Coercion

[Bill_Powers1]

[From Bill Powers (980521.0835 MDT)]

Tim Carey (980521.1820)--

You're really trying hard to win this one, aren't you? Don't celebrate
too soon.

Nope. I'm not controlling for winning Bill, how would it come to be that
you would even frame up this discussion in terms of winning and losing?

That's how I interpreted your "yippee," to mean that you had caught us in a
contradiction and were celebrating the pleasure that gave you.

There are two ways to think about this result. You can say that the child
finally learned the right behavior and no longer needs to be coerced into
showing it, or you can say that the child is producing this behavior to
keep the coercer from forcing, yelling, disapproving, or hurting the

child.

The first view is the one that your idea of coercion supports: if the
adult is not actually applying force, yelling, disapproving, or hurting the
child, and the child is doing what the adult wants, no coercion is going
on.

And now who's putting words into other people's mouths ??

Wait a minute -- that's NOT what you mean? I thought you only counted it as
coercion when the person was actually applying some kind of force to the
coercee. I thought that if the coercee spontaneously did what the coercer
wanted, so the coercer didn't have to take any action, you would say no
coercion was taking place. If that's not what you mean, I guess I don't
have any idea of what you mean by coercion. Please explain.

It's YOUR definition of coercion that says a classroom full of well-behaved
children is not being coerced. Mine says that they are (or may be)
well-behaved only because the teacher is present, ready to throw them out
of the class if they misbehave. They are spontaneously doing what the
teacher wants them to do, and they are being coerced, in my view, even
though no force is being applied right at the moment. The force WILL be
applied if they ever should behave differently from what the teacher wants
to see, and they know it.

I have said all along that ONE of my criteria for coercion to exist would
be that the coercer had intentions for the coercee to behave in a
particular way. THE OTHER criteria is what the intentions of the coercee
are. If they have the same intentions as the coercer it's not coercion, if
they do then it is.

That's what I said. If the coercee intends the same behavior that the
coercer wants to see, the coercer will obviously not apply any force to the
coercee. You don't push a child back in line if he's already in line. So
when the coercer sees the child spontaneously doing what the coercer wants
to see (indicating that the child's reference level for the behavior is the
same as the coercer's), the coercer exerts no force on the child. I have
described exactly the same situation you just described, yet you say I'm
putting words in your mouth. Please explain what the difference is.

So in your example above you would simply ask the
coercee or apply some version of the Test to see what they were controlling
for.

I don't see what's so objectionable about that idea.

Who said anything about "objectionable?" You seem to be jumping to the
conclusion that if I say coercion is going on, I'm saying it SHOULDN'T be
going on. The only thing I'm objecting to is using coercion and then
claiming that it's not happening. Or, if that word is the hangup, using
force or the threat of force to overpower a weaker, smaller control system
in order to control its behavior and then saying you're not doing that. If
you're going to do that, don't claim you're not doing it. If you are going
to do it, analyze it correctly. That's all I'm after.

Under this picture, the child is behaving
correctly

only out of fear of what the adult will do if the right behavior is not
shown.

How do you know that the child is acting out of fear? There is a huge
assumption going on here that if someone is bigger and stronger then the
only reason anyone weaker would do what they wanted them to is out of fear.

This is a ridiculous question. If I say, "Suppose the 4-horse wins, what
should I do with the money?" do you reply, "How do you know the 4-horse is
going to win?" I'm saying that under my interpretation (that is, IF the
children know the teacher will apply force to them whenever they deviate
from a certain behavior -- do you understand "IF?") it is likely that the
children would be behaving well to keep the teacher from applying force to
them, not because they had somehow agreed that they liked behaving that
way. It does not follow from that that any particular bigger and stronger
person would threaten to use or actually use force on the children, so they
would feat him or her. However, a bigger and stronger person (i.e., an
adult) who DOES threaten or use force is likely to get compliance from the
children because they want (I am assuming) to avoid having force used on
them, which is what I mean by saying they "fear" it.

Please read what I'm saying carefully. I'm trying to say it carefully, and
you should read it the same way.

Best,

Bill P.

[bnhpct]

[From Bruce Nevin (980521.1414 EDT)]

Cool. The Test identifies a variable p(X) within the Tester that the Tester
perceives as being in the shared environment. Physics supports the
assertion that p(X) = qi a variable in the shared physical environment; PCT
supports the assertion that qi = another perceptual variable p(X) within
the tested control system. (This is what I meant when I said a successful
model is adequate warrant for that assumption.)

Bill Powers (980521.0924 MDT)]

Bruce Nevin (980521.0000 EDT)--

I thought the Test identified a perceptual variable p within the Tested
control system. I see now that it might identify a measurable variable qi
corresponding directly to p, but more probably it measures an environmental
variable as perceived by the observer which is presumed to correspond
directly to qi.

The variable qi IS a perception in the observer. It is only hypothetically
and theoretically a variable in the environment between organisms (the
theory being the set of physical sciences). And it only theoretically
corresponds to an unobserved perceptual signal inside the other person
(that theory being PCT). The Test does NOT identity a perception in the
other person. It identifies a perception in THE OBSERVER which the observer
is testing to see if it is under control by another organism. PCT explains
how qi can be under control by supplying a model claimed to represent the
internal organization of the other organism, the one acting on qi and
presumed (subject to test) to be sensing it. But none of the elements of
that model is directly observable (although, given technological advances,
they will some day be observable).

