Another conference that starts today is the Convergence conference at the Perimeter Institute. The concept is explained by Neil Turok here. His point of view is one I’m very much in sympathy with:
Turok explains that the “large bandwagon” of the last 30 years has not found experimental support. The bandwagon in question is the Standard Model of particle physics established in the 1970s, which, he says, people have been elaborating ever since. “Grand unified theories, supersymmetry, string theory, M-theory, multiverse theory,” he lists. “Each is not particularly radical, but is becoming ever more complex and arbitrary.”
To illustrate the lack of experimental support for these ideas, Turok describes how many people were hoping string theory would represent a radical development; but since string theory – as currently interpreted – leads to the multiverse, Turok describes it as the “least predictive theory ever”.
Indeed, experimental support has not been found for other extensions of the Standard Model either. “We have discovered the Higgs and nothing else,” says Turok, “yet the vast majority of theorists had been confidently predicting WIMPS (weakly interacting massive particles) and supersymmetric particles…Theorists are walking around in a bit of a stunned silence.” He adds that it could turn out to be right that all sorts of other particles are needed along with the Higgs – but that thought seems to be misguided.
“My view is that this has been a kind of catastrophe – we’ve lost our way,” he says. “What we need are ideas as simple and radical as in the start of the 20th century with quantum mechanics.”
So what might these ideas look like? Turok explains how observations have shown that the universe is simpler than we ever expected – in contrast to our theories, which are becoming ever more complex.
It’s great to see this, very encouraging to think that a more reality-based point of view on fundamental physical theory may finally be emerging. My only criticism is that the program doesn’t seem to include any mathematics or mathematicians (except perhaps in the context of history and Emmy Noether). If you have a structure you don’t understand, which is unusually simple, with surprising explanatory power, you might want to ask mathematicians about it…
The schedule is here, there’s a blog here.
Update: You can follow a lot of what is going on at this conference on Twitter, here. For example, I was glad to hear about this comment from Dimopoulos
There is no difference that we know right now … between the story of divine intervention and the multiverse.
It’s great to see a conference on fundamental physics where the multiverse is coming in for some appropriate skepticism.
Update: One thing to say about the multiverse, it does provide an excuse for an endless number of popular articles mulling it over. Just today, there’s Caleb Scharf and George Johnson.
I would like to know please what structure do you have in mind when you say “If you have a structure you don’t understand, which is unusually simple, with surprising explanatory power, you might want to ask mathematicians about it…”
Isn’t risky to ask mathematicians about the subject because all they will want to do is first put your question into context by inventing all sorts of axioms and, hence, complicate the subject further and drive it farther away from reality without any deeper insight on the physics?
I am most impressed with Neil Turok. I like the fact he is in charge of such an influential institution but he doesn’t “play it safe” by going along with the mainstream – which must be the temptation. He is prepared to go against the flow and take some risks.
I also suspect he is correct – which always helps.
You haven’t talked to many mathematicians… They come in all kinds, with all kinds of interest and expertise. Some might behave as you suggest, all you have to do then is ignore them, you haven’t risked much.
I was of course oversimplifying the situation. There is already a very active area of interaction between mathematicians and physicists both interested in fundamental physics. My suggestion was just that Turok and Perimeter might want to encourage that by bringing some of it into this sort of event. You can take the attitude that you don’t think mathematicians know anything that could be of use, but given how stuck physicists are, I don’t think they should be so quick to convince themselves this isn’t a potential route to progress.
If you want an example of a mathematician who might play a useful role at a conference like this, one would be Graeme Segal. Recently on this blog, see
there was some discussion of a talk by him on Wick rotation. He’s a mathematician who has thought very deeply about this, and it’s not at all impossible that this could give an insight into the nature of time that physicists would not otherwise come up with. He knows a great deal about quantum field theory, and is not at all someone who is going to make things more complicated rather than less.
If reconsidering our basic ideas is going to be necessary to move forward, to what extent do you think this will include some of the fundamentals of quantum field theory itself, as opposed to just the standard model?
Peter, I never understand this frequent wish you make for mathematicians to come in and sort out physics. One of the most damning criticisms of string theory (and other “Standard BSM” approaches) is the lack of experimental support and the movement towards anthropics, landscapes, multiverses and the like.
