If you haven’t seen it already, the front page of the July 5th issue of Nature has an amusing picture inspired by the “Many Worlds” interpretation of quantum theory. (I’ve said it before, and I’ll say it again. One world is enough for me, thank you very much.) Included in this issue is an essay on the many worlds interpretation:
All three approaches have their adherents, but for what seems to be a growing number of physicists, especially those working in quantum information and cosmology, it is Everett’s alternative that wins out…
Hmm, well we can run an experiment, err, I mean a social science experiment, aka a poll. Okay so these categories are subject to interpretation. For a rough guide here is wikipedia’s interpretation of quantum theories page.
[poll=2]
Also in this issue is an article Many lives in many worlds by Max Tegmark arguing for the many-worlds interpretation as well as an article Surfing the multiverse by Gary Wolfe describing many worlds and science fiction.
Finally, and perhaps more exciting than all this many worldliness, in this universe it appears that low enegy acoustic plasmons at metal surfaces have been observed. Word as to whether this observation has been made in other universes awaits the construction of a quantum computer 😉
I like that you have both an “Other” and “None of the Above”. Is this some kind of subtle joke?
I have to vote for either “Other” or “None of the Above”. Not sure how to decide on which 🙂
I don’t think we understand quantum theory, yet.
Of course, I really agree with Asher Peres’ comment after a similar vote: “And who here believes that te laws of Nature are decided by a democratic vote?”
I voted for “non of the above”, because I think that QM does not require any interpretation. The goal of a good physical theory is to predict results of measurements as soon as preparation conditions are specified. QM does exactly that and does it brilliantly. All these “interpretations” are supposed to give us this “good feeling” that we understand what happens with the system while we are not watching. Even if we think we understand this, we have no way to verify our understanding experimentally. So, it doesn’t count as science.
It’s too bad you didn’t have an “I don’t care, I just calculate when necessary” box. Or, are those of us agnostics just supposed to select “Copenhagen”?
I kind of agree with Jon, and so checked Copenhagen.
But is the pot calling the kettle black here? Quantum Computing has contributed to the revival of interpretations-fever by its continual focus on the weirdness of quantum physics in its popularizations.
For example, the continual use of the word “entanglement” in your discipline, even when the term is not needed, helps to drag us further into the morass of interpretations.
Well what I meant was “Other” = “something that does not fit into any of the above categories” like, for instance if you have your own pet interpretation. Whereas “None of the Above” = “all interpretations stink, I don’t like any of them” Or more bluntly “Other” means your a crackpot, whereas “None” means our agnostic.
In retrospect I should have put a shutup and calculate. In light of the above, I think “shup and calculate” goes best in “None”.
“And who here believes that [the] laws of Nature are decided by a democratic vote?â€
Who here believes that the laws of Nature include an interpretation of quantum theory? 😉
“Quantum Computing has contributed to the revival of interpretations-fever by its continual focus on the weirdness of quantum physics in its popularizations.”
I know where you’re coming from, but I have a slightly different read. Part of this is that quantum computing stresses a particular take on interpretations which is that it focuses on information/computational resource quantification. For example a lot of work which most people call interpretational these days focuses on theories which share similarities with quantum theory but are different. And that these differences lead to differences in computational complexity/information resource tradeoff/etc. is studied. I guess part of the hope is that this broadening of the space of theories considered might lead towards picking out the unique features of quantum theory and hence give some sort of “explanation” of quantum theory.
I have a strong love for the word entanglement and would love to hear examples where it is used incorrectly!
What about the Relational interpretation?
My own bias and ignorance has relegated the relational interpretation to the other bin. I guess I shouldn’t say “other”=”crackpot”, doh.
I don’t entirely agree that theories are restricted to what we can observe and predict. Models have their interpretation, and it doesn’t need to be minimal. (Though it should be minimized.) And in that sense I think the “non-interpretative” alternative makes an interpretation as well.
QM interpretations are different because there are contenders that can’t be distinguished (yet) and it is difficult to decide which one is “better” or even minimized. AFAIU Tegmark notes the MWI down as minimized because two ordinarily axioms can be interpreted as results, which beats the agnostic alternative. But other persons judge it differently, and there is no right or wrong.
