Hurkyl said:
This sentence is a very good description of what it means for an interpretation to assert that the wave-function to correspond ontologically to a real-world system.
So, I find it quite odd that you suggest you are doing exactly the opposite...
There's a key difference. By saying the wavefunction encodes the information to understand something, I have asserted an epistemological meaning to the wave function. MWI rests on an ontological meaning to the wave function, and that I pretty much view as folly, as it hasn't worked yet in the interpretation of any physical theory.
(but, at least, you do put "to understand" in there -- many people who argue viewpoints like yours do not, or even explicitly decry the idea of using the wave-function to understand reality)
Yes, I do think the "shut up and calculate" camp goes too far in making that assertion. None of them ever really do that anyway! The fact is, we don't just use physics to calculate, we use physics to understand also. But I'm saying, that understanding always stays in our heads-- it never graduates to being something that is actually going on somewhere else. The latter seems to be a key element of the MWI, at least the way it is often used.
(aside: if you don't admit to using quantum mechanics to shape your views about reality and you're telling the truth, then what are you using?)
Quantum mechanics can shape our views about reality without MWI. It certainly shaped Bohr's and Heisenberg's views of reality, did it not?
Ignore the question about "taking something literally". MWI is merely the negative answer to the following two questions:
- Is "collapse" under time evolution part of the "holistic information required to predict and understand a system"?
- Is there any "holistic information required to predict and understand a system" that is not encoded in the wave-function?
I agree that those two questions warrant negative answers, but I see all
interpretations of quantum mechanics (other than deBB) as giving negative answers to those questions. (So I don't like deBB as anything but an exercise in demonstrating the fundamental limitations in either deterministic or "intrinsically random" ontologies. It should be used to show why ontology is impossible, not to argue for determinism.) I can't say exactly how Bohr would have answered them, but here's my guesses:
1) collapse is a concept required for us to connect our macroscopic perceptions of reality with a theory that works on the microworld. It is not part of the holistic information required to understand the system, because it deals with an aspect of the system that we never understand and are simply not equipped to understand.
2) there is no holistic information not included in the wave function, but we must recognize that "information" is on our side of the divide-- it is part of our epistemological task of understanding the system, it is not part of the system itself. So again, the wave function is all about our interface with that system, and the informatics required to achieve a successful interface.
Of course. My point is that collapse-based interpretations encourage improper usage -- indeed the resolution to each of the issues I mentioned boils down to getting the student to stop applying the main method of the interpretation too early.
But look at what MWI encouraged-- quantum suicide.
As I recall, Einstein didn't have problems with entanglement -- he just didn't believe that it corresponded to reality.
I would call that having a problem with it. But we agree-- Einstein felt QM could not be the fundamental truth, even though it worked. MWI says it is the fundamental truth. Bohr says there's no such thing as the fundamental truth, at least not anything we have access to.
No -- it collapse is a provably (essentially) impossible result of the unitary evolution of wave-functions.
And there's the problem-- it is "provable." Truth is never provable, what is provable is logical connection to a set of axioms. So yes, unitary evolution is logically connected to a set of axioms that are used to evolve a wave function. The connection to reality is what we are talking about-- the
ontology of the axioms. You can only argue that because they work, the axioms must be the correct ontology. To that I would raise two objections:
1) there is no logical connection between a working axiom and a true ontology, and
2) the history of physics delivers a harsh assessment of the argument that good postulates are true axioms.
MWI has decoherence, and the analytical method of separating a mixed state into a linear combination of states.
True, but it has no idea how to connect those mixed states with what we actually observe in individual cases. So it must either invoke a concept of ontological randomness, which is bogus because randomness is an effective model not an ontology, or it must come to grips with what consciousness is, which we are presently ill-equipped to do (again I mention quantum suicide).
If a wave-function decoheres and doesn't subsequently collapse, it's not a Copenhagen-esque approach.
I guess I don't understand what you mean here-- collapse in Copenhagen is not the evolution of the wave function, it is the manual re-assessment of the wave function based on new information. Copenhagen sees a wave function as an expression of knowledge, with no ontological status at all, so it seems the wave function as evolving unitarily when there is no change in information status, but in other ways when we do have new information (like the outcome of an experiment). Thus, when I say Copenhagen is fine with decoherence, I mean that Copenhagen applies decoherence in the opposite direction as MWI-- MWI says the wavefunction collapse occurred after the decoherence (either by some random selection agent or some nonexistent model of how consciousness is being generated in the various branches), and Copenhagen says decoherence is what you get when you apply the new information that is being acquired in a gradual rather than sudden way (taking the perceptions of the consciousness as the fundamental reality and forcing all information to be consistent with it).
You misinterpreted me -- I meant "If there are two physicists analyzing a quantum system, you won't be able to tell which one is using MWI and which one is using CI". (except, of course, on the off chance one starts talking about the system actually collapsing)
I agree with your quoted remark, but I don't see how it adjudicates the options here. All interpretations are fine with that statement, it's essentially the reason these are interpretations and not theories.
Your rephrasing looks like an entirely different point. My point is "We have our experiences with how reality behaves. Can Quantum Mechanics reproduce that behavior?"
I was saying what I view as the source of "quantum weirdness," which is a bit different from what you see as its source. That may well be why we feel differently about MWI! (I don't really favor any interpretation, by the way, I see them all as inevitably flawed and would prefer to pick and choose from among them to make any particular point as needed. But I prefer the Bohr approach when it is used to foster skepticism about any ontological interpretation as being "the truth" of the matter.)
It is not weird that quantum mechanics can reproduce our experiences, what is weird is that the way it does it is to deviate so strongly from our common interpretation of our experiences. This should be a lesson to us about the flaws in all kinds of interpretations, rather than as evidence that a new one has replaced an old one! That's the key point I'm stressing-- the lesson of history should not be "the old was wrong, now we have it right", it should be "everybody always thought that, from Aristotle to Newton to Einstein to Weinberg-- isn't it time we got the real message, which is that all our theories and their interpretations have the ways in which they work for us, and the ways they fail us."
It's at least partially a rather serious and important point -- some interpretations are wholly incompatible with the idea that quantum mechanics could apply in those domains, and leads people to scoff at (or even blatantly ridicule) attempts by others to consider the topic.
I am sympathetic to your point here that the "cat paradox" and "Wigner's friend" are often misinterpreted, leading people to dismiss macroscopic extensions of quantum mechanics. I think the problem is simply that the important distinctions between epistemology and ontology are not enforced, but MWI makes that same mistake by suggesting an ontology when all it can really support is an epistemological understanding of the quantum domain. When interpretations are predicated with "to understand what this theory is telling us about our reality, although the theory never actually asserts this is true, you can imagine that...". This is along the lines of saying that you can understand what GR is telling us about the history of the universe by "imagining that space itself is expanding", without ever actually asserting that space itself actually is expanding, which is probably pretty close to nonsense. (The next theory might give some actual meaning to that claim, or it might expose it as nothing but a handy mnemonic, such is the history of physics.)
The bottom line is, physics theories are epistemological entities that generate ontological constructs along the way, but the theories are never judged by those constructs, they are judged by their epistemological successes. What's more, the ontological entities generated along the way are often not unique within a given theory (witness Newton's use of "forces", and Maupertuis's use of "action", as one common example). No one argues over whether there really are such things as forces when we teach all our undergraduates that there are-- this is simply because we are lazy about our distinctions between what is an epistemological success and what is an ontological claim on reality. That crucial imprecision is at the heart of all the debates about quantum mechanical interpretations, in my view.
