KKLT Smackdown

I was dubious of the value of a new “sci.physics.strings” newsgroup when it was first proposed, but now must admit it seems to have been a great idea It started up a week or two ago, and quickly someone asked the seemingly innocuous question of how many different possible vacuum states were expected in string theory. This is a hugely controversial issue among string theorists, largely because recent evidence is that the number is definitely astronomically large, and this makes it very unlikely that current ideas about string theory can ever be used to predict anything about the real world.

A lot of the discussion revolves around the “KKLT” proposal for constructing a large number of these vacuum states. The acronym is the initials of the authors, three of whom are at Stanford: Shamit Kachru, Renata Kallosh and Andrei Linde. Also at Stanford is Lenny Susskind, who has been spending the last year or so going around giving talks on the “Landscape of String Theory”. It’s hard to believe this, but Susskind’s claim is essentially that the lack of predictivity of string theory is a good thing, since it allows so many possibilities that anything can happen. One can then invoke the “Anthropic Principle” to explain why the world is the way it is. It seems that Susskind is even writing a book about this wonderful “discovery”.

Amazingly enough, the thread about this on sci.physics.strings, entitled “Conceptual question”, has brought a public attack on the “Stanford propaganda machine” by a well-known European string theorist (Wolfgang Lerche), a detailed defense of his ideas by one of the KKLT authors (Kachru), contributions from the inimitable Lubos Motl from Harvard, and, while I was writing this, a defense of the anthropic principle from Joe Polchinski just appeared, which attacks the “cult of monovacuism” embodied by David Gross and Ed Witten.

Thanks are due to the creators of this newsgroup. Pass the popcorn!

This entry was posted in Favorite Old Posts, Multiverse Mania. Bookmark the permalink.

8 Responses to KKLT Smackdown

  1. Peter says:

    Hi String Theorist,

    The argument that maybe the things we don’t understand about the standard model are artifacts of history and not something a more fundamental theory can predict deserves a longer response than I have time for right now. It’s an important issue and I’ll try and write up some thoughts about it soon.

    You’re not answering the question I’ve been trying to get an answer to from string theorists: what exactly do you expect string theory to be able to ever predict?

  2. D R Lunsford says:

    Comments to comments by Anonymous Stringer:

    “Just because string theory is a “theory of everything” does not mean it can be used to calculate everything! Let me explain by analogy.”

    This is like absurdist fiction. No other comment required. One invokes the “prima facie preposterous idea” argument with success.

    “Do you think that Newton’s theory of gravity is vacuous because it cannot predict the number of planets in the solar system, or calculate their distances from the sun (0.39, 0.72, 1.00, 1.52, etc A.U.)?”

    This is an absurd analogy. What Newtonian gravity comprehensively succeeds at is explaining orbital motion and dynamics, and that is all that is required. The proper analogy would be a version of string theory that, given a proper limiting argument, showed that weak interactions did not conserve parity, explained the tiny mass of the neutrino, gave a theoretical justification for the phenomenology of the Higgs mechanism etc.

    “One of the hallmarks of scientific progress is that frequently, what we had thought were interesting scientific questions turn out to not be scientific questions at all.”

    This is more absurdist fiction. The same questions – what is motion, what is the structure of matter, etc. – are as self-evident and vital today as they were in the time of Democritus. There are no real questions of science that later turn out to be misguided – what DOES exist is a clear history of trying to explain away the uncomfortable facts, as in mechanical models of the ether.

    “Nowadays, of course, we understand that there is no reason to require physics to be able to calculate the number of planets, or their orbital periods or sizes. They are what they are.”

    This is sheer sophistry and an embarrasing confusion of phenomenology with theory.

    “This is why I asked you what you thought string theory should be able to calculate. And, as I expected, the things you mentioned are all things to be observed, not things to be calculated.”

    What can be calculated can only be determined after a theoretical model is in place. There is no real model in string theory other than the on-the-face-of-it preposterous idea of fibrills in an umpteen-dimensional wonderland without any bearing on history or reality.

    In fact the goal of any subsuming theory should be to explain the previous one by a limiting argument that not only shows why the last one failed, but how the new one points to new phenomena to be expected. All the previous successes in physics are like this.

    “It is no more “necessary” for the gauge group of the universe to be SU(3)xSU(2)xU(1) than it is “necessary” for there to be four gas giant planets in our solar system.”

    This is totally wrong-headed. Particle physics is phenomenology that will one day be explained theoretically when the excresences of this generation are forgotten. The gauge group of the Standard Model is simply the encoding of the phenomenology – this is not a criticism, rather praise for experimentalists and field theorists to get at least part of it unmistakably correct. There is more truth in one generator of this “arbitrary” group than in all the decades of windmill-tilting of stringers.

    “Similarly, when you say “Calculate for me the fermion mass matrix” it sounds the same to me as “Calculate the masses of the planets.”

    More absurdist fiction. A real encompassing theory, such as the one that explains the relation of electricity to magnetism, always includes dynamical explanations of previously phenomenological inputs. The masses of the Fermions are the LEAST things to expect from a good theory. And in fact, there already exists a good theory of Fermion masses based on analysis of Clifford algebras – so clearly this is not going to be the hardest problem.

