June 28, 2010

Many-Worlds Incoherence

First, I understand how thinkers felt driven to MWI to avoid what they considered an otherwise insoluble (?) quantum measurement problem. But MWI is IMHO grossly fallacious and full of hand-waving snow-jobbery. The honest way is to admit we can't solve something, not to put forth a fallacious pretense because many thinkers just can't stand the alternatives. The key difficulty is in way things "split" and the attitude towards why and how we find quantum statistics. What we find is statistics, not a superposition of states. The statistics are based on squared amplitudes.*

So how does MWI deal with that? Well suppose there are two states, showing amplitudes: 0.6i |1> + 0.8 |2>. (Sometime I'll get around to improving the symbols here.)  When we look we find either |1> or 2>, at respective probabilities 36% and 64%. But we started with two states, and no statistics. In WMI, neither states just "goes away" and the true outcome in the mulitverse  has no preference - hence no genuine statistics of e.g. a real card game (heh, well a "real card game" as we imagine it in a classical world or collapsian world!) But there "are" only two states, even if we keep both and just somehow decouple them from each other. How oh how to get 36% of one outcome and 64% of the other? A mere split wouldn't do the trick, since I'd have equal chance of ending up in either outcome. After all, the amplitudes are just intensities, not actual sets of things. (And if they were, how many would there be in each set? Some arbitrary number? But if infinite such as Aleph null, no way to have the ratio.)

So the MWIers double-talked their way into some mystical, inexplicable sort of thickness of the wavefunction that somehow stands in for real statistics (like, authentic frequentism) that they call "measure."  Everett imagined that one's "subjective probability" of ending up in a world depending on this "depth" (which BTW is not the amplitude, but its square.) We see an illustration on p. 172 of the silly and fawning book Schrödinger's Rabbits by Colin Bruce. Huh? That's just doubletalk. What the heck is that? If there's no unique self but both states |1> and |2> continue to evolve, then it isn't like the chance that one "self" will end up in world A or world B. They can't explain what they mean by that. Real quantum experiments do build up patterns the same as are generated by classical sources of unique outcomes.

Furthermore, this sort of "measure" is not the authentic probability measure in statistics. Maybe that fools readers about MWI into thinking a sort of inherent, hand-waving mystical equivalent of real probability (of how many of one v. the other) is mathematically rigorous and kosher. But I can't see any way to explicate the phony MWI "measure" and explain what's it's proponents claim is going on. For example, in Rabbits we see this quote:
"He has his own take on the question, does measure require large, maybe infinite numbers of each world-line to generate the correct probability ratios. For him, measure has no more meaning than it is postulated to have. You could perhaps (very loosely) think of it as a kind of tag attached to each world-line with a percentage value written on it, but certainly not in terms of huge stacks of each world-line."
Huh? A "tag" in which you just wave a wand and say "there aren't really more of one than the other, but it's as if there were." How?! This is an abomination, it's intellectual irresponsibility.

Another big complaint I have about MWI is about where in the chain of events, does the separation of states come about? Briefly here, I may have more later: In MWI, observation is not special. So, that juncture should not be the specific instigation of the separating of the states. Now consider a Mach-Zehnder Interferometer with photon entering and "splitting" at BS1. So shouldn't the photon state "separate" at the first beamsplitter instead of after recombining at BS2? Why not? That's supposedly a choice between taking the lower leg or the upper leg, isn't it? But if that happened, then the lower and upper paths are "separated" - so why should they be able to interfere later?  Yet we know there's interference.  I realize that issues of of decoherence complicate that, but it's food for thought. (I also have big gripes about the idea that decoherence explains collapse. Read more elsewhere on this blog, or Google for decoherence + "circular argument."

Schrödinger's Cat is still wanted, dead and alive. Reports of his liberation from uncertainty are greatly exaggerated.  Is there an answer?  I don't think so. I believe the universe is just not always amenable to our modeling hopes and capabilities.  We should have the humbleness to accept that could be the case.  It sure beats coming up with disingenuous "models."

*(Well, squared moduli, but the complex numbers just show relative phase. We can represent phases between real AC currents with complex numbers if we want to, it doesn't suggest anything about how "real" the currents are or not. So it's a canard to say "the WFs aren't real, because they "are represented" by complex numbers. Uh, no - not "are represented", that's just a choice. And if the WFs aren't real, what the hell is really there?)

