Saturday, December 21, 2013

Alan Guth: a talk at Hunter College

Alan Guth of MIT was one of the nine inaugural winners of the $3 million prize. He was awarded for his cosmic inflationary theory.

In May 2013, he gave a 80-minute talk at the Hunter College, CUNY, New York:

The video was posted on the YouTube channel of the Milner physics prize that you're invited to look at – it's full of acceptance speeches by the winners, aside from long talks by Polchinski, Linde, and others.

Alan was introduced as a postdoc from 1979 who was visited by a muse and biked to his office at Stanford or SLAC. Then he won the Milner Prize a year ago. How did it feel? He had $200 on his account on the previous day; and $3,000,200 on the following day. Alan had to clarify this common misconception. In reality, he only had $3,000,188 because he was charged a $12 wire fee, an event that made him really upset. ;-)

Guth starts his lecture by basic explanations of the rudimentary big bang theory. He said something like
Our whole universe was in a hot dense state,
Then nearly fourteen billion years ago expansion started. Wait...
The Earth began to cool,
The autotrophs began to drool,
Neanderthals developed tools,
We built a wall (we built the pyramids),
Math, science, history, unraveling the mysteries,
That all started with the big bang!
And so on. Afterwards, he would talk about miracles, the repulsive gravity allowed by Einstein's theory, and inflation – the event that gave the bang to the bang and naturally produced (i.e. the cosmic inflation theory has explained) the amazingly uniform *and* flat Universe we see. He spent some time with the multiverse. His own contributions to inflation are mostly independent of the multiverse but Alan is a moderate multiverse believer of a sort. It may be a good idea to hear the story about such theories from non-zealots like Alan.


  1. Apparently the new PC term for "retarded" or "disabled" is
    "diversely abled" ---we are becoming an ever more ridiculous species.

  2. "because he was charged a $12 wire fee, an event that made him really upset." Rather than foster brilliance we allocate for its suppression. Obamunism is not merely a Beltway lobotomite clown car of incompetence and criminality, or obdurate somnambulant legislative stupidity, it is compulsory degradative egalitarianism.

  3. Luboš, does the experimental violation of Bell's Inequality exclude ALL conceivable hidden variable theories, or simple a broad subset of them?

  4. Hi, the experimental violation of Bell's inequality excludes all *local* theories of hidden variables. But Bell's theorem is just a tiny portion of scientific evidence. Other experiments pinpoint quantum mechanics and exclude pretty much all hidden variables theories including non-local ones.

    The proof of that claim is somewhat general and having the limited physics level of rigor. However, there are much more specific and rigorous ways and experiments to exclude very large classes of nonlocal hidden variable theories, too - see e.g. Zeilinger et al. 2007:

  5. I think that James knows the difference between can't and don't know. He seems intent to grab things out of context so to restore a reality that is long gone. Of course I think but can't know that he knows more than he will allow himself to accept. What I can't know is how he decides when he will superpose antagonizing and complimenting on this blog.
    I loved the Russell quote.

  6. Oh yes, that one from Alan Guth I have already seen, it was nice :-)

    But it is a good idea to check the FPP sites from time to time for other nice talks popping up ...

  7. You are over your head here, Justin. God really does play dice; Einstein was wrong and so are you. Give it up.

  8. My twin brother is also a retired professor of chemistry and he willingly admits to a limited understanding of QM. He does, however, grasp the basics far better than you or your uncle.
    You are not retarded but you are unwilling to learn something that has been perfectly clear for almost nine decades.
    Lubos is simply expressing his frustration with your bullheadedness. I don’t blame him.

  9. Good try, lucretius, but you are probably butting you head against a wall.

  10. By “a philosophical problem” you mean that you just don’t like the way QM works; that’s all. Tough shit.
    Lubos has spent an ungodly amount of time trying to educate you folks, far more than I would have the patience for. If he explodes in exasperation that is the inevitable result of dealing with closed minds.

