[Martin Taylor 980301 12:28]
Bruce Nevin (980227.1206 EST)]
Martin Taylor (980227 03:20)--
How can you both embody something and at the same time have the
something be outside you? The two concepts seem incompatible.
I think you know the answer to this, at least intellectually.
What you know is in your perceptual universe.
Let's take this as a starting point. What it implies is that "what you
know" is limited to the set of your perceptual signal values at the moment.
It implies that your perceptual input functions, your output functions,
and the linkages among ECUs are not part of "what you know." Those
properties of your control hierarchy are "what you embody," not "what
you know." I'm happy with that.
But I'm not happy with the implication that "what you know" is limited
to _current_ values of perceptual signals. It seems to violate the idea
of episodic memory (memory for some specific happening, that can be
recovered more or less at will), and of creative imagination (which I
suppose can be identified with control operating on imagined states,
so perhaps that can be absorbed into "current perception").
I have this problem with a lot of what is written in psycholinguistics.
The authors assume that when people conform to the grammatical regularities
of their language, they "know" the rules. If that were the case, why has
it taken thousands (two, at least) of years of diligent enquiry, which
as yet has not achieved general agreement, to determine how the "rules"
of a language should be described? If everyone who "uses" the "rules"
"knows" those rules, why has it so far proved impossible to put them in
generally agreed terms for even a single language? Why is it a field of
professional academic enquiry?
People can say (often) whether a particular utterance in a particular
context disturbs a perception that people (should) utter things that
are appropriate in context, but they often can't say why an utterance
"feels wrong." The "knowledge" of language use is embodied, not "known."
It's not a part of one's perceptual universe unless one has been schooled
to make it so.
Agreements are remarkable achievements. On the one hand, it has been argued
that agreement is the most sound form of verification; on the other hand,
by our craving for agreement we seduce ourselves into absurd delusions.
The approach of PCT to both embodiment and "knowing" is that the (unknowable)
real world exists. It determines the effects of actions on perceptions.
The perceptions that are retained in reorganization are those that are
amenable to control, _and_ for which control results in side-effects
that keep intrinsic variables in a good state. The real world determines
which perceptions those are. It is not surprising that there is a bedrock
layer of perceptions about which we (almost) all (almost) agree.
The "real world" includes all manner of other living control systems (LCSs).
We control our own perceptions best when the effects of control by other
LCSs is at worst not disturbing to our controlled perceptions, and at
best protects our controlled perceptions from disturbance. It is not
_necessary_ that our perceptions agree in kind with those of other LCSs,
but if other LCSs do have perceptions that agree both in kind and in
value with our own, it is easier to establish each for the other what
_actions_ (i.e. the outputs of perceptual control systems) are likely
to ease each other's control. If to ease another person's control is
something for which you are controlling, then this communcation will
help.("You hold this post and I'll hit it with this mallet" is much
easier to get across if the imagined perceptions of "post" and "mallet"
and the imagined result of hitting are common to both people.) If
easing another person's control is something for which many people
control much of the time, the intercommunicating group is likely to be
reasonably stable. If it isn't, then the individuals are more likely to
reorganize until the society falls into a pattern in which it is.
We
deny the reality of someone's perceptions when we cannot replicate those
perceptions ourself, or cannot imagine doing so, and we have no way of
distinguishing when this is scientific verification and when it is the
stupid prejudice of pundits saying meteors cannot be of heavenly origin
because rocks obviously can't fall from the sky, so you must have been
wrong about what you saw, buster. Or when it is wishful thinking about
controlling others codified as theories of linear causation in psychology.
It makes no sense to deny that a person has a perception X, but it does
make sense to deny that perception X corresponds to anything in the
"real world." The perceptions of which you speak are those derived from
interactions among people, typically linguistic interactions which always
relate to the control of imagined variables (at the "program" level?).
Imagined perceptions can be controlled without reference to the "real
world." We can agree about such perceptions, easing communication, without
either party being able to test them against reality. Other people are
quite likely to have quite different perceptions, even ones to which the
same labels are given, and these imagined perceptions may behave quite
differently when controlled by the different individuals.
Control in imagination involves "environmental" feedback paths in
imagination. Unless those feedback paths involve only bedrock perceptions
based on feedback through the "real world," there's no reason to believe
that the same set of perceptions would be stable against reorganization in
different individuals. But within communicating communities, they are
likely to be, because if reorganization changes the perceptual function
in one individual, communication with the rest of the community becomes
more difficult (i.e. more error --> further reorganization). It's when
an individual from one community tries to communicate with an individual
from another that problems occur, especially in relation to perceptions
of different kinds that are lumbered with the same label.
The disagreements that we see on CSGNET do not bely my claim that we crave
agreement. Agreements at the level of our discourse are trivial in
comparison to the multitude of scarcely ever reconsidered assumptions we
need in order to talk at all, assumptions that we reaffirm in the course of
any cooperative endeavor, not least conversation. And of course the basic
existential affirmation:
Yes, I see you. I see you doing that.
I think this is what I am saying. But is it? I can't know, can I?
Martin