Strominger Interview

There’s an interview with Andy Strominger in the Calcutta newspaper The Telegraph. Strominger was presumably in India for the string theory conference there this past week.

The thing I found interesting about the interview was how skeptical the interviewer was, repeatedly asking about whether string theory might not be wrong. Perhaps at least some members of the media are starting to get a clue.

This entry was posted in Uncategorized. Bookmark the permalink.

81 Responses to Strominger Interview

  1. Chris Oakley says:

    “we know that General Relativity is correct, because it emerges out of string theory”

    Is this true? How can a theory that makes no predictions have a classical limit?

  2. Arun says:

    Chris W.:

    The point is well taken about how exactly theories are subsumed into other theories – how the wider theory explains the successes of the lesser theory and specifies under what conditions the lesser theory fails.

    But if you ask your friendly neighborhood string theorist, and it seems inconceivable to them that string theory could be extendible in a similar way. I’ve already quoted to that effect. Perhaps it has to do with the converse of something else you wrote: “Any such theory that has empirical content also has the potential of being contradicted by observations.”

    To take up your example of the anomalous acceleration detected by Pioneer, I guess the SuperStringers’ attitude would be – we know that General Relativity is correct, because it emerges out of string theory, so these experimentalists who propose spending hundreds of millions of dollars on a spacecraft experiment are stupid hacks who should not be paid to be scientists.

    Maybe I’m being unjust.

  3. Quantoken says:

    Lubos erased my post on his blog which declared a fatal fraud in a paper he published, claiming that I do not understand mathematics enough to be able to find his mistakes and that I am not going to say what it is.

    The fact is he could not find his own mistake even after I gave him a hint. Of course I am going to point out exactly what his mistake is, I said I will disclose it after Christmas. And of course Lubos is the one who does not know math. Do you have any idea what is elliptic curve, Montgomery multiplication, CRT, etc in mathematics? Don’t know, do you?

    Doesn’t matter. Mathematics matters only where it matters to the actual physics. Your mistake in the paper, Lubos, is NOT trivial math or computation error. Your mistake is you used the wrong math in the wrong physics.

    Would you tell me what is the value of Hawking Entropy or the Schiwaltzchild radius of one single hydrogen atom? Or would you tell me what is the temperature of the electron that circles the proton in a hydrogen atom?

    Being an idiot like Lubos is, he will go right away and copy the formulas from textbook and plug in numbers to calculate. I am sure he will double check his math to make sure he did the calculation right. But still he will give you the wrong answer!!!

    The point is the physics is no longer right when you push the physics/math formulas beyong where they should be used. You really can’t calculate the Hawking Entropy of a hydrogen atom using that a quarter Planck Area formula. Because the hydrogen atom is too light to form even the smallest possible blackhole. And in a blackhole of any size, there could not possibly be any atom. You can’t use another formula to calculate Hawking Entropy of a hydrogen atom either. The concept of Hawking Entropy simply do not apply to hydrogen atoms.

    Now, I have given you enough hint why you made a fatal mistake in your original paper. Go figure. I will tell you after Christmas, if you can’t figure out by then.


  4. Quantoken says:

    Ah ha, Wow!

    Lubos was totally wrong in his ARXIV paper deriving the quantum of area 4*ln(3). See his paper:

    The whole thing was wrong for a very simple reason. That’s a mistake that he really should not have made if he has the least of physics instincts. At least I never expected a guy who was a Harvard assistant professor make such a mistake. The paper is 25 pages long but it took me no more than 5 minutes to find where he made a fatal mistake.

    Now, Lubos, go find a spider hole first. Because once I tell you the mistake you made, you would be so ashamed of yourself, that you want to dig yourself a hole to hide in.

    Had you shown a little bit respect to me before. I wouldn’t have choosen to disclose your mistake publicly, and would instead email you in private. Or at least post it in your Blog only.

    Oh well. Too late, you are going to be really embarrased this time. To give you a chance to save face. I will give you a hint first, see if you can find your own mistake before I tell you.
    The hint is: what is the numerical value like of frequence Omega that you are talking about in your paper?

    I will tell you where you made a fatal mistake in your paper, after Christmas.

    Mean while, every one, Merry Christmas!


  5. Chris W. says:


    I’m sorry for not responding more promptly to your comment. First of all, let me quote your main points:

    The idea is, I suppose, that we can only show that a physical theory corresponds to reality to a certain precision within a certain domain. We can never rule out the possibility of a theory with wider validity, just as Newtonian gravity was replaced by and subsumed within General Relativity. So proof of a theory is not possible.

    But the Theory of Everything (or at least Superstring Theory) is different. It either applies to the real world entirely and wholly, or not at all. And here is what a professional physicist, Lubos Motl, wrote on his blog : “I am always a bit puzzled by Andy’s [Strominger] statements that string theory is “just another step” – what sort of other step that goes “beyond” string theory but does not invalidate it is Andy thinking about?”

    I would amend your first remark to the following: “We can never be sure that all testable consequences of a theory will be confirmed by observation; proof of a theory is not possible. Our past experience gives us reason to hope, however, that a theory that conflicts with observation will eventually be subsumed within a theory of wider validity that explains these failures, and specifies the conditions under which the failures will occur.”

