Coercive Help

[From Rick Marken (980615.0850)]

i.kurtzer (980615.0000) --

Therefore, ALL interactions about a common CV will fall under
coersion given a suitable gain difference.

Yes. I agree.

Does that not seem strange?

Yes. It does not seem strange.

Does that make for a very strange taxonomy with helping as a
subclass of coersion?

Maybe it's surprising. But I wouldn't call it "strange". Many of
the interactions that people call "helping" are clearly examples
of coercion (as I have defined it). For example, suppose I want
to help you become a better person by forcing you to spend more
time in school and less time rolling around on your skate board.
Am I helping you? Or am I coercing? I can call what I'm doing
"helping" but you (as the victim of my "help") know that I am really
coercing. Agents of the government will arrest me if I smoke dope
in the privacy of my own backyard. They think they are helping me
by keeping me away from the evils of drugs. Of course, they are
really just coercing me into not taking the drugs I want to take
(I am actually not interested in taking drugs but I have seen
this kind of "help" given to friends who _were_). I'm sure you can
think of many more examples.

It's not rare for "coersion" to be done in the name of "helping";
it would be nice if the RTP program could avoid this hypocricy.

In PCT, real "helping" would involve making it possible for the
helpee to control what s/he wants to control. Any direct efforts
to help the helpee bring a controlled variable to it's reference,
however, would almost certainly lead to conflict (due to control
system "parallax"; the helper sees the controlled variable from
a different point of view than the helpee so the helper is likely
to be trying to help bring the controlled variable to the "wrong"
reference state).

I think PCT has some interesting implications regarding how we
might go about trying to help people. Maybe a discussion of
helping, from a PCT perspective, would be helpful;-)

Best

Rick

···

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

[From Rick Marken (980615.0850)]

i.kurtzer (980615.0000) --

> Therefore, ALL interactions about a common CV will fall under
> coersion given a suitable gain difference.

Yes. I agree.

> Does that make for a very strange taxonomy with helping as a
> subclass of coersion?

Maybe it's surprising. But I wouldn't call it "strange". Many of
the interactions that people call "helping" are clearly examples
of coercion (as I have defined it).

Now i agree that people can casually mislabel a social interaction while a few
a-holes systematically mislabel to disguise their intentions. However, the
question is not casual mislabellling, but taxonomic logic. By your definition
ALL instances of helping---whereby both ECS's perceive and affect a common CV
and their is a gain difference operationally defined as different enough---are
instances of coersion as helping is a subclass of coersion. This is what i
am asking and from above i will assume you agree to.

For example, suppose I want
to help you become a better person by forcing you to spend more
time in school and less time rolling around on your skate board.

Well, from the "forcing you" is implied that i was not so much interested
which implies that my intentions are important in defining that interaction.
For example,

Am I helping you? Or am I coercing?

Why are you asking this? If you know that there is a suitable gain difference
between two controllers of the same enviromental variable then there HAS to be
coersion BY YOUR DEFINITION. And moreover helping is not in an either/or
relation to coersion but is a subclass of it (by your definition) .

I can call what I'm doing

>"helping" but you (as the victim of my "help") know that I am really

coercing.

how would i ---the one whose intentions are irrelevant--- really know
anything? Are you saying that only the stronger can really define the
relation, but only the weak really know what that relation is? That's some
pretty wierd S-and-M sociology.

It's not rare for "coersion" to be done in the name of "helping";
it would be nice if the RTP program could avoid this hypocricy.

Again, by your logic this is not a necessary contradiction rather "helping" is
one type of "coersion". You still have the option of changing that position.

In PCT, real "helping" would involve making it possible for the
helpee to control what s/he wants to control. Any direct efforts
to help the helpee bring a controlled variable to it's reference,
however, would almost certainly lead to conflict (due to control
system "parallax"; the helper sees the controlled variable from
a different point of view than the helpee so the helper is likely
to be trying to help bring the controlled variable to the "wrong"
reference state).

Rick, PCT/real "helping" does not almost certainly lead to conflict because of
"control system parallax" . Given "control system parallax" , yes. but
sometimes people can come to consensus quite quickly and accurately. This
includes M.Duggins' Master thesis "When is helping, helping" as well as
potions of Trey Gann's and Mark Lazare's work. Actually, Mark if your still
out there, tell me.

One more time. You say that "In PCT, real "helping" would involve making it
possible for the helpee to control what s/he wants to control." But if there
is a gain difference then real helping is also real coersion by inheritance.
I think this wierdness is a real problem.

i.

