On Mon, Dec 2, 2019 at 8:08 PM Bruno Marchal <marc...@ulb.ac.be> wrote:
> On 29 Nov 2019, at 00:50, Bruce Kellett <bhkellet...@gmail.com> wrote: > > On Fri, Nov 29, 2019 at 1:27 AM Bruno Marchal <marc...@ulb.ac.be> wrote: > >> On 26 Nov 2019, at 22:39, Bruce Kellett <bhkellet...@gmail.com> wrote: >> >> On Wed, Nov 27, 2019 at 12:27 AM Bruno Marchal <marc...@ulb.ac.be> wrote: >> >>> On 25 Nov 2019, at 22:53, Bruce Kellett <bhkellet...@gmail.com> wrote: >>> >>> Because, the wave-function itself is non-local -- it contains entangled >>> particles that are widely separated in space. That is the definition of >>> non-locality! >>> >>> >>> I am not sure. I use “non-locality” for “FTL physical influence”. >>> >> >> That is just an abuse of language. Non-local means "not local", i.e., not >> all in one place. >> >> >> Then even Newton Universe is non local. >> > > Yes Newton was aware of this. > > > >> Some attempt has been made to replace the term "non-local" with the term >> "non-seperable”. >> >> >> Yes, notably d’Espagnat. It avoids the confusion with the Eisnsteinian >> non-locality, which requires FTL (cf the “spooky action at a distance”), >> which must exist in QM + the assumption of a unique universe. >> >> I think we can all agree that the singlet wave function is non-separable >> -- it cannot be written as a simple product of two terms, one referring to >> each particle. >> >> >> Yes, we agree on this. >> >> I maintain that it is also non-local, in that the two particles are at >> different locations (locales). Non-local can have no other meaning in >> ordinary linguistic usage. >> >> >> I invite you, and Alice, and I give you an envelop to each of you. You >> are told that one contain a piece of paper with O inscribed on it, and the >> other with one. Then you go in different galaxies, say, and open it. Once >> you see 0 (res. 1) you know that Alice will see 1 (res. 0). This seems non >> local in your sense, where most would agree that in this case, there is no >> “non-locality” issue. What I claim is that in the Everett theory, all >> non-locality are of that type. >> > > That non-loclality has a common cause explanation. Like Bertlmann's socks, > there is no mystery here. The problem is with entangled systems, where > non-separability means non-locality that has no common cause explanation, > even in many-worlds theory. > > > I doubt this. The MWI reduces the non-separability of the probabilities > into an equivalent with Bertlmann’s socks, still keeping the violation of > Bell’s inequality justifying the appearance of non-locality. > The devil is in the detail. And you have still not provided any detail. In the MWI, some particles can be entangled but without implying any >>> possible FTL when we do measurement on them, except from the local point of >>> view, due to our ignorance of all terms of the wave. It means simply that >>> Alice and Bob belongs to the same branch of history/reality. >>> >> >> The trouble with this hope is that it no local account of the EPR >> correlations been realised in any coherent mathematics. Bell's theorem >> rules it out: no local hidden variable account of the EPR correlations is >> possible in any theory, whatsoever. It is a no-go theorem; it proves a >> negative -- something is impossible. Many-worlds does not subvert Bell's >> theorem. >> >> >> That is right. But the violation of Bell’s inequality entails FTL only >> when one world is assumed, with well defined outcome for all measurement, >> or put in another way, assuming a unique reality, with one Bob and one >> Alice, but Bell’s reasoning does not prove FTL influence in The >> many-worlds, where all outcomes are obtained, and propagate between diverse >> Alice and Bob locally, leading to the apparent violation of Bell’s >> inequality, but without FTL. >> > > Bell did not assume a collapse. His is a mathematical result, where the > only assumption is locality. As usual, if you think there is a local > explanation of the EPR correlations in many-worlds, then produce it. > > > We differ only on the way we interpreted the wave and the worlds. The > singlet state is … local! It does not entail any correlation between the > Alices and the Bobs. It enforces only that the Alices and Bobs can meet > only their corresponding correlated partners, among the infinitely many > Alices and Bobs (most of them being not accessible from each others). > The singlet state is non-separable, and hence non-local when Alice and Bob are separated. The rest of you comment here is without meaning. You have to flesh it out, and your reluctance to do so convinces me that you cannot. You are just hoping for a miracle. > I think it is becoming generally accepted in the physics community that >> the entangled state is intrinsically non-local: acting on one part of it >> affects the rest, even