From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!olivea!spool.mu.edu!sol.ctr.columbia.edu!zaphod.mps.ohio-state.edu!menudo.uh.edu!mtecv2!pl160988 Tue May 12 15:48:27 EDT 1992
Article 5345 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!olivea!spool.mu.edu!sol.ctr.columbia.edu!zaphod.mps.ohio-state.edu!menudo.uh.edu!mtecv2!pl160988
>From: pl160988@mtecv2.mty.itesm.mx (Ivan Ordonez-Reinoso)
Newsgroups: comp.ai.philosophy
Subject: Re: The intentionality of corpsicles
Message-ID: <5683@mtecv2.mty.itesm.mx>
Date: 30 Apr 92 22:52:19 GMT
References: <g681iB1w164w@PDaXcess.techbook.com>
Organization: I.T.E.S.M. Campus Monterrey
Lines: 61

In article <g681iB1w164w@PDaXcess.techbook.com> mongo@PDaXcess.techbook.com (Mongo) writes:

:              A Thought Experiment (Realized).

[Introductory stuff deleted]

:OK, suppose we enhance them to build sophisticated  internal
:models  of  the world around them (and themselves!), and the
:ability to transfer such models from shelter to shelter.  No
:problem,  you say, still a formal, externally specified sys-
:tem. Only intentional with respect to the intentions of  the
:designer  -  _derived_ intentionality.  In fact, no enhance-
:ment of function, you might claim, can ever give it anything
:but _derived_ intentionality.
: 
:If any of you agree  with  that,  then  you  deny  your  own
:intrinsic intentionality, because we are an evolved species.
:We evolved because we protect and carry forward our  genome.
:Every  drive  we  have, we have because it enhances the sur-
:vival of our genome. Every ability we have is the result  of
:a genetic algorithm computing over millenia. We, it could be
:argued, cannot even  claim  the  _derived_  purpose  of  the
:mobile shelters above, for we are not the engineered product
:of an  acknowledged  intelligence,  but  the  product  of  a
:thoroughly syntactic algorithm (genetic) interacting with an
:environment.

This reminds me of Richard Dawkings' "The selfish gene". You are making
several assumptions, but the most evident is that the evolution process
from which we originated is computable. I work with genetic algorithms,
and my experience from what they can and cannot do tells me that they
are nothing but a pale shadow of the real thing. I don't think that
computers (as they are now, this is, formally equivalent to Turing
Machines) will ever be able to evolve the way life evolves, because life
may have formed as a result of non-computable phenomena.

:To  carry  this  thought  experiment  beyond  it's  original
:instantiation  in Dennet's 'The Intentional Stance', suppose
:that the mobile shelters, developing  their  models  of  the
:world  without  our supervision, discover by accident that a
:rock smashed through a porthole significantly  raises  their
:available  energy  (by  killing  the  corpsicle  inside, and
:turning off the refrigerator). Soon, all of  the  corpsicles
:are  dead, (though still carried about, due to a programming
:atavism) and the shelters continue merrily along,  replicat-
:ing themselves, and improving their models of the world.

It would be very funny if we found a way to kill all the genetic material
in every cell of our bodies. Anyway, I admit this is not a good analogy,
since the 'corpsicles' in your example have no functionality. But the
people who designed the mobile shelters need to be very stupid to
include such a complex surviving mechanism, and exclude the idea that
the main function of the shelters is to protect them.

Machines can duplicate, and they can evolve. But they will always be
machines, no matter how complex they become. They will always be
equivalent to Turing Machines. Unless somebody proves that life is
totally computable, your thought experiment proves nothing.

Ivan Ordonez-Reinoso
pl160988@mtecv2.mty.itesm.mx