It is interesting that p(X) is a controlled variable within one control
system but controlled by another. When the Test identifies a controlled
perception, it is a case of applied coercion. The Tested control system is
the coercer overwhelming disturbances introduced by the Tester. The tester
might have the capacity to overwhelm the Tested system's control, but the
Tester volunteers, as it were, to be coerced. Having the capacity to
overpower another is not sufficient, to coerce one must also have the
intention.

  Bruce Nevin

[R_Marken]

[From Rick Marken (980521.1200)]

Bruce Nevin (980521.1414 EDT) --

When the Test identifies a controlled perception, it is a case of
applied coercion. The Tested control system is the coercer
overwhelming disturbances introduced by the Tester. The tester
might have the capacity to overwhelm the Tested system's control,
but the Tester volunteers, as it were, to be coerced. Having the
capacity to overpower another is not sufficient, to coerce one
must also have the intention.

Yes.

This insightful, clever and novel observation about The Test
shows me that you know how to apply PCT and that you know what
Bill and I mean by "coercion".

Thanks

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[Bill_Powers1]

[From Bill Powers (980522.0627 MDT)]

Bruce Nevin (980521.1414 EDT)--

It is interesting that p(X) is a controlled variable within one control
system but controlled by another. When the Test identifies a controlled
perception, it is a case of applied coercion. The Tested control system is
the coercer overwhelming disturbances introduced by the Tester.

Your comment shows me it's a little more complicated than that. It's this
ambiguous word "behavior" again. Do we mean the action (qo), or the
quantity that is manipulated by the action (qi)? Let's try this again using
only the language of PCT: qi and qo, controlled quantity and action.

(1) In the Test, the Tester's action applies a small force to qi, which the
tester can feel as muscle effort and skin pressure. If there is a control
system, the Tester can identify an action of the other person which changes
by enough to provide an equal and opposite force on qi, so qi does NOT
change appreciably. The Tester is not trying to control the
formally-identified qi. The Tester wants to see the other system's qo vary
as the applied test-force varies, and also to see that the Tester's action
has almost NO effect on qi. In the Test, the disturbing force or effect is
kept small enough that the OTHER system maintains good control of qi. The
Tester is really controlling for a feeling of resistance from the other
person, not for an actual change of the controlled variable. The tester
wants to feel some effort from his own muscles, but only a small amount.
That sense of effort tells the Tester that the other system is pushing or
pulling back. For other types of input quantities, the Tester wants to
perceive a modest amount of whatever is being used as a disturbing
variable, and specifically _not_ a change in qi. If qi changes
significantly, the applied disturbance is too large.

(2) In a conflict, the person who was the Tester in the previous paragraph
now wants to _control_ qi, not just apply a disturbing force to it. The
ex-tester has a reference level for qi picked independently of the other
person, so in general it will differ from the reference level for qi that
the other person has. This is the familiar simple two-way conflict, in
which qi will come to some virtual reference level and the two output
quantities will become very large and will oppose each other. We usually
assume that the people in the conflict are about equally matched in ability
to produce effects on qi via their actions.

(3) When coercion takes place, we have the conditions of a conflict as
defined in the previous paragraph, but the ex-tester is now able to produce
far more output (qo) than the other person's maximum output. If we watch
the onset of the interaction in slow motion, we see at first an ordinary
conflict. The first person's action (the coercer's action) applies a force
to qi, and the other person's counteracting force rises along with it, so
there is no net effect on qi. As the opposing forces increase, eventually
we reach the maximum force that the second person can produce. At this
point the opposing forces are still in balance, and qi is still unaffected.

Now we enter the coercion regime. The first person's output continues to
rise while the second person's output remains constant at the maximum that
can be produced. Now the first person's increase in action changes qi in
direct proportion to the excess of output over the other's maximum output.
This continues until qi matches the reference level of the first person.

This would be a good place to introduce the "universal error curve." It
seems, from observation in many circumstances, that when a living control
system is faced with a very large disturbance, larger than it can oppose,
it does not maintain the maximum possible effort it can produce (as a
simple servomechanism would), nor does it even try to do so. Before it
becomes exhausted, it gives up. It's as though the error signal caused by a
disturbance is a two-valued function of the actual discrepancy between
perception and reference signal. For small disturbances causing small
errors, the error signal rises as the increasing disturbance causes the
perceptual signal to depart from the reference signal in magnitude. The
output rises accordingly. This rise continues to the point where something
less than the maximum possible output is being produced. Beyond this point,
a further increase in the disturbance causes the error signal to flatten
out and then start to decrease again. As a result, the opposing output
action starts to _decrease_ as the disturbance continues to increase even
further and the actual discrepancy between perception and reference
continues to increase.

Depending on the characteristics of the output function and the
environmental feedback function, this reversal in the relationship between
disturbance changes and output changes may simply produce a decline in the
output, or the positive feedback that now exists may cause the output to
collapse to a level close to zero. To a small or a large degree, then, the
control system faced with a very large disturbance appears to give up, to
relax and allow itself to be completely overwhelmed. This can be explained
in various ways (for example, another control system that turns down the
loop gain when efforts to resist the disturbance reach a dangerous level),
but we can just take the universal error curve as an empirical phenomenon
and include it in the model of conflict and coercion.