Do you think mathematicians, not well known for their solid grounding in reality, are likely to bring physics back to earth?
Yes, I think new ideas not just specifically about the Standard Model, but about the kind of quantum field theory (gauge +spinor fields) used in the SM may be needed. More specifically, ways to think about this kind of quantum field theory as constructing representations of certain infinite dim symmetry groups (or whatever the right generalization of a representation is…). In physicists language: new ways of exploiting symmetries in QFT.
Anthropics, landscapes, the multiverse etc. are the kind of thing physicists come up with, not mathematicians. What mathematicians are very good at is knowing the difference between what they understand and what they don’t, and not fooling themselves about it. In addition, there undeniably are some very sophisticated mathematical ideas being used in the Standard Model. While physicists are to some extent trained to know something about these, there are people who are actual experts on them, and they tend to work in math departments.
The historical parallels aren’t exact of course (and never are) but it seems like you are waiting for the 21st century equivalents of Jacobi, Poisson and Hamilton to come and formalise existing physical theories and link them to existing mathematical structures. I would prefer the 21st century Plancks, Bohrs and Heisenbergs to bring new physical theories, rooted in experiment, to explain old and new inconsistencies. Perhaps you think the current experimental state of the art isn’t leaving enough clues for prospective Planks but, in my opinion, the list of known unknowns now is at least as rich as it was in 1900.
Eventually of course the 19th century formalism turned out to be useful for the 20th century revolution and I certainly wouldn’t want to claim mathematicians interested in physics are useless, but I don’t think it’s debatable that the real progress came from physicists and would have happened with or without the previous century’s mathematical developments. Mathematicians are likely to have good ideas about the mathematics of Quantum Field Theory but progress in physics (and that is what we care about right?) is more likely to come from people who understand error bars and know the values of some fundamental constants.
Yes, I think the lack of clues in available data is the main reason conventional paths to progress are not working in physics. In particular I disagree that there are now known unknowns anything like what was available in 1900. At that time there was already a huge amount of spectroscopic data about the energy levels of atomic systems, and it was this data that made possible Bohr and Heisenberg’s advances.
The combination of cosmological data (CMB, dark matter, dark energy), particle data (neutrino oscillations, old and forthcoming LHC data), theoretical questions (colour confinement, strong CP, black hole information) and null results (proton decay, horizon problem) seems as strong a list as was available at the beginning of the 20th century. I am sure at the time it looked equally impossible to discover anything useful from a hodgepodge of spectra. Perhaps I am just more optimistic than you.
Dear Dr Woit
I was astonished that you were pleased by the absurd statement made by Dr Dimopoulos. “There is no difference that we know right now … between the story of divine intervention and the multiverse.”
I appreciate you wish to avoid theological discussion, but my point is more about evidence than anything else. Thus I remind you that a large number of independent witnesses have reported occurrences of Divine Intervention, in the New Testament, at Fatima in 1917, and elsewhere.
Of course, you may find the accounts of these, (or any other witnesses) to be non-credible, But of course, the accounts have been found to be credible by billions of people. And more importantly, the accounts of witnesses is direct empirical evidence, which is a lot more than your Multiverse promoting colleagues can offer.
Tammie Lee Haynes
I think Dimopoulos was referring to questions of the origins of the universe or of the laws of physics, not of human experience. Sure, in terms of the latter, there’s more evidence of divine intervention than of other universes with different physics.
That said, any further theological discussion is strongly discouraged…
“We have discovered the Higgs and nothing else,” – Turok
That’s incorrect, I don’t know why people say that.
Dark energy has also been discovered in the past thirty years. Dark matter and massive neutrinos may have as well, depending on where you set the discovery date.
