Revisiting Aspect’s third experiment

I often wonder how many people who work in quantum foundations have actually read Aspect’s papers describing his famous experiments, particularly his third.  In the ensuing years it seems to me that we have tried to bury the implications of non-locality in this paper beneath layers and layers of often impenetrable mathematics and philosophical hyperbole.  As Aspect says in the paper,

In this experiment, switching between the two channels occurs about each 10 ns. Since this delay, as well as the lifetime of the intermediate level of the cascade (5 ns), is small compared to L/c (40 ns), a detection event on one side and the corresponding change of orientation on the other side are separated by a spacelike interval.

The italics are mine. The key word here, to me, is corresponding. If the detection event on the one side and the change in orientation on the other are connected in some way and they are spacelike separated, I fail to see how we can so blithely rule out a violation of Einstein causality.  While you may beg to differ that it is not “blithe,” it appears as such to an empiricist.  If you insist on talking about abstractions like C*-algebras, then you’re opening another can of worms and asking me to believe in the physical existence of these algebras.  In other words, the more abstract we get, the more we really need to decide whether these mathematical “objects” truly are real or not.  Otherwise, it just looks like a lot of overblown rhetoric to me.

Advertisements

12 Responses to “Revisiting Aspect’s third experiment”

  1. ” layers of often impenetrable mathematics”
    Huh? I thought you were the one advocating for the use of category theory to understand quantum computing. What happened? Did they kick you out of bed for eating crackers?

  2. quantummoxie Says:

    I personally believe that there is a way to use category theory that is not impenetrable, though this method is not often employed by category theorists. It is best outlined in the first part of this paper by Jamie Vicary.

  3. I suppose it would be a bit pedantic of me to note that none of Aspect’s experiments closed the detection loophole.

    In any case, I don’t think that the possibility of a violation of Einstein causality has been “blithely ruled out”. It is pretty well accepted amongst people who advocate realist interpretations of quantum theory (and as I’ve said before, if you’re not looking for a realist interpretation then I’m not sure why you have a problem with QM interpretation in the first place). For example, read Tim Maudlin’s book, which is pretty much the bible of nonlocality on the philosophy side of things. He is extremely clear on just this point. There is also the fact that it is called “nonlocality”, which is pretty strong admission of this point.

    You know, I think the problem is that you go to conferences where there are a lot of physicists and qinfo people who are a bit wishy-washy on the meaning of nonlocality. Go to a serious foundations meeting, which has philosophers and everything, e.g. the New Directions conferences in Maryland, and you’ll find plenty of people who get this point.

  4. quantummoxie Says:

    LOL, Matt, you may be right. I wrote this after John Baez made some comment to me about this on the nForum (and, don’t get me wrong, John’s a nice guy).

    I would love to go to New Directions or other such conferences (and would prefer them over QInfo ones), but they all seem to be invitation-only and no one ever invites me (though Chris Fuchs did say he’d ask Jeff Bub to add me to the list for next year).

  5. I have had two extended discussions with experimental physicists, regarding experiments that observed quantum “jumps”. The results were as follows.

    Dan Rugar wanted to know “How the the Stern-Gerlach effect really work?” This question arises in the context of MRFM experiments that continuously observe the orientation of a single electron spin. The practical stakes are considerable, since classical and quantum models of the signal-to-noise ratio is the IBM experiments very different results for these two cases.

    In the IBM case, the experimental data turn out to be consistent with quantum orthodoxy. And that quantum orthodoxy can be expressed via three different unravelings: (1) the IBM spins jump discretely (batrachian unraveling), (2) the IBM spins diffuse in state-space (ergodic unraveling), (3) the IBM spins are dragged by backaction (synoptic unraveling).

    So there is no sense asking whether the IBM spins “really” jump as opposed to “really” diffuse — that’s not a well-posed experimental question. In particular, if Alice wants to unravel spin-measuring experiment in terms of quantum jumps, that’s cool, but Bob is never obligated, by any amount of Alice’s data, to unravel an spin-measuring experiment uniquely in terms of jumps.