The use of this universal error curve in the control system models makes
the model of coercion even more realistic. Suppose you challenge a child to
a tug of war. Assuming you can get the child even to try, when you start to
pull on your end of the rope, at first the child will pull back. But quite
soon your pull will exceed the maximum pull the child can (or is willing
to) produce, and the child will stop pulling and let go of the rope. Only a
very rare child would continue pulling as hard as possible while you drag
him or her along. Most children, when seized by the wrist and pulled along
by an impatient parent, will resist momentarily and then stumble along at a
much lower token level of resistance. It's not only humiliating to be
dragged along despite the largest effort you can produce, it's extremely
tiring. So this universal error curve and its deduced effects are
convincingly real, to me.

Having the capacity to
overpower another is not sufficient, to coerce one must also have the
intention.

Not the intention to overpower, of course -- just the intention that qi be
exactly at its reference level. That's all that's necessary for a control
system to produce as much force as needed. The force is produced by the
error signal and the output function.

What may well be intentional, at a higher level, is adjusting the
sensitivity of the output function so a small error will produce a large
action. One increases the importance of an error, as I have previously
defined importance. The standard HPCT model doesn't include the ability of
a higher control system to act by changing loop gain in a lower system, but
we've talked about it enough to assume that some day we'll have a working
model that behaves like this. If the loop gain is set high enough (and the
control system remains stable), enough force will be produced to assure
that the final error is small.

Best,

Bill P.

[bnhpct]

[From Bruce Nevin (980522.1323 EDT)

Bill Powers (980522.0627 MDT)--

Thanks. That clarifies the boundaries of your definition of coercion quite
a bit.

Has the universal error curve been measured in living organisms, or is it
an explanatory construct (a very elegant and simple one!) in models?

The coercer controls qi, which happens also to be qi for the coerced system.

I had thought that qi for the coercer is qo for the coerced system, that
is, I thought that the coercer simply aims to oppose the coerced system's
behavioral outputs. This was my interpretation of the statement that the
coercer has a reference level for the behavior of the coerced system. The
response to this is that at some point there is some qi for both systems.
If Janey is reaching for a pencil her qi has something to do with her hand
and the pencil by means of changing her arm configuration. If mother puts
her hand on the inside of her elbow, preventing her from reaching for (she
thinks) a cookie, her qi at that level is janey's hand grasping the cookie
(and her reference level for the corresponding perceptual signal is zero).
But descending the hierarchy of means to their respective ends, Janey's qi
and mother's qi is rotation of the upper arm from the shoulder (as part of
extending the arm).

The appearance that the coercer focusses attention on the specific outputs
that are a disturbance to a higher-level qi is an artifact of the two
systems both controlling some lower-level qi as means to their respective
higher-level aims.

I note that in the view that you have articulated, it is immaterial whether
or not the coercer perceives that the coerced system is controlling qi, or
even perceives that the coerced system exists. John and Billy approach
opposite sides of a swinging door. John perceives that the door appears to
be sticking a bit and pushes harder; Billy is unable to push through the
door. Though accidental, this is coercion nonetheless. In the view that Tim
is articulating, if I understand rightly, it would be necessary for the
coercer to perceive the coerced system and intend to overpower it.

I think it is these two preconceptions--that the coercer controls the
other's output, and that the coercer must perceive the other and intend to
overpower their output--that makes it difficult to understand that the PCT
account is adequate. Also clouding the issue, insofar as we feel that
bullying is blameworthy, there is the belief that retribution
(counter-bullying) is called for, and, due perhaps to one's imagining
retribution, denial of guilt.

  Bruce Nevin

[R_Marken]

[From Rick Marken (980522.1310)]

Bruce Nevin (980522.1323 EDT)--

I had thought that qi for the coercer is qo for the coerced
system

I think qo is typically a qi for a lower level system. So,
for example, when the mother tries to control Janey's arm
position (qo) while Janey is controlling for picking up a
pencil (qi) she is actually controlling a controlled variable
(qi) because arm position (qo) is a controlled variable (qi)
that is being controlled as the means (qo) of controlling
for the pencil (qi). Damn levels;-)

John and Billy approach opposite sides of a swinging door. John
perceives that the door appears to be sticking a bit and pushes
harder; Billy is unable to push through the door. Though accidental,
this is coercion nonetheless.

Yes.

In the view that Tim is articulating, if I understand rightly, it
would be necessary for the coercer to perceive the coerced system
and intend to overpower it.

I don't remember whether or not this was part of Tim's definition
of coercion but it was not the bone of contention. The bone was
Tim's contention that the coercee had to _feel_ coerced (or, at
least, had to have a reference for qi that differed from that
of the coercer).

I'm not quite ready to sign up to the idea that coercion cannot
be accidental (as in your door pushing example above). If coercion
could not be accidental (unintended) then the RTP teacher could
deny the use of coercion because coercion is certainly not
_intended_ in RTP. I think the RTP program itself is proof that
coercion occurs even when it is not the coercer's intention to
overpower (coerce).

