or because, at least in some cases, their knowledge of string theory helps them to do something.

But I still think that every string theorist more or less agrees what is "certainly" string theory, what is "partly" string theory (or "inspired by" string theory or "generalized" string theory), and what "is not" string theory at all. My feeling is that there's too much work that "is not" string theory at all and too little work on the truly stringy, foundational issues. I blame the anti-string crackpots and the atmosphere they have created to be the main reason why it's so. But of course, I don't have any reliable proof that they are the main reason. In the past, the activity in "purely stringy" topics has temporarily decreased even though there were no Swolins and Šmoits around.

Barry says that "string theory" is wrong also because string theorists work on things not linked to explicit strings; he suggests "formal particle theory" or "fundamental theory" etc. Well, these terms are also being used but they just mean something else. They are more general. Someone doing conceptual work involving quantum field theory may be a "formal particle theorist" but he doesn't need to be a "string theorist" at all. So I don't see any problem at all. What I implicitly see behind Barry's proposal is to try to make string theory itself invisible as if it were a taboo. If that's so, I surely oppose this suggestion as aggressively as I can. It's not a taboo. It's really the main gem of the "formal particle theory" in the recent 20, 30, and maybe 50 years. But it's a gem, not the whole thing.

Also, sort of independently of that, Barry suggests that "string theory" – even if it is really string theory – is too big a subject and it should be split. So he wants to talk about dualists, mirror symmetrists, supergravitists, D-branists, and similar people. This is ludicrous. People study dualities or mirror symmetry or supergravity or D-brane etc. But those topics usually heavily overlap so they can't quite define where they belong. If they need to describe their research in more detail, they can do it by adding some extra words. But there is no need to coin these new -ists words that sound like ideologies (feminism, Nazism, communism, environmentalism).

Moreover, I don't like what Barry says for another, albeit related, reason. He really proposes to create overspecialized small boxes where people "sharply" fit. I think it's completely counterproductive. This would really be a way to encourage unoriginal derivative work. String theory is a rich subject but it is not so wide to justify the near isolation of its subfields or the non-existence of contributions that cannot be categorized. "Interdisciplinary" (this is a silly overstatement) papers within string theory are surely important and new subdisciplines and intermediate disciplines within string theory may and do regularly emerge. It's really one of the goals of the research. Barry's proposals are proposals to create exactly the kind of stagnating environment that he claimed to have hated!

Also, the term "string theorist" is fine enough to be useful for the laymen or other scientists. They wouldn't really understand how a supergravist differs from a D-branist.

Lisi's writing is bogus

Barry also mentions that Lisi's writing is incomplete or bogus according to (almost?) all string theorists. Good that it's being accepted that people agree about this simple point. But Barry hasn't ever read Lisi's paper. I find such a remark surprising. Not because Lisi's paper is truly important to be read. It's not.

But the amount of attention that was dedicated to Lisi's stuff among the "physics fans" of some type has been so intense that I can hardly imagine that I would resist the temptation to look what this stuff was all about. And of course, TRF reviewed that paper long before all the media. I had an expectation that it's probably wrong – a naive text by someone who doesn't know many important graduate-school-level things about particle physics – as soon as I read the (ambitious) title or whatever I saw at the beginning.

On the other hand, I have always been attracted to papers that had a chance to be groundbreaking (whether they are explicitly presented as stringy papers or not). Each of them could be an important new destination, a station we need to visit before the holy grail sometime in the future. The next string revolution could be ignited by someone who superficially looks like Lisi – or at least someone who likes to surf on Hawaii. ;-) I simply had to see Lisi's paper myself to be sure whether or not there was something promising in it; other people's testimonies wouldn't satisfy me. It seems to me Barry couldn't have known about this excitement about the possible game-changing discoveries. He probably saw just the asphalt everywhere, indeed, which is why it may have been wise for him to abandon the field.

But other young people thankfully haven't and many of them keep on seeing the green scenery, castles, butterflies, and rainbows. I wish them at least as much good luck as I wish to Barry.

No comments:

Post a Comment