Wednesday, December 09, 2015 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Polchinski and string theory to the rescue

Joe Polchinski was invited to the Munich conference about the "falsifiability under the light of string theory". Because of his being tired of conferences or a hospitalization (in that case, I wish him to have doctors who safely come to his rescue!), he was the only participant who attended the meeting through a wormhole at where he posted

String theory to the rescue,
a wonderful philosophical essay with a story, an essay of the kind that should be winning all the "quantum gravity essay contests" except that the organizers almost never manage to receive contestants that would be this good. Joe had to ask a speaker to present the paper – a random Nobel prize winner, David Gross, was good enough for the job. Because he likes to convey other people's ideas so nicely, next time, someone could perhaps hire someone like David to spread some rumors from the LHC, too. ;-)

Polchinski's basic point is that quantum gravity seems hard at first sight, and for two reasons:
  1. the Planck length is incredibly short;
  2. and random processes could have decided even about some "universal" features of Nature (like they do in the multiverse).
However, string theory is the shocking good news with the capacity to change the pessimistic expectations.

Polchinski's first hero is Max Planck who introduced the natural Planck units and fully realized how far-reaching their status was. All extraterrestrial civilizations have inevitably invented the exact same units. On the log scale, Polchinski showed how people were getting closer to the Planck scale which is still very far, anyway.

If Planck has happened to have gotten the result of \(10^{-19}\) meters as the Planck length, we wouldn't be sitting in Munich and whining about falsifiability. Instead, we would be anticipating the wonderland of experimental quantum gravity. (Also, we wouldn't be alive because the long evolution on planets requires large enough, long-lived stars and stars are only large if and because gravity is very weak at the level of individual particles.)

Polchinski is a practical physicist so he is not using adjectives such as "post-empirical". That's a very sensible attitude. But the ultrashort value of the Planck length has some implications, anyway. The task seems difficult but there's one cool thing Planck couldn't have seen, the discovery of string theory. He mentions that string theorists are the physicists who have been and who may be great in numerous other disciplines of science and deals with the silly criticism that "they are not sufficiently like Einstein". Surely when it comes to the unity of physics and the ability to influence all of theoretical physics, they are very similar which is why string theory remains so vital even when the amount of data is so scarce.

When it comes to string theory itself, it has about five mostly surprising virtues:
  1. solution to the short distance problem
  2. uniqueness of the dynamics
  3. unification of physics and geometry
  4. gauge-string duality
  5. insights on quantum mechanics of black holes.
In the first point, he compares gravity to the weak force and discusses why the "smearing" is needed to cure the nonrenormalizability and what it means. Planck couldn't have expected that the theories – ways forward – would be so restricted.

However, Polchinski quickly notes that he doesn't want to give negative arguments (why people work on string theory rather than loop quantum gravity, for example – the latter sucks) but positive ones (why they work on quantum gravity at all) and the remaining four virtues are more relevant for that positiveness.

The second point is that string theory's dynamics is unique. When one demands the general universal principle (which we formulate in too technical, context-dependent words so far, he later adds, however), there is nothing that may be adjusted about string theory at all. General relativity seemed to have this virtue as well – the equivalence principle was the master principle. But people realized that the (more relevant, in the technical sense) cosmological term may be added; and, more importantly, the higher-derivative terms shouldn't be truncated, either, because the separation from the Ricci-tensor two-derivative term is artificial.

Quantum field theory has tons of choices on fields, gauge symmetries, renormalizable, and nonrenormalizable interactions. No string-like uniqueness to be found there.

The dynamics-geometry link has first been pointed out by general relativity – the physics of gravity is about the spacetime curvature. This has to hold for all of physics, too, and string theory gives real beef to this general proposition that was just a belief decades ago. Mirror symmetry is a deep mathematical insight with a new physical explanation. Polchinski says that this geometry-physics link is less impressive from his viewpoint because his background is pragmatic, not too mathematical. (E.g. Cumrun Vafa would place this virtue at the very top.)

The gauge-string duality is presented as another step in the physicists' ability to find the common origin in seemingly diverse phenomena – a newer analogy of Maxwell's unification of electricity and magnetism (with a bigger generality in mind, we could mention Darwin's theory of common ancestors, too). In the gauge-theory description, strings are emerging from the same building blocks as hadrons etc.

The fifth point is about the quantum physics of black holes – he mentions Strominger and Vafa and the information puzzle and avoids his bullšit claims about firewalls and the incompleteness of AdS/CFT in the main text (except for a reference). OK, Polchinski views 1,3,5 as very important warriors on the pro-learning team while 2,4 are just remarkable.

On the other side, we find the demons who try to make the learning harder or impossible. The multiverse is a major terrorist. Einstein's and similar equations are beautiful but they have many solutions. Calabi-Yau manifolds are among the compact solutions but the total number of solutions to string theory is even higher than the number of Calabi-Yau topologies. He sketches the inflationary reasons why we might live in the multiverse and frames the positive cosmological constant as positive empirical evidence in favor of the multiverse (via Weinberg's argument).

Fast food co-exists with Smetana's The Moldau nicely. A flash mob in the Vaňkovka Gallery, Brno, Moravia, CZ.

Polchinski says that Weinberg's prediction for the cosmological constant debunks the claims that the multiverse is not predictive. But even if we ignore this single vague prediction, it's still true, he stresses and I also like to say, that it is not up to us to decide how many features of Nature are due to random events.

He offers an amusing and sensible Bayesian calculation of the odds that we live in a multiverse: 94%. The non-multiverse possibility starts fairly at 50% (the prior) and is halved thrice down to 6.25% – because
  1. non-multiverse theories we know seem to predict too high a cosmological constant so it's necessary but a bit unlikely to assume that there are better non-multiverse theories with reasonable predictions for the cosmological constant
  2. the cosmological constant is nonzero (I am a bit confused why he counted this comparison twice but not a big issue)
  3. our candidate theory of quantum gravity seems to make the multiverse more likely than not.
The multiverse may still go away, he reasonably claims, much like other 2-sigma (95% confidence) bumps. You may disagree with Polchinski's figure 94% but there is no valid argument that would eliminate the multiverse at a near-100% certainty. I agree with that.

But I would still get a lower result than 94%, perhaps close to 50%, because I would also add negative contributions to the probability of the multiverse, especially the fact that "the value of all observable constants relevant in the whole universe was so far always calculationally reduced to a smaller number of parameters controlling a more fundamental, shorter-distance theory". It seems "more likely than not" to assume that this experience may be extrapolated by another step (or steps) – this extrapolation is analogous to Polchinski's argument #1 but has the opposite effect – which is why the probability of the multiverse that reduces things to randomness should be approximately halved. And I would stress that even if randomness has played a role in determining the observed vacuum or its low energy parameters, it wasn't necessarily the only "actor".

116 years ago, Planck introduced the units. 116 years in the future, in 2131, people will have the full theory of quantum gravity based on string theory, Polchinski forecasts, along with some predictions whose content can't be predicted by Polchinski in 2015. Don't give up too early – after all, (not only) Polchinski is trying to bring the year quantified as 2131 closer. ;-)

Add to Digg this Add to reddit

snail feedback (0) :