  11. Yes. I appreciate Lubos' position and the time he is spending in explaining theoretical physics to people who either cannot or do not have patience or time to go through technical papers. Only thing I was talking about was that it is interesting that great physicists (in past) like Schrodinger (perhaps including Feynman also) and (present) t'Hooft , Weinberg and several others are interested in foundations of QM and regard the subject as still open. This statement does not take anything away from Lubos' arguments. His opinion is perfectly OK for me. I still hope to learn theoretical physics from his blog. Of course, I am interested also in metaphysics, specially at my age!!! But that is another story.

  12. My pacifism has nothing to do with quantum mechanics. I was pointing out that it is mean to pick on someone so peaceful. There's no need to hurl nasty insults at someone who has always been so nice.

    This will probably be the last post I make on this blog maybe forever, though I will continue to visit.

    I just want to say one last thing to you Gene. You say I'm wrong and God does play dice. I'm not sure if you've ever read anything I've ever written, but I say over and over again that God does play dice, that there can't be any hidden variables as envisioned by Einstein. It's the consequences of the no go on hidden variables that I'm interested in. You should read some of Bell's papers on the subject. Here's a good book:

    He fully accepts no hidden variables, no signalling faster than light. He's bothered by other things. Insult him and me if you will, but please try to stop hurling wrong criticisms of us.

    This will be my last post (which Motl can interpret as a Christmas present).

  13. I don't think you should be insulted. However, you have misread, misquoted and ignored the content of so many for so long that it is understandable that they lose patience with you, IMHO.
    Anyway I think it would be a shame if you don't choose to get past the mid 20th century which is really the last time your ideas did not get crackpot awards automatically. My undergraduate major was chemistry but I am supposing your uncle was not my professor because mine taught me right off

  14. The talk was very nice indeed, and Edward Witten explained how closed string field theory works in principle so nicely and patiently :-)

    Now I know all I need to know ... ;-P ;-)

  15. Hi Lubos,

    Around 20mins in, Guth describes his 2nd miracle: he says that the energy of the universe is zero. He says that the positive energy of matter is cancelled by the negative energy of gravity.

    This sounds wrong to me.

    But what do you make of Guth's comments about energy in GR?


  16. So is Guth right that the energy of the universe is zero?

  17. Yes, sort of. He likes to say that the Universe is the ultimate free lunch.

    More precisely,

    in cosmology (general/compact situations in GR), there is no objective enough conserved nonzero quantity. The energy conservation - which is needed to make the term "energy" meaningful - holds as a result of the time-translational invariance of the laws of physics.

    The general background is not time-translationally invariant, so there is no corresponding conservation law. Alternatively, GR is invariant under these translations but they are a gauge symmetry, so the corresponding conserved quantity exists, but it is identically zero (gravitational field's contribution cancels the matter's contribution).

    Yet another alternative is to define some energy that is conserved but nonzero but the value is so heavily dependent on the choice of coordinates that one can't derive anything about the final state out of this quantity's conservation law.

    At any rate, the nonconservation of energy in cosmology is something that is easily seen on the radiation and/or dark energy, and inflation is producing an exponentially growing space with a constant energy density "out of nothing", so it creates the energy of all the visible Universe pretty much out of nothing at the beginning.

  18. kashyap vasavadaDec 25, 2013, 2:26:00 AM

    Hi Lubos: I would like to understand this statement "Alternatively, GR is invariant under these translations but they are a
    gauge symmetry, so the corresponding conserved quantity exists, but it
    is identically zero (gravitational field's contribution cancels the
    matter's contribution)." mathematically. How do you show that E=0 under gauge symmetry? If it is involved, can you give a reference?

  19. Thanks for your interesting response; and I agree with your comments about energy in GR.

    In his talk, Guth claims explicitly that energy is conserved in GR, and that the positive energy of matter is cancelled by the negative energy of the gravitational field.