Turok doesn’t say “We have discovered the Higgs and nothing else in the past 30 years”, you’re putting together two things a couple paragraphs apart, taking them out of the context of what he is saying. The “30 years” is a reference to the history of speculation about GUTs, SUSY, string theory, M-theory (40 years is actually more like it at this point). The “and nothing else” comment about the Higgs isn’t associated with a date, and refers to those speculative ideas. Put differently, I don’t see him saying that dark matter and dark energy are not discoveries, but that they are not ones of the sort that were predicted by GUTs, SUSY, string theory, M-theory, the multiverse (unlike nonexistent LHC SUSY WIMPs).
dear Radioactive, the situation is different from ≈100 years ago. Then, matter was not understood and a lot of data was accumulating. Today, only Dark Matter remains not understood, and so far we see Dark Matter only trough its gravity.
Then, mathematical physicists had grasped some aspects of classical mechanics later relevant for quantum mechanics, but Heisenberg and Schrodinger invented quantum mechanics from data, without knowing these mathematical instruments.
One similarity is maybe that 100 years ago most physicists believed in aether and tried to insist on it despite negative experimental results…
Apart from the absense of BSM physics, one striking fact has been learnt in recent years: the Higgs mass appears to sit exactly at the border of stability. 1 GeV less and the SM vacuum would be unstable, 2 GeV more and the SM would be boringly safe in the stable region. The authors of http://arxiv.org/abs/1307.3536 calls this the principle of living dangerously. I think this may be an important hint, and in any case it is pretty much the only hint that experiments has provided.
M, I strongly disagree. Kelvin’s famous statement “There is nothing new to be discovered in physics now. All that remains is more and more precise measurement.” was made in 1900. They had a pretty good grasp of nature and the spectacular successes of Maxwell’s electromagnetism, thermodynamics and the kinetic theory gave them good grounds for confidence.
I would draw parallels between the discovery of the Higgs and Einstein’s Brownian motion papers. That is, difficult to detect particles, already used with great predictive success, whose existence was believed by most physicists and awaiting experimental confirmation. Or perhaps between blackbody radiation/cosmic expansion – experimental anomalies which fit awkwardly with our current theories (EM+thermodynamics/GR+QFT), whose resolution seems like a bit of a kludge Planck quantization/inflation.
Of course the analogy is not exact in every sense. The problems nowadays are certainly harder because we know more, so there are more constraints and we have picked most of the low hanging fruit already. But my point is that Kelvin had a pretty deep mathematical understanding of classical mechanics which turned out to be completely useless for making progress in clearing up the one or two ‘loose ends’ of 19th century physics. Moreover these loose ends ( dark matter, dark energy, strong CP and the others I listed above ) turned out the unravel the whole of physics.
I agree with M that the situation then was the atomic scale was a huge unexplored territory, with physicists having a lot of tools at hand to learn more about it. One can have a different point of view about this, but in any case I think maybe physicists should just get over the idea that huge revolutionary progress like that of the period 1900-1925 is just around the corner. This is a stock claim at every conference like the Perimeter one, has been my entire career. One should be an optimist that progress is possible, but maybe it’s a naive idea that all that is needed is a new bright young Einstein/Bohr/Heisenberg to take a fresh look at the data and get some insight from that that will change everything and throw our current theories in the trash bin.
Peter, huge revolutionary progress would be nice, or any progress at all. Though what we think of as progress is probably very different. I would prefer a ‘Bohr atom’ to an elaborate and beautiful theory linking QFT, geometric Langlands and the Monster group but which doesn’t explain any of the currently known experimental anomalies. Fortunately these things often come hand in hand.
It’s always fascinated me why anyone thinks the Fermi paradox is a paradox. Fermi especially. I mean he clearly knew that the speed of light is a barrier to civilizations finding each other? There could be 100 advanced civilizations within a 100 light year ball around us, and we would never know. It’s a pity Mars wasn’t just a little bigger, we might have had obvious life right next to us.
@ Radioactive, 3:48 am
Regarding the quote attributed to Lord Kelvin, the attribution is disputed, and the earliest known claim for the quote is dated 1988, at least according to Wikipedia:
Pingback: Multiverse skeptic on physics conference | Uncommon Descent
Radioactive, besides Kelvin’s famous remark about the future of physics, many theoretical physicists were well aware that something was not quite right about classical physics. Lorentz and Poincaré, for example, were hard at work trying to reconcile the dynamics of the then recently discovered electron to the microscopic Maxwell-Lorentz electrodynamics, long before Einstein’s 1905 paper on special relativity. (Give a read on Poincaré’s “Sur la Dynamique du Electron” on the Internet that you will see how close he came to discover relativity.) Or the Balmer formula, known since 1885 but that would have to wait for 28 years to be derived by Bohr! J. J. Thomson tried to invent a lot of classical models in order to describe the Hydrogen atom using only classical physics and his beloved electron.
