Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A physicist who bets that gravity can’t be quantized

quantamagazine.org

Post reply on HN
Print view View original

271–280 of 378 posts

Re: A physicist who bets that gravity can’t be quantized

#271

Post by zmgsabst »

xmcqdpt2 wrote: post #260
nocoolnametom wrote: post #182

Earlier quoted context omitted.

When you say superdeterminism are you referring to something like Pilot Wave theory, where what appear to our measurements as probabilistic yet random interactions are merely expressions of a more complex yet non-random underlying system that we cannot, as yet, measure? (I don't even know if that's the proper description of the hypothesis.)

Superdeterminism is simply the idea that all quantum experiments results could have been known before performing them, assuming a perfect knowledge of the state of the universe. It's one of those things that could be true and explain all of QM but also is kind of a cop out. "Of course your two detectors are giving correlated results, they were tightly coupled 13.8 billion years ago and now they are forever linked lik…

How is it a cop out?

We know two things:

- universe was small enough everything was tightly coupled

- non-local phenomena occur

Insisting that our beliefs reflect an ideology (eg, you can segregate off portions of reality to study in isolation) which seem contrary to observed reality (eg, the points above) is religion — not a scientific investigation of the universe.

Re: A physicist who bets that gravity can’t be quantized

#272

Post by drdeca »

pravus wrote: post #212
krastanov wrote: post #206

Earlier quoted context omitted.

Any reason you are specifically picking the Higgs field in your question? I am asking, because this sounds a bit like you are riffing on a common misconception that the Higgs field has something to do with gravity, which is not the case. The interaction with the Higgs field is the reason some (only some) of the particles have a mass, but explaining gravity does not need to have anything to do with the Higgs field. Bu…

My understanding is that the Higgs field should be simpler than gravity because it's just a static value whereas gravity is SU(1)? At least to me it seems logical that if the Higgs is quantized, surely more complicated fields would be as well?

SU(1) is the trivial group. SU(n) is the group of unitary nxn matrices with determinant 1. There is only one 1x1 unitary matrix with determinant 1, and in fact there is only one 1x1 matrix with determinant 1 period, namely, the 1x1 matrix whose only entry is the number 1.

So, to associated something with a symmetry group of SU(1) would be effectively the same as not associating it with a symmetry group.

Re: A physicist who bets that gravity can’t be quantized

#273

Post by zmgsabst »

consilient wrote: post #269
xmcqdpt2 wrote: post #260

Earlier quoted context omitted.

Superdeterminism is simply the idea that all quantum experiments results could have been known before performing them, assuming a perfect knowledge of the state of the universe. It's one of those things that could be true and explain all of QM but also is kind of a cop out. "Of course your two detectors are giving correlated results, they were tightly coupled 13.8 billion years ago and now they are forever linked lik…

> Superdeterminism is simply the idea that all quantum experiments results could have been known before performing them, assuming a perfect knowledge of the state of the universe. That's just determinism. Superdeterminism additionally posits that the results of those experiments are all correlated so as to make it appear to us as if local hidden variable theories were false. Pick the settings on two polarimeters for…

Alternatively, non-local correlations (in the style of anyons) persist from the early universe.

And we’re just measuring those.

Re: A physicist who bets that gravity can’t be quantized

#274

Post by astrange »

meindnoch wrote: post #226
eigenket wrote: post #207

Earlier quoted context omitted.

That is certainly true. In fact you don't even need rational numbers, it is entirely possible that there are a finite number of positions on the universe, for example. Nevertheless physics seems to work very nicely when expressed in the language of calculus. Everything from Schrödinger's equation to the Einstein field equations, and from classical mechanics to the standard model of particle physics are expressed in t…

Unless the resolution of the rational model is fine enough to be indistinguishable from the continuous solution.

That would be very fine indeed since the CMB doesn't eg seem to be pixelated and we don't observe numeric instability anywhere.

Re: A physicist who bets that gravity can’t be quantized

#275

Post by drdeca »

zmgsabst wrote: post #271
xmcqdpt2 wrote: post #260

Earlier quoted context omitted.

Superdeterminism is simply the idea that all quantum experiments results could have been known before performing them, assuming a perfect knowledge of the state of the universe. It's one of those things that could be true and explain all of QM but also is kind of a cop out. "Of course your two detectors are giving correlated results, they were tightly coupled 13.8 billion years ago and now they are forever linked lik…

How is it a cop out? We know two things: - universe was small enough everything was tightly coupled - non-local phenomena occur Insisting that our beliefs reflect an ideology (eg, you can segregate off portions of reality to study in isolation) which seem contrary to observed reality (eg, the points above) is religion — not a scientific investigation of the universe.

First, as a religious person: no, contrary to what you seem to be saying, dismissing superdeterminism isn't "religion".

Second, dismissing superdeterminism is not, as you seem to say, "contrary to observed reality".

The belief that one is capable of forming ideas that bear any resemblance to reality, is justified on account of, in order for one's beliefs or lack-thereof to have any use, it would have to be true.

Things like the no-speed-up theorem (which I admit I'm not super familiar with), and other things of that sort, lead me to expect that, while I don't have a full argument for this, that it isn't possible for computational complexity reasons, for the early universe to be such that it, in effect, encodes predictions of what future measurements people will make, in a way that makes signals determining what measurement directions get used, correlated in the way that superdeterminism requires.

Re: A physicist who bets that gravity can’t be quantized

#276

Post by Iwan-Zotow »

srlowe wrote: post #177
consilient wrote: post #138

Earlier quoted context omitted.

> Hmm, so doesn't the fact that this particular value makes the math simplified not also imply some kind of meaning? No. There are plenty of other sets of constants you can choose to set to 1, inducing other length scales. Only dimensionless constants have physical meaning in isolation; dimensioned quantities are meaningful only with respect to each other.

Thanks for the explanation. I had to google "dimensionless constant" lol. Maybe it is all just math at the foundation after all!

Fine structure constant, famous alpha ~1/137. It also means that you could multiply or divide what we know as Plank length and it still would be a "valid" unit to be a minimum length entity

Re: A physicist who bets that gravity can’t be quantized

#277

Post by javajosh »

xmcqdpt2 wrote: post #267
javajosh wrote: post #202

Earlier quoted context omitted.

I'm already convinced[0] the effect is small, if it exists at all. However I thought of another experiment that might be easier to perform, to see if there is any effect at all: take two crystals, as similar in depth as one can make them, with two different isotopes, and measure the difference in index-of-refraction. Silicon (28 and 30?) would probably be good for this, as would laser interferometry. (Maybe the Gravi…

Isotopes are not 100% chemically equivalent and they form chemical bonds with different vibrational resonances because the vibrating masses are different. You can see those differences in spectroscopic line shapes including in refraction (which is related to absorption by the Kramers-Kronig relation). My guess is that those effects would swamp out any relativistic ones by a lot, and because you can't tune the isotopi…

Perfectly applicable paper, thanks! I skimmed it, but I didn't see an explanation for the lower refraction index for the heavier isotope. It had the lower RI at all wavelengths they looked at. I think it's useful to ignore the "phonon resonance" wiggle in the middle of their tested range on page 4, around 600nm, if looking for some other effect. (it's crazy that an electron that weighs like 10^-31 kg could actually jiggle a crystalline lattice to the point you can see it at all, but I suppose there's a lot of electrons, and Tacoma Narrows bridge reminds me of how powerful resonance effects can be).

Re: A physicist who bets that gravity can’t be quantized

#278

Post by audunw »

eterevsky wrote: post #48
contravariant wrote: post #39

Earlier quoted context omitted.

> possibly even without creating a particle accelerator the size of the solar system. Sure, you just need to entangle two objects big enough to exert a noticeable amount of gravity on one another but somehow do not interact gravitationally with the rest of the set up. Anyway, let's try a cat sized object first, then we'll finally know if Schroedinger had a point.

If they are close enough to each other and far enough from everything else, then why not. The smallest objects for which we measured their gravitational interaction weighed just 90 mg ( https://arstechnica.com/science/2021/03/researchers-measure-... ). The biggest object put in quantum superposition weighed around 1 mcg ( https://physics.aps.org/articles/v16/s45 ).