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[Tim_Carey1]

[From Tim Carey (980523.0655)]

[From Bill Powers (980522.0627 MDT)]

This would be a good place to introduce the "universal error curve." It

Thanks for this post Bill. It was really helpful and your concept of the
universal error curve clarified for me where a lot of your previous
comments on the school system were coming from. I guess the universal error
curve phenomenon is what makes it so stuff to separate cooperation and
coercion between two people when there is a power imbalance with these
people.

Cheers,

Tim

[bnhpct]

[From Bruce Nevin (980522.2121 EDT)]

Rick Marken (980522.1310)--

I think qo is typically a qi for a lower level system. So,
for example, when the mother tries to control Janey's arm
position (qo) while Janey is controlling for picking up a
pencil (qi) she is actually controlling a controlled variable
(qi) because arm position (qo) is a controlled variable (qi)
that is being controlled as the means (qo) of controlling
for the pencil (qi). Damn levels;-)

I won't disagree with your restatement ;->

The bone was
Tim's contention that the coercee had to _feel_ coerced (or, at
least, had to have a reference for qi that differed from that
of the coercer).

You and Bill have both indicated how it is very unlikely for one control
system to have the same reference for qi (at all levels) as another, at
least not for very long; longer at higher levels of the hierarchy. If so,
then Tim gets that much for free.

Sometimes the tact of teaching is communicating the high-level aim and then
not coercing the lower-level means but waiting until, with error, the
student is trying to find some way that works better. That old "teachable
moment." Make sure qi is the same at the higher level, and don't enforce qi
at lower levels--yet. Some are better at following examples than others. On
the other hand, traditional pedagogy in Japan has the teacher e.g. grasping
the pupil's hand and arm and moving them in the correct manner of making
brush strokes; in which the pupil acquiesces as means of learning the right
moves.

I'm not quite ready to sign up to the idea that coercion cannot
be accidental (as in your door pushing example above). If coercion
could not be accidental (unintended) then the RTP teacher could
deny the use of coercion because coercion is certainly not
_intended_ in RTP. I think the RTP program itself is proof that
coercion occurs even when it is not the coercer's intention to
overpower (coerce).

What's more, you would have to say that "the system" (of regulations, etc.)
was doing the coercing of which the teachers were innocent.

Janie is coerced when she pulls her hand out of the cookie jar upon hearing
her mother's footstep in the hall. Her mother is controlling a perception
"cookies before dinner" with a reference level of zero, even though her
control of this perception is not currently being disturbed, since she has
not seen Janie snitching a snack. Janie controls the perception "cookies
before dinner" with a somewhat higher reference level. The interesting
thing is that, in order to imagine her mother controlling it at zero
reference level she must herself control it at zero (in imagination) in
parallel with her control at a higher reference level (through effectors
into the environment). This is interesting because it puts the coercion
inside her little head, and because it is maybe an insight into one kind of
learning.

We set different reference levels for controlled perceptions in different
social situations. When mother is around, cookies before dinner == 0, when
mother is not around the level is a few notches higher. Kids in a classroom
with the teacher present learn to control some perceptions that we might
call comportment at levels that are different from the levels that they set
for these perceptions when the teacher is not present, when they are on the
playground, or in the cafeteria, or on the school bus, or at home, or in
church, and so on.

Kids find out what is appropriate comportment by trial and error, of
course. Consider now when Janie takes a cookie before dinner, deliberately
doing so in a way that her mother can see. Maybe she forgot how close to
dinner it is, and she's hungry. Maybe she thinks it's not yet that close to
dinner time, and her mother will help her to calibrate her reference level
for that perception. Maybe she is being provocative (her mother might even
say this), that is, maybe she is Testing -- what is the reference level for
this perception, it seems to vary, what is the rule here, really?

Does this remind you of communications on CSG-L? It does me.

  Bruce Nevin

[Bill_Powers1]

[From Bill Powers (980523.0323 MDT)]

Bruce Nevin (980522.1323 EDT) --

Has the universal error curve been measured in living organisms, or is it
an explanatory construct (a very elegant and simple one!) in models?

It's a phenomenon, as far as I know. Even a dog on a leash shows the
giving-up effect when you drag it away from a delicious smell. I haven't
yet tested any model of it formally in an experiment. The mechanism of the
"universal error curve" could be an actual nonlinearity in a comparator or
output function, or an action of a higher-level system that reduces a
reference signal as a mean of avoiding attempts to produce amounts of
effort that might damage the organism. Or both. It seems to be a fact that
organisms give up when overwhelmed by disturbances -- at least higher
organisms. That's more or less the connotation of "overwhelm." The
universal error curve is just one possible way of modeling this effect.

The appearance that the coercer focusses attention on the specific outputs
that are a disturbance to a higher-level qi is an artifact of the two
systems both controlling some lower-level qi as means to their respective
higher-level aims.

That's how it works out, I guess. Coercion is most dramatic when the
controlled variable is something the person does or says, but I suppose it
applies to all controlled variables, like getting good grades (you're
grounded until you get a C) or learning to play the piano.

I note that in the view that you have articulated, it is immaterial whether
or not the coercer perceives that the coerced system is controlling qi, or
even perceives that the coerced system exists. John and Billy approach
opposite sides of a swinging door. John perceives that the door appears to
be sticking a bit and pushes harder; Billy is unable to push through the
door. Though accidental, this is coercion nonetheless. In the view that Tim
is articulating, if I understand rightly, it would be necessary for the
coercer to perceive the coerced system and intend to overpower it.

