![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Afshar's "which-way" experiment
Apr. 30th, 2004 09:14 pmOh floof. The Google cache of Cramer's PowerPoint on Afshar's experiment failed to make clear to me the most crucial part, which Chad Orzel just explained: Afshar measured which slit the photons had gone through after his wires detected the interference minima!! This makes the experiment quite unmysterious.
Well, of course the result he got would be the result he'd get. Any interpretation that said otherwise would be pretty stupid, and Cramer's claim that many-worlds and Copenhagen say this is based on strawman versions in his own head.
First, Afshar's wires measured an operator which was not quite the photon-momentum operator, but, as![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
bram said, left unmeasured a whole degenerate subspace of states that still encoded nearly all of the which-way information. Then the subsequent optics extracted the which-way information with high accuracy.
From a collapse-based perspective, what happened was that part of the state vector collapsed at the wires, but not all of it. And what didn't collapse was enough to produce the eventual which-way result. From an Everettian perspective, some of the coherence of the wave function was destroyed by correlation with the state of the wires, but much was left intact. This sort of thing has to happen even for much simpler experiments to work, such as one in which you split a beam with a beam splitter and then play with the optics of just one of the two legs.
Granted, some early versions of Copenhagen back when nobody knew much of what was going on might have been stymied by this. Modern QM interpretations typically talk about operators and downplay the old mystification about wave-particle duality that Cramer seems to be hung up on. And Everett's relative-state approach certainly would have no trouble with it, though somebody's subsequent garbled version of many-worlds might, for all I know.
(Of course, from Cramer's perspective I'm surely just being as slippery as he predicted people like me would be, and thinking up an ad-hoc excuse to save my beloved every-interpretation-but-his, to which I am irremediably wedded.)
Well, of course the result he got would be the result he'd get. Any interpretation that said otherwise would be pretty stupid, and Cramer's claim that many-worlds and Copenhagen say this is based on strawman versions in his own head.
First, Afshar's wires measured an operator which was not quite the photon-momentum operator, but, as
bram said, left unmeasured a whole degenerate subspace of states that still encoded nearly all of the which-way information. Then the subsequent optics extracted the which-way information with high accuracy.From a collapse-based perspective, what happened was that part of the state vector collapsed at the wires, but not all of it. And what didn't collapse was enough to produce the eventual which-way result. From an Everettian perspective, some of the coherence of the wave function was destroyed by correlation with the state of the wires, but much was left intact. This sort of thing has to happen even for much simpler experiments to work, such as one in which you split a beam with a beam splitter and then play with the optics of just one of the two legs.
Granted, some early versions of Copenhagen back when nobody knew much of what was going on might have been stymied by this. Modern QM interpretations typically talk about operators and downplay the old mystification about wave-particle duality that Cramer seems to be hung up on. And Everett's relative-state approach certainly would have no trouble with it, though somebody's subsequent garbled version of many-worlds might, for all I know.
(Of course, from Cramer's perspective I'm surely just being as slippery as he predicted people like me would be, and thinking up an ad-hoc excuse to save my beloved every-interpretation-but-his, to which I am irremediably wedded.)