From: ram@zedat.fu-berlin.de   
      
   Michael Cole  writes:   
   >In my humble opinion, the \_fatal flaw_/ of many worlds is   
   >that it is metaphysics, not science.  That interpretation of QM is   
   >untestable. It is fun to think about, but it is a question of philosophy   
   >that is outside the province of science.   
      
     Is the MWI a different theory ("fifth axiom") or just   
     an interpretation ("an irrelevant adornment")?   
      
     I'd say the MWI is not a additional "fifth axiom",   
     but rather a /removal/ of some rules!   
      
     Traditional QM goes something like:   
      
   D0 - A system A is described by a space H_A.   
   D1 - States are vectors (rays) of this space.   
   D2 - They evolve according to a unitary time evolution.   
   D3 - Composite systems are described by tensor products.   
      
   M0 - Observables are represented by operators.   
   M1 - The Born probability rule gives probabilites for outcomes.   
   M2 - After a measurement, we observe one of those outcomes.   
      
     But D2 and M2 are contradicting: When states evolve   
     unitarily, there is no need for M1 and M2. Which processes   
     are deemed "measurements", so that we have to use M2 and M2   
     for their evolution, and which aren't, so that we have to   
     use D2 for their evolution?   
      
     To get rid of this over-determination of physical processes   
     by two different systems of rules, Everett removed M0-M2.   
      
     This is not so much like an addition of an interpretation,   
     it is more like a /removal/ of some interpretations!   
      
     So, MWI is just D0-D3 without M0-M2.   
      
     The rules D0-D3 /have/ been tested many times (double slit,   
     Mach-Zehnder), and we just have to conclude that a particle   
     /does/ "take both paths" sometimes and that the vector-addition   
     of state vectors to form a superposition /is/ an accurate   
     description of nature. In this sense, MWI /has/ been tested.   
      
     We also can see that the addition of M0-M2 can lead to   
     overdeterminations of measurement processes, because they   
     also should evolve according to D2. So, MWI /has/ been tested   
     and adding more rules (M0-M2) than MWI (D0-D3) leads to problems.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)