    More precisely, both the successes and failures of a theory are observations—empirical facts about the results of tests of the theory. Such facts call for explanations. We don’t want to dismiss the success of classical celestial mechanics (for example) as meaningless or inexplicable. Indeed, part of the problem that Einstein set himself was understanding such successes, once he realized that he must reconsider the intertwined foundations of Newton’s mechanics and Galileo’s relativity principle in order to reconcile them with Maxwell’s electrodynamics.

    Many physicists and philosophers of the 19th century were ready to believe that Newton’s mechanics and theory of gravitation constituted a “theory of everything”, with electrodynamics relegated to the status of a continuum mechanics of the mechanical ether. The necessity of subsuming 19th century physics into a deeper and broader framework only became clear in retrospect, as did the crucial roles of the speed of light and the quantum of action, and the irrelevance of the classical idea of an ether.

    So in response to your second point, nothing is fundamentally changed by considering a theory as a candidate for a “theory of everything”, ie, a theory that is presumed to constitute a shared basis and explanation for everything we have observed about the universe, including the successes of any previous theories. Any such theory that has empirical content also has the potential of being contradicted by observations. Of course, one can try to interpret the theory so as to evade the contradictions, but one always does so at risk of destroying its empirical (as opposed to formal or mathematical) content. The most profound discrepancies are those that leave one unable to offer any scientifically defensible explanation within the framework of the theory. Either we accept them and admit that the theory fails to be both all-encompassing and scientific, or we set about undermining the theory’s scientific status in order to preserve its metaphysical authority. (This is not to say that metaphysics can or should be avoided under all circumstances, which is itself a pernicious attitude.)

    It is worth reviewing the efforts of John D. Anderson, Slava Turyshev, and their collaborators to characterize the Pioneer spacecraft trajectory anomalies as a reminder of what a fundamental discrepancy with theory might look like, and how much effort is involved in eliminating auxiliary hypotheses that might explain it. (Anderson, Turyshev, and Michael Nieto are now advocating a new mission to the outer solar system designed specifically to investigate this problem. The Pioneer spacecraft have ceased communicating; Pioneer 10’s last contact was in January 2003.)

  6. I am afraid that string theory involves so many
    wrong assumptions that there is no hope of making
    progress by deforming algebraically a little bit. It would be better to take a more philosophical attitude, and look carefully for the strengths and weaknesses of string theory and how to possibly cure them. Here are some opinions about what “going beyond string theory” might mean.

    1. Wrong realization of super-conformal invariance

    Appropriately generalized super-conformal invariance is very probably something shared by any future theory. Personally I see the realization of conformal invariance at two-dimensional string world sheets un-necessarily restrictive and as the fatal flaw of the string theory.

    The way out of dimension 2 is based on simple observation: light-like 3-surfaces of Minkowski space (or any 4-D space with Minkowskian metric signature) allow generalized conformal invariance by their metric 2-dimensionality and the dimension 4 for space-time becomes unique.

    Also generalized symplectic structure is possible for light-like 3-surfaces, and the group of isometries for light-like surface consists of combinations of conformal transformations and compensating local scalings in light-like direction and is thus infinite-dimensional. Thus light-like 3-surfaces are really incredibly beautiful structures and I find difficult to understand why string theorists refuse to take them seriously. Also a new view about super-conformal symmetry emerges and no space-time super-symmetry (sparticles) is predicted.

    2. Spontaneous compactification and Kaluza-Klein
    philosophy as the second fatal mistake

    Personally I experience spontaneous compactification and Kaluza-Klein approach as something absolutely ugly. To me it is a purely ad hoc attempt to extend the explanatory power of
    string model beyond its natural limits. The accompanying ad hoc assumption is that of space-time super-symmetry.

    Instead of trying to make 4-dimensional classical world to 11- or 2-dimensional, I would prefer to start from a real problem of general relativity. In general relativity the basic conservation laws due to the isometries of Minkowski space are lost, and one can ask how to combine the good aspects of special and general relativities. This is achieved by identifying space-time as 4-D surface of some higher-D space of form H=M^4xS, M^4 Minkowski space. Poincare invariance lifted to H. S can be fixed to S=CP_2 by requiring standard model symmetries.

    In this framework the higher-D space cannot be nor need to be dynamical since the dynamics is at the level of space-time surface, and the horrors of landscape are avoided.

    3. Infinite-dimensional existence is unique

    M-theory has led to inflation of all kinds of deformations and variants of finite-dimensional geometry. There is however a different direction that could be followed in the attempts to generalize the notion of geometry.

    Taking the uniqueness of physical existence as a guideline, one ends up with the notion of infinite-dimensional geometry. Already loop space Kaehler geometry is unique as Dan Freed showed in his thesis. The requirement that Riemann connection does not lead outside the tangent space forces infinite-dimensional isometry group and makes Kaehler metric of loop space unique. There is still a problem: although Ricci tensor is finite, curvature scalar is infinite, a clear symptom that basic objects cannot be 1-dimensional.