By the way, Einstein’s hypothesis of the light quanta and his relativistic energy-momentum relations would be established experimentally only in 1923 with the Compton effect. Even Bohr expressed doubts about the reality of photons before that. So if something like photons or E=mc^2 took almost a decade in order to be proved experimentally, it seems reasonable that SUSY and related ideas should take even a longer period.
About the usefulness of pure mathematicians in guessing new laws, I am inclined to agree with Peter. Remember that pure mathematicians like Gauss, Riemann, Ricci and Levi-Civita created the mathematical structures that Einstein would use later to realize his ideas on gravitation and discover GR. Indeed, I read somewhere that both Einstein and Marcel Grossman studied Levi-Civita’s “Absolute Calculus” carefully around 1910.
“What we need are ideas as simple and radical as in the start of the 20th century with quantum mechanics.”
I wonder if this is a fair statement. There were completely unexpected empirical discoveries that preceded the “simple and radical” ideas, no? Discreetness was forced on Planck by the data and by the complete lack of fit with classical theory. It’s well known he didn’t like this “radical” revision of physics, and even resisted when Einstein took his ‘s ideas further.
Then, when the next steps were taken, again based on new data coming from atomic spectroscopy and the very early particle physics, the people that came up with the “simple and radical ideas”, Heisenberg et al suffered much resistance from the determinists like Einstein and Schrödinger, that continues today with all the debates about “interpretations” of quantum mechanics.
What really new radical discoveries have their been? The expansion of the Universe and the accelerated expansion come to mind. But most things after the 1920s generally fit the frameworks that emerged by that point.
Maybe if physics turned to other phenomenon it doesn’t currently include in its scope, similar progress could be made. For example: how is living matter different from nonliving matter? What the heck is mind and consciousness? In spite of the tons of smoke and obscurity surrounding these topics, they are “low hanging fruit” for a radical revision of our understanding of the universe. These phenomenon are part of our experience of the physical universe, but right now they are completely outside the bounds of physics.
Thanks for allowing me to express my views.
It is clear that we need a more predictive theory than string theory. As far as the quantization of gravity is concerned, there are already simpler theories which achieve this, for example, spin-foam and spin-cube models, and these models do not need supersymmetry in order to construct a finite quantum gravity theory. As far as the explanation of the elementary particle spectrum, I believe that one must find a generalization of the concept of symmetry, which means to generalize groups and supergroups. Such generalizations already exist, for example, quantum groups and categorical generalization of groups, called 2-groups, and the goal is to see whether these structures suffice or we need something new. This will also require to make an appropriate generalization of the space-time structure (i.e. to generalize a smooth-manifold).
I definitely agree with Don, although as a biophysics student I’m a bit biased.
The fact of the matter is that many regimes of nature stubbornly resist an organized theoretical description. The magnitude of the problem is highlighted by how little we need to move away from typical physics to find systems where textbook physics becomes surprisingly impotent; just look at the semi-classical mess that is a membrane protein to find something which, while technically a physics problem, demands numerous novel methods to study.
More importantly information theory is slowly becoming a more generally accepted framework for thinking about and tackling hard problems. In many regards the problems with protein dynamics are problems of dimensionality reduction; what is the most minimalistic way to describe the entire problem? Working on a variety of problems could spur developments in information theory which could have kickbacks for more traditional challenges (although I’m an outsider so I don’t know how deeply related information theory is to fundamental physics, although the observer/reality relationship and cutting edge concepts like the black hole information paradox suggest that it could be).
I came across this talk of Feynman (argument from authority!) with his ideas about exactly these issues – mathematics in physics and how to make discoveries.
He makes some interesting points; in particular, I think that the most optimistic take on the relationship between mathematicians and physicists is that mathematicians are typically useful after the fact. I can think of only one major exception and it’s an outlier: general relativity. Otherwise the process of pioneering physics is only encumbered by attentiveness to axioms and extreme rigor. Once you’ve figured out the rules organizing and packaging them as the mathematicians like to do becomes more imperative.
Actually I thought it was kind of embarrassing to watch Feynman pontificate about something he clearly knows nothing about. The idea that what mathematicians do all day is engage in extremely rigorous discussion of axioms, in very general cases, with no interest in special cases, no idea of the “meaning” of what they are doing, is about the same as the idea that physicists all spend their time wearing white lab coats and standing in front of a bench with a piece of experimental apparatus on it.
I wonder if you are reacting to some of the idiosyncrasies of Feynman’s communication. For instance I don’t think he was saying that something like a ring has no meaning, but was actually referring to levels of abstraction.
To him I would guess the equation x + 5 = 7 has no “meaning” until it is specified that x represents the number of dollars one must add to 5 to get 7, even though the equation still means something in the abstract sense.
It’s probably not too productive to argue about what Feynman thought but I suspect it is unlikely he was unaware of the fact that mathematicians work on specific cases considering he worked with people like Kac on occasion as well as more mathematically oriented physicists. At any rate Peter is it not true that a substantial amount of modern mathematics research remains interested in theory building from axioms and high levels of rigor? If we take it for granted that he conflated these individuals with all mathematicians, we can still discuss whether or not this mathematical subculture is relevant to physics I think.
Like physics, mathematics is a huge activity with a wide range of people doing a wide range of things, with many of those things having various relations to physics. I don’t see the point of debating stupid caricatures of what mathematics is, what mathematicians do, and what the relations to physics are, even if they’re coming out of the mouth of Feynman.
One of the reasons I posted the video was Feynman’s assertion that physicists have to care about ‘blocks of copper and glass’. One of my problems with SUSY, strings or multiverses is the failure of the community to abandon or drastically modify these theories in the face of failed predictions or failure to make predictions. Peter, your desire for more mathematicians in physics seems to run opposite to this. There are a great deal of different types of mathematicians but they are as a group, it’s fair to say, not concerned with blocks of copper and glass and not likely to care very much if something beautiful, like string theory, has no connection to reality. And this has been your lament about physics for the last 10 years.
Unlike a lot of people, my criticism of string theory has never been that too much attention is paid to mathematical beauty and not enough to experimental results, and I think the physics community is making a big mistake if they decide that is the lesson of the string theory fiasco. The current explanation for how string theory works, the landscape, is something of unparalleled ugliness, which only a physicist, not a mathematician, would find appealing.
The underlying problem is the fact that there are very few hints from experiment about how to do better than the Standard Model. It just is not true that there are lots of such hints and they are not being paid attention to. It has been clear now for a long time, long before the LHC results, that string theory unification predicts nothing. The problem is a failure to acknowledge that the internal problems of the theory are deadly, instead creating a bogus pseudo-scientific connection to experiment (the landscape).
What mathematicians are expert at is understanding exactly what the properties of a mathematical framework are, what works and what doesn’t. They have a strong culture of abandoning ideas that don’t work. Working in a situation where you don’t have experimental clues, but have to rely on the coherence of the ideas and models you’re studying is exactly what they are experts at, and I think this could be helpful. I don’t see the argument that they’re going to do worse than the “physical” path that has brought us to the multiverse.
Contra @Mol, I thought that Einstein’s hypothesis of the light quanta was established by Robert Millikan’s 1914 experiment on the photoelectric effect.
@Douglas J. Keenan,
I have before me Robert L. Weber’s delightful anthology A Random Walk in Science, copyright 1973. On page 191, is a selection entitled “Clouds, 1900” from Lord Kelvin: “The beauty and clearness of the dynamical theory, which asserts heat and light to be modes of motion, is at present obscured by two clouds.
I. The first involves the question, How could the earth move through an elastic solid, such as essentially is the luminiferous ether?
II. The second is the Maxwell-Boltzmann doctrine regarding the equipartition of energy.”
The citation is “Slightly condensed from Philosophical Magazine (6) 2 1 (1901).” The editor adds: “Kelvin could certainly recognize the important clouds. One needed relativity, the other quantum theory, to blow it away.”