Comment by ThePhysicist | Hacker News Reader

Comment by ThePhysicist | Hacker News Reader

Parent Full thread

The article is of course oversimplifying, but if I understand correctly the irreversibility in the qubit system comes from the fact that they assume you don't know the states of the qubit chain that transforms the first qubit from B_1 to B_2. If you'd know that the state would be pure and it would be easy to reverse the entanglement, it's just that if you look at the state of the first qubit alone that it's mixed.

I actually envisioned a similar thought experiment that could be done with a very complex quantum computer (say millions of qubits) to test whether the universe is actually keeping tabs of all these states or if it gives up at some point and produces irreversibility. But I guess if you'd formulate it like that the academic community wouldn't take you very seriously anymore. I'd wager Deutsch's actual motivation might be along this line, as he seemed very interested to find a way to test the many worlds interpretation against other interpretations of quantum mechanics.

Deutsch's position is actually that the many worlds interpretation has already been tested and has passed the tests so far, whereas rival theories have been refuted.

See https://www.sciencedirect.com/science/article/pii/S135521981...

Deutsch is wrong.

http://blog.rongarret.info/2019/07/the-trouble-with-many-wor...

To summarize the argument that the blog post makes:

The multiverse interpretation accounts for cases where the probabilities of each distinct observable outcome are equal according to the Born rule. But it cannot account for cases where the Born rule says there are non-equal probabilities for the possible outcomes.

This is a fairly common problem which critics of the multiverse interpretation raise. See [1] for another example where a critic raises that problem in more detail (with examples).

Deutsch and others have responded. In short, the reply is that the mathematics of the Schrodinger equation describes a density of universes in the continuum which is the multiverse, not a finite set of distinct, interacting single universes. This undercuts the critics' argument. See [2] for an example of this kind of response.

[1] https://arxiv.org/pdf/0810.2657.pdf [2] https://www.pnas.org/doi/pdf/10.1073/pnas.1324238111

FYI, I am the author of the blog post.

> This undercuts the critics' argument.

No, it doesn't, for all of the reasons given in the blog post and the subsequent discussion. The Born rule is an observable fact of nature, and you cannot derive it from the Schroedinger equation, Duetsch and Wallace's claims to the contrary notwithstanding.

Maybe you can help me understand your argument better. If the Schrodinger equation describes a density of universes, doesn't that undercut the argument in the blog post?

Edited to add my understanding: if the cardinality of the multiverse is uncountable, one has to talk about densities of branches. In that scheme, the problems which branch counting introduce do not arise.

> If the Schrodinger equation describes a density of universes,

It doesn't (see below) but even if it did...

> doesn't that undercut the argument in the blog post?

No. It's really very simple: the SE is linear. The BR is not. You can't derive a non-linear result from a linear equation.

The only thing that is even a tiny bit complicated about this is figuring out where the Deutsch-Wallace argument fails, because Deutsch and Wallace are not idiots and they hid their questionable assumption very well.

So... why does the SE not describe a "density of universes"? It's because the SE does not describe universes at all! Deutsch and Wallace specifically acknowledge this, which makes their argument all that much more untenable. The SE describes the behavior of a function called "the wave function". The wave function is not a description of anything physical. The domain of the wave function is not physical space, it is configuration space. This is the reason entanglement seems weird to us, because we live in physical space and the wave function does not.

In order to get anything physical out of the wave function -- particles, universes, anything resembling our classical reality -- you have to add some additional assumption, and any additional assumption that you add will be in some sense isomorphic to the Born rule. There is no way around this. Trying to derive the BR from the SE is like trying to prove Euclid's fifth postulate.

> The wave function is not a description of anything physical. The domain of the wave function is not physical space, it is configuration space. This is the reason entanglement seems weird to us, because we live in physical space and the wave function does not.

This is circular: you're assuming that realism about the wave function is incorrect to argue against the (tragically misnamed) many-worlds interpretation, but the only substantive claim the MWI makes is that realism about the wave function is true.

> you're assuming that realism about the wave function is incorrect

No, I'm not assuming it. I simply observe that the SR by itself does not describe the outcomes of experiments. And I'm not even talking here about quantum experiments. The SR cannot by itself even account for the fact that we live in a three-dimensional space or that there is an arrow of time. And it certainly cannot account for the Born rule, which is also an experimentally observed fact.

> the only substantive claim the MWI makes is that realism about the wave function is true.

Yes, I know. The problem with this is that it is violently at odds with observations that every human makes every moment of their lives, and it does not account for the Born rule.

Other than that there's no problem with it.

> I simply observe that the SR by itself does not describe the outcomes of experiments.

No formalism describes the outcome of experiments by itself. There always needs to be an interpretation attached, and our choice of interpretation is always underdetermined by observation.

> The problem with this is that it is violently at odds with observations that every human makes every moment of their lives

Like all other interpretations of all other physical theories, MWI isn't and can't be at odds with our observations alone, because our observations alone aren't enough to imply any particular relationship between observations in general. And like all other interpretations, it is and must be at odds with competing interpretations, and in particular with the folk-physics model we're naturally inclined to use. But this means very little, because folk-physics is wrong, and this far from the paleolithic savanna, it's wrong in ways that matter. Intuition is evidence, but it's not particularly strong evidence, and it's certainly not proof.

> and it does not account for the Born rule.

Sure, it still needs a separate account of subjective probability. But so does everything else: it's just a little harder to notice in the classical regime, since the relevant probabilities are always either 1 or 0.

> No formalism describes the outcome of experiments by itself.

That is so wrong as to border on the ridiculous. The quantum mechanics formalism, i.e. the SE plus the BR together describe the outcomes of experiments. No interpretation is required. You just turn the crank on the math and predictions pop out. There is a reason that "shut up and calculate" is a thing.

The only reason people talk about interpretations at all is that the structure of the math is violently at odds with our intuitions and subjective experience, and this leaves some people feeling queasy and wanting an explanation. There's nothing wrong with that. But it has absolutely nothing to do with the power of the formalism to make predictions, which is completely independent of any interpretation. (And this is true of classical mechanics as well.)

Ultimately the MWI is saying “believe the math - configuration space is physical reality”. So you are on solid ground when you give arguments that are mathematical, however the metaphysical stuff about the self is basically talking past the MWI, which specifically does not acknowledge that as valid form of argument. I also agree with the post above that arguments about “branch counting” are bunk because you cannot count in a continuous space (see: banach-tarski and related paradoxes).

No, the MWI says to believe half of the math and to ignore the other half. The problem with that is that the second half is what allows the first half to make experimental predictions. Without the second half, there is no reason to believe the first half.

Your reference [2] is somewhat suspect. As far as I know, it is not generally accepted in the physics community that the Born rule can be derived directly from the Schrodinger equation. In fact, it seems generally accepted that such a derivation does not currently exist, and MWI has to stick it in as an additional postulate in some form (or otherwise give an auxiliary argument for it, none of which seem generally accepted either).

Further, the author of that paper, Frank Tipler, is a notorious nutjob: https://en.wikipedia.org/wiki/Frank_J._Tipler.

I don't have time at the moment to review the paper in detail to locate any potential flaws, but I wanted to flag this for others. I think there is substantial reason to be skeptical.

I agree that Tipler is far off the deep end on on some topics. But you should check out that one paper I linked nevertheless. PNAS is not a junk journal.

I also agree there is no consensus that the problem of getting Born's rule from the Schrodinger equation has been solved to everyone's satisfaction.

PNAS publishes a lot of junk. If you're connected to the Academy you can get papers through with very light review. In fact, one can even seen this happened in the Tipler paper: "This Direct Submission article had a prearranged editor." [0]

Andrew Gelman calls it a "tabloid": https://statmodeling.stat.columbia.edu/2014/06/17/hurricanes....

I will check the paper out though.

[0] https://www.science.org/content/blog-post/no-more-prearrange...

Edit: I read the paper. It is borderline crank work. There are decent arguments for the Born rule in MWI but this is not one of them. (I like David Wallace's book, for example.)

Care to elucidate some critiques of the paper?

Care to outline a precise argument for me to respond to? I'm honestly not sure one can even be extracted from pages 3 through 5, and I don't want to attack a strawman.

One basic problem is that using symmetry arguments to justify the assignment of subjective probabilities (cf. page 4, right column) is going to fail when there isn't a symmetry available, for example for irrational probabilities.

A deeper issue is the implicit assumption that one can coarse-grain into discrete "effectively distinguishable universes" for the purpose of doing these counting arguments. There are significant conceptual difficulties doing this, and this claim is much stronger than the bare MWI. Wallace discusses this at length; see for instance the citations in the blog post you responded to.

Perhaps it occupies the gray area between "wrong" and "not even...". O:-)

Reply on news.ycombinator.com