Labels:

27 Comments:

Anonymous Anonymous said...

Thanks for writing this up.

What do you think about David Deutsch's claims that quantum computing "proves" MWI? I believe he said something like, "If MWI isn't true, then explain how a quantum computer can be getting answers in polynomial time"?

28/6/10 15:02  
Blogger Neil B said...

Well, I don't have a simple or deep answer but looking at the explanation of quantum computing (e.g. Wikipedia) the conceptualization seems independent of whether there "really are" other worlds. The "formalism" explains in the abstract why the results are what they are, so once again it seems we are stuck with wondering how to "interpret" the machinery for getting right answers. And again: so, how many other "worlds" are there; as I noted?

28/6/10 20:19  
Anonymous Anonymous said...

Thanks for providing answers to this lone anonymous questioner!

Let me see if I understand you clearly, though. When you say "The "formalism" explains in the abstract why the results are what they are, so once again it seems we are stuck with wondering how to "interpret" the machinery for getting right answers.", do you mean that the results of quantum computing are compatible with the usual wide variety of interpretations (and possibly more)?

Mind you, I'm not a MWI advocate. I just wanted to make sense of the claims, and had stumbled over your blog in the past.

28/6/10 20:32  
Blogger Neil B said...

Basically, as best I gather - yes. The operation just as easily refers to the state space and how they collapse under rules about entanglement and squared moduli etc., there is no explicit need to imagine literalism of the other outcomes. (Deutsch is basically just saying, "how could it ... " Well, how could a lot of things ...)

And again, MWI doesn't "coherently" so to speak even rigorously explain just how to divvy up the other worlds.

Some people claim there are true experimental ways to distinguish. See my post Decoherence Interpretation Falsified? in which I propose a test of whether decoherence can lead to a literal mixture of photons instead of a continued superposition. It got argued about a lot. Note that DI is not identical to MWI but they cross-pollinate. Finally got acceptance that my result would obtain (experiment not done yet) but not agreement on whether I was right about implications.

BTW if you think I've got some clever points you could help by linking at various blogs and fora ... ;-)

28/6/10 21:00  
Blogger Neil B said...

This
article
by Adrian Kent at ArXiv takes aim at MWI in similar and more expert vein. He says of various attempts to justify MWI:
This article reviews some ingenious and interesting recent attempts ... and explains why they don't work. I like this quote:
It is argued that Wallace's strategy of axiomatizing a mathematically precise decision theory within a fuzzy Everettian quasiclassical ontology is incoherent.

28/6/10 22:03  
Anonymous Anonymous said...

Thank you for the paper reference, I will struggle to absorb it.

In the meantime, I have another question or two for you. Ones more amateur.

Doesn't the MWI theory commit one to the view that, whenever scientists are studying quantum theory, other scientists in other universes are necessarily interfering with their experiments? And that these scientists basically 'pop into existence' when the experiment starts, exact duplicates conducting the same experiments, and that we can't make sense of our answers without their results being factored in?

Is it ever explained how or where these other universes come from? Or is this just accepted as a fundamental rule, that universes like this just pop into existence all the time?

29/6/10 00:13  
Anonymous Allen said...

If the laws of physics govern the universe then, since we are part of the universe, they also govern what we experience and think - and therefore what we can know about the universe.

If these physical laws are deterministic, then they absolutely determine what we know. In this case, everything that transpires (including the discovery of our current scientific theories) is a fixed and inevitable consequence of two, and only two, things: the universe's initial state and its causal laws.

If the physical laws are probabilistic, then these laws still set the framework of reality, but within that framework there is a random aspect to the actual flow of events. In this sense, the universe is similar to poker - the rules of poker are stable and unchanging, while the randomness of the shuffle adds an element of unpredictability as to which cards you are actually dealt. So the origin of every thought and experience can be traced back to the interplay between the fixed rules and the random "shuffles".

In either case (deterministic or probabilistic) the laws of physics act to "generate" our experiences of, and thoughts about, reality.

Given that this is the case, is it reasonable to assume that our generated experiences tell us something true about the generating process? Why should they? What justifies the belief that they do? What would require that the generated experiences only be experiences *of* the process that generates them? Isn't it just as plausible that our generated experiences reveal nothing about the generating process?

