This week’s New Yorker has an article about the controversy over string theory, written by Jim Holt, with the title Unstrung. On the web-site there’s also a link to Woody Allen’s 2003 humorous New Yorker piece on string theory, Strung Out.
The New Yorker article pretty much gets the story right, although the description of the Bogdanov affair isn’t completely accurate. The Bogdanov papers were about quantum gravity, but were not string theory papers (although they claimed to be motivated by string theory, and at least one referee described their results this way). Holt also describes members of the Harvard string theory group as unsure whether the papers were a fraud or sincere, which does correspond to an e-mail that circulated at the time. However he doesn’t mention that at least one member of the Harvard string theory group to this day not only believes the Bogdanov papers were written sincerely, but considers them to be serious scientific research (an opinion shared by very few others).
Holt accurately describes Smolin’s book as more accessible than mine, then chooses a very good example of an “indigestible” sentence from my book:
The Hilbert space of the Wess-Zumino-Witten model is a representation not only of the Kac-Moody group, but of the group of conformal transformations as well.
That is an example of some of the very advanced material I tried to include in a few places in the book. It’s the precise expression of the mathematical relationship of representation theory and QFT that has been worked out in recent decades in two dimensions, exactly the thing that I would argue we should be trying to understand in the physical case of four dimensions. To the extent that the book contains a positive argument about alternatives to string theory, my decision was not to over-hype it, but to try and explain a point of view about the history of the relation of mathematics and quantum field theory that implicitly leads to this way of thinking.
Also out today is an article by JR Minkel on the Scientific American web-site entitled That’s Debatable: Six Debates at the Frontier of Science. The first of the debates listed by Minkel is Is String Theory Unraveling?, and it’s largely about the landscape. It includes a couple quotes from me, as often the case a bit abbreviated to make them sound even more provocative than I intended…
Update: The usual sensible commentary on the New Yorker review from Lubos. Holt is a “cretin from the garbage bin of the journalistic colleges”, I’m the “black crackpot” (due to the color of the cover of my book, Smolin is the “blue crackpot”). Lubos reports on the reaction to the review from “one of the leading physicists of the current world” (presumably one of his colleagues):
What’s wrong with these people? Why don’t they choose f***ing instead of writing about things that they don’t like and they don’t understand?
Update: The story has made it to Slashdot.
This is a pretty disappointing article. Take
“Nowadays,” one established figure in the field has said, “if you’re a hot-shot young string theorist you’ve got it made.
for example. The quote is from a number of years ago. These days, if you’re a hotshot young astrophysicist or phenomenologist, you’ve got it made. String theory jobs are getting harder to come by.
And, regarding Friedel’s e-mail about the Harvard string theory group, frankly I just don’t believe it. Pretty much everyone I know who looked at the papers agreed they were nonsense. I find it almost impossible to believe that the Harvard faculty wouldn’t agree. Even Lubos’s ravings on the subject are to the extent that there might be some ideas in there, not that the mathematical content of the papers is at all coherent.
I’m a bit tired of all this talk about how we should cultivate ‘seers’ or ‘valley crossers’ or perhaps ‘people Smolin likes’ without any concrete suggestions on how we change the current incentive system in physics. I don’t think there is this great division of people between the mindless string theory computation-o-trons and the deep thinking everyone else. What there is is a lot of smart people working under a strong incentive to develop a long publication record. If you want people to spend more time on hard questions, you’re going to have to find a way to have it not hurt their job prospects.
Peter,
You write frequently about representation theory and QFT, so might it not be a generally good idea to write something pedagogical for would-be physics grad students before they prematurely commit themselves to any specific research direction.
In fact, I was thinking it may be even more helpful to write a ‘Resource Letter’ along with the topic-intro and post both on the arXiv. This may help to disseminate your idea(s) to a wider reader base, especially given the climate of doubt which seems to have settled within the string theory community.
Adieu,
Stefan
Aaron,
Do you really think that a young grad student or post-doc who came up with something new about string theory that got a significant amount of attention would have trouble finding a job now? That’s not my impression, but maybe I’m wrong.
