[extropy-chat] Is Many Worlds testable? - Randall

[scerir]

Randall Randall:
> Really?  What simple experiments?

(I think I've already posted this here.
Not sure though. Maybe in another world!)

MWI is said to be 'ontological'. That is to say, 
quantum states are seen as physical, real, 'ontic' 
states. 

(There are other interpretations, similar to 
the MWI, like the MMI or many minds, the 
original 'relative state interpretation' by 
Hugh Everett III, and the 'many histories'.
Not to mention Feynman's paths, at least when
their 'looping' in time seem to be a parallel 
universe [1]. But in the following I'd say 
something about MWI only).
  
 In MWI the quantum states become real, evident, 
as 'relative' states, as states which are 'relative' 
to specific (and let me say also non reversible) 
measurement apparata ... 

|a>|detector_a> 
                + |b>|detector_b> 
                                  + |c>|detector_c> 

Each quantum state, of a quantum superposition, is
entangled with a specific state of the measurement
apparatus (a macroscopic pointer in general, 
or whatever).

And here comes the 'branching', the 'splitting', 
or the creation, by decoherence or whatever else, 
of the multitude of different worlds.

It is possible to show (at least imo) that the MWI, 
when considered as an 'ontological' theory, is 
inconsistent, or 'not even wrong'.

Imagine a source s, inside a box, emitting, 
at time t0 = 0, a charged particle, with velocity v 
(v << c). There is no essential problem here (due 
to the HUP) about the emission at time t0 = 0
(this time t0 is known by the observer).
 __________________ 
|                  |
|                  |
|                  |
|                  |
|        s         |
|                  |
|                  |
|                  |
|__________________|

When the charged particle hits one of the faces
of the box, it produces a sharp flash of light,
because each face is made of scintillating stuff.

By seeing the flash of light, the observer realizes 
that the particle has reached a face of the box,
at a certain time t2 > t0 

Now we put, inside the box, at a distance r from
the source, a little scintillating surface m. 
 
 __________________ 
|                  | 
|                  |
|           m      |
|           |      |
|        s  |      |
|           |      |
|                  |
|                  |
|__________________|
  
There are, of course, chances that the charged particle
hits the little scintillating surface m, and not
a scintillating face of the box. 

In this case m would produce a sharp flash of light, 
at time t1 = r/v (v is the velocity of the charged particle). 
And the observer will see this flash (at time t1 = r/v, 
since v << c). We have, of course: t0 < t1 < t2.

The quantish description would be a superposition like
   c1 |charged particle hits m at time t1> + 
   c2 |charged particle hits a face of the box at time t2>

Upon measurements, according to the MWI, the branching
occurs. You can see only one of the two component above.
(The other you can see the other component in the other
world).

Given the geometry of the setup, the chances that 
the charged particle hits a scintillating face of the box 
are greater than the chances that the charged particle 
hits the little scintillating surface m. 

Now let us follow this specific experimental evolution.

At time t0 = 0 the source emits the charged particle
(and the observer knows that).

At time t1 = r/v (where t0 < t1 < t2) the observer
does not see any flash of light. 
Thus he realizes that the charged particle did _not_ hit 
the scintillating surface m. 
Hence he also _realizes_ he is living now in a world 
in which the component
   c1 |charged particle hits m at time t1>
has not been reificated.
Hence he also _realizes_ he is living now in a world
in which the component
  c2 |charged particle hits a face of the box at time t2>
will be reificated at the future time t2 > t1.

Since the MWI picture of the 'splitting' is based
on measurements, and since here, at t1, no measurement
whatsoever occurs (the charged particle does not hit 
the scintillating surface m at time t1; the charged
particle will hit a face of the box only at time
t2 > t1) we can say that MWI is not an 'ontological'
interpretation. [2]

Any solution, or hint, is wellcome. 

s.
 

[1]
David Deutsch:
"Like many purported alternative 'interpretations', 
the plausibility of this one relies on equivocating 
about whether these 'loops backwards and forwards 
in time' exist in reality or not. If they don't, 
then the 'interpretation' suffers from the inconsistency 
of attributing observable events to causes that never  
actually happened. If they do, then the loops are 
simply parallel universes in a pointless disguise."

[2]
If Damien is reading ... yes this is the usual weird
role of the parameter 'time' in QM. (A strong Everettista 
would say, perhaps, that the splitting occurs at time t1,
even with no measurements performed at time t1,
because the global Schroedinger function, or the
local Schroedinger function, _knows_ everything,
i.e. it already knows that the charge particle 
will hit a face of the box at time t2.)