George Levy wrote:
> The correct conclusion IMHO is that consciousness is independent of
> time, space, substrate and level and in fact can span all of these just
> as Maudlin partially demonstrated - but you still need an implementation
> -- so what is left? Like the Cheshire cat, nothing except the software
> itself: Consistent logical links operating in a bootstrapping reflexive
> emergent manner.
Surely this is the 'correct conclusion' only given that one first
*accepts comp*? We can show that a maximalist comp position (like the
UDA argument) cannot depend on 'computationally independent' (i.e. as
distinct from 'computationally substituted') physical supervention, at
root because such supervention can be shown to be arbitrary. That is:
any computation can be implemented in arbitrarily many physical
implementations that may incidentally be *interpreted* as being
computationally equivalent, without this having the slightest effect on
what actually occurs in the world.
In other words, given physicalism, comp can only be a metaphor, relying
entirely on physics to do whatever work is entailed in
acting-in-the-world. For supervention to be true (as it may be)
consciousness would have to map to, and co-vary with, *specific*
physical processes, that happen incidentally to be capable of
interpretation as computations. This is simply entailed by what we mean
by physicalism - all complex processes *reduce* in principle to unique
physical events. Conversely, for comp to be true, the 'physical' must
*emerge* from recursively nested computational operations - i.e. the
reverse explanatory direction.
This disjunction is in itself is an extremely powerful result with
profound, and as yet unresolved, consequences for AI and the study of
consciousness. But as to which is true (or neither for that matter) we
can only follow the consequences of our assumptions here - 'proof'
requires empirically falsifiable prediction and experiment.
David
> List members
>
> I scanned Maudlin's paper. Thank you Russell. As I suspected I found a
> few questionable passages:
>
> Page417: line 14:
> "So the spatial sequence of the troughs need not reflect their
> 'computational sequence'. We may so contrive that any sequence of
> address lie next to each other spatially."
>
> Page 418 line 5:
> "The first step in our construction is to rearrange Klara's tape so
> that address T[0] to T[N] lie spatially in sequence, T[0] next to
> T[1] next to T[2], etc...
>
> How does Maudlin know how to arrange the order of the tape locations? He
> must run his task Pi in his head or on a calculator.
>
> Maudlin's reaches a quasi religious conclusion when he states:
>
> "Olympia has shown us a least that some other level beside the
> computational must be sought. But until we have found that level and
> until we have explicated the relationship between it and the
> computational structure, the belief that ...of pure computationalism
> will ever lead to the creation of artificial minds or the the
> understanding of natural ones, remains only a pious hope."
>
>
> Let me try to summarize:
>
> Maudlin is wrong in concluding that there must be something
> non-computational necessary for consciouness.
>
> Maudlin himself was the unwitting missing consciousness piece inserted
> in his machine at programming time i.e., the machine's consciouness
> spanned execution time and programming time. He himself was the
> unwitting missing piece when he design his tape.
>
> The correct conclusion IMHO is that consciousness is independent of
> time, space, substrate and level and in fact can span all of these just
> as Maudlin partially demonstrated - but you still need an implementation
> -- so what is left? Like the Cheshire cat, nothing except the software
> itself: Consistent logical links operating in a bootstrapping reflexive
> emergent manner.
>
> Bruno is right in applying math/logic to solve the
> consciousness/physical world (Mind/Body) riddle. Physics can be derived
> from machine psychology.
>
> George
>
>
> Russell Standish wrote:
>
> >If I can sumarise George's summary as this:
> >
> >In order to generate a recording, one must physically instantiate the
> >conscious computation. Consciousness supervenes on this, presumably.
> >
> >Maudlin say aha - lets take the recording, and add to it an inert
> >machine that handles the counterfactuals. This combined machine is
> >computationally equivalent to the original. But since the new machine
> >is physically equivalent to a recording, how could consciousness
> >supervene on it. If we want to keep supervenience, there must be
> >something noncomputational that means the first machine is conscious,
> >and the second not.
> >
> >Marchal says consciousness supervenes on neither of the physical
> >machines, but on the abstract computation, and there is only one
> >consciousness involved (not two).