Smolin and I do both make concrete suggestions about how to change the incentive structure. To get them taken seriously though will first require that leaders of the field acknowledge that there’s a problem. Besides positive incentives, I also fear that negative incentives may be part of the solution. If people decide they won’t be able to get a permanent job by producing a long list of unambitious papers on trendy subjects, maybe they’ll do something else. I would love to see the NSF/DOE start to take this issue seriously and start a discussion about their role in the current incentive structure and how to change it.
I don’t have a problem with Smolin’s “valley crossers” analogy, but I will admit that I’m congenitally dubious about “seers”.
The e-mail in question doesn’t refer to “Harvard Faculty”, it refers to the string theory group as a whole, and presumably the discussion being reported was largely among postdocs. Lubos was one of them at the time and, given his later behavior, I wouldn’t be at all surprised if his comments were one of things the writer had heard and was referring to.
Stefan,
Yes, I would very much like to write something expository about QFT and representation theory, starting with something basic about QM and representation theory. This spring it looks like I’ll be again teaching the second half of our graduate course on representation theory. Last time I did this I wrote up some preliminary notes which are on my web-page. This semester I hope to have time to improve those, and add some more on the topic of the relation of QM to the subject.
Let me know when you finish the work.
Do you really think that a young grad student or post-doc who came up with something new about string theory that got a significant amount of attention would have trouble finding a job now? That’s not my impression, but maybe I’m wrong.
Depends on how new it is. I can tell you that I know a fair number of long term post-docs who have done very good work, but are deeply worried about their job prospects.
Smolin and I do both make concrete suggestions about how to change the incentive structure.
Can you please tell me what they are, then? I think it’d be a lot more helpful than just continually saying how much string theory has failed to do. I’d love to hear some really specific ideas.
Lubos was one of them at the time and, given his later behavior, I wouldn’t be at all surprised if his comments were one of things the writer had heard and was referring to.
And so again Lubos somehow becomes representative of the entire field? I swear, if Lubos didn’t exist, you guys would be forced to invent him.
Aaron,
Lee has various suggestions, I don’t have his book at hand. I assume one of his main ones would be to basically implement what he has been doing at the Perimeter Institute on a larger scale, in more places.
Among the things I’ve suggested that people should consider are:
1. Giving people directly out of graduate school longer term postdocs (e.g. 5-6 years), so they have more than 1-2 years in which to come up with something for their next job.
2. Graduate student birth control, bringing the ratio of Ph.Ds to jobs to something reasonable, so that the job market is not so insanely competitive and people are more likely to feel that they can have a future in the field even if they don’t work on the latest, hottest topic.
3. Senior theorists need to stop putting students to work on the latest, trendiest string theory topic, encourage their students to work on a wider variety of things. At the same time they need to change their standards for hiring postdocs and junior faculty, making it clear to applicants that they want to see original ideas, not the same thing everyone else is doing.
4. The NSF/DOE should explicitly admit that particle theory research is in trouble, give guidance to people reviewing proposals that copycat proposals on the latest string theory topic will not be funded, and emphasize that priority will be given to diversity, that proposing to do something different will be a lot more likely to get you funded. This applies to grants for workshops/conferences, as well as grants to individuals and theory groups.
As for Lubos, who said he’s representative of the entire field? Unfortunately he does represent the Harvard string theory group, at least he often claims that he represents their opinions. If this is not the case and they’re unhappy with him, I’ve heard nothing about this from any of them, directly or indirectly. They’re the ones who, after dealing with the guy for a couple years as a junior fellow, decided he was the best young person available in the field and offered him a tenure-track position.
Peter,
I had always kind of thought that the overproduction of graduate students was a means of keeping wages low and therefore reducing the number of full time professors one needed to work in a particular department.
I am not sure to what extent this reasoning is sensible but the overproduction of science PhD’s probably lowers wages in all fields for which science PhD’s are eligible.
As I understand it, the growth in wages of professors has not kept pace with the growth in wages of other middle class professionals such as doctors and lawyers.
I think rather than graduate birth control, there should be more simple truth from the beginning of the system to the end of the system. Isn’t it fair to say universities overhype the potential job opportunities and leave it to the student to eventually stumble into the rather brutal truth?
Didn’t you have to discover for yourself that there were no jobs for nonstring theorists? I find that kind of system fundamentally unfair. You could have probably started thinking about your transition to math years earlier and perhaps had a smoother transition.