    The space of 3-surfaces in M^4xCP_2, which possesses a maximal group of isometries, is the TGD inspired guess for the world of classical worlds as the unique arena of quantum dynamics. This space would be a union of infinite dimensional symmetric spaces with vanishing Einstein tensor labelled by zero modes having interpretation as classical, non-quantum fluctuating degrees of freedom characterizing the space and size of 3-surface. This formal analog for landscape is in key role in TGD based quantum measurement theory.

    M^4xCP_2 is a very promising candidate for imbedding space. The canonical transformations of VxCP_2, V light cone boundary, act as isometries of this space and super-conformal invariance in generalized sense appears naturally in the construction of the configuration space geometry.

    Configuration space gamma matrices in fact have identification as fermionic generators of super conformal algebra so that both fermionic statistics and super-symmetry have purely geometric interpretation. An important deviation from string models is that super-generators do not correspond to Majorana spinors (this leads to dimension D=10 or 11 in string models) but carry quark and lepton numbers.

    4. Tangent spaces of space-time and imbedding space as number field like structures

    H=M^4xCP_2 is the only option allowed by Particle Data Table and the construction of configuration space geometry favors strongly this choice. One might however argue that the preferred role of M^4xCP_2 should have some deeper, perhaps number theoretic, explanation.

    The dimensions 4 and 8 for space-time surface and imbedding space of course bring in mind quaternions and octonions, and CP_2 labels the quaternionic sub-algebras of octonions just like CP_1 labels the complex sub-algebras of quaternions. The first guess is that space-time could perhaps be thought of as being surface for which tangent space corresponds to quaternionic, or more generally associative, sub-space of octonions at every point so that classical dynamics would reduce to number theory. The problems relate to the Euclidian signature of the metric. Also the non-commutativity and non associativity are questionable features.

    Perhaps a better guess is that the tangent space of space-time corresponds to H_4, the algebra and “almost number field” of 4-D hyper-complex numbers. The tangent space of imbedding space could in turn be regarded as the space H_8 of 8-dimensional hyper-complex numbers.

    Quite generally, hyper-complex numbers have dimension 2^n which is the same as the dimension of Clifford algebras spanned by gamma matrices, and the natural requirement is that units allow a representation in the Clifford algebra of the space in question.

    a) For H_2 the commuting units are 1, e1 with e1^2=1.

    b) The commuting units of H_4 are 1, e1, e2 and the product e1e2, all of them having squares equal to 1.

    c) For H_8 the units are 1,e1,e2,e3 and their products. One can decompose H_8 number to two parts h= h_1+e_3h_2, where h_1 are H_4 coordinates, the “real” and “imaginary” parts of h.

    The number-theoretic norm of hyper-complex number h in H_m, m=2^n, is expressible as the product of all m conjugates of h and is the m:th power of Minkowski length squared of m momentum. Lorentz group leaves the norm invariant. The failure of number field property manifests itself as the existence light-like hyper-complex numbers having no inverse and forming an ideal of H_m. Remarkably, the spectrum of momentum squared is integer valued in the restriction to hyper-complex integers: nothing but the stringy mass squared spectrum.

    From the p-adic viewpoint the hyper-complex primes are of obvious interest, and the construction of infinite primes generalizing the notion of finite hyper-complex prime is especially interesting since the correspondence of infinite primes with the states of super symmetric arithmetic quantum field theory becomes very concrete.

    5. M4xCP_2 as a unique 8-D hyper-complex manifold which is a homogenous space?

    Hyper-Kaehler manifolds generalize the notion of complex manifold to the quaternionic manifold. This means that tangent space allows the representation of quaternionic units 1, I_1,I_2,I_3 as metric and antisymmetric covariantly constant tensors defining symplectic and Kahler forms.

    In hyper-complex case similar idea works. The basic idea is that the generating units e_i are representable as contractions of covariantly constant antisymmetric tensors with sigma matrices, which commute and have squares equal to unit matrix. The 2-forms associated with generating units define what can be regarded as a generalized symplectic structure.

    M^2 is the simplest hyper complex manifold and e_1 corresponds in light-like coordinates u=t+z, v=t-z to antisymmetric tensor with a non vanishing component E^1_uv=1. The square of the contraction with Sigma_uv gives unit matrix in Clifford algebra. E_uv defines a Minkowskian analog of symplectic structure since its square is metric (rather than -1 times metric).

    In case of M^4 the coordinates u=t+z,v=t-z,x,y corresponding to the decomposition M^4=M^2xE_2 to longitudinal and transverse degrees of freedom are preferred. The appropriate generalization of these coordinates to what I call Hamilton-Jacobi coordinates play a key role in the construction of solutions of field equations in TGD framework.

    The units e_1, and e_2 correspond to antisymmetric tensors E^1 and e^2 with non vanishing components E^1_uv=1, E^2_xy=i. iE^2_xy defines symplectic and Kaehler structure in E^2. The contractions of E^i with sigma matrices obviously commute. Note that the flatness of M^4 is absolutely essential prerequisite for having two covariantly constant 2-forms. e_1e_2 is represented as a production of contractions of E_1 and E_2 with sigma matrices.

    In the case of M^4xCP_2 one additional covariantly constant 2-form E_3 is needed and CP_2 Kaehler form defines it. M^4 coordinates correspond to the real part of H_8 coordinate and CP_2 coordinates to its imaginary part. Of course, more general choices of coordinates obtained by applying generalized canonical transformations leaving the forms E_i invariant.

