?

Log in

No account? Create an account

Previous Entry | Next Entry

upcoming colloquia

John Preskill, the Director of the Institute for Quantum Information, is going to speak at our colloqium this week. He recently won his 1997 bet with Stephen Hawking that information is never destroyed by black holes. He bet both Stephen Hawking and Kip Thorne in 1997 that black holes don't swallow information, they just sort of slip it under their tongue for a while until they eventually have to cough it up again. Hawking conceded on July 21, 2004, but Kip Thorne apparently still isn't completely convinced enough to concede.

And while we're on the subject of information disappearing, this is yet again illustration that Everett's interpretation of quantum mechanics (many worlds, many histories, quantum multiverse) is the only sane way to look at it. While all other interpretations of quantum mechanics, if they talk about information at all, predict that information is destroyed during certain processes (the collapse of the wavefunction, or the interaction of the large with the small), Everett's is the only one which predicts that information is never destroyed. Lest I make it sound like the settlement of this bet has direct consequences on how to interpret quantum mechanics, I should admit that even in Everett's interpretation it's still possible for information to disappear from a single branch of the multiverse (some of it simply spreads out and flows into the other branches) and the type of information loss they were really talking about in the black hole scenario was just that type of disconnection of information from the rest of the observable universe. So really, there is no falsification either way here. But nevertheless, it still illustrates the utility of believing in the rest of the quantum multiverse. It's very similar to the concept of potential energy: potential energy was invented originally because we noticed that all other forms of energy we could measure were always conserved, aside from a missing piece that depends on position. So we just made up a term "potential energy" to describe the missing piece. It makes sense to believe in it because it makes the conservation laws simpler. The same is true for the other branches of the multiverse: they make the theory much simpler because then you can assume conservation of information, just like we assume conservation of energy. No spooky appearing and disappearing of stuff!

Interestingly enough, the settlement of the bet is also good evidence that string theory is right, or at least useful. I'm much less dogmatically in favor of string theory (mainly because I know so little about it) compared to the many-worlds-interp, but I still think this is neat. Even though there hasn't been any directly observed experimental consequences of string theory, it was able to solve the problem of "what is the microscopic origin of the black hole entropy" before the traditional particle physicists or cosmologists were able to. And from what I hear, Hawking's non-stringy explanations of the black hole entropy were in large part inspired by what the string theorists, such as Juan Maldecena, had already figured out by using string theory. If the theory can get to the answers first, then surely it's useful enough to be called physics, in direct opposition to those who claim it's either pure mathematics or philosophy.

At any rate, Preskill knows a ton about quantum information theory, which is my absolute favorite subject in physics. So I'll definitely be attending this week's colloquium. Two weeks after that, Nima Arkani-Hamed is coming from Harvard to talk about his new paper on the string theory Landscape. That should be good as well, since I'm starting to toy with the idea of maybe doing something with the Landscape for my research.

Comments

( 5 comments — Leave a comment )
dankamongmen
Feb. 19th, 2005 09:54 pm (UTC)
Very cool! Let us know how it goes; QIT is very badass.
onhava
Feb. 19th, 2005 10:49 pm (UTC)
this is yet again illustration that Everett's interpretation of quantum mechanics (many worlds, many histories, quantum multiverse) is the only sane way to look at it.

That's a rather strong statement. I guess I'm not sane -- I see no need to buy into any "interpretation" of quantum mechanics. When do wave functions collapse, anyway? Quantum field theory is a completely coherent framework, with no need to postulate any collapsing wave functions or multiple universes or any other nonsense, as far as I'm concerned.
spoonless
Feb. 19th, 2005 11:27 pm (UTC)
When I say "Everett's Interpretation" I mean the simple statement "Quantum Mechanics is unitary at all scales". If you agree with this statement, then you are an Everettist. If you think that it's sometimes unitary but at other times the wave function either collapses, or undergoes some sort of slow collapse-like process that involves a non-unitary evolution of the wavefunction, then you are a non-Everettist.