    “There are still plenty of people clinging to the modern day equivalent of Bode’s Law or Kepler’s solids, believing that there might be a unique consistent quantum field theory which therefore is necessarily the one describing our universe. But I think this point of view will soon be widely seen as outdated.”

    Who are they? The only physicists who seem manifestly Dadaist to me, are you people. If you only knew how foolish are the things you claim, you would desist and try something else.

  3. String Theorist says:

    Hi,

    I think it is very important to discuss the issue of what kinds of things we should and should not expect to be able to calculate within string theory. Just because string theory is a “theory of everything” does not mean it can be used to calculate everything! Let me explain by analogy.

    You seem to think that string theory is vacuous because (among other things) it cannot predict the gauge group, or calculate the fermion mass matrix.

    Do you think that Newton’s theory of gravity is vacuous because it cannot predict the number of planets in the solar system, or calculate their distances from the sun (0.39, 0.72, 1.00, 1.52, etc A.U.)?

    This is a fair analogy: Newton’s gravity is a theory of everything relevant to the dynamics of non-relativistic gravitationally interacting bodies (such as planets), while string theory is a theory of everything relevant to the dynamics of elementary particles (such as gauge particles and fermions).

    One of the hallmarks of scientific progress is that frequently, what we had thought were interesting scientific questions turn out to not be scientific questions at all.

    For example, the ancients thought there were five planets, and spent a great deal of intellectual effort theorizing why there HAD to be precisely five planets. Once it was realized that the Earth also orbits the sun, Johannes Kepler tried to explain the existence of precisely six planets (and to explain the sizes of their orbits!) by inscribing the five Platonic perfect solids between the orbits of successive planets.

    The discovery of Uranus in 1781 relegated the Platonic solid idea to the history books. Now the best theory was Bode’s law, originally pointed out in 1766 by Johann Titius, which predicted the location of Uranus almost spot-on! Unfortunately, Neptune didn’t fit into the picture very well.

    Nowadays, of course, we understand that there is no reason to require physics to be able to calculate the number of planets, or their orbital periods or sizes. They are what they are.

    This is why I asked you what you thought string theory should be able to calculate. And, as I expected, the things you mentioned are all things to be observed, not things to be calculated.

    It is no more “necessary” for the gauge group of the universe to be SU(3)xSU(2)xU(1) than it is “necessary” for there to be four gas giant planets in our solar system.

    Similarly, when you say “Calculate for me the fermion mass matrix” it sounds the same to me as “Calculate the masses of the planets.”

    There are still plenty of people clinging to the modern day equivalent of Bode’s Law or Kepler’s solids, believing that there might be a unique consistent quantum field theory which therefore is necessarily the one describing our universe. But I think this point of view will soon be widely seen as outdated.

    String Theorist

  4. Peter says:

    The list of things an improvement on the standard model should be able to predict is well-known: choice of gauge groups and representations in the SM, gauge couplings, Higgs sector parameters (including fermion mass matrix), Newton’s constant (for instance as ratio of Planck/GUT scale), cosmological constant. Presumably the standard model breaks down at some energy scale; one should be able to say what that is and predict what new phenomena occur above that scale. Especially impressive would be a prediction of one of the still unmeasured SM parameters (the Higgs or neutrino masses).

    The closest thing I know of to a framework that can do any of this is the supersymmetric GUT framework, where one can get a prediction of one gauge coupling in terms of the two others that works at the 10% level. Unfortunately this comes with a lot of baggage, including another hundred or so undetermined parameters of the supersymmetric extension of the standard model. If one is pursuing this idea about unification, these are on the list to be predicted.

    I’m curious how string theorists look at this. What are the prospects now for using string theory to make any predictions of the sort listed above?

  5. String Theorist says:

    Hi Peter,

    OK, I’ll bite. You wrote:

    Last year as a joke I was telling people that the way things would end up is that string theorists would simply declare victory, announcing that string theory was the final TOE even though it was inherently incapable of calculating anything.

    What exactly would you most like to see us (try to) calculate? Maybe you could give us a 10 most wanted list?

    String Theorist

  6. Peter says:

    Hi Thomas,

    I did notice that Lerche was on the “multivac” side of the discussion; that he was just annoyed that others were claiming credit for it.

    On the sci.physics.strings newsgroup only Motl seems willing to defend the “univac” side (presumably Gross and Witten still haven’t given up, but are unlikely to take to posting on a newsgroup).

    Last year as a joke I was telling people that the way things would end up is that string theorists would simply declare victory, announcing that string theory was the final TOE even though it was inherently incapable of calculating anything. This year this doesn’t seem to be a joke.

  7. Thomas Larsson says:

    It seems that Wolfgang Lerche does not object to the AP per se. Rather, he complains that people who proposed the AP back in 1986/87 (himself??) did not get credit for it.

  8. D R Lunsford says:

    Heh, made me look. Without this blogodatum I would have probably avoided it.

    -drl

Comments are closed.