On 26 Mar 2015, at 22:13, meekerdb wrote:
On 3/26/2015 8:40 AM, Bruno Marchal wrote:
On 25 Mar 2015, at 16:35, Stathis Papaioannou wrote:
On Wednesday, March 25, 2015, Quentin Anciaux <[email protected]>
wrote:
2015-03-25 12:25 GMT+01:00 Quentin Anciaux <[email protected]>:
2015-03-25 12:09 GMT+01:00 Bruce Kellett
<[email protected]>:
Quentin Anciaux wrote:
Le 25 mars 2015 07:27, "Quentin Anciaux" <[email protected] <mailto:[email protected]
>> a écrit :
> Le 25 mars 2015 07:23, "meekerdb" <[email protected] <mailto:[email protected]
>> a écrit :
> > On 3/24/2015 11:18 PM, Quentin Anciaux wrote:
> >>
> >> Le 25 mars 2015 05:08, "Russell Standish" <[email protected]
<mailto:[email protected]>> a écrit :
> >> >
> >> > On Wed, Mar 25, 2015 at 12:25:04AM +0100, Quentin Anciaux
wrote:
> >> > > Le 25 mars 2015 00:11, "meekerdb" <[email protected] <mailto:[email protected]
>> a écrit :
> >> > >
> >> > > When rerunning the program with the recorded initial
input, by hypothesis
> >> > > the second run must be as conscious as the first when
the inputs came from
> >> > > the 'real' external world... The program itself can't
tell as it receives
> >> > > exactly the same inputs... Not similar inputs but
*exactly* the same. So
> >> > > either the second run is as conscious as the first or
none are.
> >> >
> >> > Or there is precisely one sequence of conscious observer
moments no
> >> > matter how many times it is rerun (or recorded and
replayed, whatever).
> >> >
> >> > Cheers
> >>
> >> Then in this case physical supervenience is false...
> >
How so? Supervenience doesn't forbid different substrates from
producing the same supervening effect. In this case it would be
two different instances of the physical process producing the same
conscious thoughts.
If it's different instances both moment are conscious... Not only
one... The how many time it is run is important as by physical
supervenience, it's the physical token that generates
consciousness. So if ypu say that it doesn't matter how many times
you run the cpnsciuous able program with the correct inputs,
Because there is only one conscious moment
then you reject physical supervenience.
I do not think this follows. Consciousness supervenes on the brain
states. It does not matter if these are instantiated in brain
wetware or in an accurate record of these brain states on a film
or in a computer memory. It is the states (or sequence of states)
that makes up the conscious experience. If the record is exact,
then replaying it reproduces exactly the initial conscious
experience (as Russell points out), not some other experience.
Yes... that's what I said... replaying it N times under physical
supervenience means you have N times the conscious moment
supervening on the substrate *in realtime* (exactly the same
conscious moment) but it is instantiated N times, not only once...
(when I say realtime, it's not that the inner time of the
conscious moment should be one to one with the external time where
that conscious moment is supervening, but that the conscious
moment exists at the same time it is running) (as Russel seems to
say).
Correction as Russel seems to say there is only one conscious
moment... how many time you run it... well under physical
supervenience you have N times exactly the same conscious
moment... but each run is as real and existing as the other... and
there is not only one... saying there is only one is rejecting
physical supervenience.
If my mind is being run on two separate computers, I can't know
which one of the two, and I can't say that my last remembered
moment was run on one or other or my next anticipated moment will
be run on one or other. If one computer stops it makes no
difference to me and if a third computer running my mind comes
online it makes no difference to me. So effectively there is only
one conscious moment. Under physical supervenience, stopping all
the computers stops the conscious moment.
I am OK. I think Quentin is arguing in the reducto ad absurdum part.
In a sense both Russell is righ (there is only one 1p-experience),
and Quentin is right: we can attribute consciousness in each
running (but then if we attribute it to the physical activity
token: we get the absurd conclusion: playing records and real-time
consciousness supervene on a static film, etc.
What happens is that consciousness here and now exist by the
existence of the computation (and thus in arithmetic), and the
probability of this or that differentiation has to take into
account all the running in arithmetic (and not all playing
records). here the running are the computations, or the triple
(universal number, program, a number of steps), and the records are
the Gödel number of those computations. Both are realized in
arithmetic, but the computation are realized in some standard
interpretation of arithmetic, and the Gödel numbers are only
syntactical description of them so that the machine and us
know which computation we are talking about.
Russell used the ...-1-1-1-1-view, but in MGA (and UDA) we need to
use some 3-1 views, or 3-1-1 views.
Consciousness depends only on the existence of the (relevant)
computations, but the relative stability of the local consciousness
flux depends on the relative proportion of histories/dreams, and
for this we need to consider the 1-views attributed to person, but
incarnated in some 3p activity (program executions), and thus we
need to use (implicitly) some 3-1 view.
If 100 computers, physically distinct, "run my
consciousness" (simultaneously or not), it is the same
consciousness, and the 1p is unique, but I will prefer that 1% of
the running differentiate into an hellish experience instead of 99%.
This seems like branch counting in MWI. If one branches in a
hellish experience while 99 continue on the same non-hellish path
then there are only two streams of consciousness. So does your FPI
tell you you have probability 1/2 of experiencing hell or does the
amount of physical instantiation determine the measure?
Good question.
Locally, it will be the number of physical instantiations (susceptible
to be distinguished in principle).
But globally it depends on the measure on the computations, or
arithmetical instantiation, or realization, making up your history/
histories (it depends on the true, machine-independent, relative
proportion of relative states in arithmetic).
Yes, there might be too many. That is the question addressed and
eventually translated in arithmetic (which gives the tools: S4Grz1,
Z1*, etc.)
I insist that I am not presenting computationalism as a solution, but
as a problem.
With Materialism there is a big problem: consciousness.
With computationalism there are two big problems: consciousness and
matter.
Consciousness is made easier, as it becomes a non definable attribute
of agent, and they are invariant for some functional digital
substitution, but the price of this is that matter will have to be
justified by invariants related to that machine's first person views
or consciousness.
Bruno
Brent
For extracting the physics, this 3-view is needed, or at least
helps, I think. In modal logic, those are given by multi-modal
expression, like []<1>[2]P, for example, which admits a non
ambiguous interpretation in arithmetic, as they are all defined
from the Gödel arithmetical 'Beweisbar' predicate ([]p).
Bruno
--
Stathis Papaioannou
--
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 http://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 everything-
[email protected].
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
--
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 http://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 http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