(OK, some of the kookier ‘interpretations’ are wrong.)
Any information about the results of this poll in the other universes?
s.
” Once, at the afternoon tea, in the Institute E.Teller tried to explain to Bohr why he thought Bohr was wrong in thinking that the historical set-up of classical concepts would forever dominate our way of expressing our sense experience. Bohr listened with closed eyes and finally only said: ‘Oh, I understand. You might as well say that we are not sitting here, drinking tea, but that we are just dreaming all that.’ ”
‘The Copenhagen Interpretation’, by C.F.von Weizsaecker, in ‘Quantum Theory and Beyond’, Ted Bastin ed., Cambridge U.P., 1971.
“would love to hear examples where it is used incorrectly!”
– I don’t know if it is incorrect so much as violating Occam’s Razoer. I’m not a quantum computerer, so I don’t have precise references, but he kind of thing I am thinking about happens when I see writing about NMR processes as the basis for quantum computing. The quantum mechanics of NMR pulse sequences and their relaxation processes have been discussed for years without ever once using the word entanglement. Suddenly when these same processes become of interest to the QC community the word appears.
Being a natural cynic, I suspect use of the entanglement lexicon is tightly coupled to grant proposals and other external funding proposals, particularly from potential private sector physics enthusiast patrons.
I still think that for most appealing, I take Bohm’s pilot wave. It’s deterministic and it’s easier to explain. Don’t mind Bell non-local inequalities, that’s for people who want to spoil the fun… Many-worlds is rather Deustchesque, I like it a lot too.
Now seriously, it doesn’t matter, I think there are some things that, within our lifespan, are just impossible to elucidate.
I’m convinced that in some universe the MWI is the hands down winner.
“in some universe the MWI is the hands down winner.”
I’m tempted to say “Last time I looked it was Copenhagen” but I’d better not.
Hey Tom S.
Oh yeah, I hear you. This is especially interesting since most NMR experiments don’t have entangled states (too close to the maximally mixed state.)
There is, I think, a lot of bandwagoning in quantum experiments. I don’t find this too troubling as long as the experiments are interesting. It seems that physics has a long long history of such bandwagoning. Eventually, of course, the bandwagon will leave town: either as a quantum computer or as failed endeavor.
Oh, hey look someone else voted for the transactional interpretation besides me 😉
My problem with Copenhagen is that it doesn’t say what “counts” as a measurement, and as such is incomplete. You can’t “shut up and calculate”, because there are border cases where the rules aren’t clear as to how you should calculate. Is this interacting system “classical enough” yet? Meh. Everett at least makes some answer with “it never is, but it can look like it, and it’s a more-or-less smooth transition.”, though that’s not terribly solid either.
Travis, it comes down to problems to do with self identity and why it is desirable to be alive in the first place. From my point of view, the main problem with your scheme is that the “me” in the other worlds is not the same “me” that is in this world. He’s had different life experiences, different thoughts, etc. If I kill myself, my consciousness isn’t going to magically jump into one of the other worlds. I’ll just be dead. There will be some other fellow in some different part of Hilbert space who is very much like me, and very rich, and still alive, but it isn’t me.
If that argument isn’t convincing to you, try this: If there is a single universe, my goal might be to stay alive for as long as possible (and be as happy as possible while doing so, or whatever).
For many worlds, it makes sense to me that the goal should then be to maximise my total expected lifespan. ie. the sum of all the lifespans in all of the many worlds weighted by the mod square of the amplitude of each world. If I kill myself in one world, that clearly isn’t heaping my cause.
