Saturday, November 08, 2014 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Nima's public lecture at Perimeter

Nima Arkani-Hamed is the world's leading particle phenomenologist, at least among those under 45 years and my wonderful ex-colleague (and co-author). He's been mentioned in many contexts on this blog. More recently, he became a highly mathematically sophisticated theorist, too. That's why we would read about the twistor minirevolution that culminated with things like the amplituhedron, too.

But a day ago, he gave a popular lecture at the Perimeter Institute in Waterloo, Canada. Because it's a popular lecture, it spends more time with things that other people have discovered, and not just Nima's own research. You may see the whole 85-minute lecture (the first rectangle), along with the people's comments (the second rectangle), here:

Perimeter Institute Now Streaming (latest live broadcast is here)
He was supposed to talk for only hour only, and then take questions. In reality, all of these 85 minutes were just his talk.

Like in many other popular talks, Nima describes the twin revolutions of the 20th century physics, relativity and quanta, and how they led to quantum field theory as our current "theory of nearly everything" (I think that this is Lisa Randall's term). And if you have listened to many popular talks, you finally get a chance to get a bit further.

There are Feynman diagrams and Nima describes the simplicity of the Standard Model based on particles and the corresponding fields. Conceptually, the picture is very simple. Why have the people missed it for centuries? Because something that looks like "details" from the short-distance viewpoint completely obscures the common origin of all the particles and forces we know.

He tried to focus on the interesting philosophical points associated with the Standard Model. Why did the things end up being so simple? Why are there just cubic vertices (well, there's also the quartic Higgs self-interaction)? Why is the menu offering the allowed spins so limited? He gets to quantum gravity, mentions why it's hard to write consistent, convergent formulae for the graviton scattering, why the string-theory-based amplitudes are the only ones we know that it work.

Nima also tells you that it's wrong to imagine that general relativity is an irreducible primordial theory generating all the wisdom – as well as special relativity as its limit. Instead, general relativity may be derived from more elementary starting points – from special relativity and gravity (and that's pretty much how Einstein did that, although in a less straightforward way). Elevators and carousels aren't really needed; the GR is a mathematical consequence of the (equivalence principle obeying) gravity and special relativity. Quantum mechanically, this produces spin-2 particles, the maximum possible values for massless (or light) elementary particles.

Units earned some special discussion. He says that he's never been quite a master of units and he confirms this point when he says that "2 meters" means "2 times a stick somewhere in Paris". In reality, it has meant 2/299,792,458 light seconds (light second is the path that light makes in a second) where a second is some number of periods of a specific atomic radiation (used in atomic clocks). And even before the speed of light was fixed to be 299,792,458 m/s, one meter was defined as some number of some wavelengths of a radiation, not from a French stick. And before that and before a short period of the length of a stick, one meter was 1/40,000,000 of the meridian etc. (In a sentence, Nima reveals that he does realize the new definitions.) It's stupid to use human-centered units such as the Parisian sticks and weights because it's expensive for our extraterrestrial friends to buy air tickets to France.

OK, some Nima presents some natural units and dimensional analysis to relate the weakness of gravity, the highness of the Planck scale, etc. He heuristically derives the inverse square laws from quantum fields. Now, why the spins are limited? He introduces helicities and says that it's very informative to focus on spinning massless particles decaying to two lower-spin massless particles at angles. By symmetries, he shows that the 3-particle interaction depends on no kinematic continuous parameters. It's a (coupling) constant. That's simple and helps us to analyze the 4-particle interaction. Now you have everything you need, you go to the grad school and learn to perform half a page of algebra. You find out that it may only work for spins 0, 1/2, 1, 3/2, 2, otherwise the function of energy and angles is inconsistent (with the Lorentz symmetry etc.).

You choose how many spin-0 and spin-1/2 particles there are, you choose how many spin-1 particles by picking a gauge group. The spin-3/2 haven't been seen but they shouldn't be missing and their number is up to 8, and the spin-2 particle is unique, the graviton. Then he talks about extrapolations to higher energies, experiments, and so on.

Why the quantum gravity is hard? How hard it is to produce consistent formulae for graviton scattering? How does it break down in extreme conditions? None of the proposals work – but the Virasoro-Shapiro formula (closed string counterpart of the Veneziano amplitude – a notorious crackpot has called it the "Veneziano amplitude" and be sure that the Veneziano amplitude does not describe gravity because it applies to open, not closed strings; the graviton expression differs by an extra polynomial prefactor from the V-S formula). It's a ratio of generalized factorials, no ghosts are there, it just looks great.

