Sean Carroll has posted a passionate defense of the Many-world interpretation to quantum mechanics.
I have often talked about the Many-Worlds or Everett approach to quantum mechanics — here’s an explanatory video, an excerpt from From Eternity to Here, and slides from a talk. But I don’t think I’ve ever explained as persuasively as possible why I think it’s the right approach. So that’s what I’m going to try to do here. Although to be honest right off the bat, I’m actually going to tackle a slightly easier problem: explaining why the many-worlds approach is not completely insane, and indeed quite natural. The harder part is explaining why it actually works, which I’ll get to in another post.
Carroll’s description is well done, and I recommend reading the full post. My only concern is his characterization of the Many-worlds interpretation as inevitable. As I’ve written here before, I personally view Many-worlds interpretation as a candidate for reality, but I remain unconvinced that we’ve reached the point where we can move it from candidate to settled.
The conclusion, therefore, is that multiple worlds automatically occur in quantum mechanics. They are an inevitable part of the formalism. The only remaining question is: what are you going to do about it? There are three popular strategies on the market: anger, denial, and acceptance.
Or we can simply admit that we don’t yet have unique evidence for any of the interpretations. No anger or denial necessary, yet.
There are other silly objections to EQM, of course. The most popular is probably the complaint that it’s not falsifiable. That truly makes no sense. It’s trivial to falsify EQM — just do an experiment that violates the Schrödinger equation or the principle of superposition, which are the only things the theory assumes.
I think spreading superposition that we can’t observe, is an assumption. True, if we falsify the Schrodinger equation or superposition, we’ve falsified the Many-worlds interpretation, but haven’t we also falsified every other interpretation? To consider Many-worlds falsifiable, don’t we need to be able to, at least in principle, uniquely falsify it?
Occam’s razor often surfaces in these discussions. But Occam’s razor seems like a tough call on these interpretations. Is mathematical parsimony the same thing as ontological parsimony? I’m not sure of the answer, but dismissing the concern as silly seems unjustified.
A part of me wonders if there’s any real harm in people concluding that the Many-world interpretation is true, since its truth or falsity seems to have no bearing on the rest of the world. But this comes back to what should count as settled science, and I can’t see how a concept that is not testable in any foreseeable manner should count. And, as other physicists have said, there is a danger of accepting this interpretation and prematurely ceasing to look for what might be the real explanation, one that might turn out to open new doors.
SelfAwarePatterns 
“My only concern is his characterization of the Many-worlds interpretation as inevitable. As I’ve written here before, I personally view Many-worlds interpretation as a candidate for reality, but I remain unconvinced that we’ve reached the point where we can move it from candidate to settled.”
Agree 100%. But, I would like to go one step further.
Interpretation by definition does not alter the validity of the subject topic, that is, the issue of right or wrong of an interpretation is not superbly important. But, there are good, bad, better and the best interpretations. Only when one interpretation becomes the ‘gateway’ to new physics, it will become a better or the best interpretation.
Currently, there are many different interpretations for this quantum-measurement issue. As this post is about discussing Sean Carroll’s view, I will thus talk about only two interpretations: Copenhagen interpretation (CI) and Many worlds interpretation (MWI).
First, what are the essences of these two interpretations?
One, Copenhagen interpretation (CI): this is in fact a ‘two-worlds’ interpretation. That is, there are two different worlds: the classic world and the quantum world. When this two world collides, the quantum world ‘collapses’. CI admits that these two worlds are not unified.
Two, Many worlds interpretation (MWI): this is in fact a ‘single-world’ interpretation. In addition to the fact that quantum particle is in a quantum world, the apparatus which does the measurement is also sitting inside of this quantum world (also in a superposition state). That is, both parts of the measurement sit in the same world, and this is a view that classic/quantum are unified. But, without giving an actual ‘unification’ mechanism, this is a ‘faked’ unification, without beef. The fact is that this MWI is not a ‘gateway’ to a new physics. Although Sean is advocating the Multiverse, the MWI has nothing to do the Multiverse and cannot provide any help for it.
The MWI is a great attempt to provide a classic/quantum unification image, but no cigar. Hilary Putnam (a prominent philosopher) discussed this quantum-measurement issue at his blog. One view is a true ‘single-world’ interpretation. That is, the electron (quantum particle) is as a solid marble similar to the apparatus, while both of them sit in a ‘single-quantum-world’ (see, http://putnamphil.blogspot.com/2014/05/the-measurement-problem-in-qm.html?showComment=1402087925047#c6965420756934878751 ).
Is this new interpretation any better than the two above? This can be decided by a ‘beauty-contest’. Is this new interpretation a ‘gateway’ to new physics? Can it provide answers for any known open-questions of today in physics?
LikeLike
Thanks Tienzen! I fear most of that discussion at Putnam’s blog was hopelessly above my head, but I didn’t realize he had started a blog. Just subscribed to it.
I agree that some interpretations are better than others, however I tend to think the obviously bad ones have fallen by the wayside over time. The ones we still talk about are the ones that are hard to dismiss.
I do agree with Putnam that settling on one interpretation is premature. Until we have evidence unique to one interpretation, physicists should continue thinking about this and looking for ways to test these interpretations. Until then, which interpretation you prefer is a philosophical conclusion.
LikeLike
I observe that he writes down equations that talk about “multiple states” and then after a few paragraphs starts using the phrase “multiple worlds” and pretends not to notice what he’s done.
LikeLiked by 1 person
SelfAwarePatterns: “… and I can’t see how a concept that is not testable in any foreseeable manner should count.”
I must disagree with you on this one.
In recent years, many people try to go around the falsifiability issue or simply abandon it for variety of reasons, such as, the failure of M-string theory, the failure of SUSY and the failure of multiverses. I strongly disagreed with their cases because of their using the wrong arguments.
