[From Rick Marken (930324.0900)]
Bruce Nevin (Wed 930324 08:44:38) --
It is possible to know what someone is doing by watching what
they are doing, when they are doing something that is socially
standardized.
When I say that you cannot tell what a person is doing by watching
what they are doing, I mean that you cannot tell what perception a
person is controlling by watching the outputs (or observable side
effects of the outputs) that are involved in the control of those
perceptions. This applies whether or not these outputs (or side
effects) are considered "socially standardized" events or not. A
concrete example: can we agree that "stopping at a red light" is
a socially standardized event? Your statement above suggests that
if I see someone stop their car at a red light I know that this
event is "what they are doing" -- by which I mean, the intended
result of their actions or (in PCT) the perception that is under
control. But it is certainly possible that the person has not noticed
the red light but, instead, has seen a robbery in progress in the
store behind the light. Coincidentally, just as the light turns
red, driver stops the car -- not to produce the socially standard-
ized result "stop at red light" but to produce the socially quite
non-standard result; try stop the robbery.
The meanings (nonverbal perceptions) that I associate with the
words and words-in-relation may differ from those that you
associate with them, and probably do. Nonetheless, we assume
that the differences should be matters of detail.
This is a big assumption. You might make it; but it seems clear
to me (especially from these net discussions) that the differences
in the meanings (non-verbal perceptions) evoked in different people
by the very same words can be quite a bit more than a matter of
detail.
(This is the
fundamental assumption of science: that the universe is knowable,
consistent from one part of it to another and from differing
points of view.)
I don't see what this assumption has to do with language. I think
the success of science results from its non-dependence on language;
agreement comes from experimental demonstrations -- finding that
you can reliably produce particular perceptions by operating on
the world in a particular way. We do have to try to communicate
(using language, to some extent) how to operate on the world in
order to produce particular perceptions reliably -- anyone who has
taken a science lab knows how difficult this communication process
can be. But once one is able to translate the linguistic descriptions
into appropriate (non-verbal) control systems in oneself (if one
is ever able to do this, and this is not guaranteed -- like me in
chemistry lab) then one can stop talking and start doing science.
This works not only in science but in others areas where language is
used (at least in the beginning) to try to communicate control
skills -- like in teaching a kid how to drive or do math. The short-
comings of language become quite apparent in these situations -- and
we sometimes resort to physical demonstrations and what not.
When you are using language, you are producing outputs as part of
the process of understanding what was said.
I would say "I'm producing outputs as part of the process of
producing perceptions that match my references for those
perceptions (perceptions corresponding to "understanding" among
them, perhaps); the references are the embodiment of 'what I
want to say'"
There is no need to
invoke references for uncontrolled perceptions to account for
your discomfort at someone saying things about a cherished
concept that you would regard as error if you said them yourself.
I hope my diagram cleared up why they are necessary. It is also
worth recalling that, if a perception as under a reasonable
level of control, there is not much error. I can't control my
perceptions of what others say and these are the perceptions that
tend to be "disturbing" -- in the Tuckerian sense, that is, they
result in error. I imagine that this is because I have a reference
for these uncontrolled perceptions.
And I hardly think it is the case that your concept of PCT is an
uncontrolled perception for you!
I can only control what I can affect via the environment. I can
affect what I perceive myself saying about PCT; so I can
control my concept of PCT as it results from my own actions; but
I can't control what I perceive to be your concept of PCT.
Having a reference for a perception is not enough to make it
a controlled perception. It must also be true that the
perception is a joint function of the environment (d(t)) and
my effects on that environment (o(t)). [After reading your next
post I have had to amend this and admit that, it is possible
to control when o(t) is imaginary].
Bruce Nevin (Thu 930325 08:36:48)--
I am glad you read the first sentence of my (Wed 930324 08:44:38).
Any agreements or disagreements with the rest of it?
Both. See above.
On your revision of my diagram; I see is that you have removed
the effects of my outputs on the environment.
What is indicated here is a "shadowing", in imagination, of the
other's utterance in the environment.
I think you would also have to change the controlled variable, p(t),
to something like "g(o(t) = g(d(t))"-- that is, the higher
level variable is controlling for "shadowing" by generating
an imaginary perception (o(t)) to match what is being said, d(t).
I suppose this is a good exmaple of controlling completely
through the use of imagination. So I guess I am wrong about
control only occuring when the system has an effect on the
environment. Here is a case where p(t) = imagined(o(t)) + d(t).
So I take back what I said in the paragraph above -- I was wrong
(how's that for non-arrogance). You can control using only
imagination.
There is a problem indicating the relation between f[o(t)] and
meaning[o(t)] with a labelled arrow if that means, as it usually
does, a relation between two levels of the perceptual hierarchy.
I was obviously skipping a lot of steps is the hierarchy. The
point is that somehow the nervous system converts strings of
written or spoken words into perceptual signals; at one level
these signals a preceptions of the word patterns themselves. At
other levels they are perceptions of what is represented by
the word. My little perceptual function boxes collapsed many
layers of perceptual processing into a simple input -- output
function; words in -- meaning (in the form of a variable magnitude
perceptual signal) out. What that signal "means" is embodied in
the function that converts words into perceptual signals.
Does ANYBODY get it yet, or will this be ignored too?
I don't know what I was supposed to get? But I think I now agree
that people can control variables using imaginary output. I just
think that if they do it alot they might run afoul of their
intrinsic references.
Best
Rick