Live data from Hacker News

A physicist who bets that gravity can’t be quantized

quantamagazine.org

281–290 of 378 posts

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

#281

Earlier quoted context omitted.

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.

I wouldn't even assume that discrete theories must have a resolution that can be detected.

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

#282

Out of curiosity, what's the quantum angle limit? If position is quantized and all fundamental particle are isotropic, there must be a minimum "angle of turn" for all objects, and by extension, a "maximum angular accuracy" for something like, say, orienting the face of a macroscopic object to face a particular direction. How many of those fit in a circle?

I don't think position is likely to be discrete, but,

while I also don't really think that angles are discrete, angular momentum *is* discrete, coming in multiples of hbar, the reduced Plank constant.

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

#283

Earlier quoted context omitted.

If gravity is not quantized, would that not mean that it has infinite information (the accuracy to represent its values to infinite digits of precision), and thus cause a black hole due to such high information density?

IIUC, quantized doesn't mean finite, it just means discrete. Energy being quantized in bound states in quantum mechanics means that the eigenvalues are discrete, but a state can still be any linear combination whatsoever of the eigenstates. And since spacetime is continuous, it's determines by its values on a dense subset, in particular a countable sense subset, which would make both sets of possibilities, gravity an…

What determines the amount of information? Is it the total values the gravity could have had with unlimited mass? The number of values that are less than the one it is observed to be? All the values in the probability distribution for some uncertainty thing?

If you have a carbon atom, does it have unlimited information because it theoretically could instead have been some other number of atoms?

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

#284
post #248

Earlier quoted context omitted.

Keep in mind in GR singularities don't really make sense either: i.e. when you pass the event horizon of a black hole, under GR the rules say that you must always been traveling towards the singularity. But the corollary of this is that it's not actually possible - mathematically - to arrive at the singularity (because then you'd be moving parallel with it rather then towards it). So while we can define what happens…

Can they reach the center? And how long would that take? I ask because time will slow down for them as they accelerate towards the center.

Your first question is the problem: they must always be moving towards the center - which implies they must be getting closer to it. Because if they can never actually reach the center, then in 4D spacetime they're no longer moving towards it - they'd be traveling parallel with it.

Which would imply that singularity isn't a singularity - i.e. the "hole" it makes would in fact be highly curved spacetime in one dimension, but completely flat in another (i.e. it would be a cylinder).

Which creates a hole lot of weird infinities in the system: i.e. a black hole becomes a finite bounded volume on the outside, but contains an infinite amount of space on the inside (since no matter how much you distort spacetime, the molecules of whatever falls in can just pull themselves together more tightly provided the distortion doesn't happen too quickly).

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

#285
It certainly feels wrong for spacetime to be quantized, because it’s fundamentally no linear and geometric in a way that other things aren’t, and quantized theories like string theory tend not to be inherently background independent, which seems really wrong.

At the same time, we know that a basic semiclassical + Everettian model of the world does not correctly model reality - if we run a quantum-linked cavendish experiment, we don’t observe the gravitational effects of superposition. I.e. each branch of a quantum superposition behaves as if it is in its own spacetime. We can even run a Bell experiment linked to gravitational detectors and observe that the results are consistent with gravity being quantized.

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

#286
post #68

Earlier quoted context omitted.

What would such an experiment look like? If it makes it easier, perhaps an experiment that's even harder (or much, much harder) to realize but easier to explain.

It's pretty simple. You basically have two suspended mirrors on wires and you shoot lasers at each mirror. Due to gravity there will be an interaction between the mirrors. There are then two things you can probe. First, the correlation between the two reflected lasers, and if you can achieve the extreme temperature requirements, entanglement between the two reflected lasers caused by the gravity. Now, the former does…

You might also be interested in Bose et al. (2017), "Spin Entanglement Witness for Quantum Gravity" https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.11... aka [hep-th] https://arxiv.org/abs/1707.06050 which proposes a way to "certify gravity as a quantum coherent mediator, through simple spin correlation measurements".

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

#287

Somewhat off topic, and somewhat related: I don't think there actually is such a particle as a graviton. General relativity says that you can't tell the difference between being unaccelerated, and being in free fall. But in one case you have no gravitons coming in, and in the other you have gravitons. I can change whether gravitons are there or not by a (general) relativistic transformation. But that's not possible.…

"General relativity says that you can't tell the difference between being unaccelerated, and being in free fall"

I think this is a misunderstanding. Special relativity says you can't tell the difference between free fall (in a gravity well) and inertial acceleration with the example of a guy in an elevator typical here. Both cases have acceleration.

My question: does an accelerating elevator due to motors and pulleys curve spacetime too?

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

#288

Earlier quoted context omitted.

Can you provide an example of a Lorentz boost that changes the number of particles?

It's pretty general. Just draw the worldlines for some particles bouncing off eachother, and then perform a Lorentz boost, which means choosing a new tilted spatial surface to intersect the worldlines. If you perform a boost, then the spatial surface intersect fewer or more worldlines, and worldlines which were for particles in one frame can become antiparticles in another. Consider an electron absorbing a photon at…

That example I think is tangential. The better example is free fall in a gravity well v. being accelerated up like in an elevator in space far, far away from gravitational sources

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

#290

> It’s become dogma. All the other fields in nature are quantized. There’s a sense that there’s nothing special about gravity — it’s just a field like any other — and therefore we should quantize it. I keep on citing Stephen Hawking here on HN, but it again seems very appropriate: > It would be rather boring if this were the case. Gravity would be just like any other field. But I believe it is distinctively different…

In General Relativity, matter curves spacetime, so instead of saying that all the “other fields” act within spacetime, perhaps “the metric [tensor] is determined by the matter and energy content of spacetime?” If geometry is described fully with a (tensor) field, perhaps we’re really just looking for a gauge theory but then that would be QM. Now I’m trying to imagine a tensor computing some infinite curvature supposedly necessary in GR.
Post reply on HN