For example, our experiences in a dream reveal nothing about the "real" waking world. Given the example of dreams, why do we assume that our waking experiences reveal anything true about the objective physical world?

The orderliness and consistency of our waking experiences seems to require an explanation. And since we can't explain these experiences only in terms of themselves, we introduce the additional concept of an orderly and consistent external world. In this view, our experiences are orderly and consistent because they are caused by an orderly and consistent external world. Problem solved.

BUT wait...if the "External World Hypothesis" explains the orderliness and consistency of our experiences, then what explains the orderliness and consistency of the external world?

It seems that we didn't really answer the original question, we just rephrased it. The exact same question about order and consistency still remains, but now we ask it of the external world instead of our experiences. Though for some reason adding this extra "physical" layer between us and the question strikes some of us as progress.

But if the answer is "nothing explains its order and consistency, that's just the way the external world is," then why not just make the same fiat declaration about our conscious experiences and judge the problem as solved there - without introducing all of the extra machinery of the physical world? Why not just say: Our conscious experiences exist fundamentally and uncaused. There is no reason that they're orderly and consistent, they just are.

So, to recap: We can explain our experiences, but only in terms of something that is itself unexplained. And in the process, we've introduced new questions: What is the nature of matter, energy, space, and time? And how do they all combine to generate conscious experience, when none of them have any obvious connection to conscious experience?

We can translate statements about our conscious experiences into statements about the interactions of quarks and electrons, BUT, this does not mean that quarks and electrons actually exist, or are the building blocks from which our experiences are constructed.

29/6/10 14:43  
Blogger Neil B said...

This comment has been removed by the author.

29/6/10 15:17  
Blogger Neil B said...

Allen, very thoughtful. First, I don't think it's as simple as just imagining determinism or randomness or even a combination. First, "randomness" can't even be defined logically as a pure entity since REM in classical mechanics everything is "determined", we just don't know all the details. Note also that mathematics is deterministic: you can't construct a math process that produces different results each time from the same entity. (And "random variables" are cheating, since you just say by fiat they produce random output without having the Platonic "machinery" in place to do so.)

Yet "real indeterminacy" (like in traditional QM, but not MWI!) is not a well understood or conceived concept. It is hard to explain, we can't generate it by logical models or techniques since they are deterministic (like, taking roots of integers, always the same output.) I think there's more to all this. For now, consider I can have a determined series, or a random series; or instead something with an "intrusion" to which neither a particular probability nor deterministic law can be assigned:
1, 2, 3, 7, 5, 6, 7, 8, 9, ... > ?

As for laws and explaining experience etc: we say there are laws explaining the way either things happen or we experience them; or both together. Then we can ask: why those rules instead of some other set of rules? That is the big question, more than whether the rules per se could just be directly phenomenological versus apply to a ground that we experience indirectly. And that big question leads into all that modal realism and infinite possible worlds etc. down blog.

BTW it is helpful to me to track where commenters found me, LMK if OK with you.

29/6/10 15:21  
Blogger Neil B said...

Anonymous, it isn't necessary to believe the other worlds pop into existence. Ordinary, pre-MWI QM just said, the results of QM depending on probabilities. That which MWI says splits off is just "lost" in the old QM. We don't "need" to imagine those other streams of reality, they are an interpretation by those who can't imagine how some outcomes could happen and not others. That's why the universe would generate them, if it can't just "decide" and cut out some and not others. But maybe someday Deutsch or et al will prove they are needed - but the burden of proof is on him.

29/6/10 15:26  
Blogger Neil B said...

Allen - and sure, the observations don't prove that quarks and electrons really exist (or in the manner we imagine.) Indeed, we can imagine a structure that regulates "appearances" and that we can't genuinely model or picture. Considering the collapse problem and the pitiful attempts to resolve it (giving up, a mental event, MWI, phony decoherence double-talk, etc.) I think we can't fully model the universe after all.

BTW maybe even better than Plato or Kant, is Spinoza - who understood the idea of "relative character" or what in phil-mind is called property dualism. I think only PD has the chops to reconcile experience, brain function, and causal integration.

29/6/10 15:59  
Anonymous Anonymous said...

Neil B,

I found you from some of your comments on David Heddle's blog a while ago.

Also, I know that QM doesn't commit one to believing that these universes are popping into existence, so to speak. I just was wondering if MWI did. As in, isn't that what they're really saying?

Thanks!

29/6/10 22:11  
Blogger Neil B said...

Yes, regular QM said the wave states just settle down and the unobserved one is just lost - butMWI says the waves split, as if worlds. However it is ironic since MWI says they aren't creating a new world, just keeping the "multiple worlds" of superposed wave functions (like alive + dead cat) that were already there to start with. But the difference is, in MWI those states detach and don't influence each other any more. It's a detaching of what was already in existence, not a whole new reality.

But like I say, they're wrong anyway.

29/6/10 22:27  
Anonymous Allen said...

I'm out of time tonight, I will reply to your response on this post tomorrow!

30/6/10 01:03  
Blogger Phil Warnell said...

Hi Neil,


Interesting article and one I find a little surprising as I always had you pegged as a Many Worlder. What seems to be the essential distinction that is being made here is that there is a difference between potential and possibility in respect to what can be actualized and still remain consistent with the spirit of probability. I think this is an excellent example of how interpretations like Bohm’s differ from that of many worlds as the later has all possibility on equal footing while in the former has it limited to what Bohm called the quantum potential. This potential serves as a qualifier within the theory as opposed to a quantifier which is the essential element that is missing within many worlds where there exist no qualifier as to what van be and not be.

Best,

Phil

1/7/10 06:50  
Anonymous Allen said...

I think I originally came across your blog via some comments you left on a Cosmic Variance post by Sean Carroll.

SO, why do you believe that something underlies and "causes" consciousness?

And if that is the case, then what causes the cause of consciousness? And what causes the cause of the cause? And so on?

And why that chain of causes instead of some other chain? Why any chain of events at all, instead of nothingness?

What if your life is a dream that has no dreamer? What makes this impossible? Or even unlikely?

1/7/10 23:32  
Blogger Neil B said...

Allen, you view reminds me of some deep Mahayana Buddhist thought, the "no mind" (more like "no ego") school which says there is no specific self in the midst of the conscious phenomena but only the phenomena themselves. We are "skandhas" held together by worldly desires and true liberation means letting the clay pot representing the self shatter so the pure mind can be part of the whole it was enclosed out of. Even then, there is "desire" motivating the phenomenal universe (or else, why the sorts of things we see? It's an early attempt to explain anthropic qualities.)

I had the great opportunity to study Tibetan Buddhism at UVA under renowned expert Dr. Jeffrey Hopkins, a personal friend of the Dalai Lama.

But if there's "no reason" I can't imagine this being here instead of chaos or "everything" without distinction. I can't accept "it just is" at this level. Whatever "just is" is deeper and more abstract and less a show of bits of happenings than this stuff and experience here. I don't know of course, who does?

2/7/10 16:55  
Anonymous Allen said...

Maybe reality is chaotic when looked at on larger time scales?

In an infinitely long string of random numbers, every possible finite sequence of numbers is repeated an infinite number of times. Which means that the binary code of every operating system every written is found an infinite number of times. The ASCII equivalent of every book every written occurs an infinite number of times. The JPG formatted equivalent of every painting and every photograph ever made occurs an infinite number of times. The digital equivalent of the DNA of every human who has ever lived, arranged in chronological order by birthdate, is in there an infinite number of times. And so on.

And yet, despite all of those gigantic islands of apparent order - taken as a whole the number string is still completely random.

Besides, if reality was mostly chaotic, would you necessarily even notice? Maybe your awareness is time sliced like a program running on a multitasking operating system. So you have a few seconds of conscious experience of an orderly world before dissolving into chaos for the next trillion years, until just the right configuration emerges to give you your next few seconds of awareness. Maybe the time slices don’t even execute in order. Maybe you have memories of a childhood whose time slices haven’t even executed yet? How would you know?

Boltzmann Brains on steroids...

It seems to me that everything we know tells us that we can’t trust what we know. And yet people continue to dogmatically insist that we can. Very peculiar.


Speaking of Boltzmann Brains, have you read Sean Carroll’s book, “From Eternity To Here”? I thought it was very good.

4/7/10 05:52  
Anonymous Allen said...

