WD:
> On Tue, Feb 24, 1998 at 03:35:35PM +1000, Mitchell Porter wrote:
> > Suppose I said that I said that humans are made of atoms, but that
> > in my ultimate theory, atoms are 'only approximations'. Wouldn't
> > that cause some cognitive dissonance? Wouldn't you pester me for
> > an explanation of what humans are made of, in *non*approximate terms?
> > Wouldn't you be dissatisfied if I couldn't provide one?
> >
> > The situation is analogous, except that humans are *part of* a
> > world, rather than made of worlds. How can one make sense of the idea
> > that human beings are part of an approximation?
>
> But I've already said that according to the MWI, in non-approximate terms
> we're part of the UWF.
*Where*?
Please show me in what sense even a single particle is part of the UWF.
If you point to one of the branches - e.g. suppose you tell me
that something is part of Psi by virtue of being part of Psi0 -
please explain why we need to postulate Psi, above and beyond Psi0,
since Psi0 is apparently causally self-contained.
> The concept of the many worlds is only used to
> explain why, if we're part of the UWF, we would see what we see (and not
> see, for example, macroscopic superpositions).
Again: if Psi0, Psi1, etc are the worlds, and if they are
causally self-contained, how can anything to do with the other
worlds explain anything about this one.
Maybe we can finally tie this into the probability thread. The
*only* way I can see the other worlds being relevant to explaining
anything that we observe, is if they serve as the framework for
possible-worlds questions of the form, "What were the odds that
I would observe such-and-such?"
But to pursue that line of thought, I think we need to be clear
that the worlds are causally isolated. I *think* that David Deutsch
in his book (_The Fabric of Reality_) hints that what happens here
(in this universe) receives some causal contribution from the universes
next door, but he doesn't spell out how this works. I think that
idea would require some working-out of the notion of worlds or
subsystems "splitting" and later "fusing" - whereas what we have
so far is strict causal isolation of worlds existing in parallel,
and a non-causal concept of correlation.
> > In any case, in the example of Psi = c0 Psi0 + c1 Psi1, there's
> > nothing approximate about the relationship between Psi, Psi0
> > and Psi1. What's 'approximate' is the relationship between Psi0,
> > Psi1, c0 and c1 (through the use of an epsilon-criterion to say,
> > "Below this point, there's no 'interference'"). If we turn to
> > Psi0 or Psi1 as individual worlds, is it any easier to find
> > human beings inside them than it is to find human beings in Psi?
>
> Depends on what you mean by easier. Computationally it should be easier;
> information theoretically it isn't.
I don't understand what distinction is being made here.
-mitch
http://www.thehub.com.au/~mitch
Received on Tue Feb 24 1998 - 01:05:24 PST