That's an interesting point. Consciousness of acting coercively doesn't
matter, of course, in the HPCT model, because there's nothing in the model,
yet, that acts any differently with or without consciousness. Yet when we
blame people for being coercive, I think we tend to assume that coercion is
being done knowingly, awaredly. It's hard to blame someone for something
that's done unconsciously. The whole question of consciousness is vexing to
me. Here we seem to have a model that works perfectly well without it, yet
in practice paying conscious attention seems to make a lot of difference to
our performance of a control task like driving a car. Why wouldn't it make
a difference in coercing someone, too? If it does make a difference, I
think we'd have to acknowledge that HPCT is incomplete without some sort of
account of consciousness. Heck, it IS incomplete -- this is a major
phenomenon that the model doesn't handle.

I think it is these two preconceptions--that the coercer controls the
other's output, and that the coercer must perceive the other and intend to
overpower their output--that makes it difficult to understand that the PCT
account is adequate.

It's a matter of levels, as I keep saying. If I grab your hand to keep it
out of the cookie jar, that's controlling your output relative to getting a
cookie, but it's controlling an input relative to controlling where your
hand is. If we stay within PCT terminology, there's no problem, because we
wouldn't have to use the common-sense term "coercion." We'd just say I'm
controlling the same variable you're controlling, but since I'm much
stronger than you, I determine its state and your actions have no effect on
it, relatively speaking. What one calls this situation doesn't matter. You
can call it coercion, or applying overwhelming force, or anything you like,
as long as we all know the specific relationship we're trying to indicate.
The bad effects of this situation don't follow from the name you give to
it, but from the relationship between the control systems and what the
control systems try to do about it.

Also clouding the issue, insofar as we feel that
bullying is blameworthy, there is the belief that retribution
(counter-bullying) is called for, and, due perhaps to one's imagining
retribution, denial of guilt.

Yes, exactly. But we mustn't discount the possibility that the state we
call consciousness does make a difference in how we we control anything --
even coercively.

Best,

Bill P.

[Bill_Powers1]

[From Bill Powers (980523.0410 MDT)]

Bruce Nevin (980522.2121 EDT)--

Janie is coerced when she pulls her hand out of the cookie jar upon hearing
her mother's footstep in the hall.

...

Very nice post, Bruce N. I could just feel the tension draining out of the
discussion as you laid out your analysis. What matters is the phenomenon in
all its interesting variations, not what category we want to put it in, or
what label we want to use for it. It's especially interesting that coercion
can, if used properly, become the occasion for voluntary learning. For the
child wanting the cookie, learning where Mom puts the limits may be more
important than getting the cookie at a given time. The child is coerced but
is also capable of seeing the sense in it. That moves coercion out of the
category of nasty things that are done to one and into the category of
natural phenomena.

That awakened a whole class of memories -- being coerced, but realizing
that the person doing it really thought it was best for me and wasn't my
enemy -- even if I didn't agree it was best for me. Going along with
misguided advice (temporarily) to get along.

Best,

Bill P.

[bnhpct]

[From Bruce Nevin (980524.1545 EDT)]

Bill Powers (980523.0323 MDT)--

Consciousness of acting coercively doesn't
matter, of course, in the HPCT model, because there's nothing in the model,
yet, that acts any differently with or without consciousness. Yet when we
blame people for being coercive, I think we tend to assume that coercion is
being done knowingly, awaredly. It's hard to blame someone for something
that's done unconsciously. The whole question of consciousness is vexing to
me. Here we seem to have a model that works perfectly well without it, yet
in practice paying conscious attention seems to make a lot of difference to
our performance of a control task like driving a car. Why wouldn't it make
a difference in coercing someone, too? If it does make a difference, I
think we'd have to acknowledge that HPCT is incomplete without some sort of
account of consciousness. Heck, it IS incomplete -- this is a major
phenomenon that the model doesn't handle.

I've been thinking today about one aspect of this. Let's say I'm engaged in
a tracking demo. Someone begins talking conversationally with me. As I
attend to what they are saying, I seem to loose the visual input for
tracking. My eyes wander from the target so that I am no longer perceiving
the relationship between target and cursor.

It seems as though I am using my visual perceptions in imagination mode as
part of the process of following the conversation, and there is
insufficient bandwidth for the tracking task.

This belies the principle of parallel processing. Perhaps it says there is
a limit to how many tasks may be done in parallel, or how complicated the
parallel tasks can be.

This says nothing about the subjective experience of attending -- PCT says
nothing about subjective, qualitative experience -- but it does speak to me
about the opposition between attention and distraction.

What do you think?

Spent much of the day visiting graves with my mother, in five of the six
towns on the Island (all but Gay Head, now renamed Aquinnah). Tomorrow the
graveyards will be too crowded. Don't know why the holiday was moved a week
earlier, maybe distancing the monday off from July 4 monday off, but an
unfortunate effect is that flowers were not quite so plentiful. A beautiful
day though. I hope that you are all enjoying a restful and reflective
weekend.

  Be well,

  Bruce Nevin

[Fred_Nickols3]

[From Fred Nickols (990502.0625 EDT)]--

Coercion is in the eye of the coercee, not the coercer.

