From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!uwm.edu!linac!mp.cs.niu.edu!rickert Tue Jun  9 10:05:47 EDT 1992
Article 5991 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!uwm.edu!linac!mp.cs.niu.edu!rickert
>From: rickert@mp.cs.niu.edu (Neil Rickert)
Subject: Re: Hypothesis: I am a Transducer (Formerly "Virtual Grounding")
Message-ID: <1992Jun1.012608.3756@mp.cs.niu.edu>
Keywords: computation, transduction
Organization: Northern Illinois University
References: <1992May29.152559.226@mp.cs.niu.edu> <9597@scott.ed.ac.uk> <1992May31.145204.16357@Princeton.EDU>
Date: Mon, 1 Jun 1992 01:26:08 GMT
Lines: 98

In article <1992May31.145204.16357@Princeton.EDU> harnad@phoenix.Princeton.EDU (Stevan Harnad) writes:
>Many readers still do not seem to have understood my point about
>transduction, so I will try yet another intuition pump:

  Perhaps some readers understand your point perfectly, and it is because
they understand it that they don't agree with it?

>                                                        Although it is
>an over-simplification, consider my hypothesis to be that you ARE a
>transducer.

  I hope you will not take offense at this, but part of the misunderstanding
is because you, like Searle, are fixated on a formalistic view of a
computer as a Turing Machine.

  Much of my response will be on the theme of what a computer is.  But let
me start by pointing out that a computer IS a transducer.  We think of a
computer as containing millions of digital components, and sending digital
signals over the bus.  But this is a gross simplification.  The signals on
the computer bus are perfectly ordinary analog voltages.  Although the
circuitry is specialized for digital usage, the basic components are
perfectly ordinary analog transistors.  The dynamic RAM cells are
perfectly ordinary analog capacitors.  Transduction is going on all the
time as these analog transistors detect their analog input signals and
interpret them as if digital signals.

  I realize you will say that this is all beside the point.  But it really
isn't.  It points out, amongst other things, that there is nothing special
about analog as against digital.  There is nothing particularly special
about transduction, either.  All that really matters is the input and
the output.  In man made technology, whether the transduction from
input to output occurs by purely analog means, by purely digital means,
or by a mixture of the two, is simply a matter of convenience and economics,
and there is no other essential difference.  I suspect that you will find,
as more and more is learned about biology, that nature too sometimes uses
analog methods, sometimes uses digital methods, and sometimes uses a mixture
of the two.  The choice is presumably made by evolution, but that may be much
the same as saying that the choice is based on convenience and economics.

>            If that hypothesis is correct, then there may be many
>different ways to implement you -- namely, all the different ways of
>implementing a transducer with your capabilities (TTT) -- but among
>those ways is definitely NOT one in which instead of a transducer there
>is a computer simulation of a transducer (a "virtual" transducer).

 I don't mean to sound insulting.  But your assertion about computer
simulation seems terribly dogmatic.  Besides that, it is just plain wrong.

 Consider the exchange I have been having with sharder@cogsci.ed.ac.uk
(Soren Harder) on the same general thread.  Soren maintains that you 
need feature detectors to detect crescendos, sharp attacks, etc in
the music.  Let's take a CD.  Play it through an audio system, and feed
the output of the audio system into some suitable feature detectors.
Then the audio system, and feature detectors make up a transducer.

 Suppose, though, that we design a computer program to read the CD
digital output directly, and detect exactly the same features.  This
could be done by simulation of the analog transducers.  Or it could
be done directly with some computational algorithm.  We provide our
computer with enough power to do this in real time.  What we have,
then, in our computer system, is a device for converting
the CD output into the features.  It IS a transducer.  It is not
merely a "virtual" transducer.  It is doing transduction every
bit as much as the analog system.  The fact that it is implemented
on a digital computer with a program has nothing to do with the
matter.  It has the identical input and output to the analog
transducer.  It IS a very real transducer.

>                                                                   I
>repeat, my hypthesis is that you ARE a transducer. If that is true,
>anything that is not really a transducer is not really you.

 This is the heart of our dispute.  You are interpreting my comments
as meaning that I disagree about the nature of _you_.  I don't.  Our
disagreement centers around the nature of the computer, and in my opinion
your view is far too narrow.

>                                              People keep reverting to
>the rival computational hypothesis (which I have tried to show is
>false, Harnad 1989, 1990, 1991, 1992) in which you are a computational
>core, with the transducers simply carrying information TO it ("you"),
>as our senses do to "us."

 A good part of the misunderstanding is because you yourself keep
reverting to the computational hypothesis that what happens in a
computer is due to a computational core in which syntactic
manipulation of symbols occur.  This may be a good way of thinking
about traditional payroll programs, or perhaps even expert system
programs.  It is not nearly as useful a way of looking at the
operating system of a multiprogramming system.  It is even less
useful in understanding what is happening in programs which test
aircraft designs with numerical simulations of wind tunnels.

-- 
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
  Neil W. Rickert, Computer Science               <rickert@cs.niu.edu>
  Northern Illinois Univ.
  DeKalb, IL 60115                                   +1-815-753-6940