    The conjecture is that symmetric space property plus the existence of hyper-complex structure select M^4xCP_2 as a unique candidate for H. Obviously, the existence of symplectic structure in the generalized sense is essential for the existence of hyper-complex structure. For instance, the replacement of CP_2 with symmetric space S^2xS^2 would bring in *two* additional symplectic forms whereas S^4 does not possess any Kahler form. Does the hierarchy of hyper complex manifolds contain other manifolds than M^2, M^4, M^4xCP_2, M^4xCP_2xCP_4, M^4xCP_2xCP_4xCP_8,… is an interesting question.

    6. Number theoretical classical dynamics

    Despite the fact that hyper-complex numbers are not a field, the notion of analyticity generalizes and leads to the idea that the classical dynamics of TOE could reduce to 4-D hyper-complex analyticity condition. Indeed, in string model the solutions of 2-D d’Alembert equation can be expressed as hyper-complex analytic functions of a hyper-complex variable assigned to the string world sheet with Minkowskian signature of metric.

    A very attractive idea is that space-time surfaces correspond to surfaces of H=M^4xCP_2 for which the “imaginary” part of hyper-complex valued analytic function (possibly satisfying some additional constraints such as being polynomial) vanishes:


    By implicit function theorem this gives 4-D hyper-complex number h_2 as an analytic function of h_1:

    h_2= f(h_1).

    This is an attractive candidate for a purely number theoretical realization variational principle, which should in TGD framework has absolute minimization of Kahler action as counterpart. The challenge is to prove this.

    Matti Pitkanen

  7. Quantoken says:

    Lubos said:

    “When Einstein started to propose his EPR paradoxes, he was sure that quantum mechanics had to be giving wrong predictions for these experiments. It’s not just a matter of philosophy. He believed that the quantum entanglement was impossible, and using the current terminology, he would definitely agree that Bell’s inequalities must be satisfied in reality.”

    Einstein was wrong in his judgements about quantum mechanics. But he was right on the profound philosophical believe that the physical world is fundamentally deterministic. Quantum mechanics is certainly a very successful theory in making predictions. But few people believe it is a final theory due to the difficulty of reconciling the difference between philosophy and the quantum mechanical picture. That’s why, even in 2004, people are still debating about EPR paradox and struggle to try to understand what experiments like Alan Aspect’s really tell us about reality.

    It’s far from being settled even today. Nobody can boast to understand EPR completely. If you think you know, it only shows you have learned the formulations of physics but really don’t understand it.

    Now about the famous Bohr – Einstein debate about that gedanken light box experiment. I do not why no body has pointed it out so far. The fact is both of these two great men had made serious mistakes in their great debate.

    Bohr had made a big mistake in a fraud in his arguments, which used Einstein’s General Relativity to prove that the uncertainty principle
    still holds. And Einstein, astonished that his very GR theory was used against him, did not realize that Bohr made a fraudulent usage of his theory in his arguments.

    Actually if Bohr had been right in his reasoning, he would have had discovered a way directly linking GR with the QM uncertainty principle. We wouldn’t be still struggling so hard trying to associate GR with QM today.

    Bohr’s mistake being he confused the two delta T’s. One being the deterministic and predictable time dilation due to GR effect. The other delta T being the actual uncertainty of time, which is random and un-predictable. The two delta T are totally different and unrelated. But Bohr tried successfully in confusing Einstein by using the same terminology to describe these two different delta T.

    At the end of day, the uncertainty principle still holds, and Bohr could have defeated Einstein using another simplier and correct argument, by considering that photons have wavelengths related to their energy. So it is impossible to release a photon of certain energy, if the shuttle is opened for too short a time.


  8. Quantoken says:

    Dr. Lunsford said: “My God, Lubos, gigawatt source of utter BS, has the gall to set himself up as a critic of Feynman.”

    I am sorry, Dr. Lunsford, but Lubos is not Your God 🙂 The remaining part of your statement is correct.


  9. D R Lunsford says:

    My God, Lubos, gigawatt source of utter BS, has the gall to set himself up as a critic of Feynman.

    Hows does a person like LM exist? How is it that he is able to get a job ANYWHERE, not to mention one of our best universities? He does nothing other than run his mouth – has no ideas of his own – reveals his shallowness and arrogance at every stage – is perhaps the most annoying person I’ve ever encountered online. How did this zero get a job at Harvard?


  10. Lubos Motl says:

    Future limitations of string theory.

    Arun, we exchanged a few sentences with Andy S. about it today – well, there were some more material topics, too. When you ask Andy what kind of the new steps “beyond” string theory one can imagine, he says “I don’t know”.

    It’s of course partly a matter of terminology – what you consider string theory. Is M-theory still string theory or not? I think that if you accept the terminology that string theory contains things as different as M-theory, then there are only two possibilities. String theory will be proved completely wrong, irrelevant for the real world, or it will be correct and the progress will only be within string theory itself.

    There is a difference from relativity and quantum mechanics. You know, the Newtonian mechanics contained three independent units, let’s call them meter, kilogram, second. All other quantities have units that are products of powers of these basic units.

