From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!cs.utexas.edu!uunet!tdatirv!sarima Mon Dec 16 11:01:17 EST 1991
Article 2056 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!cs.utexas.edu!uunet!tdatirv!sarima
>From: sarima@tdatirv.UUCP (Stanley Friesen)
Newsgroups: comp.ai.philosophy
Subject: Re: The Robot Reply (was Re: Searle, again)
Message-ID: <309@tdatirv.UUCP>
Date: 11 Dec 91 23:04:49 GMT
References: <2127@ucl-cs.uucp> <91338.113617KELLYDK@QUCDN.QueensU.CA> <5796@skye.ed.ac.uk> <YAMAUCHI.91Dec5235651@heron.cs.rochester.edu> <5825@skye.ed.ac.uk>
Reply-To: sarima@tdatirv.UUCP (Stanley Friesen)
Organization: Teradata Corp., Irvine
Lines: 33

In article <5825@skye.ed.ac.uk> jeff@aiai.UUCP (Jeff Dalton) writes:
|Looking at it from the inside, the input from sensors is just
|more squiggle-squiggles and squoggle-squoggles.  That some of
|them come from a camera while others are written down by a
|person -- why does adding some of the former suddenly solve
|the problem for all of the inputs involved?

Looking at it from the inside the input to the brain from sense organs
is just more dits and dahs travelling along the axons.  That some of them
come from a photoreceptor while others are derived from a vibration detector
activated by another person -- where does semantics come from here?


You see, the *exact* same arguments apply to humans.  The signals in neurons
are simply transduced sense inputs until the brain starts to process them.
And in the brain the processing is done purely through 'syntax', that is
it is based wholly on the 'shape' of the representative signals, not on any
relationship these signals might have with an outside world.  The same input
channels could be attached to different transducers, and the brain could
only tell the difference if the *structure* of the signals differed, that is
if the input no longer matched the syntax currently associated with that
channel.

We *learn* the syntax associated with each input channel and then build
higher levels of data structure on top of this to which we attach symbols,
all by means of *learned* *associations*.  The actual internal operations
of the brain are purely in terms of the neuronal signals and thier
relationships, which seems to meet the definition of 'syntactic' presented
here earlier.
-- 
---------------
uunet!tdatirv!sarima				(Stanley Friesen)



