4 ms·
Too many worlds
- jessriedel 11y agoNot really sure what the purpose of this piece is. My research concentrates on figuring out what exactly is missing from MWI to make it a complete, sensible theory, so I am very happy to entertain criticisms of it. But the author don't make much attempt to understand the arguments of MWI proponents. It's basically just MWI bashing. He mostly attacks the popular-level work of Tegmark, or other big-name MWI proponents like Hawking who don't actually publish on this stuff. Tegmark is a fantastically smart guy with deep insight on a lot of stuff, but his approach to the measurement problem simply isn't as clearly and exhaustively laid out as people who concentrate on this full time. In particular, you really need to address guys like David Wallace, for whom articulating MWI from a technical background is part of their core research. It would be like attacking sloppy language in a popular book by a world-famous cancer researcher regarding evidence for natural selection. The cancer researcher has a pretty good grasp on what constitutes evidence for natural selection, but even he can slip up and (say) fall into a just-so story if he's not careful. Attacking that guy is not a good way to engage with relevant philosophy. Anyways, I have a technical comment on this: > This picture really gets extravagant when you appreciate what a measurement is. In one view, any interaction between one quantum entity and another – a photon of light bouncing off an atom – can produce alternative outcomes, and so demands parallel universes. This isn't right in general. There definitely needs to be some sort of amplification (and hence, usage of negentropy) to get the production of branches in the wavefunction. Systems that have thermalized don't branch any more.
- drcode 11y agoThis article seems silly: Both the Copenhagen and Multiple Worlds interpretation (as far as I understand, as a non-physicist) yield the same objective answers, so deciding on one interpretation or the other is really a matter of "taste" for the physicists involved... in a way, each physicist can arbitrarily decide for themselves which interpretation they find most useful for their work. Now Mr. Philip Ball comes along, who isn't a thoretical physicist AFAIK, and says that the MWI people are doing it wrong. In this article, the only concrete argument he gives for why MWI is "wrong" is because it blurs the idea of "self", which the author finds distasteful. But science doesn't "care" about tastefulness, all it cares about is hypotheses, experiments and repeatability. Unless the author can cite an experiment that shows MWI to be invalid (or alternatively can make a formal argument as to why it represents a needlessly complex hypothesis, ala Occam's Razor) he is not basing his argument on a scientific foundation.
- meowface 11y agoYeah, this article presents attempts to prevent a philosophical argument for why the MWI is inherently illogical, but fails miserably. His point about it being an untestable theory, and one which has no particular empirical evidence, is valid, though.
- eridius 11y agoMWI isn't scientific to begin with. As you said, it's indistinguishable from Copenhagen as far as prediction and testability is concerned. The only reason to prefer MWI is for whiffy non-scientific reasons, and that's what this article is challenging.
- Houshalter 11y agoHow is any less scientific than Copenhagen? As I understand it they are both equally valid hypotheses. But people prefer MWI because it's simpler or makes more sense.
- eridius 11y agoIt's not less scientific. It's scientifically identical to Copenhagen (because it makes identical predictions with identical assumptions). But the distinguishing characteristics of MWI (the bits that make it "not Copenhagen") are non-scientific, as they are non-falsifiable.
- Houshalter 11y agoSee dkbrk's comment on this thread. MWI follows naturally from our models of quantum physics. It doesn't make any additional assumptions. It's the simplest possible model that fits the data. It's Copenhagen that makes additional, untestable assumptions. And therefore, by your logic, is unscientific. That is, Copenhagen asserts that the quantum states we don't observe, are destroyed. They have to exist, in some sense, to provide the weird entanglement effects we observe in quantum physics. The whole "the cat is both dead and alive at the same time" thing. MWI just follows naturally from these observations. Copenhagen assumes that there is only one universe, without any justification for this assumption. It assumes that the quantum states we don't observe just cease to exist. It assumes some source of randomness, when "universe splitting" is sufficient to explain observed randomness. It adds a bunch of weirdness that isn't necessary.
- kmicklas 11y agoThis article is complete garbage - full of ad hominem attacks and philosophical whining. The author rejects MWI for apparently completely emotional reasons and seems to take the idea of there being multiple worlds in a much too concrete manner.
- meowface 11y agoPhilosophical whining isn't so bad if the philosophical logic is actually sound. This article is a product of someone who knows little of either philosophy or physics attempting to use the former to refute ideas of the latter.
- jblow 11y agoI came here to post "this article is garbage" but I see I have been beaten to it 3 times by the only 3 other comments here. So.
- Houshalter 11y agoHN needs a downvote button for articles.