    For example velocity is in meters per second. Special relativity showed that length and time are really related, and Newton’s physics only holds for velocities much smaller than the speed of light. You set c=1 in relativity.

    Similarly, quantum mechanics uses hbar, and things only look classical if the typical quantities of the same dimension as hbar (like action, angular momentum) are much greater than one. Naturally in quantum mechanics you set hbar=1, and you end up with one independent unit.

    Now you’re in hbar=c=1 units of quantum field theory, and everything is in powers of a GeV, so to say. If you now include quantum gravity, it also sets G=1, and says that the previous physics only worked at distances much greater than L_{planck} etc. But in these quantum gravity units, all observables are dimensionless, and there is really no other inequality that you can violate to go beyond the reach of validity of the previous theory.

    String theory itself tells us exactly what sort of deformations you can do with its backgrounds and what sort of deformations you can’t. There just does not seem to be any way to deform “away” from string theory. It must be exactly what it is.

  11. Lubos Motl says:

    Nope, quantoken, you were probably skipping your classes of history of physics.

    When I write that Einstein did not believe quantum mechanics, it means that Einstein did not believe quantum mechanics. Your attempts to look for errors in my sentences are counterproductive. You should better look at the serious errors with YOUR thinking.

    When Einstein started to propose his EPR paradoxes, he was sure that quantum mechanics had to be giving wrong predictions for these experiments. It’s not just a matter of philosophy. He believed that the quantum entanglement was impossible, and using the current terminology, he would definitely agree that Bell’s inequalities must be satisfied in reality.

    He tried to invent dozens of possible contradictions in quantum mechanics, and all of these attempts of Einstein were wrong. (Well, one of them was also helpful in the understanding of QM, but not in the way Einstein anticipated.) Every time an experiment showed that Einstein was wrong, he invented another possible problem.

    Moreover, in the last 30 years of his life, he really ignored the new discoveries in physics – like in nuclear physics. When he was constructing his unified field theory (until his death in 1955), it was always just a unified field theory of electromagnetism and gravity because he never considered any of the details of the emerging quantum field theories seriously.

    Your comment that you don’t believe in any limitation of determinism even in 2004 does not surprise me a single bit. I’ve already learned the level of your knowledge and understanding of physics in more detail than what is necessary.

  12. plato says:


    It either applies to the real world entirely and wholly, or not at all. And here is what a professional physicist, Lubos Motl, wrote on his blog : “I am always a bit puzzled by Andy’s [Strominger] statements that string theory is “just another step” – what sort of other step that goes “beyond” string theory but does not invalidate it is Andy thinking about?”

    I can’t speak for Lubos, and as a commoner, the thinking of the journalist would fall comparatively to my domain?:) So if the confusion rests, it would rest in the lack of interpretation that is derived from the roads leading too? We acknowledge the planck Epoch? The timing in Steven Weinberg’s model of the First three minutes?

    The conceptual difference as we now look at this model I wonder indeed how such thinking has been altered.We had to assume that this universe is cyclical in nature? So where do strings fit in? Gabriele Veneziano made this point clear.:)

    It might be that the laws change absolutely with time; that gravity for instance varies with time and that this inverse square law has a strength which depends on how long it is since the beginning of time. In other words, it’s possible that in the future we’ll have more understanding of everything and physics may be completed by some kind of statement of how things started which are external to the laws of physics.

    Richard Feynman

    If such a inquistive naure is not acknowledged then how would we move beyond the boundaries which we like to hold to in the physics of approach. You needed to have these frameworks which would motivate exploration of concepts and then bring back for us, new roads for consideration. Would one quickly dispell Smolins Three road attempt at comprehension?

    Nima and others dimensionally have challenged these current views:)

  13. Arun says:

    Chris W. asks “Why in 2004 are we talking about proving physical theories?” in his criticism of Pathik Guha’s newspaper report.

    The idea is, I suppose, that we can only show that a physical theory corresponds to reality to a certain precision within a certain domain. We can never rule out the possibility of a theory with wider validity, just as Newtonian gravity was replaced by and subsumed within General Relativity. So proof of a theory is not possible.

    But the Theory of Everything (or at least Superstring Theory) is different. It either applies to the real world entirely and wholly, or not at all. And here is what a professional physicist, Lubos Motl, wrote on his blog : “I am always a bit puzzled by Andy’s [Strominger] statements that string theory is “just another step” – what sort of other step that goes “beyond” string theory but does not invalidate it is Andy thinking about?”

    If you disagree, then certainly a newspaper reporter has more excuse to be confused than a physicist. 🙂

  14. Quantoken says:

    Mr. M-* said:

    “Einstein did not believe quantum mechanics, even though he was a genius. There were fortunately many others who did. Feynman did not believe string theory, but there were others who did. You don’t need to go far – Gell-Mann was the guy who created the strong Caltech’s string theory group.”

    Why said Einstein did not believe in quantum mechanics. He certainly believed in the formulations of quantum mechanics and their ability to predict experimental results correctly. He just did not agree with the way how quantum mechanics was interpreted, philosophically. He believed in a deterministic picture of the world. Such belief is reflected by a phrase he said often: “God does not need to toss a dice.” He thought that quantum mechanics is incomplete and must be deriveable from a better theory that is deterministic.

