Gravity & Quantum Mechanics: Susskind's Letter Explained

[smodak]

In his recent letter to the physics community, Susskind says Gravity and Quantum Mechanics is inseparable. Read it here .
Frankly, I do not fully understand his letter or the claim that he is making.
What do you think?

 

[PeterDonis]

I do not fully understand his letter or the claim that he is making.

If you don't already have a lot of background in this area, you should not expect to understand it. He is assuming that a lot of theoretical claims which, as far as I know, are not definitely established (many of them can't even be tested at all experimentally), are true, and working out a particular implication. But without a detailed understanding of all those theoretical claims, it's going to be hard to make sense of this letter. I don't follow all of it myself since I'm not familiar with all of the literature he is referring to.

 

[Giulio Prisco]

He's going one step beyond ER=EPR (the two ends of a wormhole can be considered as entangled in a QM sense, and two entangled particles can be considered as linked by a wormhole), and proposing that GR and QM imply each other. A weaker version of his claim is already implied, I think, in AdS/CFT duality: a quantum field theory without gravity (CFT) is equivalent to a string theory with gravity in a space with one more dimension (AdS).

 

[Demystifier]

Is he saying that classical GR is equivalent to QM? If so, what is the GR equivalent of quantum uncertainty relations? Or of unitarity and superposition principle?

 

[Giulio Prisco]

Well he couldn't say that because all his arguments rest on AdS/CFT, which is not only quantum but stringily quantum. However, if classical GR is the low-energy effective field limit of string theory, he could say that classical GR follows from quantum strings.

 

[Demystifier]

However, if classical GR is the low-energy effective field limit of string theory, he could say that classical GR follows from quantum strings.

That's nothing new, we already know it from the 70's and 80's.

 

[Demystifier]

In the past, string theory has been accused for being not even wrong. I am afraid that some modern trends could be characterized as something even worse, like fashionable nonsense https://en.wikipedia.org/wiki/Fashionable_Nonsense

 

[vanhees71]

Ehm, doesn't this fall under "personal speculation" and thus shouldn't be discussed here, although the speculation comes from a quite famous physicist? SCNR

 

[Fra]

Ehm, doesn't this fall under "personal speculation" and thus shouldn't be discussed here, although the speculation comes from a quite famous physicist? SCNR

I agree. But it might? fit a little better into btsm?

https://www.physicsforums.com/threads/space-is-entangled-says-leonard-susskind.925412/#post-5845768

/Fredrik

 

[smodak]

I agree. But it might? fit a little better into btsm?

https://www.physicsforums.com/threads/space-is-entangled-says-leonard-susskind.925412/#post-5845768

/Fredrik

Agree. Mods would you please move it to "Beyond the Standard Model" forum?

 

[musician ilhan]

susskind is a fanatic of string theory. string theory unifies quantum mechanics and general relativity but only mathematically.
string theory is a belief that cannot be proven or disproven.

 

[PeterDonis]

Mods would you please move it to "Beyond the Standard Model" forum?

Good point, moved.

 

[PeterDonis]

doesn't this fall under "personal speculation" and thus shouldn't be discussed here, although the speculation comes from a quite famous physicist?

I don't think it's just personal speculation; the models Susskind is referring to are models that have been published in peer-reviewed papers and are used in a lot of ongoing research. But it was correctly pointed out that the topic does belong in the Beyond the Standard Model forum, so I have moved this thread there.

 

[Giulio Prisco]

doesn't this fall under "personal speculation"

Is there any physical theory that didn't start as a personal speculation? None that I can think of.

However, yes, this does belong in "Beyond the SM."

 

[Giulio Prisco]

That's nothing new, we already know it from the 70's and 80's.

I think a better wording is "some already thought it from the 70's and 80's" (also related to your following post). Perhaps Susskind would say that recent theoretical developments (e.g. AdS/CFT, ER=EPR) provide more support for the idea that classical GR follows from quantum strings.