[From Rick Marken (980615.1530)]

i.kurtzer (980615.0000) --

By your definition [of helping] ALL instances of helping---whereby
both ECS's perceive and affect a common CV and their is a gain
difference operationally defined as different enough---are
instances of coersion as helping is a subclass of coersion.

This kind of helping does lead to conflict. It's the kind where
you assume you know what a person wants to perceive and you try
to help them produce that perception. I was suggesting that another
kind of helping may be more successful; the kind where you first
try to find out what the person actually wants and then suggest
ways for them to get it. This kind of help is optional; if the
helpee rejects the help then that's fine with the helper.

Me:

For example, suppose I want to help you become a better person
by forcing you to spend more time in school and less time
rolling around on your skate board.

isaac:

Well, from the "forcing you" is implied that i was not so
much interested which implies that my intentions are important
in defining that interaction.

If your intention were to spend more time in school then I would
not have to use force, true. But I am still coercing you;
I am controlling for you spending more time in school, regardless
of your intentions and reisistance (or lack thereof).
.

Are you saying that only the stronger can really define the
relation, but only the weak really know what that relation is?

Not intentionally;-)

Me:

It's not rare for "coersion" to be done in the name of "helping";
it would be nice if the RTP program could avoid this hypocricy.

isaac:

Again, by your logic this is not a necessary contradiction rather
"helping" is one type of "coersion". You still have the option
of changing that position.

Maybe the problem is that people use the word "helping" to refer
to at least two different phenomena. One kind of helping is the "I
know what's good for you and I'll help you get it" type. I think
some of the "help" offered in RTP is of this variety. I object
to calling this "helping" because it is really just a form of
coercion. The other kind of helping is the "I'll do whatever _you_
want to help you get what _you_ want". I think this is real help.
I don't always want to do this kind of help -- I wouldn't help
Charleton Heston achieve his NRA goals, for example -- but at least
this is real help (not thinly disguised coercion).

One more time. You say that "In PCT, real "helping" would involve
making it possible for the helpee to control what s/he wants to
control." But if there is a gain difference then real helping
is also real coersion by inheritance.

This is not what I'm saying at all. In what I call "real" helping
any gain difference between helper and helpee is irrelevant. That's
because there is no attempt by the helper to bring qi to what is
thought (by the helper) to be the helpee's reference for qi. Real
help occurs when the helper offers the helpee suggestions and/or
resources that can be used _or not_ by the helpee to acheivce his
or her goals. Real help can always be rejected by the helper; the
helper's only goal is that the helpee get what s/he wants -- whether
the helper thinks the helpee _should_ want it or not. The quality of
help is not strained; it droppeth like the gentle rain from heaven
upon the helpee below.

Best

Rick

···

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

[From Bill Powers (980616.-909 MDT)]

Rick Marken (980615.0850)--]

I agree with all your comments about helping and coercion, but I fear that
they create too much error in those to whom helping is by definition a Good
Thing no matter how given or in what circumstances. If you define a behavior
as "helping" if it does in fact make control easier for another person, then
helping is normally a good thing and does not involve coercion. The problem
is that many people define their own behavior as helping without taking into
consideration whether the helpee wants to be helped, or if there is a desire
to be helped, to be helped toward the precise goal the helper has in mind.
The boy scout earns his merit badge by helping the little old lady cross the
road whether that's the way she wants to go or not. The social worker helps
the homeless person get into a shelter even though the homeless person hates
being in shelters. The general form of this kind of non-help starts with one
person deciding what is good for another person and then forcing the change
on the other person -- coercively.

By our definition of coercion (one stronger person controlling a controlled
variable of another weaker person), what creates the coercion is failure to
enquire about the weaker person's goal for the mutually-controlled CV, and
comply with it. Control by the coercer is done without caring what the other
person's goal is. In fact, for the coercer, the coercee's attempts to
maintain control are felt as minor disturbances which are overcome without
straining the coercer's capacities. The coercer does not need to know what
is causing the disturbances in order to cancel their effects. This is why
the coercee's goal is irrelevant; changing the goal will have no effect on
the state of the CV, so the coercee has no control. The flea can take the
elephant for a walk, but only if the flea chooses the same destination the
elephant chooses at every moment. The elephant, on the other hand, is not
even aware that the flea exists.

Coercion, helping, and the giving-up effect are logically connected. It's
not hard to work out the connections, given our definitions of the terms.
However, many people find the conclusions unacceptable, because they prefer
to think of themselves as helpers rather than as coercers (and that being a
helper is always good), and they prefer to interpret the giving-up effect as
grateful acceptance of the help rather than resignation to the coercion. So
rather than accept the unacceptable conclusions, such people want to change
the definitions -- for example, by making "help" always be a good thing, or
by saying that if you don't think of your behavior as intended to coerce, it
automatically becomes noncoercive.