When does it collapse? Never; that's what distinguishes Everett from the alternatives. You might object to various labels which are attached to this view such as many-worlds or multiverse, but they nevertheless all refer to the same specific belief about what happens at the physical level. As you say, QFT is a coherent framework (aside from whatever goes on at high energy scales), which is precisely why I think everything should be viewed from within it rather than having to postulating that there is an additional collapse or non-unitary, non-reversible, information-destroying process which gets involved. This type of process really isn't a part of QFT, so if you don't believe in many-worlds then you must believe there is more to the story than just QFT.

Personally, I think many-worlds is an entirely appropriate term for this belief... since unitarity implies that when I as an observer become correlated with some microscopic observable there will be a branch of the wavefunction for each outcome of the measurement, none of which can interfere with each other any longer (because of decoherence), weighted by the appropriate amplitudes. Since these multiple copies of me have decohered to the point where they have vanishingly small likelihood to ever interfere or merge together again, they are isolated enough to call them separate universes, IMO. What else would you call them?
onhava
Feb. 20th, 2005 12:30 am (UTC)
OK, fair enough -- I thought you were somehow trying to argue that the particular "many-worlds" view (or maybe I should say language) of unitary quantum mechanics was more sane than any other view (I guess I tend to espouse what people call the "shut-up-and-calculate" interpretation). I think we pretty much agree, then (modulo the quibble that you say at "all scales" whereas I would say "at all scales we have information about"; one could imagine that in the future some experimental evidence for some non-unitary process would arise, but I find it extremely unlikely).

I still don't like the terms "many-worlds" or "multiverse," as I feel like they're somehow mystifying what should be straightforward. Sort of like they're portraying the quantum world as a proliferation of classical ones, whereas if you just accept that the world is quantum it doesn't seem so confusing. But, that's not a substantial disagreement, just one of language and viewpoint.
spoonless
Feb. 20th, 2005 01:39 am (UTC)

(modulo the quibble that you say at "all scales" whereas I would say "at all scales we have information about"

Funny you mention that, because that was the one thing that I thought "I should have clarified" after I wrote it. By all scales I really just meant from small (atomic or nuclear) all the way up to large (human or cosmic) and everything in between. We don't have information about what happens at ultra tiny scales since we need to know more about how quantum gravity works first, so a bit of agnosticism there is warranted.

I still don't like the terms "many-worlds" or "multiverse," as I feel like they're somehow mystifying what should be straightforward. Sort of like they're portraying the quantum world as a proliferation of classical ones, whereas if you just accept that the world is quantum it doesn't seem so confusing. But, that's not a substantial disagreement, just one of language and viewpoint.

I agree that it's often not understood and assumed to be a lot more complex than it really is. But I wouldn't say it mystifies much, certainly not to the extent of the Copenhagen school of thought. Copenhagen is what caused people for years to say bizarre things like "nothing is real until you look at it", which is certainly not the direction I think we should be leading the public in. Everett provides a solid realist foundation for quantum mechanics which doesn't have a need for talk about observers or collapse, until it actually comes down to doing an experiment.

I also worry that people will picture a whole bunch of classical worlds stitched together when that's not really right; it's one world that's all quantum, just with different parts that can split off by becoming effectively causally disconnected. I think this is why I slightly prefer the term multiverse over many-worlds: because of the analogy to a multithreaded process on a computer where a process is allowed to fork, and can sometimes share information between threads and other times allow them to become isolated. I think this is why some of the most outspoken advocates of Everett are those who work in quantum computing. There was originally an operating system named multics, and only later did they make the one called unix; same root meaning.

Unfortunately, nowadays multiverse has also acquired a second meaning as an ensemble of separate universes (eternal inflation, landscape stuff, etc.), which I don't think is as appropriate although I can see why it got applied there too. I wish they had just kept it as meaning only the quantum multiverse and came up with different names for the others--and it's even worse when people start confusing the different types. But I can't do too much about that now, the terms are here to stay and so I use them.

Oh and by the way, when I say I consider it "the only sane interpretation" I don't mean to say that not interpretting it is insane. In fact, that's quite a practical approach which will work for the foreseeable future. I just mean that if you're going to interpret it you've got two options: either come up with something really wacky and mystic, or go with many histories or some other Everett variant.
( 5 comments — Leave a comment )

Profile

blueshirt
spoonless
domino plural

Latest Month

May 2017
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

Tags

Powered by LiveJournal.com
Designed by Lizzy Enger