    I believe in the same. Actually I already see what that deterministic better theory is. The physical world can be completely deterministic, but yet still none-predictable, because we observers are part of the universe and we can not obtain the whole information of the whole universe to be able to predict things completely deterministically. The job of acquiring and processing that amount of information would require one whole universe to do.

    There are two kinds of believes. The kind of “seeing is believing”. Which is scientific beliefs. And the kind of “accepting is believing”, like believe in God or Budha. Talking about string theoreticians believing in string theory, without seeing any experimental support. That is “believing with no seeing”, which I think can be classified into the religious beliefs.

    Such a belief look even more like religious belief. Just like you never know whether there is soul and whether there is a God, until the very moment you die, and by then it is too late for you to change your religious belief. Many of the die hard string theoreticians also probably never will know for sure whether they were right or wrong, until they have spent their whole lifes, and probably until they go to heaven and meet Einstein there, and got the answer from him. But by then it’s too later to change their beliefs.

  15. plato says:


    Even Richard Feynman was not sure. When I complete posting on, The Gravity for Instance Varies with Time, you will understand what I am saying in regards to dimension and what even Richard Feynman did not realize.

    Einstein limitations, were his views of what did not exist yet, and his refusal to accept quantum mechanics. We know different now, as Lubos mentions, as you now know, in context of what you quote from that Canto book.

    You have to bring yourself up to speed on the subject?:)

  16. Chris W. says:

    Pathik Guha’s piece struck me as a fairly sophomoric effort (and poorly edited to boot). Why in 2004 are we still talking about “proving” physical theories? Robert Geroch, in his beautiful little book General Relativity from A to B (1981), remarks early on that he can’t imagine what would constitute a proof of a physical theory. Guha’s careless formulation does nothing to clarify the issues now at stake in fundamental physics.

    One of those issues (as Feynman said) is testability and avoidance of “immunizing strategems” when confronting the results or even the possibility of empirical tests. More specifically, in quantum gravity a central problem is how to meaningfully and consistently define observables, when spacetime structure is not only dynamical but quite possibly only an approximation to some deeper structure. From Raphael Bousso (hep-th/0412197, Introduction):

    One problem with quantum gravity is that we dont know what the theory should compute. In particle physics, the most precise observable is the S-matrix. But this quantity seems ill-suited to cosmology, where the observer is not outside the system, initial states cannot be set up, and experiments cannot be arbitrarily repeated to gain statistically significant results.

    This ignorance is not especially unusual or embarrassing. It is rarely clear at the outset what a theory should compute. For example, the insight that gravity is a theory of a symmetric, diffeomorphism-invariant tensor field in itself already constituted a significant part of the development of general relativity. But once a theory is in its final form, the observables should be apparent.

    If string theory is the correct quantum theory of gravity, then whatever it computes presumably are the observables. But string theory—perhaps because it is not in its final form—has so far sidestepped the problem of cosmological observables. It defines quantum gravity for certain classes of geometries characterized by asymptotic conditions, such as asymptotically flat or Anti-de Sitter spacetimes. In these geometries an S-matrix happens to make sense, and string theory computes its matrix elements. (In the case of AdS, it computes boundary correlators, which are a close analogue of the S-matrix.)

    However, we have yet to learn how to apply string theory to cosmology or to an observer inside a black hole, with the same level of rigor as in Anti-de Sitter space. Hence, it would be premature to conclude that the S-matrix will remain the only well-defined object. It is too early to know what, if anything, string theory has to say about cosmological observables.

    Fortunately, classical and quantum properties of cosmological solutions impose significant constraints on possible observables, and may even hint at some of the principles on which a theory computing them must be based. De Sitter space is a case in point. Semi-classical analysis has provided overwhelming evidence that no exact observables exist in eternal de Sitter space—at least, none that correspond to experiments that can be performed by an observer inside the universe. This is related to the presence of a cosmological event horizon in de Sitter space, which limits the accessible information and emits pernicious thermal radiation.

    Of course, as Daniel Friedan has discussed in detail, string theory has difficulties with large distance physics in general, not just cosmology.

  17. Mubos Lotl says:

    “To move forward, Strominger argues, the string theorists should work on the problems that they are able to make a progress on. For example, he says, they should concentrate on explaining why the dark energy that is enhancing the cosmic expansion rate is so small. We arent producing ideas on that, he observes”

    Excellent! So who at Harvard is following this good advice?

  18. Dave Bacon says:

    Lubos: “(violated the copyrights)”

    Come on Lubos, you really believe quoting a small passage of a book with a reference is a violation of copyright law?

    I use all these petty little inccorect statements by string theorists such as yourself as yet more proof that string theory is wrong (that’s a joke, in case you’re flaming powers are firing up.)

  19. Arun says:

    Another Pathik Guha article:

    While it is mostly about Trieste ICTP, here is a relevant excerpt:

    The audience bursts into laughter as Prof. Alvarez-Gaume, while discussing the string theory (experts concept of a single idea to explain all physical phenomena), flashes a surrealist painting on the screen. It shows a castle on top of a huge piece of rock hanging in thin air! I want to convey the idea that the string theory has a rock-solid base, he comments, highlighting critics arguments that the theory is merely a mathematical artifact and has not been corroborated by experiments yet.