> >
> >Of course, this all applies to dreaming machines, or machines hooked
> >up to recordings of the real world. This is where I concentrate my
> >attack on the Maudlin argument (the Multiverse argument).
> >
> >Cheers
> >
> >
> >
>
>
> --------------030109040106000302090500
> Content-Type: text/html
> X-Google-AttachSize: 3516
>
> <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
> <html>
> <head>
> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
> <title></title>
> </head>
> <body bgcolor="#ffffff" text="#000000">
> List members<br>
> <br>
> I scanned Maudlin's paper. Thank you Russell. As I suspected I found a
> few questionable passages:<br>
> <br>
> <blockquote>Page417: line 14: <br>
> "So the spatial sequence of the troughs need not reflect their
> 'computational sequence'. We may so contrive that any sequence of
> address lie next to each other spatially."<br>
> <br>
> Page 418 line 5:<br>
> "The first step in our construction is to rearrange Klara's tape so
> that address T[0] to T[N] lie spatially in sequence, T[0] next to T[1]
> next to T[2], etc...<br>
> </blockquote>
> How does Maudlin know how to arrange the order of the tape locations?
> He must run his task Pi in his head or on a calculator.<br>
> <br>
> Maudlin's reaches a quasi religious conclusion when he states:<br>
> <blockquote>"Olympia has shown us a
> least that some other level beside the computational must be sought.
> But until we have found that level and until we have explicated the
> relationship between it and the computational structure, the belief
> that ...of pure computationalism will ever lead to the creation of
> artificial minds or the the understanding of natural ones, remains only
> a pious hope."<br>
> </blockquote>
> <br>
> Let me try to summarize:<br>
> <br>
> Maudlin is wrong in concluding that there must be something
> non-computational necessary for consciouness. <br>
> <br>
> Maudlin himself was the unwitting missing consciousness piece inserted
> in his machine at programming time i.e., the machine's consciouness <u>spanned
> execution time and programming time</u>. He himself was the unwitting
> missing piece when he design his tape.<br>
> <br>
> The correct conclusion IMHO is that consciousness is independent of
> time, space, substrate and level and in fact can span all of these just
> as Maudlin partially demonstrated - but you still need an
> implementation -- so what is left? Like the Cheshire cat, nothing
> except the software itself: Consistent logical links operating in a
> bootstrapping reflexive emergent manner.<br>
> <br>
> Bruno is right in applying math/logic to solve the
> consciousness/physical world (Mind/Body) riddle. Physics can be derived
> from machine psychology. <br>
> <br>
> George<br>
> <br>
> <br>
> Russell Standish wrote:<br>
> <blockquote cite="mid20061004020247.GZ11565.domain.name.hidden"
> type="cite">
> <pre wrap="">If I can sumarise George's summary as this:
>
> In order to generate a recording, one must physically instantiate the
> conscious computation. Consciousness supervenes on this, presumably.
>
> Maudlin say aha - lets take the recording, and add to it an inert
> machine that handles the counterfactuals. This combined machine is
> computationally equivalent to the original. But since the new machine
> is physically equivalent to a recording, how could consciousness
> supervene on it. If we want to keep supervenience, there must be
> something noncomputational that means the first machine is conscious,
> and the second not.
>
> Marchal says consciousness supervenes on neither of the physical
> machines, but on the abstract computation, and there is only one
> consciousness involved (not two).
>
> Of course, this all applies to dreaming machines, or machines hooked
> up to recordings of the real world. This is where I concentrate my
> attack on the Maudlin argument (the Multiverse argument).
>
> Cheers
>
> </pre>
> </blockquote>
> <br>
> </body>
> </html>
>
> --------------030109040106000302090500--
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list.domain.name.hidden
To unsubscribe from this group, send email to everything-list-unsubscribe.domain.name.hidden
For more options, visit this group at
http://groups.google.com/group/everything-list
-~----------~----~----~----~------~----~------~--~---
Received on Thu Oct 05 2006 - 12:03:03 PDT