Nima points out that as a child, one may be confused and think that people chose strings "randomly" from an infinite list of possible shapes hiding inside the particles, including fractals and little green men. But it's the other way around. They tried to assume no compositeness at all and just reached the formulae that had the desired properties. And after that, they reversely extracted the shape of the underlying objects – and they were strings. So the arguments forced them to learn about the shape; it is not true that the stringy shape was the starting assumption. This success doesn't prove that there are no other formulae that work but the success does prove that the task is amazingly constrained.

He points out that this uniqueness of string theory is no sociological conspiracy. Imagine that someone in the audience finds another way for these things to work. Tomorrow, she will become the superstar of theoretical physics. How could we have overlooked this extra loophole for 45 years? How is it possible that you, a non-member of the club, could find it? The point is that there is a way to decide "right or wrong" before any experiment is done here: even a chance to be right is extremely precious and low.

Nima says that the issues of the graviton scattering are not the deepest problems of the unification of GR with QM. To measure things (distances etc.) exactly, with a great resolution, we need huge apparatuses – but they gravitationally collapse in the presence of gravity. So the spacetime inside isn't operationally real; it can't be probed arbitrarily accurately. It's hard to define exact observables in quantum cosmology (in de Sitter space etc.).

Also, the vacuum is exciting in quantum field theory. In the presence of gravity, it also curves the spacetime; the cosmological constant problem is hard. Fine-tuning. A similar (but numerically less severe) problem with the low Higgs mass. The spin-3/2 particles and SUSY may be the solution of that, for the hierarchy problem. SUSY is like a theory of extra dimensions but anticommuting ones. Superpartners follow from that. It reduces the cosmological constant problem (from the discrepancy of 120 orders of magnitude down to 60 or so) – it cancels the fluctuations between bosons and fermions. With SUSY breaking, this solution is still inadequate.

And there's lots about the possible future discoveries at the LHC – SUSY by 2018 or so or the multiverse, if I simplify it a bit, and so on. Our time is closer to a decisive time, a revolution. The big questions are already being asked in a meaningful way so it's an exhilarating epoch to do physics.

The comment section is fun. For example, there is an 11-year-old kid from Japan excited about quantum mechanics, and many more. ;-)

Add to Digg this Add to reddit

snail feedback (18) :

reader john said...

Lubos, why Arkani-Hamed says that quantum mechanics need to modified ?

reader Luboš Motl said...

Dear John, could you please be more specific about the timing where he says it to see the context? Thanks.

reader john said...

Dear Lubos, you can start from 6:00, he says quantum mechanics `may` need to be modified, not need to, my mistake. He doesn`t explain why. But he also said similar things in a couse he had gived in perimeter:

I watched it a year ago, it is almost impossible to find where it is. But i remember that it is something like you can`t really measure things using infinitely macroscopic apparatus in principle (I tell you what I remember vaguely, details may be wrong). Also I he said something ADS/CFT may offer a solution to that measurement problem. Here measurement problem is not the usual measurement problem people talk about, but about impossibility to construct an infinitely macroscopic measurement apparatus.

reader Luboš Motl said...

I think that he explained very accurately what he means and why. Most of the previous minutes are spent with emphasizing that QM and relativity are true, so it's a bit bizarre that you focus on this.

In that segment, he said that "at least one of those two things" will have to be modified. Spacetime is doomed and it is "conceivable" that QM will have to be modified in some very advanced questions in cosmology. I agree with that although it's a speculation - no modification that would help the laws to work in quantum cosmology is really known in literature, but it's also true that no complete formulation allowing one to calculate all calculable things in quantum cosmology is known, either.

Indeed, he never discusses the "measurement problem" popular among the "interpreters of quantum mechanics" because there is no problem. The "problem" that he discusses under those words are linked to the usual well-known facts that one needs high energy/mass to probe short distances, and high masses collapse to black holes etc.

The whole AdS space is "compact" in the conformal sense but its proper volume is still infinite and the energy is unbounded which is why one may define and measure some quantities arbitrarily accurately. The gravitational uncertainties go away in AdS, a point that is made obvious by the dual, non-gravitational CFT description.

reader john said...

Thanks for your reply Lubos. I have focused on this because I knew that I have heard he said space-time is doomed and quantum mechanics also may need to be modified a couple times, I have found one of them:

See from 01:15:00

reader john said...

I think the problem is not directly linked to probing short distances. It is about cosmic horizon, i.e., you can`t have an infinite apparatus because you don`t see anything after some distance.

reader JollyJoker said...

