On 09 Aug 2016, at 01:15, Bruce Kellett wrote:
On 9/08/2016 12:39 am, Bruno Marchal wrote:
On 08 Aug 2016, at 01:26, Bruce Kellett wrote:
On 8/08/2016 1:30 am, Bruno Marchal wrote:
But in step 3, I ma very careful to not use the notion of
"consciousness", and instead a simple 3p notion of first person.
usually many relates it two consciousness and assumes that when
the guy say "I see Moscow", they are conscious, but that is not
needed to get the reversal.
Maybe that is the basis of the problem. In step 3 you seem to be
claiming nothing that could not be achieved by a non-conscious
machine:
Yes.
take a machine that can take photographs and compare the resulting
images with a data base of images of certain cities. When a match
is found, the machine outputs the corresponding name of the city
from the data base. Send one such machine to Washington and an
identical machine to Moscow. They will fulfill your requirements,
the W-machine will output W and the M-machine will output M.
This is what you are now seeming to describe. But that is not FPI.
How could the machine predict the result of the match? Give me the
algorithm used by that machine.
The machine program knows the protocol -- it knows that one copy
will be transported to M and one to W. The machines are already
physically different (different locations if nothing else), so it is
a matter of a coin toss as to which goes where. The machines do not,
however, share a consciousness, so this does not answer what will
happen with a conscious being.
You forget that we assume comp. So we are machine ourselves, and so,
for the first person points of view, it is indeed like tossing a coin,
and that's the FPI.
Consciousness is treated later. For the reversal, only the notion of
knowledge and/or first person is enough.
Otherwise your prediction is no different from predicting the
outcome of a coin toss. Think of one machine, it will be unaware of
the other, if it knows that it will go to either W or M on the
result of a coin toss....... prediction, 50/50. (But if the machine
doesn't have the protocol programmed in, it will simply answer:
"What?")
You make my point. Just apply computationalism.
The "P" in the acronym stands for "person", and if the "person" is
not conscious, it is a zombie and any output you get has no
bearing on what will happen to conscious persons.
The problem is a problem of prediction of future first person
account.
That is a problem only if you have a person -- a conscious being.
Not at all. You forget we *assume* computationalism. Without the
reconstitution in W, we have already agreed that P(M) = 1, and vice-
versa, so that the guy's consciousness is linked to its first person
experience in the usual way. So all you need is to assume that when
you teleported on Mars, seeing Mars is a (personal) confirmation of
your survival. Your consciousness and identity remains invariant by
definition of computationalism. We agreed that both copies are genuine
survivor of the duplication experience, and computationalism does not
make them able to share consciousness "here-and-now". They share only
the (non transitive) personal identity, that the memory of who they
are (here: the guy who was in Helsinki and pushed on the button).
Bruno
Bruce
The zombie machines will probably not be aware of each other, but
from that you cannot conclude that the conscious persons will not
be aware of each other, or that consciousness necessarily
differentiates on different inputs.
Well, you need the inputs being enough different (like seeing W,
resp. M) so that the machine can take notice of the difference, and
write distinct outcome in the diary, of course.
Bruno
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