- eridius 11y agoOne big complaint I have with MWI is it seems to throw probability out the window. If every possible result of a quantum calculation actually has physical reality, then doesn't that make all the states equally likely? The only solution I can think of to that is to have every quantum calculation actually spawn an infinity of worlds, of which a certain percentage of them have a given result that matches the probability of that result in the wavefunction. And that just seems to make MWI incredibly messy. There's a certain conceptual cleanness in saying that the universe splits for every decision such that all possible decisions have their own world, but saying that every possible quantum event results in an infinity of worlds (which also means that for every possible result there's also an infinity of identical worlds for that state) just seems ridiculous.
- Houshalter 11y agoImagine a ball falling down a peg board. Each time it hits a peg, it has a 50% chance of falling either direction. This creates a Gaussian distribution where it lands. It has a very high probability of landing in the exact center, and slightly less probability of landing to the right, and even less at landing to the far right, etc. See this image: http://ptrow.com/articles/Galton_June_07_files/GaltonBoard2.jpg http://ptrow.com/articles/Galton_June_07_files/GaltonBoard2.... Now imagine instead of probability, you just split universes each time it hits a peg. One universe the ball goes to the left, and in one universe, the ball goes to the right. Each possibility gets it's own universe. Then the same result happens. The majority of universes have the ball land in the middle, and very few universes have it land on the tails. But each possibility does have it's own universe. It's just some things happen more than others. If you pick a random universe, you are more likely to end up in one with the ball in the middle, then on a tail. And this doesn't require an infinite number of universes. It's a completely discrete system.
- eridius 11y agoI appreciate the example here, but this only works when every decision has 2 outcomes with a 50/50 probability distribution (or more generally, when every decision has an even weighting between all results). I don't pretend to have a good understanding of quantum mechanics, but my impression is that this model doesn't match how quantum "decisions" actually work.
- grondilu 11y ago> Here, after all, is a theory that seems to allow everything conceivable to happen. That seems inaccurate. Brian Green once joked about the existence of a universe where Sarah Palin is the president of the United states, ruling this possibility out as incompatible with the laws of physics[1]. I'm not sure why Max Tegmark often states that every decision we make causes a split of the Multiverse, for instance. Do we know for sure that the processes in our brain are quantum and not classical? They can be I guess, but I doubt they are necessarily. 1. https://www.youtube.com/watch?v=NqyBKJL1eyk https://www.youtube.com/watch?v=NqyBKJL1eyk
- RobertoG 11y agoIt seems to me that the author just don't feel comfortable with MWI. He feels threatened by it, in the same way people didn't feel comfortable with Copernic, and just decide that it has not to be right. Doesn't look like a good basis for doing science. In my opinion, MWI is the simplest explanation. In all the others you have to add strange components to make it works. In David Deutsch's "The fabric of reality" there is a good philosophic defense of many worlds. At some point he says something along the lines of: "if in the double-slit experiment photons are interfering with ghost-photons, what it means to say that those particles doesn't exist 'in reality'? "
- dkbrk 11y agoI'm not an expert in this area, so please feel free to correct me if I've made a mistake or am completely and utterly wrong. From what I understand, the Many-Worlds Interpretation is, in effect, the "default" hypothesis, insofar as that it follows naturally from state evolution according to the Schrodinger equation. The burden of proof is on the Copenhagen Interpretation to provide some additional mechanism which destroys the quantum states we don't observe. Take, for example, the double slit experiment with a single photon. If we don't measure which slit it passes through, it has the pure state which we can describe as (|left> + |right>), and you can solve the equations with space and time to see that |left> and |right> overlap after they diffract past the slits, so you get an interference pattern. If you measure which slit it passes through, the photon's states get entanged with the experimental apparatus. For example, |left> is now entanged with the left detector resistering its passing, which causes an electric current to a display, which causes photons to be emitted, which interact with the laboratory, and some of which register on our retinas, which cause our neurons to be in a different state, etc.. So the wavefunction is now: |left, measured-left, brain-saw-apparatus-say-photon-went-left, ...> + |right,...> The photon is now thoroughly entanged with the state of our brain and the entire laboratory, so when the photon diffracts after the slit the two states |left,...> and |right,...> no longer overlap and can't interfere. It's much the same thing as EPR, except rather than (|0> + |1>) getting entangled with a second particle to produce (|0,0> + |1,1>), it gets entangled with all the many degrees of freedom of the laboratory, and our brain, and the entire world. Postulating that if we observe |left, ...> then |right, ...> no longer "exists" is entirely analagous to particle B in EPR "observing" that if it is in the state |0,0> then |1,1> no longer exists.