Under these circumstances, I think this three-part discussion has gone about
as far as possible.

Best,

Bill P.

Bill Powers (980616.0930)

Coercion, helping, and the giving-up effect are logically connected. It's
not hard to work out the connections, given our definitions of the terms.
However, many people find the conclusions unacceptable, because they prefer
to think of themselves as helpers rather than as coercers (and that being a
helper is always good), and they prefer to interpret the giving-up effect as
grateful acceptance of the help rather than resignation to the coercion. So
rather than accept the unacceptable conclusions, such people want to change
the definitions -- for example, by making "help" always be a good thing, or
by saying that if you don't think of your behavior as intended to coerce, it
automatically becomes noncoercive.

The logical connections have not been presented publically; moreover, i find
it strange that one questionable position is being bolstered by an alleged
connectivity to an even more questionable position. Also, I do not recall
fuzzy warm feelings of good helpers vs bad coersions being a part of my, Bruce
N.'s or Tim's separate arguments. My gripe is the breach of taxonomy without
reason. That has not been answered.

Under these circumstances, I think this three-part discussion has gone about
as far as possible. >>

Therefore, we should remain conservative as no overwhelming reasons have been
given to break tradition. Coersion will continue to be defined according to
the intentions of all actors.

i.

that last gripe was me, Isaac, sorry i missed signing

[From Rick Marken (980616.1440)]

isaac kurtzer (980616)

My gripe [about coercion] is the breach of taxonomy without
reason. That has not been answered.

What "breech of taxonomy"? You mean you don't like our definition
of coercion (one stronger person controlling a controlled variable
of another weaker person) because it violates some classificatory
rule? What's the rule? Who made it up? We don't need no stinkin'
classificatory rules anyway, do we? :wink:

Therefore, we should remain conservative as no overwhelming
reasons have been given to break tradition.

What tradition has been broken? Coercion is control. We have
described coercion in terms of the variables relevant to
control (reference, error, output, controlled variable,
disturbance, perception). Sounds like we're right there in
the PCT tradition.

Coersion will continue to be defined according to the intentions
of all actors.

By you, apparently, but certainly not by me. Coercion never was
defined "according to the intentions of all actors", anyway -- at
least, not by Bill; and Bill _is_ PCT tradition;-)

Best

Rick

···

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

i.kurtzer (9806162300)

[From Rick Marken (980616.1440)]

isaac kurtzer (980616)

> My gripe [about coercion] is the breach of taxonomy without
> reason. That has not been answered.

What "breech of taxonomy"? You mean you don't like our definition
of coercion (one stronger person controlling a controlled variable
of another weaker person) because it violates some classificatory
rule? What's the rule? Who made it up? We don't need no stinkin'
classificatory rules anyway, do we? :wink:

Rick, i have explained several times what the custom has been. Yes, we do
need classifictory rules; internal logic as well as empirical verification.
Would you suggest willy-nilly reasons or none at all? You can't be serious.
Please refrain from smilies as they will only confuse me. Again the custom is
defining the interaction according to the intentions of all the actors. For
some reason, "coercion" does not follow this rule. As a social interaction
that violates a basic taxonomic rule, i wish to have reasons. The onus is on
you.

> Therefore, we should remain conservative as no overwhelming
> reasons have been given to break tradition.

What tradition has been broken? Coercion is control. We have
described coercion in terms of the variables relevant to
control (reference, error, output, controlled variable,
disturbance, perception). Sounds like we're right there in
the PCT tradition.

In building a social science on PCT the tradition has been to refer to all the
actor's intentions that are involved to define the interaction. That tradition
exists in Bourbon's many papers as well as three master's thesis's.

Coersion will continue to be defined according to the intentions

> of all actors.

By you, apparently, but certainly not by me.

By myself as well as Bourbon and as written by his student's in their theses.

Coercion never was
defined "according to the intentions of all actors", anyway -- at
least, not by Bill; and Bill _is_ PCT tradition;-)

If Bill is mistaken then he is mistaken. Lord knows, i think he got the best
batting average i've ever witnessed. Same with you. My Mind Reading copy is
so smeared from excessed reading, and i hock it constantly. But if he and you
have a stinker, tough. Also, I don't think the comparison between RTC and
Gulag will clear matters.

still without dagggers
i.