"When I consider the small span of my life absorbed in the eternity of all time, or the small part of space which I can touch or see engulfed by the infinite immensity of spaces that I know not and that know me not, I am frightened and astonished to see myself here instead of there...now instead of then." -- Blaise Pascal

4/7/10 18:45  
Blogger Crude said...

The ASCII equivalent of every book every written occurs an infinite number of times. The JPG formatted equivalent of every painting and every photograph ever made occurs an infinite number of times. The digital equivalent of the DNA of every human who has ever lived, arranged in chronological order by birthdate, is in there an infinite number of times. And so on.

Perhaps I misunderstand you, but all of those 'equivalents' seem to presuppose a mind. Otherwise, we just have random numbers with no meaning to them.

Or do we? What use is this "random" qualifier? These things are all possible in an "infinitely long string of arranged numbers" too. Heck, the string doesn't even need to be infinitely long if you go that route. Though that route is still optional.

So oddly, I'm skeptical of this whole "random" thing, taken in that sense. Useful for some models, I'm sure, but I find it harder and harder to take too seriously otherwise.

I think the comments about there not being anything that 'causes' consciousness is interesting. But from what I read, that's just dualism. (Not that dualism is bad, mind you. Seems a helluva lot more reasonable than materialism, or at least it did before materialism was broken open enough to do jobs dualism used to.)

4/7/10 23:07  
Anonymous Allen said...

Crude, I was responding to Neil’s comment that:

"if there's ‘no reason’ I can't imagine this being here instead of chaos or 'everything' without distinction.”

My point is that even a purely random (chaotic) process will manifest arbitrarily long periods of what appears to be order, given enough time.

This is what gives rise to concerns about the real possibility of Boltzmann Brains. Sean Carroll discusses this in his book, "From Eternity to Here", Pg. 363:

"This version of the multiverse will feature both isolated Boltzmann brains lurking in the empty de Sitter regions, and ordinary observers found in the aftermath of the low-entropy beginnings of the baby universes. Indeed, there should be an infinite number of both types. So which infinity wins? The kinds of fluctuations that create freak observers in an equilibrium background are certainly rare, but the kinds of fluctuations that create baby universes are also very rare."

In fact, if you take a materialist view, then given what we know about physics and applying the standard assumptions of statistical mechanics, by far the most likely conclusion is that all of our memories are false and that we recently fluctuated into existence from a much higher entropy state. See Here for more on this subject.

Also, my proposal that only consciousness exists isn’t really dualism, it’s a form of monistic idealism.

As for my argument that it's makes no sense to posit a cause for conscious experience (because then what caused the cause...), Quentin Meillassoux has developed an interesting idea which he calls the "Principle of Unreason", in contrast to Leibniz’s "Principle of Sufficient Reason."

I disagree with him about the significance of the "Problem of Ancestrality", and he ultimately draws different conclusions than I do from his Principle of Unreason (he labels his position "speculative materialism"), but still his ideas are very much worth reading I think.

5/7/10 01:49  
Blogger Crude said...

Allen,

Greetings to you.

No, I get the point you were getting at. I think I was talking past your point - forgive me, I've got a habit of doing that, but I mean no disrespect.

In fact, I agree with you about Boltzmann Brains and materialism (especially the MWI form of it). Then again, I don't have a high opinion of materialism.

And, I had a feeling you were getting at idealism. But my understanding of dualism (cartesian in particular) is that there too consciousness is not 'caused', as in 'caused by' some underlying thing or substrate. It's irreducible. So I wonder if, in a way, idealists and dualists may be talking about the same 'stuff' where minds are concerned, with idealists taking the extra step and adding "So we don't need that other half of the thing that makes you a dualist."

5/7/10 02:54  
Blogger Neil B said...

Allen, Crude - You've brought up interesting points about dis/order (a deliberate conflation to indicate that order may come from disorder as suggested.) However, my essential point about it - better made at my thread "Marcelo Gleiser Has a Point" - isn't about getting orderly Shakespeare from monkeys at typewriters. Sure, eventually we would. But rather the problem is what kind of physical laws would we expect to start with. IOW, where do the monkeys and typewriters come from, and who designs the keyboards. And I argue that the expectation is, nothing orderly unless a Big Boss Something is a source of order.