···

--

Regards,

Fred Nickols
Distance Consulting "Assistance at A Distance"
http://home.att.net/~nickols/distance.htm
nickols@worldnet.att.net
(609) 490-0095

[Marc_Abrams5]

from [ Marc Abrams (990502.0756) ]

[From Fred Nickols (990502.0625 EDT)]--

Coercion is in the eye of the coercee, not the coercer.

Certainly one way of looking at it. But I disagree. Coercion is usually
only a _part_ of a number of behaviors that are going on at the same time.
We experience multiple perceptions, multiple reference levels, and hence
multiple behaviors. We are never "doing" only _one_ thing. This goes for
coercion as well. Coercion is done without regard to the coercee. But
coercion is not the only thing that is going on between the coercer and
coercee,. Counter control; mighty very well be taking place simultaneously.
This is pretty complex and not at all easy to represent in a model.

Marc

[Bob_Christensen]

[From Bob C.(980619.0107 PT)]

I have been listening to this thread on coercion and have a couple of sense to
put in of my own. First I agree with the suggestion to do away with the word
coercion all together. No matter how PCT gurus or websters define coercion,
to a large extent words have meaning according to how they are used, and
coercion has pretty undesirable connotations and associations. It does not
seem a fair characteization for all instances of when one person controls a
perception which another wants to control, and I can see why people are
disturbed by the word. Such control can be done under a huge variety of
conditions. I think it is overall better to use other alternatives besides
force, but I think it is perfectly okay to control what another desires or to
constrain their choices in many situations especially with kids, and I try to
be honest with myself and the other person when I do it.

I worked in a residential home for a year and a half with some very difficult
kids and adolescents. I could go on for a long time about why I thought
control was the best alternative in a given situation, but mostly it came down
to situations where the youth did something pretty inconsiderate or harmful to
another person or to themself. We used to run a process of first confronting
the youth with specifically what was problematic about their behavior, then
asking them to take a break on their own, and warning them about the use of
force if they did not.

It seems that coercion is an appropriate word choice when someone's desires
are reasonable and some self serving powerful figure is exerting control and
limiting freedom. It seems inappropriate when someone has unreasonable or
inconsiderate motives. For example, what if someone has an obsessive
attraction to another person, a restraining order is placed on them, and the
police need to use force to remove the person when they break the order? Who
would characterize the police as coercive rather than protective? If a parent
does not buy a child every toy at a store, who would call the parent coercive
rather than merely setting a limit?

As far as RTP and working in the school system goes, I think honesty is the
primary issue. I think that it is best to just be honest about what options
are okay, what are not, and what will be done if a kid choses the latter. If
someone does not agree with the constraints of a school system then don't work
there or work to change the system in a tactful way.

I think PCT should stick to more simply descriptive/neutral terminology.
What's wrong with the word force by the way?

Bob C.

[KENNETH_J_KITZKE1]

[From Kenny (980619.1100 EDT)]

<Bob C.(980619.0107 PT)>

<First I agree with the suggestion to do away with the word coercion all
together.>

Perhaps there is a better word we could all agree on, but the concept of
coercion is fundamental to accepting PCT. For there is one exception to
"behavior is the control of perception." When some thing or some one
actually forces a behavior on you, regardless of what perceptions you want,
your behavior is not the control of your perceptions.

The assumption is that living control systems vary their behavior to
perceive what they want (the level of their internal refererence variable).
When an LCS is prevented from varying its behavior, what is happening is
not "the control of your perceptions" but "coercion" or whatever word you
want to use.

<No matter how PCT gurus or websters define coercion, to a large extent
words have meaning according to how they are used, and coercion has pretty
undesirable connotations and associations.>

True. But, Rick seems to believe that PCT must recognize coercion when it
does occur. And, under any of the definitions, it occurs more often than
what people would like to admit.

Can understanding PCT help us reduce the amount of coercion our behavior
generates? If so, then there seems to be value in having an understanding
that we can all focus on.

<What's wrong with the word force by the way?>

I personlly like the definition of coercion as when a person is forced to
behave against their will (their perceptions are irrelevant to their
behavior).

Kenny

[R_Marken]

[From Rick Marken (980619.1420)]

Bob C.(980619.0107 PT) --

Really nice post Bob. Very clear and well-written.

First I agree with the suggestion to do away with the word
coercion all together. No matter how PCT gurus or websters
define coercion, to a large extent words have meaning according
to how they are used, and coercion has pretty undesirable
connotations and associations.

This is a good point. But I realize now that the same thing
can (and has) been said about the word "control". There
have been suggestions that we rename "control theory"
because "control" has negative connotations; it sounds like
the theory is about how to control other people. Bill once
wrote a little paper called "the X phenomenon" (I think it's
reprinted in one of the LCSs) which describes the phenomenon
of control without using the word. As I recall, the paper
ends by consluding that the English word "control" is really
the only word that refers to the exact phenomenon (the X
phenomenon) that control theory explains. So we continue to
call PCT "control theory" because it is a theory of the
phenomnon that English speakers agree is pointed at by the
word "control".

I have found that, once people start understanding control
theory, the negative connotations of the word "control" vanish
for them; indeed, people start seeing that there are also
_good_ connotations of the word "control"; most people see that
it good to be "in control" of what matters to them, like the
positon and speed of their car.