  20. Doug says:

    I forgot to include my name in the previous post. My name is Doug, and I contribute here from time to time.

  21. Anonymous says:

    Has the principle of impotence, used so widely and successfully by the proponents of quantum mechanics, come back to haunt them?

    “There are three major problems,” explains Strominger. “One is the problem of finding a theory that in principle can unify all the forces and particles in Nature. The second one is consistently putting together quantum mechanics and Einsteins General Relativity. The third problem is the mathematical explanation of the black hole attribute that we computed. String theory has mathematically solved all the three problems. We dont know if Nature avails herself of the solutions provided by string theory. But, on principle, string theory is capable of resolving all three of the riddles. Loop quantum gravity hasnt solved any one of the three problems.”

    If string theory can solve these problems “in principle,” is that any less compelling than quantum mechanics’ ability to solve chemical problems “in principle?” Larson wrote, years ago, that adherents to quantum mechanics were employing the same “principle of impotence” in regards to chemical problems:

    “The physicists who are attempting to apply the latest quantum concepts to the problem have been singularly unsuccessful, if we appraise their results by any realistic standards. Some very broad claims on their behalf are often made by overenthusiastic supporters, the following from G. G. Hall being a typical example: ‘Quantum mechanics… gives the solution, in principle, to almost every chemical problem,’ but the true significance of such statements becomes apparent when Hall goes on to say, ‘Very unfortunately, however, there is an enormous gap between this solution in principle and the practical calculation of the properties of any specific molecule.'”

    What amazes me is that so few recognize that, mathematical complexity not withstanding, solutions need to be able to deliver the goods or they are NOT solutions. But even worse, now we are talking about “in principle” solutions to problems that are in themselves in principle “solutions.” Quantum mechanics is a “solution” to the Bohr model of the nuclear atom, that has become the “solution” to the behavior of high-energy debris in accelerator collisions. Of course, such “solutions” have to be given 19 or 20 parameters from observation first, and some “reasonable” limit chosen to cast it in, but given these and a few, more subtle, “adjustments,” it can deliver the goods all right, just check out the standard model.

    Well, at least it can if you ignore gravity, that is. We need a “solution” to our quantum mechanics “solution” that is a “solution” to our gravitational “solution,” as well. How inconvenient that the fabric of our “space-time solution” won’t accommodate the vacuum fluctuations of our “virtual sea solution.” If it did, perhaps we wouldn’t need a “solution” to our two, most spectacular, “solutions.”

    It should occur to us that perhaps we have more fundamental problems than unifying the forces of the nuclear atom, or reconciling GR and QM, or explaining hypothetical black hole attributes. Perhaps the problem is that we have overlooked the fundamental definition of force as a PROPERTY of motion. Perhaps, the reason we can’t get there from here is because of our focus on forces and the assumption that forces can exist autonomously, without underlying motions!

    Maybe we ought to go back to the beginning and ask ourselves, “What is motion?” If we did this, we would find ourselves confronted with another basic question, “What are space and time?” Now, Gross says that they are emergent, and Witten says that they are doomed, but what do we care what string theorists say anyway? Let them go back to the comfort and joy of basking in the beauty of mathematics.

    We all know the answer to this basic question and have known since childhood: space and time are the reciprocal aspects of motion, nothing else. Common sense tells us that combining space and time in a continuum may be something that we can think about, and it may even be fun to see what happens since now we know that motion affects time as well as space (now that’s a shocker!), but, come on, let’s face it, Nature is not likely to be out making up such an ungainly expedient.

    We know that radiation and energy are discrete. We know that motion affects time as well as space. These things we know, so it’s not much of a jump to suppose that motion, that is, the only known relationship of space and time, is discrete as well. If motion is discrete, then it follows that its two, reciprocal, aspects are discrete too, and there you have it – we are off and running with a whole new concept that offers us an incredible new approach.

    For instance, since we’ve learned that the speed of propagation of radiation relative to matter is constant, we can take that as a pretty good indication that the reason is that it’s the unit speed of motion: one unit of discrete space per unit of discrete time = a velocity of c. Wow, check out the symmetry in that baby, would ya? Wonder what happens when there are more units of space than time, or more units of time than space in that otherwise perfectly symmetrical equation of motion? What else might be lurking in there? Matter must be related to this basic motion, huh? Maybe magnetic and electric properties of matter are too, ya think?

    Well, to tell you the truth, there is incredible promise in this approach, including, but not limited to, particle masses, charges, and physical constants of all kinds, but we now return you to the continuing saga of legacy physics: the motion of strings in a quantum world where the curves of space don’t bother us, and its fluctuations happily evolve in time.

  22. Anonymous says:

    Hey Santo,

    I don’t quite agree. Feynman kind of realized that he may have been doing a silly mistake. This is what the second part of his answer is all about. I don’t quite understand how “more” can you be aware of your doing a mistake. If he knew exactly and clearly that his viewpoint were mistaken, then he would have changed his viewpoint, would not he? 😉

    I am happy that you (violated the copyrights) and cited Feynman accurately because his authentic quote is much better than my variation of it. But the content is the same.