Mind/body problem - I think consciousness is a fundamental potential of a real universe, not any operational trait that can be defined in behavioral, computational, output, "outside" terms. And maybe Spinoza had the best insight that it was an issue of relative character, as property dualists say.

The trouble with sheer idealism is, what orders our experience, what keeps track of what we put in the drawers etc? I do think it is like a big Mind, but hardly just any one person's mind (whose?) If the former, we need coordination. But I do think that can't be modeled in naive, childishly realistic terms. That's why no one can figure out "what happens to the wave function." That also tempts susceptible minds to cheat as with the decoherence interpretation or many-worlds.

BTW if you have any access to academics who might take my proposals seriously, I'm hoping to find someone who can e.g. do the experiment described in "Decoherence Falsified ..." Yeah, I have access too but the more the better.

5/7/10 09:10  
Anonymous Allen said...

Neil:

"And I argue that the expectation is, nothing orderly unless a Big Boss Something is a source of order."

And I argue that we may very well be living in a chaotic universe, but in one of the islands of order.

Even if you find a Big Boss Something to explain our order, what explains *HIS* order???

By taking that route, you haven't answered any questions - and in fact you've introduced MORE questions.

Can you really not see this?


Further:

"The trouble with sheer idealism is, what orders our experience, what keeps track of what we put in the drawers etc?"

Nothing does. As with Carroll's Boltzmann Brains, there are probably some conscious beings who have very disorderly experiences.

Infinitely many such beings, in fact. *And* there are infinitely many beings that have orderly experiences.

But they are the some cardinality of infinity, so neither infinity "wins" and is pronounced the "most likely."

5/7/10 16:44  
Anonymous Allen said...

Crude:

Right, dualism introduces more problems that it solves.

How could you differentiate between the mere *experience* of kicking a rock (as in a dream) and the actuality of physically kicking a rock?

Every property we ascribe to the physical is really just an aspect of our experience.

The "physical" serves no purpose expect to explain the order of our experiences. But then what explains the order of the physical?

Best to just do away with the physical. It's really not even a coherent concept. What is matter made of? Energy? Well, what is energy made of? Where does it come from? Why does it have the properties it has instead of some other properties. Etc. Etc.

5/7/10 16:48  
Blogger Neil B said...

Allen - yes we could be living in a chaotic universe but you have to get the point about physical laws v. what happens under various legal regimes. Otherwise you're not stating a meaningful point. I'm also getting at a very important further point in all these discussion: given what we already find, what else should we expect to be true? I say, in a world that's just "here" among modal realist PWs there isn't any expectation of greater order, such as electrons all having the same mass, charge etc. or the laws continuing to play out in like manner as before. It's kind of like, if you already played a royal flush, you don't expect more in the future. (See "gambler's fallacy.")

As for "explaining" the order of the ultimate reality, I'd say it's just a given. There just is such a reality and it orders everything else. I don't think "everything needs explaining", I know the turtle problem. So it's about all that "just plain exists" (something does, right?) and what it would be like.

As for the transfinite sets problem, look at my comments at Alexander Pruss's Blog. I think if that set problem was definitive then we'd have trouble here already.

I gotta give you credit for bucking usual trends and supporting Idealism.

5/7/10 17:21  
Blogger Crude said...

Allen,

Best to just do away with the physical. It's really not even a coherent concept. What is matter made of? Energy? Well, what is energy made of? Where does it come from? Why does it have the properties it has instead of some other properties. Etc. Etc.

Richard Conn Henry takes a similar tack (physicist with who wrote "The Mental Universe" for an issue of Nature.)

I admit, I recently had a laugh when I decided to look up the definition of energy, and there came the Feynman quote about how we don't know what energy is. And then going to the SEP to find out what physicalism is, and lo and behold, there are deep arguments about that as well, with the definition open-ended enough that panpsychism and possibly idealism (!) could be called physicalism now.

So, consider me refreshed to find an idealist online. (I know of Richard Conn Henry, Kenny Pearce (philosopher), possibly Keith Ward, and a scant few others.) Me, I'm less committal. But as our host said, congrats on bucking the usual trends. Makes these conversations more interesting.

5/7/10 22:01  

Post a Comment

Links to this post:

Create a Link

<< Home