I’m sick of hearing about Many Worlds. Nobody *really* believes in many worlds–if they did, they’d either be rich, or dead of apparent suicide. Why? Well, if you believe in many worlds, there’s a simple get-rich-quick scheme (or, for those who don’t care about material things, a way to never have to worry about grant applications ever again):
Prepare a quantum superposition with equal-in-magnitude amplitudes for each of the different basis states (for some basis that is convenient to measure in). Take your quantum superposition to the race track, and assign basis states to each of the contenders. Measure the state, and bet your life savings accordingly. If you win, great–repeat as necessary until you reach your monetary goal. If you lose, kill yourself. If you’re worried you might become squeamish at that point, you could devise a Schrodinger’s cat-style automated suicide device, and hook it up to your measurement device and sportsbook.com.
In all worlds in which you’re still alive, you’ll be rich. In any case, you’ll never have to worry about grant applications again.
This scheme is rather obvious, so I can only conclude that those who claim to believe in Many Worlds are either insincere or far from convinced.
i am too chicken to make the bet because of two failure mechanisms
1. Quantum Info Way
your adversary, Eve, secretly cnots your superposition onto her register. she then measures
and goes about classically rigging the race. you die in all worlds.
2. Experimental Way
your experiment measures correctly 99% of the time. in some small fraction of the cases, your measurement never contains the winning horse. you die in all worlds.
Travis, you need to be careful. There is always the possibility that you’d botch the job and be left brain damaged and unable to finish off the job. There is bound to be some amplitude associated with that possibility.
Richard Dawkins is in the universe where he believes in many universes and has a large ego.
Since all (correct) interpretations are indistinguishable, it is obviously something wrong with this assumption. 🙂
In this case I would suggest that most people doesn’t fancy suicide. I don’t think they will feel better to hear that 3 individuals out of 10 000 (or whatever) will not experience it.
Besides, it isn’t ‘analogous’ to being 0.03 % alive and rich, it is ‘analogous’ to being 99.97 % dead and poor. 😛
[Btw, you will loose this argument anyway: an MWI:er will conclude that in some worlds you will agree with him. Bummer! :-o]
Travis, I realise that there are “many worlds in which everything is identical up until the measurement”. But you were talking about waiting to see if you win a bet or not, and killing yourself if you lose. That isn’t exactly a single quantum event.
Look, I can see that you still aren’t satisfied with my arguments. So I’ve got something else for you to think about. Your “winning strategy” for the many world interpretation was to basically make a bet based on some quantum measurement, then kill yourself if you lose the bet. Let me suggest a different strategy, and then you can try to tell me why yours is better. My suggested strategy is to do everything exactly the same as in your strategy except that if you lose, don’t kill yourself. That way, you still get to be rich in some other of the many worlds (like in your strategy) but not die in the world in which the bet was made. After all, what is to be gained by killing yourself?
(By the way, Aaron is right. It isn’t so much “many world” as it is “one world in a superposition of different states within a quantum universe with no collapsing wavefunctions”.)
Joe: There’s only a chance of that if you use some exotic Dr. Evil-style killing apparatus. You just need something simple and fault-tolerant to avoid that possibility.
stillpoor: Keep Eve away from your apparatus, and test it thoroughly beforehand to ensure it works properly. Use appropriate error-correction techniques to ensure your state is prepared and measured correctly to whatever tolerance you require.
zevans: In addition to the many worlds in which you’ve had different pasts, there are (at the instant after you measure your state) many worlds in which everything is identical up until the measurement. This is how Many Worlds works. It’s only that otherwise-identical subset of the worlds that we care about; presumably, there would be other worlds in which you don’t even believe in Many Worlds and thus would not attempt the experiment.
Torbjörn: This “experiment” doesn’t allow us to determine whether Many Worlds is correct, but rather only whether a given person truly believes in MWI. It thus does not contradict quantum mechanics in any way. I won’t lose the argument unless the Many Worlds believer actually carries out the plan I describe; even if he does end up dead I have to conclude he truly believed in Many Worlds (and is perhaps alive and rich in another world). If the killing device is fast enough, the Many Worlds believer won’t feel or experience anything, except in those worlds in which he becomes rich.
I personally find Many Worlds to be ridiculous–postulating an infinite number of entire universes that we can’t observe seems a horrible violation of the principle of Ockham’s Razor. By comparison, it is downright sensible to postulate a single omnipotent god as the explanation of everything. Surely we can’t do science this way and expect would-be creationists to take us seriously, even if we are just talking about an interpretation.