I now think that the same is true for the word "coercion".
This word refers to the phenomenon that Bill and I are
calling "coercion". We could call this phenomenon "the Y
phenomenon" instead of "coercion" and describe it in terms
of the equations of control theory. But when we try to find
a name for this Y phenomenon, I think we will see that the
best name is the English word that already exists to describe
the phenomenon: "coercion". So why look for other nanes?

Coercion is just one kind of control phenomenon. Like
control, it is neither good nor bad in itself. It is just
something that people do to each other. Once people can
recognize coercion (just as they can recognize control -- by
looking through PCT glasses, also known as The Test for the
Controlled Variable) then they can see it as a phenomenon
rather than as a moral indictment. In fact, they will be
able to see that in many cases, coercion seems like a good
thing. You gave some nice examples of this in your post:

I worked in a residential home for a year and a half with
some very difficult kids and adolescents. I could go on for
a long time about why I thought control was the best
alternative in a given situation, but mostly it came down
to situations where the youth did something pretty
inconsiderate or harmful to another person or to themself.

Right. Coercion seems good when you are using it to prevent a
behavior that seems bad. So coercion (like control) is not
just a bad thing. In fact, coercion (like control) is neither
good nor bad; it is just something people do to each other.

It seems that coercion is an appropriate word choice when
someone's desires are reasonable and some self serving
powerful figure is exerting control and limiting freedom.
It seems inappropriate when someone has unreasonable or
inconsiderate motives.

But the same thing is happening in both cases; in both cases
a stronger person is controlling the behavior of a weaker
person. If you call this one thing ("teaching responsible
thinking", say) when you think the stronger person is justified
and another thing ("coercion") when you think the stronger
person is not justified then (I think) you are just trying to
put spin on the phenomenon. It's coercion in both cases.

The same goes for "control". We don't use diiferent words
to distinguish the control we like from the control we
dislike. Control is control, whether a person is controlling
the position of a light switch or the position of the doomsday
switch; controlling the mixture of ingredients for a cake or
for a bomb.

I think PCT should stick to more simply descriptive/neutral
terminology.

I think you can see now that this is hard to do. "Control" is
a purely descriptive term but it is certainly not neutral.
Some people will not even consider the merits of a theory
that is named after something as horrible as "control". So
what do we do? If we change the name of the theory to
something neutral ("behavior theory"?) it won't necessarily
be very descriptive.

What's wrong with the word force by the way?

I think it only describes part of what is involved in coercion.
Calling coercion "force" is a bit like calling control "maximum
output". Coercion (like control) is a process; the coercer
continuously compares percieved to reference behavior and acts
to keep perceived behavior matching the reference. Most of
the time this may require the use of no force at all (as in
the case of the lifeguard, when no one is breaking the rules).
So I still think "coercion" is the best word to use to refer
to coercion; "control" is the best word to refer to control;
"love" is the best word to refer to love...

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[Marc_Abrams5]

From [Marc Abrams (980619.1819)]

It just hit me folks. It wasn't until Bruce Nevins' post
(980619.1624) that i realized what was happening with the coercion
thread.

Both Bill and Rick stipulated that "coercion" was the result of one
control system "overwhelming" another. I believe this to be accurate.

On the other side of the court (i.e. Tim Carey, Isaac, Bruce Nevin,
etc.) The question was raised, What about the victim? What _happens_
to him/her?

Isaac pointed out the fact that in an _interaction_, _two_ people are
involved. How can you _define_ coercion when you are only defining
what _happens_ to _one_ person. Good point Isaac.

So although "coercion" is _done_ by one person and the second is
"irrelevant" _relative_ to the first. It _affects_ all parties. Bill
and Rick have focused on the coarser. So although the coercee is
irrelevant, it is not unimportant or insignificant

Marc

[R_Marken]

[From Rick Marken (980619.1830)]

Bruce Nevin (980619.1937)

coercing is precisely what PCT theorists have repeatedly claimed
that one control system cannot succeed in doing to another.

I think I once said that people could not be controlled (coerced)
some time ago. I know now that I was wrong; people certainly
can be controlled. They might be able to change their goals to
feel good about being controlled -- but that doesn't change the
fact that their behavior is being controlled (coerced). Coercion
doesn't go away if PCT is true. Sorry.

WMVY, a local radio station...
You are telling me that the fact that I don't give a shit is
irrelevant.

You got it!

The fact that I am perfectly content to let them purvey their
music programming and their advertising to whomever wants to
listen has no bearing on the matter

"Purveying their music programming" is _not_ coercion. "Purverying
their advertising" is not coercion. They are not forcing you to
listen. The FAA is coercive because it will put your sorry ass in
jail if you try to jam their signal. That's coercion (the kind
I like;-)) whether you mind having your ass hauled into jail or not.

Description and modelling of the coercer or compeller is simple.
Description and modelling of the coerced or compelled is more
complicated, and much more interesting.

How do know?

Why evade it?

Who's evading? All I'm interested in is knowing what the RTP
teacher is doing. We can get to the kids later. I think the RTP
teacher is _required_ to coerce and does so, whether the kids
want to do what the teacher wants them to do or not.