Falsifiability ‘was’ a great ‘tool’ for physics for over 400 years, but it is wrong in ‘principle’ as the ‘true truth’ cannot be falsified by ‘definition’. In order to overcome this ‘definition’ issue, anti-realism arose from two directions (pathways).
School one, from the notion that ‘final truth’ is all elusive to the conclusion that the ‘final truth’ is a non-reality. This school is based on the falsifiability ‘principle’. If a thing is not testable, it is not a reality. The fallacy of this argument is taking the falsifiability as a ‘principle’ while the ‘falsifiability’ is the issue of being inquired; that is, proven the issue with the issue in question. This school calls themselves as ‘model builders’. They are not searching for ‘truth’, and that all they do is building the models. (see http://www.quantumdiaries.org/2014/02/07/interpretations-of-quantum-mechanics/#comment-172367 ).
School two, with a BIV (brain in a vat) argument, this is a much more powerful argument. BIV attempts to crash the long established solid foundation of “cogito, ergo sum”. This BIV attempts to create a ‘mystery space (MS)’. And in this MS, no objective knowledge can be attained. Thus, if no objective knowledge is attainable, then a reality (if any) is meaningless.
This BIV argument can be voided with two steps.
1. To show that everything in a (any) MS is totally knowable if there is ‘one’ external point outside of that MS.
2. To show that there is no totally ‘isolated’ MS (without an ‘external’ point) in ‘this’ universe.
The school one (the model only school) can be voided by showing some ‘solid’ good physics which are not the results of falsifiability-tool. Facing with this kind of solid ‘examples’, Hilary Putnam (a prominent philosopher on philosophy of science himself) again quoted from Quine’s saying {much good science is untestable}, (see, http://putnamphil.blogspot.com/2014/06/a-final-post-for-now-on-whether-quine.html?showComment=1403375810880#c249913231636084948 ).
While I disagree with some wrong reasons for abandoning the falsifiability-tool, I must say that the falsifiability is wrong in ‘principle’ as a check for the true-truth. While I agree with the fact that falsifiability-tool has done a superb job in the past 400 years in physics, it is no longer useful when we have gained the solid knowledge to such height in physics as we have now. We now can do a lot not-testable good physics. I am saying this, not as a principle nor as a philosophy. Many solid examples are on the table for everyone to see now.
The wind direction has changed. The tide has changed.
LikeLike
Tienzen, I actually reject both of the schools you describe. For falsifiability, I agree that it’s far too strong a statement to say that if it’s not falsifiable, it’s not reality. But I think it’s entirely reasonable to say, if it’s not falsifiable, we can’t know whether or not it is reality.
Put aside the falsifiability language for a moment and think of it this way. What about the observable world would be different between the theory being true and it being false? If the answer is nothing, then if it is false and we think it true, how would we ever discover we were wrong?
We can’t build any technology on top of such a theory. (Success or failure of such a technology would provide falsifiability.) If we considered the theory settled science, we could attempt to add other theories on top of it, but unless those theories made a difference in the observable world, we’d simply be building a house of cards that never leads us to useful knowledge. It is essentially a dead end.
I’m not saying that every theory should be required to state how it could be falsified in order to be explored. Speculation is too crucial a component of scientific discovery to hobble it with that. But I am saying that a theory should make some observable difference in the world before we consider it settled science.
One last point, falsifiability is a different criteria than verification (which is what Putnam was discussing). The first only requires that it be possible, in principle, to demonstrate that a theory is wrong. The second requires positive empirical evidence for the theory. Karl Popper rejected verification as far too stringent because it excluded too much valid science. (Technically, given the problem of induction, no scientific theory really meets verification.) Popper proposed falsifiability instead.
LikeLike
SelfAwarePatterns: “Tienzen, I actually reject both of the schools you describe. …”
This is the most important first step. The ‘model building’ school was the most powerful one for the past 50 years. Fortunately, it has lost some of its previous glory now.
“… how would we ever discover we were wrong? …”
This is truly the most important question that every physicist should ask. This is so big an issue and can never be wholly discussed in a short comment. But, there are some prerequisite issues which must be addressed first. The following three can be the beginning of a short list.
One, totally understand the subjective/objective issue (SOI), as this is very important to know the difference between the ‘physics of nature’ and the ‘physics of human’. Yet, this SOI itself can be analyzed ‘philosophically’.
Two, do we now have some (or any) ‘established’ knowledge (such as the Standard Model, although not complete)? Are they able to become anchors as the check points for any new theory?
Question: when a framework (I prefer not to use the word ‘theory’) makes contacts to ‘all’ known anchors (whatever they are, per our consensus), why is it not ‘make a difference in the observable world’ while no other framework is able to make such a contact?
Three, the nature universe is a given, and by now we should have some knowledge about it (the anchors). If we can ‘design’ a universe and derive a set of laws for this ‘designed-universe’, can we make a beauty-contest between these two (nature universe vs designed-universe)? Can we get some knowledge from this kind of beauty-contest?
These are just a few quick ideas of ‘how would we ever discover we were wrong?’ If a framework cannot meet the simple requisites above, it is obviously ‘wrong’. For example, the M-string theory has failed to meet its mission of ‘string-unification’, and thus it cannot be a right framework for describing the ‘nature’ (although it can be something great about some mathematical structures).
This is a big issue and will not be resolved in a comment. I will put it aside for now but will definitely revisit it.
LikeLike
Thanks Tienzen. It is indeed a big issue. Whole books have been written on it. I think falsifiability remains a useful criteria for settled science, but like so much in this arena, judging whether or not a notion meets it isn’t always easy.
LikeLike