Given the obvious (to me) ridiculousness of the MWI, I refuse take it seriously unless at least one or two proponents are willing to demonstrate the depth of their belief in Many Worlds. Since I’m feeling generous, I’ll even accept posthumous arguments.
Travis: while I’m not saying that I “buy” many worlds, that’s a horrible misconstruction of it, probably because “many worlds” is a horrible name. Call it the no-collapse interpretation, and things are much clearer. There are no actual sepearte worlds, just terms in an expansion of the wavefunction that mostly no longer interact, so whose evolution can be approximated quite well by only considering each term by itself. This isn’t testable, but calling these terms “extravagant” is like calling the infinite nummber of energy levels in a QHO to describe a particle extravagant.
MWI has problems, sure — explaining the experience of the Born rule is one. But many of the problems ascribed to it are from misunderstanding it.
The real problem with most interpretations is that they don’t nicely explain continuous measurements, and these should be fundamental, with the “instantaneous” measurements being derived as some sort of limit. Zero-time measurements should require an infinite interaction energy.
If it wasn’t clear, I’m saying that the thought experiment doesn’t work as an actual experiment.
You won’t see people willing to agonize and most likely die, just so they will survive, agonized but rich, in some other worlds. It is not a realistic method for concluding if people believes in MWI or not.
And observably, you see people who does, see above. Claiming that they are mistaken (or lie?) is not supported, nor productive.
[For the record, I can support any of the useful interpretations. It isn’t anything I want to needlessly agonize over. Either. 🙂 ]
Moving on, the ‘loosing argument’ schtick was a joke – the subtext is that the MWI:er doesn’t need to concede any argument ever, because he knows that he is correct in some world. (Perhaps you can use that to conclude that MWI is faulty. Either that, or that the particular MWI:er in question is batshit insane. :-P)
Finally, without putting on the “no-collapse” hat as Aaron terms it, I want to support his explanation of extravagance.
What characterizes good theories and how to make them can be discussed ad infinitum. But it seems clear to me that we first want theories that are powerful (makes a lot of correct predictions from a small set of assumptions), then pick the most parsimonious of equal contenders because it is associated with minimizing errors and reversals. Parsimony isn’t primarily important – predictive power is.
Parsimony is not a concern for what the theories describe as results – nature must be allowed to be as complex as we observe it. For example, QFT:s describe a field with oodles of interacting particles, and the “no-collapse” interpretation describes a world with oodles of non-interacting terms.
More precisely, he knows that his antagonist agrees with him in some worlds.
Good lord I’m slow on the uptake. OK, so, Dave, why no button for Fuchs’ Bayesian “interpretation” (if you can call it that)?
Way back in April I posted some random notes of mine from two years ago on why I have a problem with MWI.
Ian, yeah I thought about that right after I posted the poll. Doh.
Ockham’s Razor cuts both ways. You want to use it to chop off the bits of reality that we cannot observe (alternate universes, or whatever), but I want to use it to chop off the bits of our theory that we don’t need (collapse of the wavefunction).
I agree that there is not really much difference between having truly many worlds and having a single world in a superposition, but the point is that with the superposition interpretation we aren’t adding anything new to the theory. We already have these notions of superpositions and entangled states, all we’re doing is applying them to macroscopic objects as well as microscopic ones. It is worth noting that all good interpretations of quantum mechanics should agree with experimental results, so it isn’t surprising to find that the MWI agrees with the Copenhagen interpretation when it comes down to measuring something.
As for the bet; I still maintain that killing yourself works the same way whether there are many worlds or not. You die. That’s all. I don’t know why are you are asserting that being dead in the world were you lost the bet will somehow be better for you. It’s true that if you kill yourself when you lose, then “you” will only observe universes in which you are rich. But the thing is, who is “you”? I currently cannot observe any of the other possible realities, not in theory and certainly not in practice. Killing myself won’t change that fact. There will be some other possible universe in which “I” am rich, and that version of me is just going to continue to do his own thing whether I kill myself in this world or not. It makes no difference to him. All the happens is that in this world, I die. So there isn’t really any point. There isn’t some kind of “super-me” who watches over all of the universes at once, keeping track of only the ones where I am alive. Each universe will have it’s own version of events, and for these sorts of purposes they are completely independent.
zevans: your first complaint could be resolved if we found a bookie willing to accept quantum states as an indication of our choice of wager. That is, you submit your choice of horse or whatever it is you’re betting on in the form of a quantum state (in this experiment you prepare in a superposition). The race takes place, the bookie measures, and you either receive a payout or die, depending on the results of the measurement.
Why kill yourself if the bet fails? Because if you’re dead, you’re not observing anything (presumably we only care about the observable world). The idea is to only observe those branches in which you get a favorable result, and “discard” the rest. In other words, post-selection.
This idea of “one world in a superposition… with no collapsing wavefunction” seems either like a semantic distinction, or begging the question. Are the different terms in the wavefunction expansion equally “real” to whatever we experience? If so, calling them alternate universes seems entirely reasonable, since there are non-interacting versions of us observing those versions of reality. If not, what’s special about the particular term we experience that makes it more real? Might we not call that property a result of “collapse”?
Ockham’s Razor gives no guidance against infinity in the form of infinite energies and such; it does proscribe the unnecessary multiplication of entities. Infinite energies seem slightly awkward, but infinitely many realities pose a much larger problem (whether you call those non-interacting states within a quantum universe or separate universes is irrelevant).
I’m surprised nobody has made the link between my
“experiment” and Scott Aaronson’s proposal for “anthropic computing” (and the implied complexity class PostBQP). That was the inspiration. Perhaps there aren’t enough computer scientists on this blog.
zevans, Travis:
This seems to capture Travis deeper concern. But as I argue the razor still doesn’t cut both ways – the part of a theory we don’t observe directly, we may observe indirectly or not at all, but we may take them as supported because the current theory works.
Cosmic variance may be a good example. AFAIU we can see from the CMBR that the universe should have at least 5 times (IIRC) the directly observed Hubble radius. Occam’s razor on observations would have us cut the implications out because they are theory-derived.
How does MWI (it is an ontological and realistic interpretation, as far as I know) explain Renninger’s ‘negative’ measurement?
http://en.wikipedia.org/wiki/Renninger_negative-result_experiment
Torbjörn, zevans: Ockham’s Razor says one should not unnecessarily multiply entities, and thus one should prefer the simplest solution that explains all the facts. Given a choice between many different interpretations of quantum mechanics (all of which give the same predictions), the Razor implies we should choose the one with fewest entities that still explains the facts. You may find “collapse” unsatisfying, but you should be deeply troubled by an infinite number of entities that are not observable.
Since there are alternate explanations (i.e., alternate interpretations), you can’t argue that we observe “many worlds”, even indirectly; we observe something for which Many Worlds is only one of many possible explanations, and by far the least economical in terms of entities.
The Razor doesn’t cut both ways here because “collapse” is only a single entity (or, at most, a small number of entities), whereas Many Worlds requires infinitely many universes. The Razor favors the simplest explanation.
If you want to provide an exhaustive refutation of all other interpretations on some other grounds, then you have a legitimate argument for MW, since it then becomes the simplest valid explanation. So far, I haven’t heard such a refutation, and until I do, MW has a lot (infinite?) of strikes against it.
zevans: this whole argument about whether one benefits from “committing post-selection suicide” would be much more interesting if we had a Penrose supporter here (i.e. someone who believes consciousness is a quantum phenomenon). As it is, I have to infer your beliefs about consciousness aren’t quite as exotic.
Travis, give me a break! Counting each of the “other worlds” in the many world interpretation as a “unnecessary entities” makes about as much sense as counting all the numbers that we don’t use in mathematics as “unnecessary entities”. These things naturally arise from the theory that we are using. We aren’t putting them in by hand.
infinitely many strikes against it indeed…
Ð’Ñе бы ничего, только комментов много почиÑтить надо.