On Fri, 13 Aug 1999, Russell Standish wrote:
> > referring to
> > > t0 |
> > > |
> > > t1 T / \ H
> > > / \
> > > t2 / / \
> > > | | \
> > > t3 Y R B
> >
> > Assume that all three branches occur (two copying events).
> >
> > Gilles Henri wrote:
> > >With the color cards, each Jane will measure subjectively a probability 1/2
> > >of yellow, 1/4 of red (1/2 H *1/2 "being chosen as Jane 1") and 1/4 blue,
> > >so again p(H) = p(T)=1/2 with the conditional probability formula.
> > >The probability 2/3 is indeed the chance of finding someone who saw H after
> > >the first experiment from a bird perspective, because duplicating
> > >introduces a bias.
> >
> > I agree that according to the approach taken by the q-su's, namely
> > that one's measure is somehow distributed among the so called
> > computational continuations of one's brain activity, the probabilities
> > would be (1/2,1/4,1/4). It is a history dependent claim:
> > > t0 |
> > > |
> > > t1 W / \ H
> > > / \
> > > t2 T/ \H \
> > > | \ \
> > > t3 Y R B
> >
> > where W=wait to show the coin to her until the second copying
> > event. Since she doesn't know when copying occurs this looks identical
> > from her perspective, but the measure distribution is (1/4,1/4,1/2)
> > according to the QS claim. Presumably this measure distribution would
> > remain the same years later.
> > I think this is already both ill-defined and anti-intuitive.
> > To extend the example suppose that to counter the unfortunate
> > demographic imbalance in China, someone figures out how to instantly make
> > a million copies of Gong Li. According to the flow of measure claim, each
> > of these copies would have just one millionth of a normal human measure.
> > So these women would practically be zombies. It would not be
> > justified to give them equal rights since they have so much less
> > consciousness. This would remain true even as life experiences give them
> > different perspectives and evolved personalities, some of them come to
> > America, etc. I think this shows how ridiculous the claim is.
>
> No - it is a mistake to equivalence measure with degree of
> concsiousness. Conciousness is either there or it isn't. Measure, does
> however, indicate what history is likely to be observed (SSA). So an
> interesting question to ask is what is the relative proportion of
> heads and tails observed by Jane in a sequence. The probability of
> being in a branch with 2^i Janes in the world, given n repetitions of
> the above experiment is given by the binomial distribution
You misunderstand. First, the experiment I talked about is
different than the one with a coin flip determining whether to copy. As I
said,
> > Assume that all three branches occur (two copying events).
in this one there are always two copying events and three Janes at the
end.
Second, no one said anything about degree of consciousness.
Measure describes the *amount* of consciousness. I should have said that
I was talking about absolute measure (not effective probability, which is
normalized by definition to obtain an effective probability distribution).
In the QS claim measure behaves in such a way. It's just the time reverse
of the QS process.
A room with 20 people in it has 20 times the measure of a room
with 1 person. But a room with 20 of these women would supposedly have
very small measure. For example, just as it is better to kill 1 person
than to kill 20, it would be better to kill 20 of these than 1 regular
person (all else being equal). In the limit as measure goes to zero we
have a true zombie.
Another example is that when a person comes into being, he
supposedly has a standard allotment of measure (1 unit). But suppose that
there already exists a being with very small measure such that the 'new'
person is a 'computational continuation' of the old being. (e.g.
one of the above women with particular life experiences.) Suddenly the
new person (supposedly) has very small measure! But if a parameter in a
model is adjusted until the old being (in this model) vanishes completely,
suddenly the measure of this new person (supposedly) springs back up to 1.
I don't think a QSer could get away with denying the reverse
process. There could be a cyclical process in which the number of copies
is varied periodically with time (with instant copying and killing
machines).
- - - - - - -
Jacques Mallah (jqm1584.domain.name.hidden)
Graduate Student / Many Worlder / Devil's Advocate
"I know what no one else knows" - 'Runaway Train', Soul Asylum
My URL:
http://pages.nyu.edu/~jqm1584/
Received on Fri Aug 13 1999 - 09:03:56 PDT