    If he realized his mistake just one piece more than what he displayed, he would have understood why string theory is the right approach.

    Happy holidays,

  23. Santo D'Agostino says:


    I posted my comment because I did not wish readers to get the mistaken impression that Feynman thought himself making a silly mistake in criticizing string theory. In fact the full interview shows that he emphatically believed that string theory is crazy.

    Of course, I agree that one man’s opinion proves nothing, and no one is suggesting that one should treat the opinions of anyone, even a great man, as God-like. However, I do strongly believe that when one is quoting someone briefly, one must do one’s best to be accurate. But perhaps I should not hold people so strongly to task in an informal setting such as this, as I would in a more formal setting such as in a published article or book.

    Wishing you all the best,

  24. Anonymous says:

    Yes but Lubos did you even address Feynman’s point about not calculating anything?

  25. Lubos Motl says:

    Santo Agostino, you’re missing the point.

    Of course that the silly people often do not think about themselves that they are silly. But it’s irrelevant what they think about themselves.

    Feynman’s genius is that he realized very well that even the other great physicists, such as Einstein, say a lot of things when they’re old – things that look stupid to others.

    He knew well that the old men are too conservative and unable to follow the new stuff too carefully. Knowing this fact about biology is helpful, but it’s not enough to follow the new developments in physics – and Feynman is an example.

    You know, I am one of the biggest fans of Feynman in the world, as my choice of the e-mail address shows much like other things. But looking to the old Feynman as some sort of God whose opinion about very complicated things matters even if there is no actual science behind it – that would be silly.

    Einstein did not believe quantum mechanics, even though he was a genius. There were fortunately many others who did. Feynman did not believe string theory, but there were others who did. You don’t need to go far – Gell-Mann was the guy who created the strong Caltech’s string theory group.


  26. Santo D'Agostino says:

    L. Motl’s “quote” of Feynman makes it sound as if Feynman thought himself silly, but that’s not the sense I get from reading the exact quote (from Superstrings: A Theory of Everything?, edited by Davies and Brown, page 193; I presume this is the source that L. Motl is “quoting”):

    “I have noticed when I was younger, that lots of old men in the field couldn’t understand new ideas very well, and resisted them with one method or another, and that they were very foolish in saying these ideas were wrong — such as Einstein not being able to take quantum mechanics. I’m an old man now, and these are new ideas, and they look crazy to me, and they look like they’re on the wrong track. Now I know that other old men have been very foolish in saying things like this, and, therefore, I would be very foolish to say this is nonsense. I am going to be very foolish, because I do feel strongly that this is nonsense! I can’t help it, even though I know the danger in such a point of view. So perhaps I could entertain future historians by saying I think all this superstring stuff is crazy and is in the wrong direction.”

    When asked what it is that he didn’t like about string theory, he replied:

    “I don’t like that they’re not calculating anything. I don’t like that they don’t check their ideas. I don’t like that for anything that disagrees with an experiment, they cook up an explanation — a fix-up to say ‘Well, it still might be true.’ …”

    He is very emphatic throughout the entire interview, and his full comments are worth reading.

  27. DMS says:

    Actually, I am a bit surprised, as in India journalists tend to be very deferential to physicists/mathematicians (they are VERY RUDE to politicians, unlike US media, I might add!). Atiyah remarked: “I have never received the kind of audience that I have in India. I feel like a pop star.”

    Perhaps the journalist has a good physics background.

  28. Fabio Lanzoni says:

    “Though the experts, not particularly fond of hypes, dont like the name that much..”

    Since when are string theory experts not particularly fond of hype? They seem to spend a good deal of their time cultivating it.

    It’s no accident this interview came from an Indian paper: this kind of skeptical journalism is not in fashion in the US. It can lead to people asking the wrong questions and upsetting advertisers.

  29. Lubos Motl says:

    This Pathik Guha (journalist) is a pretty self-confident guy. 😉 Andy Strominger explains him that string theory has no competitors. Of course that Andy Strominger knows what he’s saying, and his sentence also implies “the enterprise called loop quantum gravity is not a competitor of string theory”. Pathik Guha nevertheless says “Strominger may be not quite right”. 😉

    We like the comment of Feynman very much. “I know that it’s silly, that all good physicists who become old suddenly start to say all these ridiculous things about science. We know that Einstein did the same thing with quantum mechanics. I know that I am going to repeat the same silly mistake (with string theory), but I just can’t help myself.” 🙂

  30. plato says:

    A question

    What’s better—


    Although theres no direct evidence that string theory is correct, Strominger points out, “weve a number of signposts that its on the right track.” What are they? String theory, Strominger explains, has shown beautiful connections with various branches of mathematics.

    —or this?

    Although theres no direct evidence that string theory is correct, Strominger points out, “weve a number of signposts that its on the right track.” What are they? String theory, Strominger explains, has shown beautiful connections with various branches of mathematics.

    In reference to content. So math doesn’t matter?:)This would put mathematics in a really awkward position?

  31. Lubos Motl says:

    Andy just returned from India – he also attended Shiraz Minwalla’s wedding.

Comments are closed.