2008-04-09

The Attentiometer

The two-slit experiment. Schrödinger's cat. Heard of 'em? If not, go read up. I'll wait.

They're important concepts. They represent the greatest mystery about reality we currently face, and I just don't hear much about them. This bugs me. The two-slit experiment demonstrates that looking at the universe changes it. But it doesn't give us an adequate description of what "looking at it" means. Schrödinger's cat was an attempt by Schrödinger to make us see how bizarre were the implications of the "observation" aspect of quantum mechanics. Remember, he intended it as a reductio ad absurdum, that is, he meant to show how silly the superposition interpretation of quantum mechanics was, by showing that it led to a cat being both dead and alive at the same time, which is absurd. But nowadays lots of physicists are perfectly happy to use it as the metaphor by which they describe reality. "Yup," they say, "until you observe it, the cat's dead and alive at the same time."

Maybe if I'd taken one more term of physics before giving up, I'd know why what I'm about to suggest is just plain wrong. I hope somebody will explain it to me, and be gentle. Here's my thought:

Observing a quantum event can change its result. If you stick a photomultiplier on one of the slits, the pattern on the film changes. Can we turn that around, and use the pattern on the film as a way to tell whether an event was observed?

Could we send people randomly into a room where the output of the photomultiplier shows on a screen? And make sure that if there's no one there the information is irretrievably lost? Like, line the room in black velvet? Or something? Then later we compare the films with records of when there was someone in the room, and (here's my hypothesis) lo and behold we can tell when someone was watching.

I dunno. Maybe if you add it all up informationally, you're always observing or you're never observing. Maybe your act of looking at the results is causally downstream of the events, and so always counts as an observation. We know already that the observation's collapse of the quantum state is extratemporal. Or maybe something like Penrose's "one graviton effect" rule holds true, and it's purely a matter of how much stuff gets bumped by the event.

But, dammit, we can see the difference in the films. Can't we? Isn't there an epistemic asymmetry there? A handle to grab on to?

Maybe there's a catch-22 guaranteeing that you can never gain utility from decoherence. They say that you can't use entanglement to make an ansible; maybe it's a similar kind of situation to that. I don't grok it; I didn't study hard enough in Differential Equations. But, if so, is there any way to rigorously prove that that's the case?

And could somebody please convince this layman, so I can stop dreaming?

Please, can we settle this question?