Off topic (haven't had time to watch the video)

Fairly tight bounds on DM interaction strengths vs mass with some assumptions on how it affects pulsars,

reader Uncle Al said...

No Equivalence Principle (EP) violation for 1.74 solar-mass 465.1 Hz PSR J1903+0327, 1.05 solar-mass star, 95.17-day binary system: –15.3% vs. -0.000369% gravitational binding energies; 1.8×10^11 vs. 30 surface gees, 2×10^8 gauss vs. 5 gauss magnetic fields; compressed superconductive protons and superfluid neutrons vs. hydrogen plasma; extreme isospin and lepton number divergences; pulsar 11% of lightspeed equatorial spin are EP violation-inert for orbit, periastron precession, and gravitation radiation orbital decay.

Test spacetime geometry with geometry. Compare single crystal test masses with all atoms in self-similar homochiral helices, left-handed versus right-handed. Given alpha-quartz, twice {20 grams as 4 single crystal test masses} compare 6.68×10^22 pairs of opposite shoes (pairs of 9-atom enantiomorphic unit cells).

Nothing prohibits EP violation in geometric Eötvös experiments. All observables exactly cancel except geometry - zero wiggle room. Do it.

reader Tony said...

Great lecture! He is radiating passion and fun.

reader Justin said...


I'm curious if you agree with Nima at around 20:00 min that Quantum Mechanics + Special Relativity = QFT --> General Relativity. For example, do you think without Einstein's work in 1916 theorists would have found General Relativity almost as a corollary of quantum field theory as Nima suggests.

reader Luboš Motl said...

Of course that I agree with that. QM + special relativity imply that QFT is a good effective theory of the resulting theory at long distances, and when spin 2 particles are allowed, (quantized, particle-based formulation of) general relativity is a part of this field theory, too. One gets spin-2 fields, finds ghost modes, realizes they have to be removed by a gauge symmetry, and finds that the diff symmetry is the only symmetry that works.

Yes, of course that GR could have been found in this way for the first time. In reality, QFT began to emerge in the late 1920s and early 1930s and key insights about gauge symmetries, renormalization, renormalization group etc. began to be clarified just decades later.

reader Peter F. said...

Amazing optimistic enthusiasm promising the ultimate theoretical physics take on What Is going on - and so amazingly simply conveyed to interested enough ordinary people!

Will it be found that, somehow, the principle of super symmetry can be stretched so far that it can be used to super-symmetrize to the effect of canceling out all quantum-cosmological conundrums by as if comb and evenly plait together the probability amplitudes of stringy dimensions’ entangled in the roots of black holes’s hair?

To me "the multiverse" is everything that (you) Lumo says it is as far as I can digest what (your) Lumo’s brain (bring) brings to (your) his blog.
But beyond 'my Lumo-aligned attitude' I am inclined to tentatively intuit that the multiverse is a: ‘timelessly eternal imaginary energy producing ignitions of occasionally bio-evolutionary energy patterning trends [or "only extremely rarely" bio-evolutionary such - whatever ratio of those infinitely repeated "ignitions" one can refer to or 'can not even wrongly refer to' in this infinity involving context! %}];

And, that at least those trends that are bio-evolutionary in character are ‘so truly infinite’ in the intricacy of their evolutionary patterning that the multiverse will in principle not even once (even given infinite "astrophysical attempts") repeat an evolutionary patterning trend.

reader Giotis said...

Amazing talk; with people like Nima around I feel safe for the future of theoretical Physics.

He explained deep physical concepts in a clear way using a Physics language laymen can understand without all the stupid metaphors people often use.

off topic: BTW Lubos if not already you have to watch this (the sock puppet is Einstein) :-)

reader kashyap vasavada said...

Great talk by Nima. Is there a review article or a blog which explains why spin stops at J=2 and higher values are not possible?

reader RAF III said...

Try this:

reader Luboš Motl said...

Yup, I've watched all A Capella Science videos. Cool.

reader kashyap vasavada said...

Thanks. I must have missed it when it came.

reader HAMMAknowsHeDoesntKnowHeKnows said...

Arkani-HAmmeddi is a bulshit to the power of infinity. read how Woit saw totally through this young buffoon who got $3 mil. and now is selling cheap excuses like " oh well we can safely assume we have no f.... clue about anything but it's better than to know we have no f.... clue about anything " . we know lubos' motivation: he seems to fall for old acquaintances and the riches, so not sure who to pitty more, Ahmaddi or Lubos