Hi Peter,
Senior theorists need to stop putting students to work on the latest, trendiest string theory topic,
That does not only go for string theory. In general, senior theorists should not dominate over the next generation by applying selection criteria that primarily support their interest, instead of criteria that support the most talented researches – even if those might have been working on topics they don’t find exciting at all. The problem is that this goes on long after graduation. Given the fact that the average postdoc is aged around 30, hiring institutions should trust in the candidates ability and responsibility to find promising research fields on their own.
Besides this, its kind of funny that the article picks out exactly the same sentence about the Kac Moody group that I mentioned. I also found the following remark ‘actually, this is a serious over-simplification […]’ very amusing 🙂
Best,
B.
It is true that Smolin’s book is designed for a more general audience. What is amazing is that Woit’s book, despite being fairly technical, is selling enough copies to be number 2 in physics book sales on Amazon after Smolin.
And this concept that the next big advance in physics is going to come from a graduate student is a little iffy. The idea is based on observations of the past, but as time has gone on it has become more and more difficult to reach the frontiers of physics. This alone would suggest that older researchers are more likely to make the advance.
Smolin had an interview where he said that his “seers” and “craftsmen” wasn’t as good as someone else’s “peak climbers” and “valley crossers”, where the peak climbers push the old ideas to their limits, and the valley crossers abandon old peaks for new ones.
TheGraduate,
It’s extremely difficult to get universities interested in the idea of reducing the number of Ph.D.s. The faculty like having large graduate programs so they can teach advanced courses, the universities like having huge numbers of people competing for jobs, they’re basically flooding their own labor pool.
The situation in particle theory is rather unusual for a scientific subfield, with a much worse imbalance between numbers of Ph.Ds. and jobs than any other subfield that I know about. The situation is much much better in mathematics.
I certainly went into graduate study in particle theory knowing that the job situation was quite bad, and then, as now, I assume that most departments and advisors let prospective particle theory Ph.Ds know this (although if they can’t figure it out for themselves, one wonders how bright they are…). String theory was a non-issue when I was a grad student (I finished my thesis and started a postdoc in mid-1984). I can’t say that I was surprised to have trouble finding a job after my first post-doc. I’d say that string theory just turned what would have been a difficult situation into an impossible one, and it was just as well that it made it much more sensible to look for a job in math than to hang around trying to find not very good positions in physics.
Pingback: Unstrung « physics musings
Bee,
Maybe because I’m older than you I’m less of the opinion that the problem is inherently in the judgment of senior people. As far as pushing people into working on bad ideas about string theory (or other trendy subjects), I just don’t see that it’s mainly the older people doing this.
Funny that you chose the same sentence as Holt (although quite possibly he reads your blog…). My book has gotten a lot of criticism from people for some of the more technical things I put in it, but I’m still quite glad I did this. The material is there to provide something new and challenging for almost every one, and Chapter 10 in particular is an effort to get down on paper the story of the successful interaction of math and physics over the last three decades. Not everyone is going to be able to follow this, but I still think it’s a worthwhile effort.
I suppose I’d feel differently if hardly anyone was buying the book because of this. As it is, it seems to be selling all right, with Smolin’s book doing better because it makes many of the same points pitched to a wider audience. This situation is fine with me.
Peter,
I finished reading your book yesterday and I had a few questions. So is S-matrix theory the origin of the string idea in string theory? Do you view string theory then as the modern face of S-matrix theory? Also, it occured to me that the sorts of anthropic reasoning and interest in theological tendencies that afflicted S-matrix theory at the end of it’s theoretical life seems to be the sort of thing affecting string theory now. Was this something you wanted to highlight?
Secondly, I felt like you were emphasizing group representation theory as perhaps having interesting and generally overlooked significance to HEP. Was part the goal of writing the book to encourage more curiousity in that relationship?
I cannot agree with the idea of reducing the number of graduate students as some sort of solution to the problem of excessive competition among would-be professors. Some areas of research allow students to go on to productive careers outside of an university.
andy,
I think most likely if the major universities coercively reduced graduate students then lower tier and fringe universities would start offering PhD’s to any people that still wanted to persue graduate degrees. I think this is already happening. It would just happen more.
The longer postdoc thing might be helpful. I’m not sure that just won’t lead to people churning out lots of papers at the same place rather than multiple places, however. I’d be more inclined to enhancing a sense of job security, but that may be my own anxieties speaking.
The reason, I should say, why string theorists encourage students to work on popular topics is that they know that that’s how jobs are obtained. Perhaps it’s a viscious cycle, but in the current environment, it’s good advice.
As for NSF proposals, I’ve never been involved in one, but in areas of theory are they really directed at specific research proposals? For one, I thought that it was often entire groups that were funded out of these grants. It doesn’t make much sense to me to tie these grants to specific research directions given that things can change rapidly.
As for Lubos, who said he’s representative of the entire field?
Well, somehow we go from Lubos thinks the Bogdanovs might not be completely and utterly full of crap to “even the Harvard string-theory group was said to be unsure” (in that wonderfully weaselly passive voice) to, apparently, the implication that string theory is barely distinguishable from nonsense.
Isn’t it fair to say universities overhype the potential job opportunities and leave it to the student to eventually stumble into the rather brutal truth?
Every place I applied told me that job prospects were crap.
instead of criteria that support the most talented researches – even if those might have been working on topics they don’t find exciting at all
How does one determine the ‘most talented researche[r]s’? That’s the rub, after all. Otherwise, we’d just give them all tenured jobs at 18 and let them do whatever they feel like.
TheGraduate,
S-matrix theory was an approach to the strong interactions that led to the discovery of the Veneziano amplitude, and that led to string theory. The ideology behind S-matrix theory was that QFT could never give a theory of the strong interactions, so had to be abandoned. This turned out to be wrong. Another part of the S-matrix ideology was that symmetries were not fundamental, this also turned out to be wrong. My suspicion is that present day string ideology has the same problems: maybe there is a QFT that gives quantum gravity, and maybe to make more progress on fundamental QFT you need new and better ways of exploiting symmetries. Representation theory is precisely how one exploits symmetries, one point of the book is that this has been the right approach in the past, even when people called for it to be abandoned, and it may yet be the right approach for the future.
andy,
Particle theory is the only area where I am suggesting reducing the numbers of Ph.Ds. I’ve heard all the arguments about why the current system that produces 5-10 times more people trained to do particle theory research than there are jobs doing such research is a good one, but I don’t agree with them. I think a system where it is much harder to get a particle theory Ph.D., but where the people who get ones have a better shot at getting a permanent research job if they do interesting research would be healthier than what we have now.
Aaron,
NSF proposals don’t just say “We’re Prestigious U. and have done good stuff in the past, send cash”, they include descriptions of the research directions people in the theory group intend to pursue. If word got out that the NSF was not looking kindly on certain research directions, that would have a big effect. In the past, I think the NSF has for good reasons been unwilling to ever do this, leaving things up to peer review.
As you know, the reference to the Harvard string theory group comes from an e-mail sent by someone visiting there at the time (I don’t know who this was, do you?). Maybe he or she was making this up or exaggerating, but the fact that one of the Junior Fellows there at the time still can’t see that the Bogdanovs are full of crap, and his colleagues thought he was a promising young genius and gave him a faculty job lends a certain amount of credence to the story.
As you know, the reference to the Harvard string theory group comes from an e-mail sent by someone visiting there at the time (I don’t know who this was, do you?).
In the e-mail the Bogdanovs sent to many people (so I think I can safely call it public knowledge, especially as I put it on my webpage at the time), it was quoted as follows:
Again, I wasn’t there, so I can’t speak of anything that happened there, but given that everyone I know who looked at the papers was ably to discern the problems, what I can do is express my disbelief of this particular e-mail.
The e-mail was forwarded by Laurent Freidel to various people Undoubtedly he, like I, found it rather amusing. But I still don’t know where he got it from.
Peter,
Concerning the limiting of Phd’s, I guess I was thinking if particle theory became any more elite, guys like Einstein and Witten probably wouldn’t make it in. The former not being well rounded academicly at least in his youth and the latter being a history major. It seems to me that the natural way that people would make it more elite is to start expecting even higher grade point averages, more research at the undergrad level, even more stellar recommendation letters etc etc. In other words, the kind of measures that in large part probably describe many of the string theorists and seem to have contributed heavily to why string theory has come to dominate.
Well, since this anonymous comment has now made it into at least one book and an article in the New Yorker, perhaps the writer of the e-mail could come forward and explain what they meant by it.
Until that happens, I don’t have much reason to believe that it bears much resemblance to reality and a lot of reason to believe that it doesn’t.
TheGraduate,
I don’t think Einstein or Witten would have had much trouble meeting a significantly higher standard than the minimum now needed to get a Ph.D. But, sure, the main problem is that the current system is incentivizing and rewarding behavior very different than coming up with good original ideas.
Here is a couple questions I had while reading NEW.
1. String theory in the first simplest version implies that the universe has 26 dimensions. Is it possible to explain in a few sentences or a couple of paragraphs why? (I heard many years ago a short 1-2 slides explanation in a half-popular talk but I completely forgot it.)
2. In the book there is a distinction (for very successful theories from physics) between “convergent series”, “useful divergent series”, and “useless divergent series”. Is there any formal distinction between the last two types? Can it make sense in math?
3. The era before QCD and the standard model is described at least as confused and chaotic as the situation in string theory today. Maybe even more than today. (Peter push the analogy by telling also there about a physist who combined physics with eastern philosophy.) Yet people who promoted these unsuccessful but very dominant theories were not ask to admit failure; moreover, they did not fail: conceptual and technical ideas from these unsuccessful theories turned out useful later, and students of these scientists had crucial role in developing more successful and completely different theories. This looks like a good model to proceed, no?
Gina,
1. Away from d=26 quantization introduces an “anomaly” in the symmetry of conformal transformations of the string worldsheet. Basically this means that the quantum string theory is more complicated and the metric on the world sheet becomes a dynamical variable you have to deal with. People study these “non-critical” string theories also. There are various different calculations that give you the 26, I don’t know of any simple physical explanation for it.
2. Some divergent series are “asymptotic” approximations to some function, which means that, at a fixed order, the truncated series is a better and better approximation to the function as the expansion parameter gets small (even though, at fixed small parameter, as you go to higher order, the series sooner or later diverges). The perturbation series for QED is supposed to be such an asymptotic expansion. This kind of divergent series can be quite useful, giving very good approximations.
3. Actually, the bootstrap program did fail as a theory of the strong interactions, and was pretty much killed off by QCD. Many of the people who worked on it certainly admitted that it failed, other die-hards never did admit this, but stopped being taken seriously by most theorists. Certainly some things learned from this were ultimately useful, but the program failed to do what it was intended to, and sensible people admitted this.
Maybe the superstring phenemenon in theoretical high-energy physics is NOT an aberration in the current world of science,
but
is an example of a more general phenomenon in which a given approach obtains by successful lobbying a dominant position with respect to funds and jobs, and then continues to expand, cancer-like, whether or not any real productive results are ever obtained.
For example, consider AI (artificial intelligence) and a 25 September 2006 article by Andrew Orlowski in the Register at http://www.theregister.co.uk/2006/09/25/bogus_ai/ which article says in part:
“… MIT’s Professor Marvin Minsky’s … linguistics-based approach to AI, symbolic AI, triumphed in the grants lotteries over an approach which preferred to investigate and mimic the neural functions of the brain. Minsky’s non-stop publicity campaign helped ensure his AI lab at MIT was well-rewarded while neural networks starved. …
The Japanese … poured hundreds of millions of dollars into their own AI project, called Fifth Generation. They may as well have buried the money in the Pacific Ocean. Two decades later there are no intelligent robots, and “intelligent” computers are a pipe-dream. …
For the past week reports have again confidently predicted intelligent computers are just around the corner. Rollo Carpenter[‘s]… chatbot Joan won an annual AI prize for creating software that most resembles a human … Here’s Joan, cited at the MIT’s Technology Review … :
Judge: So what you’re saying is that you are not just human but THE smartest human?
Joan: Yes, why, what did you think I was saying? …
Judge: How many degrees in a triangle?
Joan: 180 degrees.
Judge: Is it possible for more than 180 degrees to be in a triangle?
Joan: Probably not. You can change your opinion, but not your constitution.
Judge: It’s not a question of opinion, it’s a matter of physics and maths.
Joan: A five-ounce bird could not carry a one-pound coconut. …
Occasionally Joan will ask the interrogator if they are a robot, or chide them for being stupid. These are all pre-programmed rhetorical tricks. They may bore or bamboozle an interrogator, but this is no indication of intelligence. …”.
It seems to me that the Minsky-worshipping AI community is similar to the Witten-worshipping SuperString community in at least the following ways:
1 – using PR-type publicity to get and maintain dominance in funding and jobs;
2 – starving competing approaches;
3 – no real results after two decades;
4 – calling critics “stupid” (AI using a computer program that is merely, as the Register article says,”a database of conversational snippets”, while SuperStringers are using a Harvard physics professor).
Tony Smith
http://www.valdostamuseum.org/hamsmith/
PS – I am NOT saying that “artificial” intelligence is not possible (for example, maybe it could be achieved by closely studying the way the brain works and applying the results of those studies, or by using arrays of quantum computers, or by a combination of both such approaches, etc). I AM saying that it appears that the Minsky-type approach seems to have many of the same dysfunctional characteristics as SuperString Theory.
we’d just give them all tenured jobs at 18 and let them do whatever they feel like.
I agree with this. I think this was the goal, if any goal can be claimed, of the university uprising in Paris 1968 (also Madrid 1965 and others): never finish the university career, keep always being an student (and get food and shelter).
Also Peter, er, Kropotkine suggested to limit the job requeriments at a maximum of five hours per day, so the rest of the time can be dedicated to science or other leisures. This could be a solution of some appeal for the people forced out of academy but still interested: to try to arbitrate for non-research but part time jobs in the understanding that the rest of the day -say, the evening- one can keep research activity holding the same rights (travel, computer access, conferences, use of desk, coffee and blackboards) that a full time research professor.
Peter,
You say job prospects for math graduate students is very good, but how do number of applicants in math grad programs compare to that of applicants in physics programs. Perhaps there are lot less students applying for math programs? However, I think situation in physics (no. of jobs and no. of wanna-be physicists) is much better than humanities where apparently one needs to be already quite accomplished to get into graduate programs.
a,
I didn’t say job prospects for math Ph.Ds are very good, just that they’re quite a bit better than for particle theorists. I don’t know of any significant difference in how hard it is to get into physics and math graduate programs. I also know of no evidence that it is harder to get into graduate programs in the humanities. Again, what is at issue here is the very bad job situation in one particular subfield of physics, particle theory. The situation in math and in other subfields is mostly much better.
a,
The job situation isn’t that great for math PhD’s as far as I understand it. The AMS has yearly reports so if you google a little bit you should find articles in the Notices about it.
There is a general oversupply of science PhD’s and this keeps wages down not just for professors but for post docs and graduate students and research assistants both in the universities and in industry.
I think this state of affairs is not something the universities really have any incentive to address in any way whatsoever.
I know that people will eventually stop being interested in science as the bad career prospects become more generally known. The computer science departments have already had catastrophic losses in popularity.
To All:
Does anyone here happen to know the current prospects for (any of the following):
a) computational science / mathematical programming;
b) software engineering / consultancy;
c) computer science;
d) computer hardware / electrical engineering and related;
e) biophysics;
f) physics-related Wall Street type jobs (is it called ‘actuarial science’ by any chance)?
Stefan
Stefan,
I got lucky and am in my 2nd year in a tenure track position in biophysics. From my experience during my job search it seems to me that, if your goal is a tenure track position at a research university, the job market is pretty awful across the board. Part of this may be due to the fact that the biological sciences flood the market with Ph.D.’s to an even greater degree than physics. During 3 rounds of searching (3rd time’s a charm), I interviewed at Chemistry, Physics, Biochemistry, Biophysics and Molecular Biology departments (all told I must have sent out >200 applications). On average there were ~300 applicants for every position and I’m told that this is normal.
Overproduction of Ph.D.’s is a problem in all of the sciences. Personally I think that part of the answer is to ecourage the creation of research options other than the holy grail of the tenure track position. Some of the most talented people I know simply want to do research and don’t want the hassle of running and funding a whole group. As it is there are very few places for these people. The current system essentially forces you to find a tenure track job within a certain amount of time after your dissertation or else leave academic science entirely. I’d like to see the creation of more quasi-permanent research associate type positions. That way a person could pursue research and enjoy a reasonable degree of job security without landing one of the few available faculty positions. I suspect that science has lost a great deal of talent due to the “up or out” philosophy. And let’s face it, the skills that make you good at building and funding a research group are not necessarily correlated with the skills that allow you to do important and original science.
Stefan,
Concerning the outlook for engineering, I found this webcast very interesting: http://mitworld.mit.edu/video/321/
(I hear one can hire 10 Indian engineers for the price of one American one.)
Andrew Wiles (algebraicnumber theorist) and Edward Witten (mathematical physicist) should write an article together about their work and lives as academics who have had a profound impact on their subject. There’s a really good book on becoming a mathematician called A Mathematician’s Survival Guide: Graduate School and Early Career Development which I’ve read at the UW Mathematics Research Library. I really like Dr. Woit’s biography of Edward Witten in his book exspecially the idea that he might be from outerspace, because he disappeared on Dr. Woit.
Timothy,
I certainly think it would be great if Witten would write a memoir about his professional life, that would be fascinating. The part in the book about Witten possibly being from a superior race of non-humans was a joke (I think…)
Patrick,
I think all the signals seem to suggest that the US is planning to buy its science from China and India one day. Everything seems to point to significant overfunding of ‘stuff’ (super computers, huge new buildings etc) and significant underfunding of people.
Tim:
The survival guide is excellent. Krantz is interesting … kind of ideosyncratic … naming some of his hypothetical people in the book after porn stars etc … last information that I saw on him, he was working on the mathematics of plastic surgery
Dr. Woit,
I liked your reason for the thought because it sounded reasonable and it was a little fun part to read. A mention of a person thought to maybe teleporting when no one was thought to be watching in a non-fiction book about the wonderful connection between physics and mathematics and the problems with string theory was a great thing to put on paper especially since it was Witten.
What I don’t understand about Witten is why he believes that he can conclude that strings and extra-dimensions do exist and that tring theory is testable. To me it he is taking Einstein’s view that General Theory of Relativity was just too beautiful to be wrong and applied it string theory which can be made to say just about anything. Couldn’t someone have overlooked the mathematics and missed something like what happened with Andrew Wiles when Nick Katz was reviewing his argument for the construction of an Euler system to give the precise upper bound for the size of the Selmer group, in 1992? – http://www.pbs.org/wgbh/nova/elegant/view-witten.html
If Witten writes a memoir about his professional life, I hope he talks about how he was able to conclude that strings really exist.
Timothy,
I’ve never heard Witten claim that string theory in its present state is testable, or that he is sure strings exist. As far as I know, his point of view is that string theory is an idea worth pursuing for various reasons, in its present state it may very well be wrong, but there is enough interesting structure to what has been learned about string theory to believe that it is somehow connected to reality.
There may very well be something crucially wrong in the standard arguments about string theory. Unfortunately the theory is not precisely defined the way Wiles’s argument was, so you can’t just go through it and look carefull for holes in the argument. The problem is that there are lots of possible holes…
Maybe the Simons model is the way to go: make some money by deploying the natural smarts characteristic of the profoundly numerate, then get down and do what you want. (Did Gibbs do this?) All told, though, this looks like a harsh option; to be at once both Gauss and Brunswick, or Hardy and Ramanujan? Maybe science will revert to the hobby status it had 200 years ago. That, at least, would weed out the supernumerary doctors among us.
The hilarious part of all this is that right after hearing a bunch of academics complaining about how hard academic jobs are to obtain because of excessive competition, they go into what crappy jobs they are in terms of pay and work hours. You’d think that a freshman class in economics would correct their thinking, but I figure that what they really need is a couple years hard work in the real world.
“NOVA: It seems like the standard criticism of string theory is that it isn’t testable. How do you respond to that criticism?
Witten: One very important aspect of string theory is definitely testable. That was the prediction of supersymmetry, which emerged from string theory in the early ’70s. Experimentalists are still trying to test it. It hasn’t been proved that supersymmetry is right. But there is a very precise relationship among the interaction rates of different kinds of particles which follows from supersymmetry and which has been tested successfully. Because of that and a variety of other clues, many physicists do suspect that our present decade is the decade when supersymmetry will be discovered. Supersymmetry is a very big prediction; it would be interesting to delve into history and try to see any theory that ever made as big a prediction as that.”
“NOVA: Do you think it’s possible that string theory will turn out to be wrong, or at least some branch of knowledge that just isn’t connected to nature?
Witten: I guess it’s possible that string theory could be wrong. But if it is in fact wrong, it’s amazing that it’s been so rich and has survived so many brushes with catastrophe and has linked up with the established physical theories in so many ways, providing so many new insights about them. I wouldn’t have thought that a wrong theory should lead us to understand better the ordinary quantum field theories or to have new insights about the quantum states of black holes.
The question reminds me a little bit of the question about interpreting fossils. When fossils were first explored 100 or 200 years ago, some people thought they were traces of past life that had survived in the rocks and others thought that they had been placed there at the creation of the universe by the creator in order to test our faith. So I guess string theory might be wrong, but it would seem like a kind of cosmic conspiracy.”
– http://www.pbs.org/wgbh/nova/elegant/view-witten.html
Dr. Woit has Witten said anything about your book?
Timothy,
I didn’t realize that Witten had ever made that definite a claim string theory was testable by looking for supersymmetry. It would be interesting to know if he still would claim this, will also be interesting to see what he has to say after the LHC results are in, if they don’t see supersymmetry. He could try and weasel out of his claim by saying that the supersymmetry breaking scale might just be higher than can be observed at the LHC, but that kind of non-straight-forward slipperiness doesn’t normally seem to be his style.
Maybe he will finally admit that string theory doesn’t work if the LHC doesn’t see supersymmetry.
I sent him an early version of the manuscript, he wrote back with some comments, basically saying he disagreed with me, but that there were a lot of books out there he disagreed with. He also gave some of the standard arguments for supersymmetry and string theory, but he didn’t claim that string theory was testable by looking for supersymmetry. This was now a couple years ago, I haven’t discussed this with him recently.
Okay now I’m just confused. Is Witten saying that if he assumes that strings exist then one of the conclusions is that there are extra-dimensions? If so then why does he implie that the existence of strings is very likely because otherwise it would seem like nature was playing tricks with us?
Timothy,
If you want to know more about the standard story about string theory that Witten is one of the main people responsible for promoting, you should read one of the many books on the subject, like Brian Greene’s, which answer questions like yours.
Timothy,
You’re kind of taking over this comment section, please stop.
Thanks, Peter, for the explanations. On the third point you wrote
“Actually, the bootstrap program did fail as a theory of the strong interactions, and was pretty much killed off by QCD. Many of the people who worked on it certainly admitted that it failed, other die-hards never did admit this, but stopped being taken seriously by most theorists. Certainly some things learned from this were ultimately useful, but the program failed to do what it was intended to, and sensible people admitted this. ”
Let’s take the analogy between string theory and the “bootstrap pogram” a little further. Judging from your attitude towards string theory, it looks that you would have expected proponents of the “bootstrap program” to admit failure even *before* an alternative successful theory (QCD) came along. This seems unreasonable.
It looks that the “bootstrap program” was a reasonable line of attack before QCD and had important contributions in any case. (You can thus say that Chew and the bootstrap people’s were “false” but they were not “wrong” in the sense that these people did, as scientists, the right thing.)
Moreover, a theory “like” QCD as the solution
for strong interaction was something people considered.(In NEW you described how Gross almost succeeded to prove this task is impossible… just before he proved it being possible.) So for the case of strong interaction there was a reasonable alternative possibility of some quantum field theory being successful. Some people declared prematurely) this possibility as dead (from reasons that were and probably still are of some interest and value), but they were wrong.
On the other hand for the case of string theory there isn’t really any “shelf” alternatives so this makes the situation even more favorable to string theory.
Peter,
Could you take a moment to define the ‘bootstrap program’ ? If I’m not mistaken bootstrap is a bit of bayesian statistics which also seems to be the thinking behind the landscape. Can you discuss the relationship between the bootstrap program and the Landscape if there is one?