    The reason for Bob’s skepticism has a paradoxical aspect: as the SNR of Alice’s spin observation data is made higher-and-higher, seeming to show the reality of quantum jumps more-and-more clearly, Bob becomes steadily less convinced in the reality of quantum jumps. The reason is that in Bob’s non-jump unraveling, Alice’s increasing SNR is creating stronger backaction effects … so Bob explains the jumps in Alice’s data as experimental artifacts that are being induced by Alice’s strongly-coupled spin measurement process!

    Now, an MRFM cantilever can be viewed as a bosonic mode, which is coupled to a two-state spin that can be viewed as an atom. So all the preceding formalism applies equally to photons and optical cavities.

    So in bench-top optical experiments, are photons really emitted discretely? Precisely as in the IBM MRFM experiments, that is not a question that has a unique answer. Rather, jump and non-jump unravellings will always yield the same answer … after what can be a very long calculation.

    The reason the calculations are long is purely technical. Whenever precision timing is attempted, and high-efficiency detection is specified, then no matter whatever dynamical unraveling is chosen, an in-depth calculation of cavity QED effects always is required.

    I discussed this point at some length with John Cramer, in the context of his on-going retro-causality experiments. John took the very reasonable point-of-view, that it made sense to take measurements first, and then do the in-depth cavity QED calculations.

    At the end of the day, it is reasonable to embrace the ultra-orthodox view that cavity QED is very likely right, and if so, John Cramer’s experiments—no matter how ingeniously configured—will not observe retro-causal signaling.

    But still, Cramer’s experiments are well worth doing … because at a minimum, they help us gain deeper understanding of the mathematical aspects of quantum unraveling.

  6. quantummoxie Says:

    Interesting. Well, I’m not saying that causality is necessarily dead. I guess I’m just loathe to commit to any, one thing (e.g. that it’s the only thing). Plus, we may be limited in our ability to even answer this question definitively.

  7. LOL … here’s what our QSE group calls “a confection”.

    Suppose we arrange for Schrödinger’s Cat to do John Cramer’s retrograde causality experiment … the whole apparatus being inside his cat-box … and we ourselves do *not* watch the Cat during the experiment … but only examine the Cat’s lab-book later on … hmmm.

    Wouldn’t it be perfectly consistent for the Cat to observe (and record in the Cat’s quantum notebook) retrograde causality? Provided of course that the Cat does nothing to exploit the non-causal information so acquired (and neither does anyone spying on the Cat).

    Because the Cat is so constituted as to never cheat, we will call him Cramer’s Virtuous Cat. So seemingly it is a well-posed question: what mathematical obstruction (if any) in orthodox quantum mechanics blocks Virtuous Cramer Cats from observing—but never exploiting—retrograde causality?

  8. quantummoxie Says:

    I will confess to not being familiar with Cramer’s proposed experiments. Is there a reference?

  9. There are some references on the Wikipedia page on retrocausality, but no article that I know of.

    But heck … John Cramer’s retrocausal experiments ain’t proposed … he’s busy doing them!

    Aye laddie … now that’s the true scientific spirit.

    Designing and analyzing these experiments is definitely not trivial, as textbooks tend to gloss-over technical issues that are key in real-world experiments.

    Personally, I’ll be utterly flabbergasted if these experiments see any retrocausal effects …. but to be fair, Prof. Cramer has pointed out to me that mathematicians, theorists, and engineers have often been astounded by experimental data … and definitely I have considerable respect for that point-of-view.

  10. quantummoxie Says:

    So where is John Cramer located anyway? Hmmm. I am very interested in hearing about his work.

  11. John Cramer is in the physics Department here at the University of Washington … next time I see him, I will ask what’s up.

  12. quantummoxie Says:

    Please do. I checked out his website on the experiments, but he hasn’t updated them in three years.

Comment (obtuse, impolite, or otherwise "troll"-like comments may be deleted)

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: