irrelevance of victims

[bnhpct]

[From Bruce Nevin (980615.1209 EDT)]

Boxed in by logic, is what it looks like.

Rick Marken (980612.0820)--

Here is a diagram of what I think of as coercion:

             r
             >
       p --> C -->
       > > Coercer
      >s> >o>
----------------------------
       > qo
       > >
       qi<-------- Environment
       >
       qo' (d)
----------------------------
      >o'| Coercee

------------------------------

qi is a variable controlled by both the Coercer and
the Coercee. The Coercee's effect on qi (via qo') is
just a disturbance to qi from the point of view of
the Coercer. The same is true for the Coercee; the
Coercee's effect on qi is just a disturbance to qi from
the point of view of the Coercee. Since the Coercer is
much stronger than the Coercee, the Coercer keeps
qi under control; the deprives the Coercee of any
control over qi. Note that this analysis requires no
knowledge of the intentions (r') of the Coercee. All
we have to know is that _both_ Coercer and Coercee have
effects on qi _and_ that the Coercer is much stronger
(higher output capability) that the Coercee.

You have said that coercion is going on even when the victim is compliant
or has given up, or even when the victim doesn't care about the state of qi
one way or another, since if it did ever attempt to affect the state of qi
the coercer would overpower its outputs. So o' and qo' are irrelevant too.
That's good, I suppose, because if you include o' and qo' then you have to
include some r'. (You could derive the value of r' from qo if you could
factor out all other sources of disturbance. )

So all you need to model coercion is a model of the coercer plus the
stipulation that its outputs are strong enough to overpower all comers. Put
a black hat on your model and lean it against the bar. "I can lick any
control system here!"

Is there victimless coercion? According to Bill, merely having output power
strong enough to coerce is victimizing. The presence of weakness makes one
coercive.

There are some alternatives that require thinking on more than one level. A
victim who is studiously complying with the perceived intentions of the
coercer, or studiously avoiding disturbing the coercer's control (not the
same thing) is doing so as means of controlling other perceptions. Perhaps
doing so does not disturb any perceptions that the coercer is controlling
(avoiding trouble). Perhaps it would or will ("passive-aggressive
behavior," guerrilla warfare).

Normally, the thwarted control was means for some other end, and the victim
may be able to find other means. If you fail to model what the victim is
doing, aren't you missing something? At the least there's the birth of new
motives. Finding alternate means. Getting even. Coercing others in turn.
And so on.

In the rare but real case where a coercer aims merely and precisely to
thwart the intentions of the victim you must model the victim as well as
the coercer.

   Bruce Nevin

[R_Marken]

[From Rick Marken (980615.1440)]

Bruce Nevin (980615.1209 EDT)

You have said that coercion is going on even when the victim is
compliant or has given up, or even when the victim doesn't care
about the state of qi one way or another, since if it did ever
attempt to affect the state of qi the coercer would overpower
its outputs. So o' and qo' are irrelevant too.

qo' is _always_ relevant; it is a disturbance to qi, the
variable controlled by the coercer. A disturbance is still a
disturbance, even when it's value happens to be 0.

That's good, I suppose, because if you include o' and qo' then
you have to include some r'.

Why? qo' is a disturbance to qi. Why do I have to say more? When
I'm modeling a person controling the position of a car (qi) on
a gusty wind (d) day, I don't have to specify the climatic
conditions that cause the wind to blow (or not), do I? The same
applies with coercion; coercion is control of a behavioral
variable (qi) that is influenced by the outputs (d) of another
control system.

So all you need to model coercion is a model of the coercer plus
the stipulation that its outputs are strong enough to overpower
all comers.

Exactly.

Is there victimless coercion?

Sure. If the coercer happens to be forcing qi to exactly the
state you want, then your not a victim; you're one lucky guy.

According to Bill, merely having output power strong enough to
coerce is victimizing.

Not at all. Bill very clearly said that coercion is only possible
if you are strong enough to coerce; but just being strong enough
does not mean that you _will_ coerce.

If you fail to model what the victim is doing, aren't you missing
something?

Sure. You're missing an understanding of the victim. But this has
nothing to do with understanding that the coercer is coercing. You
can see that coercion is happening without knowing anything about
the victim except that it is a living control system.

In the rare but real case where a coercer aims merely and precisely
to thwart the intentions of the victim you must model the victim as
well as the coercer.

To do this, the coercer must model the victim. But we still don't
need to model the victim to see that the coercer is coercing.

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 (980615.2235 EDT)]

Rick Marken (980615.1440)--

[...] if you include o' and qo' then
you have to include some r'.

Why? qo' is a disturbance to qi. Why do I have to say more? When
I'm modeling a person controling the position of a car (qi) on
a gusty wind (d) day, I don't have to specify the climatic
conditions that cause the wind to blow (or not), do I? The same
applies with coercion; coercion is control of a behavioral
variable (qi) that is influenced by the outputs (d) of another
control system.

qo' is not just any disturbance, like a gust of wind. You would not say
that the car driver is coercing the wind. Why? Because the wind does not
intend to influence the direction of the car. For you to model coercion you
must model the presence of thwarted control. Otherwise you cannot
distinguish between coercion and resistance to inanimate disturbances.

So all you need to model coercion is a model of the coercer plus
the stipulation that its outputs are strong enough to overpower
all comers.

Exactly.

You can only stipulate this with respect to the output capacity of the
victims. In addition, for any but the most simple brute-force forms of
coercion, the skill and cleverness of the would-be victims might be
relevant. No, even for brute force: it is not by superior output capacity
(muscle strength) that an expert in martial arts maintains control in the
face of a would-be coercer. "If he wants to occupy that space, I move and
let him occupy that space. If he happens to be falling, I let him fall." Is
that coercion? Is the falling brute a coercer because of his superior
strength? By your logic he is a coercer despite his failure. His output
capacity is far greater than that of this wiry old man (with the black belt).

In the rare but real case where a coercer aims merely and precisely
to thwart the intentions of the victim you must model the victim as
well as the coercer.

To do this, the coercer must model the victim. But we still don't
need to model the victim to see that the coercer is coercing.

Without the presence of countervailing control (thwarted) you can only
demonstrate control. Coercion is not simply control in the face of
disturbance. It is control despite conflict. Like conflict, coercion is a
relationship between more than one control system.

The claim that the teacher is coercing even when the students are obedient
is the box forged by specious logic. The teacher is controlling, resisting
disturbances. When the resisted disturbances include countervailing control
by a student, or countervailing side effects of control by a student, then
the teacher's control becomes coercion. The claim was that the teacher's
readiness to coerce and the readiness of others to coerce in support of the
teacher makes the school a "coercive system." That may be; it is an
otherwise undefined term. When you are driving a car, resisting bumps,
curves, and gusts of wind, you are not coercing. If the bumper is pressing
against a pedestrian and you press the accelerator, you are coercing.
Likewise if you squeeze your car in front of another, confident that the
other driver would control for avoiding an insurance-liable rear-end
collision, that is coercion. The fact that you *can* do such things with
your car does not mean that you are always coercive whenever you drive. Not
even if a pedestrian is standing with his leg touching your bumper.
Coercion is not an attribute of the coercer, it is a relationship between
the coercer and a victim of coercion.

  Bruce Nevin

[R_Marken]

[From Rick Marken (980617.1530)]

Bruce Nevin (980617.1714 EDT)

Why does r2=r' vary over time? Wouldn't you say it is because of
some higher level of control?

Yes.

The victim's control at that higher level is being coerced too,
isn't it?

No.

Me:

can you at least see that the lifeguard (like the RTP teacher)
is _controlling behavior_ -- and is doing so even when there
is no resistance to that control?

Bruce:

Then you have to include in the model what the coercer is
controlling: behavior. The victim does not cease to behave...

Me:

Is that a "yes' or a "no"?

Bruce:

Model the victim's behavior (control of perception) and show that
this is what the coercer is controlling, and it's a yes.

I did model the victim's behavior in [Rick Marken (980616.2230)].
My model shows that the victim's behavior is irrelevant to
determining whether or not a strong controller is controlling
the behavior of a weak victim. So you must agree, then, that what
the lifeguard and the RTP teacher are doing is controlling behavior?
Yes?

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 (980617.1714 EDT)]

Rick Marken (980616.2230)--
[Re: Coercive help]

The graph of r1 and r2 against qi is a great improvement. Here's the
corresponding extension of your diagram of a model of coercion. changing r1
to r' to conform to your (980612.0820):

              r1
              >
        p --> C -->
        > > Coercer
       >s> >o>
......................................
        > qo
        > >
        qi<-------- Environment
        >^
        >>
        >----------qo' (d)
......................................
       >s'| |o'|
        > > Victim
        p --> C' -->
              >
              r' <== cannot determine r' by Test

r1 is
person 1's reference for the state of qi; r2 [=r'] is person 2's
reference for the state of qi.

r2
  r2
    r2 r2 r2
     r2 r1 r1 r2 r2
   r1 qi qi r1 r1 r2 r2
r1 qi r2 qi qi r1 r1 r2 r2
qi r2 qi qi r1 r1 r2 r2
time-->--------r2-----------qi-ri-qi-r1-r1----------------r2---r1-----
               r2 qi qi qi r1 r1 r2 qi
                  r2 r2 qi r1 ri qi r2
                      r2 r2 qi r1 qi r2
                          r2 r2 qi r2
                              r2 r2

Why does r2=r' vary over time? Wouldn't you say it is because of some
higher level of control? The victim's control at that higher level is being
coerced too, isn't it? Now, suppose they find other means of controlling at
the higher level. We would expect a reduction of gain on the victim's
control of a perception of qi then wouldn't we?

The value of r1 might even continue to vary; we currently have no way of
knowing that. You would conclude from that that they were still being
coerced with respect to the state of qi. But controlling the state of qi
was only means of controlling some higher-level perception, and they are
now using other means for controlling that perception. So at the higher
level they are no longer being coerced, and at the lower level they are no
longer controlling a perception of qi (because they are now using different
means) and are no longer being coerced.

You might say that the forced change of means is coerced. But your model
can't show that. Your model with the victim represented just by an output
signal o' as in your (980612.0820) is unable to show any of this.

You have to include the victim. Coercion is not an attribute of a control
system, it is a relationship between control systems. It is a limiting case
of conflict, in which one side controls and the other side cannot control.

There are serious methodological problems with including the victim. But
first ...

By your definition of coercion, person 1 is not coercing whenever
r1 = r2 and (hence) qi = r2. This occurs at two instants in
time during the course of this interaction.

I'm surprised to see you drawing such conclusions from instantaneous values
of a continuously varying control variable, where the previous value is
among the causes of the current value, and the current value is among the
causes of the next. But more importantly, the variation of r2=r' results
from control of a different perception, presumably at a higher level. That
control is never free from coercion, because it is control of that
continuous variation, not of any instantaneous value. (Same response to
Bill's dog walking example.) You could raise this question only because you
have left out of the model the coerced control system, the victim.

So, by taking the
intentions of both parties into account, we find that coercion
can come and go instantaneously in two person interactions.

I disagreed with this above. But coercion can come and go with the
intentions of the victim as well as with the intentions of the coercer. "I
will prevent you from going to Slobbovia" says Jeremy to Julep. "I have a
court injunction, a restraining order, and the power of the sherriff's
department and the National Guard to back me up." "Where's Slobbovia?" says
Julep, having no intention of going there and no means to do so anyway.
Because Jeremy and his minions are keeping watch, you tell me that Julep is
coerced. You tell me that the kid in the shorts hanging out with friends,
who doesn't like the water, can't swim, and has no intention of getting
anywhere near the edge of the pool, is being coerced by the monitoring
lifeguard who wants to make sure no street clothes or twisted knickers
enter the pool. Do tell.

The point is, so long as you leave the victim out of the model you can't tell.

I don't want you to abandon the effort to model coercion; I do want you to
include the victim in your model. Is that unreasonable? I know we don't
know how to determine what the victim would be controlling if she could (we
can't determine the value of r'). We don't know how to do that for conflict
either, I think. The Test works only when control is successful. Are these
not serious methodological problems?

The intentions of the victim may be inaccessible by current methods; and
they may have no effect on the variable that the coercer is controlling; it
does not follow that they are irrelevant.

Richard Kennaway (980617.1824 BST)--

Yes, it would be best not to give a special technical PCT definition to a
word that already has a well established meaning.

Rats. There go "behaviour", "control", and "perception".

Really? "Control" is a core concept without which there is no PCT, and in
its definition/description "behavior" takes on PCT-specific meaning. Not
sure why "perception" is PCT-specific, I'll take your word for it--oh, I
guess you mean talking about perceptions as neural signals?. "Coercion" is
neither a core concept nor entailed in the definition of a core concept.
Are you just being contentious, or do you have a serious issue here?

Rick Marken (980617.1045)]

Me:

can you at least see that the lifeguard (like the RTP teacher)
is _controlling behavior_ -- and is doing so even when there
is no resistance to that control?

Bruce Nevin (980617.1305 EDT)

Then you have to include in the model what the coercer is
controlling: behavior. The victim does not cease to behave...

Is that a "yes' or a "no"?

Model the victim's behavior (control of perception) and show that this is
what the coercer is controlling, and it's a yes. Leave the victim's control
loop out of the model, and I don't know the answer. Neither do you.

  Bruce Nevin

[bnhpct]

[From Bruce Nevin (980617.2105 EDT)]

Rick Marken (980617.1530)--

The victim's control at that higher level is being coerced too,
isn't it?

No.

Let's see. I'm controlling drawing a large circle repeatedly on the board.
You constrain vertical movement of my arm so that it's shadow keeps the sun
out of your eyes. The horizontal movement and some vertical movement of my
arm is unconstrained. One element of lower-level control is coerced, my
control of vertical arm movement. So is the higher-level control of drawing
circles. So this is yet another description that I have offered in support
of my claim that the higher level is being coerced too. You gave no
explanation for your "no". Did you have a reason, or do you just know ex
cathedra that the higher level is not coerced?

Model the victim's behavior (control of perception) and show that
this is what the coercer is controlling, and it's a yes.

I did model the victim's behavior in [Rick Marken (980616.2230)].

A plot of reference signal values is a model now? And how did you determine
the values of r2 for the victim of coercion?

My model shows that the victim's behavior is irrelevant to
determining whether or not a strong controller is controlling
the behavior of a weak victim.

I agree that, if the victim is attempting to control what the coercer is
overwhelmingly controlling, then the victim is being coerced. That is what
your plot of r1, qi, and r2 shows. What if the victim is not attempting to
control a perception of qi? You claim that is coercion also, but you
haven't shown it.

So you must agree, then, that what
the lifeguard and the RTP teacher are doing is controlling behavior?
Yes?

If indeed the behavior is there to be controlled. If the behavior is not
there, it does not matter how many bystanders are present, their behavior
is not being coerced.

To Test I suppose you must prevent the coercer from observing the victim's
behavioral outputs. (Note that here we are no longer including the case
where the victim's control just happens accidentally to be a disturbance to
some qi whose value the coercer is controlling, but where the victim does
not intend to control a perception of qi.) Distract the lifeguard, call the
RTP teacher out of the room, feed a false video signal into the cameras on
a street corner so more than the permitted 2 or 3 people can gather there.
If the victims are controlling the affected perception, the effects of
their control will be observable when the coercion is removed.

You can't be sure you have removed their perception of being observed.
That's a more complicated aspect of the victim that you haven't modeled,
what they imagine. Worse yet, what they have internalized as a kind of
self-coercion. If we get into Big Brother scenarios, then we have to
include such things in the model. Claiming that ineffectual = irrelevant
and not to be modelled is a copout.

  Bruce Nevin

[R_Marken]

[From Rick Marken (980617.2030)]

Bruce Nevin (980617.2105 EDT) --

First, let me say (before I forget) that I would like to hear your
(and anyone else's) reply to Richard Kennaway's (980617.1640 BST)
_excellent_ question from earlier today:

This all started with the observation that in the RTP, when a
student disrupts twice in class, they are made to leave the
classroom whether they want to or not, and that the recommended
form of words for use by the teacher, "I see you have chosen
to go to the RTC" is a lie. Does anyone dispute that observation?

Now, on to your post:

You said:

The victim's control at that higher level is being coerced too,
isn't it?

And I said:

No.

And now you say:

You gave no explanation for your "no". Did you have a reason, or
do you just know ex cathedra that the higher level is not coerced?

You made this statement (that the "victim's control at that higher
level is being coerced too") with respect to my little graph
showing qi being controlled relative to r1 but not r2. Higher
levels had nothing to do with that demo; so my "no" was just a
polite way of saying "Get real"

I agree that, if the victim is attempting to control what the
coercer is overwhelmingly controlling, then the victim is being
coerced. That is what your plot of r1, qi, and r2 shows. What if
the victim is not attempting to control a perception of qi?

It's still Bill and Rick - style coercion

You claim that is coercion also, but you haven't shown it.

What's to show? All I'm showing is what I mean by coercion. For me,
coercion is control of some behavioral variable (qi, qo or even
some irrelevant side effect of control) which _nullifies_ (if
necessary) the intentions of the coercee.

Me:

So you must agree, then, that what the lifeguard and the RTP
teacher are doing is controlling behavior? Yes?

Ye:

If indeed the behavior is there to be controlled. If the behavior
is not there, it does not matter how many bystanders are present,
their behavior is not being coerced.

Remember, it's a perceptual _variable_ that is controlled. The
behavior that is controlled is a _variable_. In the case of the
lifeguard, the controlled behavioral _variable_ is "wearing
shorts in water". This variable has two possible values: true and
false. The lifeguard's reference for this variable is "false". So
the controlled behaviorl variable is always "there"; it is either in
the state "false" (in which case the lifeguard needs to take no
action to change the variable) or in the state "true". The same is
true of the RTP teacher: the controlled behavioral variable is
"degree of disruption in the class". This controlled variable is
always _there_ ; the RTP teacher is always controlling for it,
even though she usually doesn't have to do anything to keep that
variable in the reference state.

So, the correct, non-waffling, PCT answer to my question:

So you must agree, then, that what the lifeguard and the RTP
teacher are doing is controlling behavior?

was simply YES!!!

You're making this far more complex than necessary. Why?

Best

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 (980618.0345 MDT)]

Bruce Nevin (980617.1714 EDT)--

I don't want you to abandon the effort to model coercion; I do want you to
include the victim in your model. Is that unreasonable?

I don't understand why you think that the victim is NOT in the model of
"coercion," whatever that is. The victim has always been in the model --
_our_ model. Perhaps you missed the reference to what model is being talked
about when we say that the coercer does not take the coercee's goals into
account. That means that the coercer does not need a model of the coercee
(though he may have one, probably erroneous); we, of course, who understand
the consequences of ignoring the coercee's desires, DO need to propose a
model including both partipants in the interaction, although we do not need
to know the coercee's goals to explain his behavior (which is totally
determined by the coercer).

I know we don't
know how to determine what the victim would be controlling if she could (we
can't determine the value of r').

Of course we can -- just remove the coercer and see what the coercee does.
Let the coercive teacher leave the classroom and listen to the noise level,
watch the spitballs fly. Of course what you observe is subject to
interpretive problems, but that's nothing new, or insurmountable.

don't know how to do that for conflict
either, I think. The Test works only when control is successful. Are these
not serious methodological problems?

No, I don't think so. By applying disturbances that permit one side of a
conflict to prevail (the experimenter temporarily nullifying the other
side), or by providing circumstances that avoid arousing the conflict, each
side in turn can be observed as if operating by itself. Remove the
disruptive child from the classroom and observe what the teacher does;
remove the teacher from the classroom and observe what the child does. Put
them together and observe the conflict. I don't see any unsurmountable
methodological problems here.

There's a "logical fact" that I think is being overlooked in this
discussion. If (A and B) or (A and not-B) implies C, then the state of B is
irrelevant. If A is the coercer's goal for the coercee's behavior
(controlled variable C), and B is the coercee's goal for the same behavior,
then the coercer's goal is all that needs to be taken into account to
explain the behavior C. The statement "A and B implies C or A and not-B
implies C" is exactly equivalent to "A implies C." B cancels out of the
logical equation.

Best,

Bill P.

[Bill_Powers1]

From Bill Powers (980618.0424 MDT)]

Bruce Nevin (980617.2105 EDT)--

Rick Marken (980617.1530)--

Let's see. I'm controlling drawing a large circle repeatedly on the board.
You constrain vertical movement of my arm so that it's shadow keeps the sun
out of your eyes. The horizontal movement and some vertical movement of my
arm is unconstrained. One element of lower-level control is coerced, my
control of vertical arm movement. So is the higher-level control of drawing
circles.

You have another hand, don't you? You can ask someone else to draw the
circle for you, can't you? You can put off drawing the circle until the sun
goes down, can't you? Constraining a lower-level control process does NOT,
in general, constrain the higher systems that use it.

Best,

Bill P.

···

[Tim_Carey1]

[From Tim Carey (980619.0610)]

[From Bill Powers (980618.0345 MDT)]

There's a "logical fact" that I think is being overlooked in this
discussion. If (A and B) or (A and not-B) implies C, then the state of B

is

irrelevant. If A is the coercer's goal for the coercee's behavior
(controlled variable C), and B is the coercee's goal for the same

behavior,

then the coercer's goal is all that needs to be taken into account to
explain the behavior C. The statement "A and B implies C or A and not-B
implies C" is exactly equivalent to "A implies C." B cancels out of the
logical equation.

This is a logical fact _if and only if_ A and B have the same cv. How is
that possible? Wouldn't A then have to know what B was controlling for? If
B changes what he is controlling for they don't have the same cv anymore so
the above statement no longer applies. I thought in coercion A was
controlling B's _actions_ which would mean B would have to have the goal of
controlling his own actions as well. This is very odd since I thought the
fundamental principle of PCT was that we controlled perceptions _not_
actions.

So what you seem to be saying is that A and B both have the _same_
perceptions (C), again this seems very odd as I thought we all had
individual perceptions. I didn't think two people _ever_ perceived the
world in exactly the same way.

This new slant on PCT is intriguing.

Cheers,

Tim

[Marc_Abrams5]

From [Marc Abrams (980618.1729)]

Felt like jumping into this one.

Tim Carey said in response to BP:

This is a logical fact _if and only if_ A and B have the same cv.

Yep. So?

How is that possible?

Why isn't it possible? You seem to hold the view that a CV is a
_single_ entity that has very _definitive_ and _exact_boundaries. Who
said so?. Its also besides the point. If _I_ want you to _do_
something _I DON'T CARE WHAT YOUR CONTROLLING FOR_. _IF_ you _happen_
to be controlling for what I want fine you lucked out. If not tough
s__t. Simple

Wouldn't A then have to know what B was controlling for?

Why? Haven't we been here before?

If B changes what he is controlling for they don't have the same cv

anymore so

_B's CV's ARE IRRELEVANT_ Period, end of story

Marc

[bnhpct]

[From Bruce Nevin (980619.1311 EDT)]

Rick Marken (980617.2030)]

I would like to hear your
(and anyone else's) reply to Richard Kennaway's (980617.1640 BST)
_excellent_ question from earlier today:

This all started with the observation that in the RTP, when a
student disrupts twice in class, they are made to leave the
classroom whether they want to or not, and that the recommended
form of words for use by the teacher, "I see you have chosen
to go to the RTC" is a lie. Does anyone dispute that observation?

Looks like the teacher's statement of the teacher's perception. The teacher
could be lying, or deluded, or referring to a system of rules that the
student has previously agreed to, or reframing the conflict in terms of
those rules, or a number of other possibilities.

You made this statement (that the "victim's control at that higher
level is being coerced too") with respect to my little graph
showing qi being controlled relative to r1 but not r2. Higher
levels had nothing to do with that demo; so my "no" was just a
polite way of saying "Get real"

I said (980617.1714 EDT):

Why does r2=r' vary over time? Wouldn't you say it is because of some
higher level of control?

It looks like you're saying I was wrong. (I wish you had said so directly.)
Where does the variation in r2 come from then, please? Is it an intrinsic
variable?

What if
the victim is not attempting to control a perception of qi?
You claim that is coercion also, but you haven't shown it.

What's to show?

Your graph would show only r1 and qi. There would be *no* reference value
r2 within the victim corresponding to qi.

For Bill and Rick style coercion, there does not even have to be a victim
present. Everybody else has gone home. The lifeguard is prowling around the
empty pool, just waiting for someone to come on the scene wearing anything
but a bathing suit. That lifeguard, by your definition, is coercing. She is
controlling a perception and will coercively prevent anyone from disturbing
her control. In your words:

Remember, it's a perceptual _variable_ that is controlled. The
behavior that is controlled is a _variable_. In the case of the
lifeguard, the controlled behavioral _variable_ is "wearing
shorts in water". This variable has two possible values: true and
false. The lifeguard's reference for this variable is "false". So
the controlled behaviorl variable is always "there"; it is either in
the state "false" (in which case the lifeguard needs to take no
action to change the variable) or in the state "true".

Enjoy!

  Bruce Nevin

[KENNETH_J_KITZKE1]

[From Kenny Kitzke (980619.1630 EDT)]
<Bruce Nevin (980619.1311 EDT)>

<For Bill and Rick style coercion, there does not even have to be a victim
present. Everybody else has gone home. The lifeguard is prowling around the
empty pool, just waiting for someone to come on the scene wearing anything
but a bathing suit. That lifeguard, by your definition, is coercing. She is
controlling a perception and will coercively prevent anyone from disturbing
her control.>

I'm in your swimming pool! Oops, I need to get my swimming trunks on in
case Rick is the lifeguard. He is unilaterally coercive you know.

Kenny

[bnhpct]

From Bruce Nevin (980619.1332 EDT)

Bill Powers (980618.0424 MDT)--

I purposely kept the example extremely simple because of communication
wierdness currently going on. The lower-level systems are for linear
movement of the arm. Saying horizontal and vertical movement is probably
wrong, depending on orientation of the limbs and muscles--something you
know a *lot* about from the little man model--but it's a clear way of
putting it anyway, etch-a-sketch fashion. There is a higher-level system
varying the reference inputs to these lower-level systems. *That* specific
higher-level system is also coerced. Sure, at a yet higher level where my
intention is to perceive circles being drawn on the board I can choose
other means. The abandonment of thwarted control and going up levels for
which other means might be found has *not* been admitted in the discussion
of coercion so far. If my ability to get someone else to draw the circles
shows that I am not coerced, then the ability of people forbidden to gather
in groups larger than 3 to find other means to communicate, coordinate
revolution, whatever, shows that they aren't coerced either. But they are,
and at more than one immediate and lowest level, and so is the drawer of
circles.

  Bruce Nevin

[Marc_Abrams5]

From [Marc Abrams (980619.1750)]

[From Bruce Nevin (980619.1624 EDT)]

Then what? The victim does not stop controlling. We know what the
coercer does, that's easy. What the victim does next is a lot less

cut and >dried, and a lot more interesting. And it is very much a part
of modelling

coercion. To what extent can coercion be successful? How does it

fail? >What are the alternatives to counter-coercion (getting even) or
role >reversal down a ladder of relative victimehood? Can PCT tell us

something useful here? Or do we just stop with saying that by

definition at >the moment of coercion the intentions of the victim are
irrelevant?

_Great_ set of questions Bruce. Got to give this some thought.

I hope this gets some attention. To me _this_ is the _heart_ of the
matter in human interactions. Again, good job.

Marc

[bnhpct]

[From Bruce Nevin (980619.1624 EDT)]

Bill Powers (980618.0345 MDT)--

I don't understand why you think that the victim is NOT in the model of
"coercion," whatever that is.

Because so far we have seen only a model of a coercer. Rick has shown
output o' with no control loop, and (separately) a plot of a reference
signal r2, but these do not constitute a model of a victim.

The victim has always been in the model --
_our_ model. Perhaps you missed the reference to what model is being talked
about when we say that the coercer does not take the coercee's goals into
account. That means that the coercer does not need a model of the coercee
(though he may have one, probably erroneous);

I'm not concerned with putting the victim inside your model of the coercer.
Coercion, like conflict, is a relationship between control systems. It is
not an attribute of the coercer.

we, of course, who understand
the consequences of ignoring the coercee's desires, DO need to propose a
model including both partipants in the interaction, although we do not need
to know the coercee's goals to explain his behavior (which is totally
determined by the coercer).

That depends upon how far up the hierarchy the coercion extends. And *that*
begins to get at something interesting.

I know we don't
know how to determine what the victim would be controlling if she could (we
can't determine the value of r').

Of course we can -- just remove the coercer and see what the coercee does.
Let the coercive teacher leave the classroom and listen to the noise level,
watch the spitballs fly. Of course what you observe is subject to
interpretive problems, but that's nothing new, or insurmountable.

I respectfully disagree. Kids released from coercion behave differently
from kids free from coercion. The model of the victim is far more
interesting than that of the coercer, and one reason is changes in what the
victim controls, by going up to a level that is not coerced (other means
are available), by trying random alternatives, by controlling being
spaceman Spiff on an alien planet (ah, Calvin! Apologies to those who don't
see U.S. comic strips).... (You're also talking here about social
interactions among the kids, which are not modelled. Put a single student
with a single coercive teacher. When the teacher walks out does that single
student start yelling and throwing spitballs? That suggests that social
interactions can be involved in victims changing what they are
controlling.) Remove the coercer and you are not likely to see control
resume as though it had never been thwarted. But that is probably what the
model now predicts--though we can't tell because the present model only
says that whatever the victim is controlling is irrelevant.

By applying disturbances that permit one side of a
conflict to prevail (the experimenter temporarily nullifying the other
side), or by providing circumstances that avoid arousing the conflict, each
side in turn can be observed as if operating by itself. Remove the
disruptive child from the classroom and observe what the teacher does;
remove the teacher from the classroom and observe what the child does. Put
them together and observe the conflict. I don't see any unsurmountable
methodological problems here.

Not insurmountable, probably, but not simple either. See above.

There's a "logical fact" that I think is being overlooked in this
discussion. If (A and B) or (A and not-B) implies C, then the state of B is
irrelevant. If A is the coercer's goal for the coercee's behavior
(controlled variable C), and B is the coercee's goal for the same behavior,
then the coercer's goal is all that needs to be taken into account to
explain the behavior C. The statement "A and B implies C or A and not-B
implies C" is exactly equivalent to "A implies C." B cancels out of the
logical equation.

The logic is impeccable. The premisses are slippery.

If the coercer is controlling the state of environment variable C then C
(or rather the state A of C) is an aspect of the coercer's behavior. The
victim is unable to control, so it cannot be the victim's behavior.
Therefore the coercer is not controlling the victim's behavior. The coercer
is preventing the victim from controlling, so the coercer is actually
preventing the state of C from being an aspect of the victim's behavior. As
you have said for many years, it is not possible for one control system to
control the behavior of another. To do so, it would have to reach inside
and set one or more reference values in the victim.

Suppose A=B, that is, the victim is controlling the state of C exactly as
the coercer prefers. You have argued convincingly that this can at best be
a transient state. Sooner or later the victim's control will have results
that disturb the coercer's control. There are lots of interesting questions
as to what is going on in a compliant control system. How did its reference
levels get set so that its effects on the environment do not disturb the
coercer's control? How does it maintain them so? Is it doing that as means
for controlling some other perceptions? If doing so is a means of
controlling, how then can it be said to be coerced? The victim is simply
treating the dictatorial requirements of the coercer as a feature of the
environment, like a door to be opened or a flight or stairs to be climbed.

Assume C is a behavioral output of the victim. The coercer requires that
the victim sit up straight, or take off his hat when the King passes, or
look deferential, or salute in the correct manner, or recite from memory
Falstaff's soliloquy on honor. It is somewhat difficult to account these as
variables in the victim's environment (except for maybe the hat), though
they clearly are variables in the coercer's environment. The victim may
have the same goal as the coercer but not perform adequately. I have
arranged the items in that list so that they require progressively more
training and practice. All of them require a willingness to perform. The
victim may disguise unwillingness as inability. But all of this is
irrelevant, you say.

Assume A is a state of environment variable C such that the victim is
unable to control the state of some other enviroment variable D. The
coercer turns on a boom box while the victim is trying to memorize
Falstaff's soliloquy on honor. The coercer turns off the lights while the
victim is trying to sketch a portrait of the King (from a photograph). They
come into conflict over the state of an environment variable, e.g. position
of light switch. The victim loses the conflict, cannot win it.

Then what? The victim does not stop controlling. We know what the coercer
does, that's easy. What the victim does next is a lot less cut and dried,
and a lot more interesting. And it is very much a part of modelling
coercion. To what extent can coercion be successful? How does it fail? What
are the alternatives to counter-coercion (getting even) or role reversal
down a ladder of relative victimehood? Can PCT tell us something useful
here? Or do we just stop with saying that by definition at the moment of
coercion the intentions of the victim are irrelevant?

A PCT claim is that it is not possible for one control system to control
the behavior of another. Have you changed your views, and decided that not
the case after all?

  Bruce Nevin

[R_Marken]

[From Rick Marken (980619.1610)]

Bruce Nevin (980619.1311 EDT) --

Where does the variation in r2 come from then, please?

From where you said: the output of a higher level system.

It was the stuff about r2 being coerced that seemed a tad
off the mark.

For Bill and Rick style coercion, there does not even have to
be a victim present.

That's correct. A "victim" is just a source of disturbance
to the variable controled by the coercer. It might not sound as
strange to you if you said "For Bill and Rick, driving is still
happening even when there are no curves in the road". Or "For
Bill and Rick, Saddam Hussain is still a dictator even though
there are no protests or opposition to his rule".

Bruce Nevin (980619.1624 EDT) --

The model of the victim is far more interesting than that of
the coercer

That's your opinion. But what the victim controls doesn't change
the fact that the coercer is coercing -- does it? When the wind
stops, does the driver stop controlling being in the lane? When
the kids are gone does the parent stop controlling for peace and
quiet? When gravity is gone does the astronaut stop controlling
the position of her limbs? When the NRA is gone will Americans
stop controlling for being armed to the teeth?

A PCT claim is that it is not possible for one control system to
control the behavior of another.

This is not a PCT claim because it is false. One control system
_can_ control the behavior of another; that's a fact.

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 (980619.2220 EDT)]

Rick Marken (980619.1610)--

Where does the variation in r2 come from then, please?

From where you said: the output of a higher level system.

It was the stuff about r2 being coerced that seemed a tad
off the mark.

It is your claim that r2 has no effect, and that is all that I mean by
saying that it is coerced. My further conclusion is that the higher-level
control outputting r2 also has no effect since its means of control
(lower-level control of r2) has no effect; and that is all that I mean when
I say that the higher-level control is also coerced. You denied this when
you said (980617.2030):

You made this statement (that the "victim's control at that higher
level is being coerced too") with respect to my little graph
showing qi being controlled relative to r1 but not r2. Higher
levels had nothing to do with that demo; so my "no" was just a
polite way of saying "Get real"

You agree now that higher levels *are* involved in that graph though they
are not explicitly shown. What did you mean by "get real"?

For Bill and Rick style coercion, there does not even have to
be a victim present.

That's correct. A "victim" is just a source of disturbance
to the variable controled by the coercer. It might not sound as
strange to you if you said "For Bill and Rick, driving is still
happening even when there are no curves in the road". Or "For
Bill and Rick, Saddam Hussain is still a dictator even though
there are no protests or opposition to his rule".

That's right, all your "coercer" is doing is controlling perceptions, just
as a driver of a car is controlling perceptions. You have no way of knowing
whether or not that control is coercive until there is actually a victim
present and being coerced. Why? Because you have no way of knowing whether
the controlling system is going to stop being coercive and start
controlling perceptions of other control systems (that are sources of
disturbance to its control) in ways that we might call "being polite" and
"being considerate of weaker ones" and such like phrases, or perhaps start
being fearful of counter-coercion by that weak little control system's big
brother.

Bruce Nevin (980619.1624 EDT) --

The model of the victim is far more interesting than that of
the coercer

That's your opinion. But what the victim controls doesn't change
the fact that the coercer is coercing -- does it?

You really aren't reading what I'm saying, are you.

When the wind
stops, does the driver stop controlling being in the lane?

If there were no lane there, the driver would have to stop controlling
being in the lane. Driving in a lane is not control of the car, tout court,
it is control of a relationship between the car and the lane. Coercion is
not an attribute of a "coercive" control system, it is a relationship
between control systems. If the control system is by itself it cannot be
coercive; it can only control. If the car is in an open field, the driver
cannot control keeping the car in the lane. (Both are possible in
imagination of course, but that's irrelevant.)

When
the kids are gone does the parent stop controlling for peace and
quiet?

No, but that is just control. It is not coercion. And while the kids are
gone the parent might get zonked on prozac or valium, or go on a meditation
retreat, or undergo a religious experience, or just decide to try something
different, so that when the kids come back the control is obviously
non-coercive even to you. They still control peace and quiet, but they
choose other means, like maybe earplugs, or leaving the house, or (what a
thought!) developing a different relationship with the kids and cooperating
with them in finding things that they all like to do. When the disturbances
cease for a while, you have no way of knowing whether the "coercer" will
continue to overpower disturbances originating from other control systems
when those disturbances occur again. Your claim that coercion continues is
so much vapor. "Air--a trim reckoning!" All that continues during the
cessation of disturbances is control, not coercion. Control and coercion
are not the same thing.

A PCT claim is that it is not possible for one control system to
control the behavior of another.

This is not a PCT claim because it is false. One control system
_can_ control the behavior of another; that's a fact.

I won't go spelunking in the archives over the past 8 years for quotes in
email ... but ... in Chapter 17 of B:CP, in the discussion of treating
control systems as inanimate objects and the sequel about conditioning,
when Bill says that it is not possible for one control system to control
the behavior of another, you're telling me that I am misunderstanding what
he is saying? Please explain how I have misunderstood.

Immediately preceding the above quote, I said

If the coercer is controlling the state of environment variable C then C
(or rather the state A of C) is an aspect of the coercer's behavior. The
victim is unable to control, so it cannot be the victim's behavior.
Therefore the coercer is not controlling the victim's behavior. The coercer
is preventing the victim from controlling, so the coercer is actually
preventing the state of C from being an aspect of the victim's behavior.

If you grab my arm and move it up and down, overpowering my puny 90 lb.
weakling attempts to resist, you would say that this waving of my arm was
my behavior? And that my ineffective attempts to resist were not my
behavior? If you were waving a flag up and down, would that be your
behavior, or the flag's behavior? If you are treating my arm as an
inanimate object, waving it like a flag, is the arm-waving your behavior or
mine? Please explain your reasons for your conclusions, don't just say
"yes" or "no". No matter how polite you want to be.

  Bruce Nevin

[Cyvchoate]

i.kurtzer (980620.0230)

[From Bruce Nevin (980619.2220 EDT)]

spot on again!!

[Bill_Powers1]

[From Bill Powers (980619.0734 MDT)]

Tim Carey (980619.0610)--

This is a logical fact _if and only if_ A and B have the same cv. How is
that possible? Wouldn't A then have to know what B was controlling for?

No. That's the whole point of the "logical fact." A will control B's
behavior whether or not B has a different goal for his behavior, and no
matter what goal is being served by that behavior. That "whether or not" is
the "B or not-B" in the logical analysis. B's behavior will be what A wants
whether B wishes to produce the same or a different behavior: A will make
sure of that. If B controls for the same behavior (keeping quiet in class)
that A is controlling for, then A will experience no disturbance and produce
no corrective action. If B controls for some other state of that behavior
(being loud in class) A will still end up perceiving that B is quiet in
class -- by forcing B's behavior to be what A wants it to be, doing whatever
it takes. B's attempt to do something else is merely a disturbance to A, who
will take care of it automatically.

If
B changes what he is controlling for they don't have the same cv anymore so
the above statement no longer applies. I thought in coercion A was
controlling B's _actions_ which would mean B would have to have the goal of
controlling his own actions as well.

B does (when allowed) control his own actions. An action at one level is a
controlled variable at the next lower level. Are you confusing controlling a
different variable with controlling the same variable but relative to a
different state?

This is very odd since I thought the
fundamental principle of PCT was that we controlled perceptions _not_
actions.

Why does everyone insist on forgetting the H in HPCT?

So what you seem to be saying is that A and B both have the _same_
perceptions (C), again this seems very odd as I thought we all had
individual perceptions. I didn't think two people _ever_ perceived the
world in exactly the same way.

The same physical situation is behind the perceptions of A and B. The two
perceptual signals might be physically different, but if they both depend on
the same physical situation, there can be conflict.

Suppose I'm reading and you're exercising in the same room. If I turn on a
nice bright light, it helps me perceive the letters on the page, and it
makes the temperature you feel get higher. When you try to reduce your
temperature by turning off the light, you disturb my ability to read, so we
are in conflict even though we're controlling different perceptions. The
common element that causes the conflict is the on/off state of the light: it
can't be in both states at once.

When I control your actions, I do so in order to achieve some higher goal of
mine. But your actions are also under your control, and serve to achieve
some higher goal of yours. If you need to send a note to someone else and I
confiscate it every time you try, I prevent your communication with the
other person by preventing the action needed to achieve it. If we were
physical equals, you would resist and try to send the note anyway. If you
are weaker and afraid of what I might do, you might refrain from sending the
note even though you still need to send it. But it makes no difference to me
whether you keep trying to send it or give up -- the note is not going to be
sent in either case, so my goal will remain satisfied. I don't need to know
why any given person is not sending notes -- whether they've given up
trying, or simply never did want to send a note. In my class, the sending of
notes is prevented, period.

You want to say that a student who tries to send a note and has it forcibly
confiscated is being coerced, but when the student finally gives up trying
to send notes (because they're always seized), that student is no longer
being coerced. I presume you would say that the student is now refraining
from sending notes out of his or her own free choice.

I would disagree, but I don't suppose that will have any influence.

Best,

Bill P.