...Plus A Suitable Theory of Consciousness

Miguel Lerma wrote:
 The problem, as I see it, is as follows. Quantum laws and 
 classical laws are essentially different, but the world seems to 
 be in part quantum-mechanical and in part classical. There seem 
 to be only three options:
    1. Everything is classical.
    2. A part of the world is classical and the other part
       is quantum-mechanical.
    3. Everything is quantum-mechanical.
 Option 1 correspond to the theories of hidden variables, option
 2 is the Copenhagen interpretation, and 3 is the Many-Worlds
 Interpretation.

That doesn't strike me as a very helpful formulation of the
problem (nor it is anywhere near a complete list of options).
It would be better to say that although our experience seems to
consist entirely of direct observations of definite qualities,
devoid of any superpositional ambiguities or multiplicities, we
nevertheless find (at least in some special regimes of experience)
that in order to explain our observations it is convenient to
imagine that the underlying physical state of the subject entities
exists and evolves in an essentially superpositional way at all
times OTHER than when we observe it.

Now, we have a wide array of possible "options" for understanding
this odd situation.  We could focus on the words "seems" and
"convenient", noting that perhaps things are not as they seem, 
or that what is convenient is not necessarily true.  Indeed, the 
many-worlds and no-collapse interpretations are based strongly on 
the premise that "things are not as they appear", whereas hidden-
variable theories like de Broglie-Bohm assert that the convenient 
formalism of conventional QM is not necessarily fully representative 
or complete.

However, another option is to accept (at least as a working
hypothesis) that things ARE more or less as they appear (gasp!),
and that the convenient and elegant formalism of QM is accurate
and more or less complete (yawn).  How, then, can we interpret
our observations?  Is there any way of reconciling these premises
with a realistic and non-dualistic understanding?  Along these
lines, I would argue that the fundamental temporal symmetry of
our physical laws and the non-positive-definite character of the
spacetime metric are by far the most likely keys to a meaningful
understanding of superpositional behavior within a realistic
context, and yet this is not even listed among your three
options.  Speaking of which:

Miguel A. Lerma wrote:
 Hidden variables theories do not seem to have been very
 successful, so option 1 has little chance to success.

Your premise is that hidden-variable theories "do not seem to
have been very successful", and from this you conclude that they
"have little chance for success".  This isn't exactly a model of
robust Aristotelian syllogism.  There's a lot that can be said
about hidden-variable theories - such as the comment in my previous
message comparing the ontological economy of MWI with that of the
de Broglie-Bohm theory - but it might just be a distraction here.
Suffice it to say that neither your premise nor your conclusion
are conceeded.

Miguel A. Lerma wrote:
 Option 2 seems to agree with observations, and that is why it is
 so popular in the labs, but it is rather unsatisfactory under an
 epistemological point of view since it lacks of a consistent
 description of the middle ground between the quantum-mechanical
 world and the classical world. It is not even clear that such a
 description can actually be found. A measurement device may
 seem to behave in a classical way, but it is still made up
 of atoms, which are quantum objects, so in a sense the device
 is also a quantum object.

So you're suggesting that there may be some kind of "measurement
problem" lurking in the Copenhagen Interpretation?  Food for
thought.  But seriously, that's not exactly controversial.  What
IS controversial is your suggestion that the CI is the only
conceivable way of reconciling the superpositional aspects of
quantum mechanics with a single world.

Miguel A. Lerma wrote:
 Option 3 has also some problems, but it has the advantage of
 being the simplest one. No hidden variables, no collapsing waves,
 nothing but the bare essential.

When you talk about simplicity you're presumably referring to
the No Collapse Interpretation (NCI) rather than the Many
Worlds Interpretation (MWI), since the latter contains profligate
splitting and branching that is anything (and everything) but
simple.  However, the NCI secures it's simplicity only by
committing itself to a representation of events that does not
manifestly conform with our observation and experience.  At best
it might be called a framework within which one might try to
construct a viable interpretation, although no one has so far 
succeeded in constructing one along these lines.  As it stands,
we are simply asked to imagine that the apparent gap between
the image presented by the NCI and what we actually observe
(or *seem* to observe) is not a significant shortcoming of the 
interpretation.  The idea of blissful evolution in eternal 
superposition is so seductively simple that some people are
prepared to embrace it in spite of the fact that it bears no 
particular resemblence to the world of our experience.