[From Rick Marken (980616.2230)]

i.kurtzer (9806162300)

the custom is defining the interaction according to the intentions
of all the actors. For some reason, "coercion" does not follow
this rule.

In building a social science on PCT the tradition has been to
refer to all the actor's intentions that are involved to define
the interaction. That tradition exists in Bourbon's many papers
as well as three master's thesis's.

It's not a "tradition". Tom was modeling two-person interactions
where the results of the interaction depended on the intentions
of both people; in coercion, the results of the interaction depend
on the intentions of only _one_ person -- the coercer.

But if you want to model the coercion situation in terms of
the intentions of all actors, I'll oblige. Below is a graph
of the time varying intentions of two people with respect to
the state of same controlled (behavioral) variable, qi. r1 is
person 1's reference for the state of qi; r2 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

If the ASCII aligns properly what you should see is r1 and r2
varying over time. qi covaries along with r1 but not r2. r1
and r2 intersect at two points in time. At these times,
qi matches the references for _both_ person 1 (r1) and person
2 (r2).

According to my definition of coercion, person 1 is doing coercion;
qi (the controlled behavioral variable) is maintained at person 1's
reference (r1); variations in person 2's intention (r2) are irrelevant
to the state of qi. So person 2's intention is irrelevant to the
fact that person 1 is doing coercion.

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. So, by taking the
intentions of both parties into account, we find that coercion
can come and go instantaneously in two person interactions. If you
are comfortable with this definition of coercion it's fine with
me; I'll then try to stick to calling my kind of coercion "control
of behavior".

Would you be willing to agree that RTP is not coercive but that it
does involve the control of behavior?

Best

Rick

···

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

From Kenny Kitzke (980617.1000 EDT)

[i.kurtzer (980616.1440)]

i,

You, and others, have raised good questions about coercion. Bill and Rick
have answered with their view and have not budged an inch. Under their
concept of coercion, their definitions seem right to me.

Under a concept that coercion can only exist if one control system *is*
controlling the behavior of another control system, when the other wants to
behave/control for different wants/perceptions, your view seems right.

I can see it both ways. But, it would be helpful to reach a consensus
about which concept best fits PCT, perhaps at the conference?

Right now, I'm leaning for clarity in terms and understanding and tend to
favor the latter concept.

For example, if a parent insists their child eat an ice cream cone or get
spanked, that does sound like coercion to me. However, if the child would
love to eat an ice cream cone, it seems that calling the two-party
interaction coercive simply defys common understanding and language. It
makes PCT sound more off-the-wall than it is already perceived by many to
be.

The parent may want the child to eat the ice cream cone for many reasons:
feels the child is too thin, wants to have the child play quietly in his
room when they return home to repay this favor, need to assuage their own
guilt for wanting an ice cream cone, etc.,

Similarily, the child can want to eat the ice cream cone for many reasons:
likes the taste, relief from a hot day, wants to obey their parent (would
even eat spinich if it was their parent's will), etc.,

It seems to me that PCT perfectly explains what a coercer and a coercee are
doing during coercion. But to have coercian a coercer must exercise
control over the behavior of the coercee. The coercee is forced to behave
as the coercer wants regardless of what the coercee wants.

When someone in a position to coerce, does not want to control the behavior
of another, I can't see this as a coercive relationship. It can be
considered a potentially coercive relationship, just not in fact occurring
at that time.

This is why I believe that the public school system can be properly
described as fundamentally coercive all the time. School authorities are
allowed to exert coercion any time they want. When authorities do not want
to coerce, they can avoid it. When students don't feel forced to keep
quiet in the classroom, there is no coercion taking place.

I believe conflict is very different than coercion. I am looking forward
to trying to get some consensus on PCT terminology. This is essential for
progress. Using common words to represent an uncommon concept simply makes
for misunderstanding and conflict, as this thread has produced from day
one.

In search of understanding,

Kenny

[From Kenny Kitzke, 980617.1100EDT)]

Rick Marken (980616.2230) to i.

<Would you be willing to agree that RTP is not coercive but that it
does involve the control of behavior?>

No. I would agree that RTP is a fundamentally coercive system used to get
what its leaders want regardless of what students want. If the RTP leaders
do not try to get the students (against their will) to do what the leaders
want, they are not using the coercive possibility available to them.

Once a student accepts that they cannot disrupt *and* be in the class, they
may willingly go to the RTP room as the choice for what they want (control
their own perceptions), or unwillingly go to the RTP room because of
coercion.