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

[bnhpct]

[From Bruce Nevin (980619.1937)]

Rick Marken (980619.1420) --

[Regarding the repeated suggestion not to use "coercion" for what the
proposed model of a coercer does]

This is a good point. But I realize now that the same thing
can (and has) been said about the word "control".

"Control" is the core concept of PCT. There are strong reasons for using
this word under its technical PCT definition, not least of which is the
fact that every less formal sense in common usage has a clear explication
in terms of the phenomenon of perceptual control and therefore has a clear
and direct relationship to the technical definition in PCT.

"Coerce" is not a core concept of PCT, nor is it entailed in the course of
defining control or PCT. Senses in common usage cannot be reduced to the
proposed technical meaning, and it includes situations that are not
accepted in common usage. In fact, coercing is precisely what PCT theorists
have repeatedly claimed that one control system cannot succeed in doing to
another. This claim, which would seem to make it a problematic rather than
a core concept, is based upon what takes place in the control system that
is subjected to coercion. The process of the subordinate control system has
been reduced, in this proposal, to the claim that whatever it intends is
irrelevant. The suggestion seems to be that the coercive state of affairs
persists indefinitely without change. These things seem to me to be
contrary to experience. I have given many examples illustrating why; I
won't repeat them here.

I have no objection to using the term "coerce". I do expect us to identify
the phenomenon to which the word refers and then to model it. What has
happened is that a model has been created of an overpowering control system
(originally with a different purpose), the observation was made that what
the model does looks like what a coercer does, and this grew to the claim
that the relationship called "coercing" is what is being modeled.

My objection is that there is no place in this model for what, to me, is
the most interesting aspect of the relationship called "coercing", namely,
what is going on in the control system that is subjected to insurmountable
disturbances produced by another control system. I have been calling this
the victim, and that's OK as long as we understand that it doesn't
necessarily mean that the coercion is successful.

What does it take for the coercion to be successful? From the coercer's
point of view, only that it continue to control successfully. That is the
point of view taken by you, Rick, and by Bill.

WMVY, a local radio station, determines how a certain RF band is frequency
modulated. Everyplace I go here where I live the space around me is
modulated in this RF band in ways that they determine. If I tune an FM
radio to that band I hear music that they have picked. Their modulations of
this attribute of our shared environment surrounds me even here in my home,
even when I go to sleep, it even penetrates and suffuses my body. In
principle I could build my own radio transmitter and stay close to it, or
carry it with me, or I could enclose myself in a Faraday shield (I think
that would work), or create some other kind of interference, but those
steps are really beyond my ready and practicable means. In any case, I am
not doing any of those things. If I ever object to their taste in music
being imposed on all my fellow Islanders sufficiently (and somehow get the
means) to build a more powerful station, I will find out that WMVY
participates in a coercive system including the FCC and federal law
enforcement officers. Therefore, by your definition I am being coerced in
this respect. They are imposing their will on an aspect of the environment
and I cannot effectively counter it. They are stronger than I am.

You are telling me that the fact that I don't give a shit is irrelevant.
The fact that I am perfectly content to let them purvey their music
programming and their advertising to whomever wants to listen has no
bearing on the matter, I am being coerced.

This seems to me to be an absurdity. It is contrary to a fundamental
element of the meaning of "coerce", namely, that it be against the will of
the one who is coerced.

You may be right, that it is sufficient that the will of the subordinate
one be nullified--that it not matter what the lesser one wants, or whether
the nullified will be contrary to the superior, or (and I hold still, and
am still unanswered) even whether or not there be anyone at all present to
be coerced, so long as the coercer stand ready to suppress, and
effectively, any opposition. Do you really affirm all these absurd (to me)
conclusions that follow from your sparse premises?

I agree with you that it does not matter whether or not the intentions of
the coercer are just (or justified by some social standard), whether or not
the suppressed intentions of the victim are good ones or bad ones, and that
coercion ought to be described in morally neutral terms of perceptual
control. "Compel" is an alternative word that might avoid some of the
emotional freight people may associate with "coerce". But it still
describes a relationship between control systems, not an attribute of one
strong control system. Description and modelling of the coercer or
compeller is simple. Description and modelling of the coerced or compelled
is more complicated, and much more interesting. Why evade it?

  Bruce Nevin

[bnhpct]

[From Bruce Nevin (980919.2127 EDT)]--

Marc Abrams (980619.1819)

So although "coercion" is _done_ by one person and the second is
"irrelevant" _relative_ to the first. It _affects_ all parties. Bill
and Rick have focused on the coarser. So although the coercee is
irrelevant, it is not unimportant or insignificant

It is clear that during coercion (however long that lasts) the thwarted
intentions of the coerced system are *ineffective*. Whether they are
*irrelevant* depends on what the coerced system does. As Bill reiterated,
the coerced system may find other means to control an intention. The first
means are ineffective because of coercion; the shift to effective means
ends the episode of coercion, even though (by Bill and Rick's definition)
coercion is still going on with respect to the first means.

The circle drawer may use the other hand, ask someone else to draw the
circles, etc. More interestingly still, the circle drawer may provide the
coercer a sun shade so that she no longer commandeers the circle drawer's
arm for that purpose. The victim may find out what the coercer's intention
is and may be able to substitute other means for the coercer controlling
that intention.

  Bruce Nevin