On Thu, Feb 19, 1998 at 05:46:41PM +1000, Mitchell Porter wrote:
> So I stand by my original claim: the many-worlds interpretation
> has not been clearly enunciated. Crucial concepts remain undefined
> in any rigorous way.
Ok, let me give my understanding of what a "world" is in MWI. Suppose we
take the UWF and consider it as the superposition of two wavefunctions:
PSI = c_0 PSI_0 + c_1 PSI_1.
If from examining the time evolution of PSI, we can tell that there is
very little interference between PSI_0 and PSI_1 after time t_1, then
questions we want to ask about PSI after time t_1 can be answered
approximately by considering PSI_0 and PSI_1 independently (i.e. as a
mixture instead of a superposition). In this sense PSI_0 and PSI_1 can be
considered to be seperate worlds after time t_1. If PSI_0 is also the
superposition of two wavefunctions
PSI_0 = c_00 PSI_00 + c_01 PSI_01
such that there is very little interference between PSI_00 and PSI_01
after time t_2, then we can also consider PSI_00 and PSI_01 to be seperate
worlds after time t_2.
So basicly, a world is a component of the UWF that has sufficiently low
interference with the rest of the UWF after some time t that it can be
treated as if it doesn't interfere at all with the rest of the UWF after