 

[Giulio Prisco]

In the past, string theory has been accused for being not even wrong. I am afraid that some modern trends could be characterized as something even worse, like fashionable nonsense

In the sense of string theory being a set of nice mathematical ideas with no experimental evidence for, and in the sense of string theory seeming messy and incomplete (as compared to, for example, clean and consistent general relativity), I understand why one can think of string theory as not-even-wrong fashionable nonsense. But I believe some recent developments, e.g. the applications of AdS/CFT to condensed matter and nuclear plasma, indicate that the mathematics of string theory does relate to things that can be measured in the lab. Here, Susskind is proposing experiments, based on more powerful quantum computers that could be developed some day.

 

[Giulio Prisco]

Susskind says: "From the holographic AdS/CFT correspondence we may assume that observers, perhaps with human-like cognitive abilities, are possible in the bulk. Can a laboratory observer conrm the presence of such a bulk observer by doing experiments on the shell?..." This is a key point in the paper.

But given AdS/CFT correspondence, if intelligent observers exist in the bulk, they also exist in the shell. So we have two mathematical descriptions of the same observer, but the two versions of this observer experience totally different things, for example one sees a world with spin-2 gravitons and the other sees a world with only spin-1 force carriers. What do you guys think?

 

[haushofer]

I don't get this. To me, holography and its specific form the AdS/CFT correspondence, states the following: if you write down the Feynmandiagrams of a CFT like N=4 Super YM, somewhere deep down in those diagrams there is information about the structure of spacetime itself hidden. That's it. It's like a promise that a painting from a child somehow resembles a Van Gogh in a particular way. That's already amazing enough. Let's not make it mystical.

 

[Giulio Prisco]

if you write down the Feynmandiagrams of a CFT like N=4 Super YM, somewhere deep down in those diagrams there is information about the structure of spacetime itself hidden.

Yes but not the same spacetime. The CFT lives on the boundary, and has info on the structure of spacetime in the bulk.

 

[Fra]

I see the paper as a bunch of examples and arguments, which arent the "real thing" but serve to make a vision more plausible. For me the vision and thinking of howto make the arguments complete is the focus.

As i see it the vision is that that dimensionality and connectivity, is just a way of coding information. The various correspondence principles is EXAMPLES of this within current theories. Also time evolution seems connection to some kind of computation (although i am not personally fond of the terminology, i like to think in terms of random walks guided by priors, rather than "computation", but once you think abbout it they two views are similar), and thus the examples of relating computation complexity with BH expansion.

Regardless of the fate of string theory, I definitely don't think these is just conicidental notes. I think there is something profound here, that lies at the core of the problem of theoretical phyhsics: foundations of QM, QG, renormalisations etc that is yet to be revealed, understood and formalised. I think the core vision here
has nothing a priori todo with string theory.

Connectivity and metric spaces are also naturally related to information divergences and also intuitively complexity will spontaneously form as a way to encode as much information as possile, given finite capacity. And what ought to happen when two such systems interact? Well likely a negotiation resulting in an attractive force along the dimension where they are connected. Most probably a universal one, regardless of details (much like gravity). So there are many more examples that adds up and supports such a vision where the information perspective of QM likely can EXPLAIN gravity, and then i do not mean "explain" just as in string theory, but explain it in a way that is independent of string theory so that whenever you have "interacting" "computing" observers, trying to take control of their environment, a universal attraction is unavoidable. It will be implied.

This is my reading out what what susskind says with "whenever there is QM there is Gravity". But of course, that makes no sense if we are just talking about regular QM. I think we need to improved its formalism, then and only then will these connetions be clear also so mathematicians will understand them.

/Fredrik

 

[king vitamin]

Well he couldn't say that because all his arguments rest on AdS/CFT, which is not only quantum but stringily quantum. However, if classical GR is the low-energy effective field limit of string theory, he could say that classical GR follows from quantum strings.

I would say his point is more that classical GR actually follows (holographically) from quantum mechanics (without gravity) in a suitable limit. For example, if the quantum mechanics model is some strongly-interacting CFT, you expect the dual theory to be well-described by classical GR physics. He then proposes to simulate a CFT with quantum computers to effectively "test" the corresponding gravity theory.

He also goes beyond classical GR. He points out that the Newton constant G in the dual theory is not necessarily small (it depends on the CFT), so you could actually probe strong-coupling quantum gravity dynamics with such a simulation.