Albro Swift wrote:
 Most especially, there is a serious problem concerning the 
 lack of a valid probability measure for the outcomes of events...  
 there is no suitable principle of individuation to distinguish 
 "worlds" in such a way that their relative numbers can be 
 "counted" in any statistically meaningful sense.

Miguel A. Lerma wrote:
 This problem was solved time ago, see Bryce S. DeWitt's: 
 "Quantum mechanics and reality", Physics Today, Vol23, No.9 
 (Sept.1970). The probabilistic interpretation arises by studying 
 what happens when a given experiment is repeated many times. The 
 sequences of outcomes that do not follow the usual probabilistic 
 rules turn out to belong to a subspace of measure zero in the 
 Hilbert space.

DeWitt's paper contained an early attempt to shore up this obvious
deficiency in the Everett interpretation, but notice that all
the criticisms cited in my previous message take DeWitt (1970)
as their *starting* point, and they conclude that his attempted
explanation was unsuccessful, and the problem still exists.  For
more contemporary discussions (post 1970) of why DeWitt's arguments
fail, see any of the references cited in my previous message
(Polkinghorne/1984, Gibbins/1987, Hughes/1989, Penrose/1989,
Rae/1992, Squires/1994, Wick/1995, and Price/1996).


Roger Penrose wrote:
 "It seems to me a theory of consciousness would be needed
  before the many worlds interpretation can be squared with
  what one actually observes..."

Miguel A. Lerma wrote:
 What we would need is to determine which physical states of
 our brains correspond to given states of consciousness. But
 perhaps again we do not need to postulate anything, just look
 what we can derive from the postulates plus our observations.
 In the example I put at the beginning I compared the states:
 (1)  |observer sees live cat>  and  |observer sees dead cat>
 with
 (2)  |obs. sees live cat> + |obs. sees dead cat>   and
      |obs. sees live cat> - |obs. sees dead cat>
 Whatever a state of consciousness is, it seems to be more
 related to states of the form (1) than to states of the form (2).

You're mixing up three different problems here.  First, there is
the problem of selecting an appropriate "basis" for resolving the
wave function into suitably realistic "states"; second, there is
the problem of ensuring (without resort to solipsism) that multiple
consciousnesses that interact with each other are all conscious of
the same particular set of apparent outcomes; third, there is the
problem of assigning relative probabilities to the various possible
outcomes.  Statements such as "Whatever a state of consciousness
is, it seems to be more related to states of the form (1) than to
states of the form (2)" do not satisfactorily address ANY of those
three problems, i.e., such statements do not constitute a working
interpretation that clearly and unambiguously maps from the
linearly evolving universal wavefunction to our observations 
and experiences.

One could just as well take Leibniz's monadism and append a
"suitable theory of consciousness" to explain why things "appear"
as they do.  The interpretation itself becomes irrelevant, and
you've simply shifted the entire burden onto an unspecified
theory of consciousness.  You might just as well claim the
entire universe is a giant Plutonium atom plus a suitable
theory of consciousness to be defined later.


Albro Swift wrote:
 My point in citing these authors is not to argue by authority,
 or to suggest that the issue be decided by a vote, but simply
 to point out some books in which several thoughtful people have
 described reasons for thinking that the MWI is not, as it
 stands, a viable and/or useful interpretation.

Miguel A. Lerma wrote:
 I am curious. What does Stephen Hawking think about MWI?
 He seems to have a remarkable intuition about the structure
 of the universe.

Remarkable intuition?  It's odd for you to cite anyone's
*intuition* as a significant factor, since you began by
disparaging "psychological" prejudices.  What, after all,
IS intuition, if not a set of psychological prejudices?

Anyway, as mentioned in my previous post, the biggest fans
of Everett-like interpretations have traditionally been in
the community of quantum cosmologists (in fact, Everett's
original motivation was to devise a version of QM specifically
for cosmological work), and Hawking is a quantum cosmologist,
so it's hardly surprising that he has had favorable things to
say about the NCI.  Of course, his comments haven't gone
unchallenged; see the interesting debate between Hawking and
Penrose on the subject of QM interpretation and Schrodinger's
Cat in the recent book "The Nature of Space and Time", and
judge for yourself.