De Broglie/Bohm versus Everett

I have a more than passing interest in quantum mechanics and have recently been reading a number of articles on De Broglie/Bohm Theory or the Pilot Wave Theory (PWT).

Given that (from the cited article):

Is Everett correct in his assumption that Pilot wave theory violates occams razor and therefore is an inferior way of describing the universe than Many World Theory, which only requires the wavefunction?

Personally, for reasons I can’t explain I like the idea of the pilot wave theory. Many worlds concerns me for the same reason, needless multiplication of entities. Although I am aware that there are numerous variation of MWT, at least one of which requires only a single wavefunction for the whole (infinite) universe and that different realities do not really exist, but those regions of spacetime are simply too far away for us to ever contact (in a nutshell, obviously it’s not as simplistic as this).

So, I have a few questions to debate, is PWT superior to the Copenhagen Interpretation in that it doesn’t rely on a classical observer? In my opinion it is.

How does PWT stack up against MWT? There seems to be arguments in favour of both, PWT has non-local hidden variables, MWT has a wavefunction the solutions of which can either be thought of as parallel universes or separate regions of one universe.

Can either ever be fully (dis)proven?

Well, if both are genuinely interpretations of QM, then no, you can’t prove one over the other, because they give rise to the same phenomenology (otherwise, if they differed in some observable way, they wouldn’t be interpretations, but alternative theories).

As for parsimony, I think a good case can be made that the Many Worlds picture is indeed the most simple one, even though that seems, at first, counterintuitive – but what we observe in quantum mechanics necessitates the existence of superposed states, and Many Worlds essentially boils down to saying, ‘OK, let’s run with that’, while both Bohmian mechanics and the Copenhagen Interpretation introduce additional mechanisms – the pilot wave and the wave function collapse. Under the Many Worlds interpretation, the superposition of states – an empirically necessitated part of the theory – is merely allowed to proliferate, pretty much what the formalism says should happen.

I too prefer the Bohm hidden variables theory. How on Earth the many world’s interpretation could be called the Occam’s razor winner is beyond me. In many world’s, infinite numbers of unobservable universes are splitting off continuously with impunity. Occam’s razor, be damned.

Bohm’s theory is very much like classical mechanics, just with the added wavefunction. Bell’s theorem essentially tells us that we have to give up either locality or objective reality. I prefer giving up locality in a non-local hidden variables theory like Bohm’s.

Unfortunately, so far as I know, neither Bohm, nor anyone else has come up with a hidden variables theory that obeys Lorentz invariance in the relativistic limit. This is a serious shortcoming, which renders Bohm’s theory merely tantalizing but incomplete. I wish 90% of the theorists who are throwing their lives away on string theory would explore some alternative ideas, including this one.

Bohm was a brilliant man, who went a bit off the deep end in later years. Still, his ideas are worth considering.

Keep in mind that Occam’s razor only concerns itself with explanatory entities, of which Many Worlds only has the wave function, while Bohmian mechanics has the particles and the pilot wave (I’m not well versed in that interpretation, though, so correct me if I’m wrong), and the Copenhagen interpretation (at least the way most people seem to use the term) has the wave function and some classical observer to collapse it.

The way I see it, MWI only says that if you have a system in some superposition of two states, |0> and |1>, and have that system interact with the world (which probably means I properly should have written |0, world> and |1, world>, since we’re not treating both as separate), everything evolves nicely unitary, and you end up with a superposition of |0, world interacts with/‘observes’ 0> and |1, world interacts with/‘observes’ 1>; then, all you gotta do is accept this as reality, rather than bringing in additional explanatory entities to force QM to actually decide for either possibility, and discard the other, and you end up with a picture in which both worlds have to be considered equally real. (That doesn’t mean I personally think it’s the ‘right’ interpretation, though.)

Besides, just leading to a potentially infinite reality isn’t generally held against a theory – modern cosmology, for instance, doesn’t seem to have any problem with an open, spatially infinite universe; again, it’s not the theory that lacks parsimony (it’s surely on par with closed universe models), it just leads to a very rich reality.

It is worth mentioning that the notion of “infinite number of universes” is incorrect. The suggestion is for a very very large, but bounded, number of universes. Whilst this is nit picking in one way, it makes all the difference for the application of Occam’s Razor. It is only infinite in infinite time. And infinite time gets you a whole new set of cosmological and philosophical problems.

My favorite way of reasoning about the application of Occam is that it is about looking for the information content of the theory. Which is, at least in principle, something you can actually measure. Reasoning about the infomation content here is going to make one’s head hurt. But it remains important to note that it should be possible and sensible to do so in both cases. One also suspects, that the answer is a tie.

According to an article by Anthony Valentini in Physics World in November Pilot Wave Theory offers some testable predictions such as non-local signals and violation of the uncertainty principle, which could be tested using cosmic microwave background anisotropies and detection of exotic relic particles.

Are there any experimental ways of testign the predictions of MWT that cannot also be explained statistically e.g. proove the existence of parallel universes?

Isn’t the MWI definitionally untestable? If the point is that |0, world observes only 0> and |1, world observes only 1>, by definition |0, world gets no information about |1, world, right?

I feel like I’m typing emoticons. :-0

In my experience, most actual physicists prefer the “shut up and do the math” interpretation. What you attach to the math philosophically doesn’t matter.

I agree up to a point, obviously all three approaches will give the same mathematical result, because fundamentally they are just differrent ways of looking at the same thing.

However, phoilosophy aside, surely there must be an underlying “reality” (can’t think of a better word right now) that explains and possibly supercedes one or more of these approaches or possibly a different approach entirely.

Will we have to wait for a working quantum theory of gravity before that happens or will that still not resolve whether any of these approaches are correct in a physical sense?

Of course, quantum gravity or a full-fledged theory of everything could change the game, by either leading naturally to some form of quantum determinism (such as the pilot wave or Gerard 't Hooft’s proposal, in which quantum mechanics emerges as a statistical description of an underlying deterministic, but information-lossy reality), or judging the quantum formalism to be incomplete – for instance, by introducing some mechanism that leads to an objective collapse of the wave function.