Like many high-energy theorists, I think he overestimates how close we are to doing these simulations on quantum computers.

 

[Giulio Prisco]

I would say his point is more that classical GR actually follows (holographically) from quantum mechanics (without gravity) in a suitable limit.

Yes. though if the relationship is really symmetric then saying that one follows from the other seems misleading, more like two sides of the same coin.

He says “If there is anything new here it is the idea that information may pass from a laboratory environment to the degrees of freedom of a physical realization of a CFT, thereby bridging the gap between the lab and the bulk. One can enter the bulk, observe it, and go back to the lab…"

He then proposes to simulate a CFT with quantum computers to effectively "test" the corresponding gravity theory.

Here is where I find it difficult to follow him. It seems to me that a CFT simulation is a simulation, not the real thing. But Susskind says that the simulation IS the real thing. Is he thinking that a simulation on a quantum computer (as opposed to a simulation on a classical computer) is a "real," material implementation of the physics being simulated?

 

[Fra]

Here is where I find it difficult to follow him. It seems to me that a CFT simulation is a simulation, not the real thing. But Susskind says that the simulation IS the real thing. Is he thinking that a simulation on a quantum computer (as opposed to a simulation on a classical computer) is a "real," material implementation of the physics being simulated?

I suspect he means that due to the computational complexity of the boundary theory vs bulk theory, classical computers are just too slow?

Wether its the real ting that depends what you mean but i also find it strange.

OTOH i don't think we need quantum computers to test this. I rather see the observation that the "bulk representation" has a lower comuputational complexity as the key both to how small systems can "survive" in the chaos. It also explains why an evolutionary selection in favour or space.

The vision i see here
Bulk = observers information
Boundary = observers interaction with the envirroment.

Now consider the evolution of interacting observers. Clearly the difference in complexity of different duals implies a selective pressure that will select our place in the landscape. The selection is a mere race... SLOW coders don't make it... fast coders dominate.

/Fredrik

 

[Giulio Prisco]

Wether its the real ting that depends what you mean but i also find it strange.

I mean the real thing in a physical sense. I have no problem with the idea that a simulated reality can be real in its own terms, for example from the point of view of sentient observers living in the simulated reality. But here Susskind says "Instead of shells supporting conformal field theories,a more practical alternative might be quantum computers simulating the CFTs," and implies that these quantum computers would have an AdS dual etc. But what physically happens in a computer simulation is electrons moving in semiconductor circuits etc., NOT QCD-like fields, so why should AdS/CFT apply? Therefore I suspect what Susskind has in mind is that what physically happens in a quantum computer (as opposed to a non quantum computer) has a dual.

 

[Fra]

I mean the real thing in a physical sense. I have no problem with the idea that a simulated reality can be real in its own terms, for example from the point of view of sentient observers living in the simulated reality. But here Susskind says "Instead of shells supporting conformal field theories,a more practical alternative might be quantum computers simulating the CFTs," and implies that these quantum computers would have an AdS dual etc. But what physically happens in a computer simulation is electrons moving in semiconductor circuits etc., NOT QCD-like fields, so why should AdS/CFT apply? Therefore I suspect what Susskind has in mind is that what physically happens in a quantum computer (as opposed to a non quantum computer) has a dual.

I didnt think too much on that part, mainly since I am not primarily seeing this from the stringside, but maybe even just because of that (as I see these from "another side") i find it intriguing that maybe all roads lead to rome after all? Its just that only the fastest road will get to name ithe destination.

But as it see it the only other more "real thing" would be that two entangled qubit registers should then make a wormhole, and thus we can extract information from the bulk QG wormhole but qubot register correlations etc? And if we can make larget set of qubit registers rather than just one qubit, some more interesting things might be inferred from this? I have not thought a lot in that direction. I rather envision classical simulations to test the ideas, but due to complexity simulating the real thing is simply impossible. You don't have to think long until its like the hitchhikers galaxy where Earth is just a simulator.

/Fredrik

 

[fanieh]

Can someone give me any work on beyond the standard model where shapes or figures or geometry itself has quantum effects. Is this GR=GM related to it?