I think the comparison you're giving here is quite misleading. You're implying that we "only" need to scale up the objects we put in quantum superposition by tens of thousands of times. But that's not all there is to it. You'd have to put it in a kind of superposition that affects its mass enough to be measured. For instance, having the whole object be in superposition of one position, or another position several centimeters away. Or being in an extremely high energy state vs extremely low energy state, with an energy difference on the order of 90mg. Given E=mc2 that's an insane amount of energy.

Seems to me that such an experiment is borderline unthinkable with present day technology.

Re: A physicist who bets that gravity can’t be quantized

#279

Post by raattgift »

eigenket wrote: post #146
bowsamic wrote: post #46

Earlier quoted context omitted.

The problem is that it doesn't act like a mass distribution, it acts as two non spatially overlapping possibilities. I think that the only way to save classical gravity would be superdeterminism. If quantum states correspond to anything other than our ignorance, i.e. if superposition states are actual physical states of reality, then gravity will need to be quantum.

I attended a talk by Jonathan Oppenheim a while back on this subject, and my understanding based on that is that in his model something (very) roughly like this happens. You put your massive object in superposition, it interacts with space-time and "tries" to put space-time in superposition but since space-time is fundamentally classical (in his model) what happens is space-time ends up in a probabalistic mixture of…

I'm quite happy to admit that I don't know the classical-quantum CQ programme well, and certainly not well enough to hazard a genuinely informed opinion that might take the form "it's a candidate for a fundamental theory" or "it's a candidate for a better EFT". I think I can say it isn't obviously not one of those, though. Also, I'm not ready to sloganize the work beyond "it's complicated".

A good entry point to CQ is Oppenheim's [hep-th] https://arxiv.org/abs/1811.03116

"... invariant under spatial diffeomorphisms ... consistent ... completely positive, norm preserving, and linear in the density matrix ... the metric remains classical even when back-reacted upon by quantum fields ... the dynamics here, while stochastic, can leave the quantum state pure -- it is rather the classical degrees of freedom, which gain entropy."

It's a heavy couple dozen pages (part V is a mountain for mountaineers (I am nowhere near the summit) and the foothills are also hard work, though most of it shouldn't pose technical comprehensibility problems anyone who's done QFT (GKSL is prominent [short intro https://arxiv.org/abs/1906.04478>], and the Lindbladian is the generator of time-translations) and who has encountered Lagrangian and Hamiltonian formulations of gravitation, especially if they've glanced at Birrell & Davies. A crash read or refresher of Wald's QFTCS https://arxiv.org/abs/gr-qc/9509057 might be useful if like me you've mostly spent time in one of the two fundamental silos and want a bit more meat than what's in the first couple pages of Oppenheim 2018's part I. The general relativistic constraint equations are also important, and there is a good overview at https://link.springer.com/article/10.1007/s41114-020-00030-z

The meat, for me, is the potential for a better approximation than semiclassical gravity when quantum fluctuations are large: "[even if not fundamental, CQ gives] the ability to consistently study back-reaction effects in cosmology and black-hole evaporation ... [but] care should be taken, since an effective theory might violate our assumptions of Markovianity or complete positivity at short time scales or when the gravitational degrees of freedom have not fully decohered.", "This theory serves as a sandbox in which to understand issues around the quantisation of the gravitational field. After all a probabilty density \rho and the Liouville equation have a lot in common with the wave function \psi and the Heisenberg equations of motion. [and the rest of that paragraph just before (eqn 2)]".

Re: A physicist who bets that gravity can’t be quantized

#280

Post by naasking »

mr_toad wrote: post #255
CuriouslyC wrote: post #97

Earlier quoted context omitted.

If I were a betting man I'd say black holes have no singularity, but rather a core of extremely dense exotic matter (probably formed from top/bottom quarks) which we haven't detected because it decays quickly under less extreme circumstances.

I think it’s true that once an event horizon forms no physical force (known or hypothetical) can stop a singularity forming. There may be some form of very dense matter that stops large stars from collapsing to the point where an event horizon forms in the first place, but that doesn’t seem to apply to super massive black holes. For super massive black holes the event horizon grows too fast.

Singularities are a deficiency in GR, they don't really exist.
Post reply on HN