On Mon, Aug 12, 2019 at 3:48 AM Bruno Marchal <[email protected]>
wrote:
On 9 Aug 2019, at 13:15, Bruce Kellett <[email protected]>
wrote:
On Fri, Aug 9, 2019 at 7:49 PM Bruno Marchal <[email protected]>
wrote:
On 9 Aug 2019, at 04:07, Bruce Kellett <[email protected]>
wrote:
From: BRUNO MARCHAL <[email protected]>
On 8 Aug 2019, at 13:59, Bruce Kellett <[email protected]>
wrote:
On Thu, Aug 8, 2019 at 8:51 PM Bruno Marchal <[email protected]>
wrote:
If the superposition are not relevant, then I don’t have any
minimal physical realist account of the two slit experience, or even
the stability of the atoms.
Don't be obtuse, Bruno. Of course there is a superposition of the
paths in the two slit experiment. But these are not orthogonal basis
vectors. That is why there is interference.
But each path are orthogonal. See the video of Susskind, where he
use 1 and 0 to describe the boxes where we can find by which hole
the particles has gone through. Then, without looking at which hole
the particle has gone through, we can get the interference of the
wave which is obliged to be taken as spread on both holes, and that
represent the superposition of the two orthogonal state described
here as 0 and 1.
I seldom watch long videos of lectures. But if Susskind is saying that
the paths taken by the particle through the two slits are orthogonal
then he is flatly wrong. Writing the paths as 1 and 0 does not make
them orthogonal. And if they were orthogonal they could not interact,
and you would not get interference. Two states |0> and |1> are
orthogonal if their overlap vanishes: <0|1> = 0. Interference comes
from the overlap, so if this vanishes, there is no interference.
Either Susskind is terminally confused, or you have misrepresented
him.
Or maybe you are wrong. Slit one is orthogonal to slit two, as much as
spin in different direction.
When you observe which slit the particle went through, then yes -- the
slits are then orthogonal eigenstates of the position operator.
OK. But without collapse, the observation of which slit the particle
was taking is only self-entanglement, and it makes the whole history
“the particle went through slit A + me seeing the particles going
through slit A” orthogonal to the whole history “the particle went
through slit B + me seeing the particles going through slit B”.
So, like I said, the slits are orthogonal if measured. This is
irrelevant to the interference at the screen, because it is the photon
waves at the screen that interfere, not the slits. Orthogonal states
do not interfere.
Decoherence is only self-entanglement. It spread at the speed of
light, or a bit below, in the environment, making hard to fuse the
histories again, through “amnesia”, but that explains why the
superposition states are are hard to maintain accessible. FAPP, we
can forget the parallel histories, but, only FAPP!
FAPP is the way we do physics. Metaphysics is for the birds that can't
fly!
The interference comes from the fact that we get a superposition of
going through slit one + going through slit two when we send a
planar monochromatic wave on the wall with the two slits, and
don’t measure which slit the particle go through.
Yes, then the states that we are measuring are not orthogonal. You
do not get interference between orthogonal states.
That is how Susskind explains the two slit experiment in term of
entanglement. You don’t need to look at the whole video, I gave
the position of this sub-talk in the video.
Any crisp measurement, like “which slit” gives rise to
orthogonal state, which can interfere when superposed.
Which is essentially what I said -- orthogonal states do not
interfere.
In the sense you mention I am OK, but we have a slight vocabulary
problem. Not important, if you agree that measurement are
self-entanglement, so that the superposition of the orthogonal state
SlitA and SlitB, say some oblique (with sqrt(2) = 1) SlitA + SlitB is
inherited by the observer “looking” which is which.
If you do not measure which slit the photon went through, then the
superposition of slits is not broken by decoherence. But the
interference at the screen depends only on things like the wavelength
of the light, the separation of the slits, and the distance between
the slits and the screen. If you refine this calculation by taking the
finite width of the slits into account, you convolute the interference
pattern with the diffraction pattern due to finite slit width. This is
an elementary calculation in physical optics, not even requiring
quantum mechanics. But the waves at the screen cannot be orthogonal,
or else they would not interfere.