across the entire universe. >> >> >> That would mean some FTL actions, but I very much doubt this. >> > > No, there is no need of FTL. For example, in the third (2011) edition of > his book 'Quantum Non-Locality and Relativity', Maudlin shows that Flash > GRW theory, as developed by Temulka, gives a perfectly relativistic account > of the EPR correlations without any FTL action. > > > This astonishes me. If you have a link I could try to see if this makes > sense, but, to be sure, I am not enthusiast at all on the GRW theory, which > is new QM theory. If some measurement affects the rest of the Universe > instantaneously, I think that imply FTL (without signals, but still with a > physical influence). > I only know this from Maudlin's books. There is a detailed account in his recent book, "Philosophy of Physics: Quantum Theory" (Princeton, 2019). The paper that Phil referenced: Esfeld and Gisin (arXiv:1310.5308) contains a summary and discussion, even though those authors are not convinced by the theory. I must admit that Flash GRW is not an approach that I find convincing either. The ontology of this theory is the flashes themselves, the wave-function is not part of the ontology, so the collapse is entirely epistemic, and just like the "collapse" in classical probability theory. It works, but may not be convincing to everyone, since the ontology is remarkable thin! The real problem is, as ever, that despite many promises, many-worlds theory does even less well in that it offers nothing in the way of an explanation. Maudlin's recent book discusses many worlds, but he concentrates on the problems with this idea, and essentially is not convinced that many-worlds makes sufficient sense in its own terms to provide an explanation for anything. Sean Carroll, in his recent book "Something Deeply Hidden", also evades the question. His discussion is on pages 104-105. He states that Bell assumed "that measurements have definite outcomes". Essentially, this is the assumption of counterfactual definiteness that has often been proposed. Carroll then says that since many-worlds does not assume that experiments have single outcomes, Bell's theorem doesn't apply. His unspoken assumption here is that if many-worlds evades Bell's result, then it can give a local account of the EPR correlations -- they happen "because of branching of the wave function into different worlds, in which correlated things happen." (p. 105) This is similar to your claim above, but it is not an explanation. And even if Bell's theorem doesn't apply, it does not follow that the theory can provide a local account of the correlations.You still have to provide that account. The other major source I can refer to is the book "The Emergent Multiverse" by David Wallace (Oxford, 2012). This book is the most comprehensive account of Everettian ideas currently available. Wallace also ducks the issue. In sections 8.5-8.7 (pages 302-312) he gives a detailed account of the branching structure that arises if Alice and Bob do independent spin measurements at space-like separations. Then on page 310 he gives a general entangled wave-function and points out that Alice's and Bob's measurements again lead to splitting. But he then says: "In this case, the amplitudes of the four sets of branches into which C [a central, neutral observer] eventually branches are not determined simply by the separate weights of the branchings at A and B. Nor is this to be expected: as I stressed previously, in Everettian quantum mechanics interactions are local but states are nonlocal. The entanglement between the particle at A and the particle at B is a nonlocal property of the forward light cone of A and that of B. Only in their intersection can it have locally determinable effects---and it does, giving rise to the branch weights which, in turn, give rise to the sorts of statistical results recorded in Aspect's experiments." This might sound good, but again, there is no detail. What exactly is supposed to happen at the intersection of the forward light cones from A and B? There is no interaction there -- any information that is made locally present there was already present in the only relevant interactions, which are the original measurements made by A and B; any branch weights that are around are set there, exactly as in the case of non-entangled particles. Wallace started out well, but ducked out at the last minute, and he failed to give any comprehensible account that does not rely on simple magic. So the best authorities available fail to give a local account of the EPR correlations in a many-worlds setting -- they all simply duck the issue when the rubber hits the road. Just as you routinely do. Bruce -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAFxXSLQc-U9O%2BN70fUjxtj%2B-94Nbf7x5hMQeLnL4SN%3Dmn__fLw%40mail.gmail.com.