    This sounds to me a lot like the false claims that Philip Gibbs has made.

    So it seems Guth has the same misconceptions about energy in GR as others do, such as Gibbs.

    Would you agree with my assessment, Lubos?

  20. No, I find the answer that "the total energy is zero because of the precise cancellation" between the gravitational and non-gravitational energy to be a completely legit explanation why there isn't any nontrivial energy conservation law in the context of cosmology, and it was one of those I listed in my previous answer.

    The point that this "revokation" of the conservation law is the reason why one may create lots of matter out of nothing is the very same point Alan Guth is making. Phil seems to deny it - he also seems to be obsessed with some gauge-dependent vacuous definitions of energy that are *nonzero*.

  21. Dear Kashyap, the straightforward definition of the "total" stress-energy tensor in GR is zero because it is proportional to the variation

    T^{mn} = delta S / delta g_{mn}

    but all such variations vanish once g^{mn} becomes a dynamical variable: the vanishing is the Euler-Lagrange equation of motion! The vanishing "total stress-energy tensor" is the appopriate difference of the matter's stress-energy tensor and (the multiple of) Einstein's tensor i.e. the difference between left and right sides of Einstein's equations.

    The matter part of this "variation of action with respect to the metric" definition of the stress-energy tensor may be seen to be equivalent to the Noether's current, up to possible (usually absent) extra terms that are conserved tautologically, because the consistency of GR really implies that the "gauge field" g_{mn} must be a conserved current, and on dimensional grounds, it must be the the current related to energy (and time-translational energy). One may also verify the tensor for each specific set of matter fields you care about, on a one-by-one basis.

  22. kashyap vasavadaDec 25, 2013, 6:24:00 PM

    Thanks Lubos for the detailed reply. I understand the main point. I knew about gauge symmetry in connection with Higgs field (QFT) but do not know in connection with GR. Admittedly I have very little knowledge about GR in the QFT theory context. I need to brush up my GR some more!

  23. kashyap vasavadaDec 26, 2013, 4:53:00 PM

    Great answer Lubos. I don't know if you will see this question in an old blog. But I will ask anyway. If E=0 initially, then by the same argument, E=0 even now. Then what about all this talk about non conservation of energy in GR. What am I missing? Thanks.

  24. Well, if a quantity is defined so that it is equal to zero for any configuration of the physical system, then the vanishing can't be considered to be a conservation law and what it "generates" isn't a symmetry, it's just an identity. Such a "conserved quantity" is completely vacuous. One may always "find" an arbitrary number of such "conserved quantities" J_i for any set of indices "i" so that J_i = 0 but such J_i is completely nonsensical, vacuous, and it's stupid to talk about it.

    Moreover, in the case of energy, any natural enough definition of energy that is useful for anything will agree that E isn't zero today.

  25. kashyap vasavadaDec 26, 2013, 6:16:00 PM

    Ok. Do I understand that this is because gauge symmetry was valid at the time of big bang but it is broken now. So arguments do not apply. If this is right then I understand the whole issue!

  26. No, sorry, kashyap, gauge symmetries can never be broken. The point is that gauge symmetries are redundancies - they relate configurations or quantum states that are not only "analogous" but that must be considered literally identical. So a generator of a continuous gauge symmetry must annihilate the quantum states - its eigenvalue has to be zero. So this would-be symmetry acts trivially (no transformation at all) on the physical state and shouldn't be called a physical symmetry at all. In the same way, the generator, a would-be conserved charge, doesn't depend on any observables so it's just a constant, not an independent conserved quantity.

  27. kashyap vasavadaDec 27, 2013, 4:41:00 PM

    Ok. Thanks Lubos. I am beginning to see your point. I think many of us would like to read a detailed article on gauge symmetry, covering both GR and QFT ,with some math at intermediate level , If it is not around, then can you write one when you get time?

  28. But the there is a subtlety here kashyap. Isn't there?

    I mean for a general gauge transformation Λ(x) there is a subset of global transformations for which Λ(x) = α is a constant.

