[From Bill Powers (930504.0900 MDT)]
Bruce Nevin (930503.1252) --
But we assume that we do in fact have access in the present to
something of the past. Just as we assume that we have access to
something of the present "boss reality" in our present
perceptual universe, whatever its distortions. The
justification and caveats appear to me to be identical in the
two cases.
Well, let's think about that some more. It seems somewhat
different to say that our present-time perceptions may or may not
reflect actual regularities in the current environment, and to
say that the past may or may not actually affect our present-time
operation. I can see, in principle, that perceptions might be
regular functions of physical variables just as we assume they
are. But I can't see even in principle how the past can affect
the present. The temporal problem adds to the other one.
The perceptions that constitute the interpretations, attitudes,
expectations, stories, and so on, are not all created de novo
by each human being. They are learned. They are taught by
example even more than by overt precept. And they are
historically contingent, as I was saying.
But what is learned is not necessarily what is taught. I think
that my main objection is somewhere in this vicinity. Learning is
an active process that requires each individual to make an
internal effort, to call into play the internal functions that
are capable of altering organization. We are not "taught" by
someone simply pouring information into us. We must drag the
information out all by ourselves, find a way of make sense of it,
decide how to alter our own goals and actions to make a place for
what we think we understand in what someone else evidently wants
to teach us.
I do think, by the way, that the perceptions are learned de novo.
There is no other way to learn them. You have to construct
perceptions and play around with them until controlling them
doesn't produce any more objections from others. There's simply
no way to compare your perceptions with mine. Every person has to
construct a perceptual world from scratch; good thing we have the
equipment built in for doing this.
Other people present us with situations, not examples. We have to
decide for ourselves what there is about the other's behavior
that we might TAKE as an example: is it my father's wise words,
or his impatient and condescending way of delivering them that I
should choose for a reference level? A lot of what we learn to DO
has nothing obvious to do with what we are learning to CONTROL.
In any given environment, there are many actions that can serve
to control any given perception. Some of these actions are easier
to produce than others, some are more effective in achieving
control than others. We find out experimentally which actions
work best in the current environment, which includes the other
people in it. But the actions that others see are not the
variables we are using them to control. We don't care about the
actions; it's their effects in ouor own perceptions that we care
about. What the parent sees as teaching good table manners we see
as a way of avoiding being criticized. Who cares if the fork goes
on the left side of the plate? But we all care about feeling
humiliated in front of company.
... if we care about cooperative relationships with those
around us we will do so with circumspection. I think this is
in part what has been meant by suggestions like "be in the
world but not of it."
Yes, exactly. We are all "in" the world in that we learn the
outputs required to get what we want while not producing
conflicts with others. But none of us is "of" the world, because
what we get out of learning these things is not visible to anyone
but ourselves. The rat in the Skinner Box has no interest in bar-
pressing; only in what the bar-pressing causes. But the
experimenter, to whom a handful of pellets is worth nothing at
all, cares only about the bar-pressing. When the experimeter
speaks about what he has taught the rat to do, he doesn't talk
about teaching the rat to get pellets for itself. He thinks he
has taught the rat to press a bar. But the rat will continue to
press the bar only as long as it keeps producing the reward;
pressing the bar in itself has no value for the rat.
... any reference perception that is socially standardized or
calibrated by individuals for conformity to a community norm
is, by the very processes
of teaching/learning/acquiring/calibrating over time,
historically contingent.
It seems to me that you're focusing on effects and treating them
as causes. Learning to behave is caused by processes inside each
organism. The effect of this learning is for the organism to
produce actions, outputs, which are effective in controlling the
things that matter to the organism. It is the environment that
determines what actions will be effective; that is how other
people get into the process of learning. But they can only
require certain outputs; they have no influence over what inputs
a person must have in order to survive and feel well.
It's true that the customary outputs are historically contingent,
but this doesn't mean they're caused by history. It is to say
only that there was a world before this one, and a world before
that world. If there were any shaping or driving force in
history, deviations would be corrected, but they're not. If one
generation learns to produce actions that are a little different
from those of the previous generation, the pattern simply
changes; there is no tendency to restore the former pattern. The
forms of action that customarily work in producing what people
want change over time, drifting aimlessly from one set of forms
to another of equally little intrinsic value. This is true of all
the forms of action except those that involve the unchanging laws
of the physical universe. There is no historical imperative for
eating with a fork. But there is an imperative saying that if you
want to eat with a fork, you must grasp it strongly enough so
friction will keep it in your hand.
What is preserved over the generations is the set of things that
people actually do value, the things they must value to survive.
Those things are not self-evident in social customs. If they
were, PCT wouldn't be news to anybody.
ยทยทยท
---------------------------------------------------------------
Wolfgang Zocher --
Don't apologize. Your struggle with English is seldom visible in
your writing, and you have one thing to be very proud of: you are
not a monolingual dummkopf like me. I really admire all of you
people out there who can converse in English, a tongue not your
own, and are considerate enough to do so for the benefit of
uneducated foreigners.
I have received Simcon 4.5 and will check it out today. I'll
change the descriptive writeup and send it to you to see that I
have correctly understood how the new commands are to be used.
I'll also try to add the feature that Gary Cziko wants, the
labeling of curves for monochrome displays. Gary, what I'm
thinking of is pretty simple, but will save a lot of programming.
The idea is to move a vertical cursor line along the display with
the left and right cursor keys, so the user can pick places where
the curves are separated enough for clear labeling. Then by
pressing the up-down keys, the user can select variable names
from the list at the upper right of the screen to be placed on
the display. This uses the brain's capacities which would take
yards -- excuse me, meters -- of program to accomplish. It will
take a minute to place the labels, but the result will be
reliable.
-------------------------------------------------------------
Best,
Bill P.