If a student never accepts this *and* or the RTP as legitimate, it would
seem evident that either forcing a student to not disrupt the class
(gagging him there) or forcing him to go to the RTP room when they don't
want to do either, is absolutely coercive.

Kindly,

Kenny

[From Bill Powers (980617.0905 MDT)]

Rick Marken (980616.2230)

Your diagram of coercion according to Nevin/Kurtzer et. al. is exactly what
I was going to try next, only I was going to use tbe example of walking a
dog in different compass directions. Coercion would exist for all directions
of walking except for those brief moments when the coercer wanted the dog to
walk in the same direction that the dog wanted to walk, plus or minus one
milliradian (or nanoradian, or whatever your definition of the "same"
direction is).

I like your suggestion that we (you and I at least) drop the term "coercion"
and just talk about one person controlling a variable that someone else is
also trying to control. Obviously we can't relate that simple "taxonomic
category" to the set of circumstances that others mean to indicate by the
word coercion, but it is a useful category nevertheless. It's a conflict
situation in which one person is so much more powerful than the other that
that person simply controls as if the other were not there. Apparently
there's no English word for that situation.

Best,

Bill P.

]

···

i.kurtzer (9806162300)

the custom is defining the interaction according to the intentions
of all the actors. For some reason, "coercion" does not follow
this rule.

In building a social science on PCT the tradition has been to
refer to all the actor's intentions that are involved to define
the interaction. That tradition exists in Bourbon's many papers
as well as three master's thesis's.

It's not a "tradition". Tom was modeling two-person interactions
where the results of the interaction depended on the intentions
of both people; in coercion, the results of the interaction depend
on the intentions of only _one_ person -- the coercer.

But if you want to model the coercion situation in terms of
the intentions of all actors, I'll oblige. Below is a graph
of the time varying intentions of two people with respect to
the state of same controlled (behavioral) variable, qi. r1 is
person 1's reference for the state of qi; r2 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

If the ASCII aligns properly what you should see is r1 and r2
varying over time. qi covaries along with r1 but not r2. r1
and r2 intersect at two points in time. At these times,
qi matches the references for _both_ person 1 (r1) and person
2 (r2).

According to my definition of coercion, person 1 is doing coercion;
qi (the controlled behavioral variable) is maintained at person 1's
reference (r1); variations in person 2's intention (r2) are irrelevant
to the state of qi. So person 2's intention is irrelevant to the
fact that person 1 is doing coercion.

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. So, by taking the
intentions of both parties into account, we find that coercion
can come and go instantaneously in two person interactions. If you
are comfortable with this definition of coercion it's fine with
me; I'll then try to stick to calling my kind of coercion "control
of behavior".

Would you be willing to agree that RTP is not coercive but that it
does involve the control of behavior?

Best

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

[From Rick Marken (980617.0800)]

Kenny Kitzke (980617.1000 EDT) --

For example, if a parent insists their child eat an ice cream
cone or get spanked, that does sound like coercion to me.
However, if the child would love to eat an ice cream cone, it
seems that calling the two-party interaction coercive simply
defys common understanding and language.

Why? The parent is still _insisting_, right? Coercion refers
to what the coercer does; not to what the coercee feels. The
dictionary says that coercion refers to _nullification_ of the
will of the coercee. That means that the will of the coercee
is _irrelevant_ to the coercer. The coercee will do what the
coercer wants _whether the coercee wants to do it or not_.
If a parent insists that a child eat an ice cream cone, then
that child will be made to eat an ice cream cone; whether the
child wants to eat it or not is irrelevant to the parent.

Best

Rick

···

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

i.kurtzer (980617.1200)

[From Rick Marken (980616.2230)]

i will think about your scenario.

i.kurtzer (980617.1205)

[From Bruce Nevin (980617.1131)]

i.kurtzer (9806162300)--

>In building a social science on PCT the tradition has been to refer to all
>the
>actor's intentions that are involved to define the interaction.

A problem: we can't determine the intentions of victims of coercion while
they are being coerced. The Test becomes somewhat complicated: remove the
coercion and Test, Test before or/and after coercion, ask what they would
be doing if they could (very problematic), probably other strategies I
can't think of. >>

Interesting point. I find this a methodological quandry as well, but to me
this thread is not about methodology per se, but about principles used to
demarcate types of interactions.

i.

[From Bruce Nevin (980617.1131)]

i.kurtzer (9806162300)--

In building a social science on PCT the tradition has been to refer to all

the

actor's intentions that are involved to define the interaction.

A problem: we can't determine the intentions of victims of coercion while
they are being coerced. The Test becomes somewhat complicated: remove the
coercion and Test, