Is Science Near Its Limits?

The past two days I’ve been at a conference here in Lisbon organized by the Gulbenkian Foundation on the ostensible topic of Is Science Near Its Limits? The Gulbenkian is probably the most well-known and best-funded cultural organization in Portugal, and it includes a world-famous museum housing the wonderful art collection of its founder, Calouste Gulbenkian, who made his fortune in the oil business early during the last century.

The conference was extremely well-run and well-attended, filling a large lecture hall where there was simultaneous translation of the talks into Portuguese. It was organized by literary critic, writer and polymath George Steiner, who gave the introductory talk. I hadn’t known that Steiner had originally started out studying mathematics, but was discouraged from pursuing a career in the subject at the University of Chicago by Irving Kaplansky, which led to his turning to the study of literature and philosophy. Steiner had quite a lot to say provocative to scientists, including questioning whether they had been able to justify to the public the large sums of money being spent on the LHC, and characterizing the lack of testability of string theory as strong evidence that science had hit a limit beyond which it could not progress.

On the whole the rest of the speakers actually didn’t have much to say about limits of science, taking the standard view of most scientists that their own field had a bright future, with no limits in sight. The final talk of the conference did return to the limits issue, with John Horgan giving an uncompromising defense of the thesis of his 1996 book The End of Science (although he did allow that possible advances in neuroscience such as the decoding of a neural code, could be as revolutionary as previous advances). While the scientists in the audience took Steiner’s attacks in stride, partly because he was our host, they were less charmed by Horgan, who got a rather hostile reaction from many of them. I hope he’ll write about his point of view on the conference at his blog, or discuss it in one of his Bloggingheads discussions with George Johnson.

I was one of the few other speakers discussing the question of limits, with my talk emphasizing that particle physics is now in a new, different environment than that of the past, one in which progress, even revolutionary progress, is possible, but much more difficult. A written version of my talk is available here. I was paired with string theorist Dieter Lust, who gave a presentation of the case for string theory unification and the Landscape. We were introduced by Gustavo Calstelo Branco of the IST, who emphasized recent advances in our understanding of neutrinos. Also speaking in another session was Luis Alvarez-Gaume of CERN, who gave a very upbeat talk on the prospects for particle physics, taking the point of view on string theory that, like any idea, string theorists will give up on it if it doesn’t work out. He already sees a diminishment of interest in string theory among particle physicists, with people moving instead towards subjects that promise some sort of interaction with experimental data. The three of us were brought together later for an interesting small and very lively discussion of the issues surrounding string theory and recent media attention to it. This was taped, and may appear in some form or other in the future.

Update: There’s an entertaining conversation between John Horgan and George Johnson about the Lisbon conference now up at Bloggingheads.

This entry was posted in Uncategorized. Bookmark the permalink.

61 Responses to Is Science Near Its Limits?

  1. mclaren says:

    One important point which it seems many folks have overlooked is Dr. Woit’s focus on the fact that some of the biggest outstanding problems in current particle physics have little to do with larger accelerator energies.
    Einstein’s theory of gravitation remains a fundamentally classical geometrical model, whereas quantum theory is quite different. Einstein’s tensors do not involve expectation values, there is no uncertainty principle involved, and above all GR is a continuous theory — quanta are discrete.
    Arguably the biggest unresolved problem in modern physics involves finding a way to mathematically unify these radically incompatible mental models of reality. Every time someone has tried, the math has blown up.
    This suggests several possibilities. First, we need a lot more and a lot better math than we currently have. Considering the immense sophistication of current mathematics, that’s non-trivial. And we probably also need some radical new conceptual leap, something different in kind from current mental models of the universe, in order to successfully unify GR and quantum theory. One of the most powerful criticisms I’ve seen of string “theoy” is that it merely tacks on extra dimensions to current mental models. I.e., it takes pointlike particles and draws ’em out into circular strings in n dimensions. That’s not a fundamental conceptual leap. Planck’s conjecture that energy comes in packets did represent a wild leap beyond 19th century mathematical models of radiation, specifically, beyond Maxwell’s equations.
    John Baez asked:
    “As for being “near the limits of science”, does anyone here actually think we are?
    The phrase “limits of sicence” was certainly poorly chosen for the conference. But yes, we are probably at or near the limits of terrestrial accelerator-based particle physics today. Future progress in particle physics is likely to come from astronomical sources. One advantage: such experiments will likely prove much less expensive than building accelerators, since they mostly involve dumping large amounts of ultrapure liquids of various kinds (cleaning fluid, mineral oil, et al.) into big underground tanks surrounded with photodetectors.
    We have clearly not approaches the limits of science in other fields. Molecular biology, materials science, astronomy, etc. remain wide open.

  2. Michael Gogins says:

    I read “String Theory and the Crisis in Particle Physics” with great interest and appreciation. I am not sure that the cutting edge of physics already has raised the bar of intellectual difficulty over Newton and Einstein, but surely, it is not an unreasonable idea. At the very least, the difficulty of relating theory to experiment has obviously increased.

    So, what kind of changes should be made in institutions, to support the “sit in an attic for 7 years and just think, like Wiles” kind of research that Woit is pointing to?

  3. Peter Woit says:


    At various times I’ve suggested various ideas one might want to consider. It should be clear to anybody in the field of particle theory what the current incentive structure is: it is determined by how NSF and DOE grant proposals are evaluated, and by how hiring decisions are made. Everyone involved on both sides of these decisions knows how they are being made, and that the incentives are heavily on the side of not working on long-term ambitious projects that may not pan out.

    If one wants to get serious about changing this, one has to

    1. Make working on such projects more rewarding: announce that grant decisions will be made preferentially for such projects, have hiring committees tell candidates that they are looking for people to hire working on such projects.

    2. Make routine work on research directions that have had huge investments but not worked less rewarding. Even a rumor that an NSF grant panel had decided to reject grant proposals in active but failed subjects such as, say, string phenomenology, would have a huge effect.

    As far as I can tell, the people with power to affect these decisions have reacted to the current crisis in particle physics not by discussing what can be done about it, but by getting defensive and attacking anyone who brings up the subject. Current thinking seems to be to just deny everything and hope that the LHC will solve all problems. Maybe it will, we’ll see. At this point, maybe I’m wrong, but I see no hope of anything happening or any serious discussion taking place until a few years from now, after the LHC results are in.

    I’d be happy to hear that I’m wrong about this, and that there are people out there taking this problem seriously and trying to do something about it.

  4. Thomas Love says:

    Researchers with ideas are going to pursue those ideas whether or not there is grant money available. People who follow the money should not be taken seriously as researchers

  5. Peter Woit says:


    This isn’t about money, it’s about jobs. For younger people, whether or not they can get a grant affects whether they can get a job or stay in one (get tenure). For people with tenure, a large part of their grant is used to pay students and postdocs. No grant, no job, if not for them, then for younger people they’d like to work with.

    No one with a brain goes into this business to make money, but whether or not they can make get a full-time job that allows time for research is crucial.

  6. anon. says:

    ‘No one with a brain goes into this business to make money, but whether or not they can make get a full-time job that allows time for research is crucial.’

    The predictable lame response of a string theorist would be:

    ‘Wrong! Take Einstein, just a patent examiner in 1905, who had no problems despite having no faculty tenure at all.’

    The Einstein claim is of course slightly misleading because the development of General relativity, 1915, which is really advanced mathematically, is another story and he would not have had the time and the helpful friends necessary to focus on the mathematics of tensor analysis without getting tenure:

    ‘Up to this time [1911] Einstein had used only the simplest mathematical tools and had even been suspicious of the need for “higher mathematics”, which he thought was often introduced to dumbfound the reader. However, to make progress on his problem he discussed it in Prague with a colleague, the mathematician Georg Pick, who called his attention to the mathematical theory of Ricci and Levi-Civita. In Zurich Einstein found a friend, Marcel Grossmann (1878-1936), who helped him learn the theory; and with this as a basis, he succeeded in formulating the general theory of relativity.’

    – Professor Morris Kline, Mathematical Thought from Ancient to Modern Times, Oxford University Press, 1990, vol. 3, p. 1131.

  7. Pingback: THE FARCE OF STRING THEORY « Æther Tracker

  8. Jack says:

    Is Science near it’s limits? Maybe, maybe not. Is Science near the limits of John Horgan’s imagination? It would appear so.

  9. Gphillip says:

    I think it’s clear that the “easy” discoveries have been made. Not to take anything away from the great minds of the past, but theirs was a simpler time in a world ripe for scientific exploration. Then monumental discoveries could be made in a small lab with a couple assistants. Now, we must invest hundreds of millions on apparatus like the LHC and employ hundreds of physicists and engineers for years to hope for a monumental discovery. This should come as no suprise. Science is nowhere near it’s limits, but we are getting ever so close to ours.

    In another century the scientific method will be so “picked over” it may take trillions of dollars of investment and thousands of scientists and engineers working for centuries to hope for a major discovery. Of course, all branches of science do not hit this wall of diminishing returns at the same time, but they all must pass their prime eventually. High Energy physics is just the most fundamental of sciences and will hit the wall first. All the other sciences will follow, eventually yielding fewer of natures secretes for ever more resources.

    In the end, how much of all there is to be known can be illuminated by the scientific method? Perhaps one percent? Perhaps one thousandth of one percent? To know that, we would have to know how much isn’t known, and that’s probably something we can never know.

  10. Jack says:

    Part of it’s about imagination, as I said before, but part of it’s about patience. In Mathematics, mathematicians have learned that it is not unusual for a problem to be so hard that it will go unsolved for many hundreds of years. Some problems unsolved today are more than 2000 years old. Physicists have not experienced, over the generations, the experience of being unable to make any progress on a problem over such a timescale. Now they have been struggling with certain questions for 30 years or so, and it is making many people, journalists and physicists, uncomfortable.

    Of course, in mathematics there are Lemmas. If you can’t solve a problem, work on a related problem which you can solve. Progress can be made step by step in this way. Eventually, someday, you realise that the easier problems you kept working on have become Lemmas in the proof of the problem you couldn’t solve before. It is not clear to me whether there is any analagous thing in physics. In mathematics you are spoilt for choice in terms of problems of all difficulties, easy, medium, hard, and incredibly hard, whereas in physics the main problems are more clearly mapped out and more easily summarized. This may seem like a good thing because things may seem simpler, but the downside might be that you have less leverage to make progress.

  11. Gphillip says:

    I generally separate math and science when discussing this issue. Math can always be abstracted to deeper levels. Some would say that the more abstract, the less it represents anything in the real world. Others would say the real world IS math (see “The Computational Universe” by Seth Lloyd). But I doubt that math will ever hit a wall where it takes ever larger number of people and huge expenditures to continue advancing. Math is after all just an abstraction of what we consider to be the real world. I believe that there is a subtle difference where science uses the scientific method to illuminate the secrets of nature, and math is just one of many tools employed in scientific pursuit. Math can also represent many things not associated with the scientific method, like accounting. In any case, all sciences will run their course in time. Either because they become almost completely defined with few meaningful questions left to answer, like Thermodynamics, or because the remaining questions would take unsustainable resources to pursue, like High Energy Physics. I’d guess Chemistry has probably just barley past it’s peak, but lots more remains to be discovered. Biology doesn’t seem to have quite reached it’s peak yet, and we have probably just scratched the surface of computer science. But for math, I see no reason it can’t toyed with until the end of time. That’s just my own humble opinion based on my anecdotal observations though. I haven’t seen any formal studies of the topic, though that would be an interesting.

Comments are closed.