  29. kashyap vasavadaDec 27, 2013, 6:07:00 PM

    Are you talking about global and local gauge symmetry? Can you clarify? I thought global symmetry was not that useful.

  30. kashyap vasavadaDec 29, 2013, 3:55:00 AM

    Finally I got enough courage to listen to Witten’s talk. I will have to listen to it perhaps five times before I understand it.I have a preliminary question, quite late for this blog. Schwinger’s method is certainly
    very elegant method to use for Feynman diagrams. Is it used to actually calculate Feynman diagrams?

  31. Dear Kashyap, Schwinger parameterization is a method that every other good QFT course teaches!

  32. kashyap vasavadaDec 29, 2013, 1:31:00 PM

    Dear Lubos: Well. I learnt QFT in ancient times when it was just enough to use Feynman rules mechanically!! Then I went into other areas like Biophysical NMR and much later into Quantum Optics. So this stuff is new
    for me. Anyway this is very interesting
    talk and lot of it is understandable by people who have only rudimentary ideas of theoretical physics. I liked the take home message that QFT can be interpreted as a one dim GR and ST as a two dim GR. Thanks for posting this.

  33. kashyap vasavadaDec 29, 2013, 1:35:00 PM

    I forget to add somethink. I suppose one can make g as higher dimension to get membranes etc. Is it?

  34. You can use the Feynman rules "mechanically" - the integrals are the same integrals all the time, after all - but to evaluate these integrals in terms of some simpler ones, you either need a brute numerical calculation or some clever tricks. If you follow the latter, the Schwinger parameters pretty much appear during the calculation, anyway!

    No, it's one of the important points that this nice spacetime theory obtained from interacting d-dimensional objects only works for d=1 i.e. the world sheet dimension D=2, not for higher values. Membranes and so on do appear in string theory but they're not equally fundamental as strings. The same method applied to strnigs doesn't produce any other theories like a "membrane theory". String theory is absolutely unique.

    For membranes and higher-dimensional objects, their internal D-dimensional world volume theory will always contain some components of the world volume metric tensor that won't be decoupled. So it will be a naively quantized theory of gravity with physical graviton polarizations, and such a theory suffers from UV problems and nonrenormalizability by itself! So you might think that you're curing these short-distance problems in the spacetime but they do reappear in the world volume.

    Strings are the only objects for which both the spacetime short-distance problems and the world volume (world sheet) problems are under control - and it's due to the ability of the 2 diffeomorphism parameters plus 1 Weyl scaling of the metric to locally eliminate all 3 components of the D=2 symmetric metric tensor.

  35. kashyap vasavadaDec 30, 2013, 2:12:00 AM

    OK. Thanks. If I understand, finiteness is a great virtue of ST. More people should appreciate this point. In QFT
    one has to go through renormalization procedure which some people regard as
    mathematically unsound and artificially implemented.

  36. kashyap vasavadaDec 30, 2013, 6:28:00 PM

    Hi Lubos: I am beginning to understand this stuff slowly.
    Question: To go from QFT to ST Witten makes real t into complex tau. Have people tried hyper complex (such as quaternions) where there are more than one complex numbers to go to higher dimensions. I am
    almost sure; someone must have tried, since these are very smart mathematicians. There may be still other difficulties you mention above.

  37. Hi Kashyap, the dimensions of the world sheet and the moduli space are complex because there is a doubled number of real parameters and there is a natural complex structure on these real dimensions.

    The second real parameter from the "length of the neck" is the "twist angle".

    Quaternions here would be just equivalent to 4 real dimensions. The theory in these 4 world volume dimensions would be non-renormalizable like the usual 4D GR without extra structure.

    But there actually exist proposals that the extra quaternion-like structure could make membrane-with-two-time-dimensions consistent, and be equivalent to string theory.

    Nothing like that is known that would include octonions AFAIK.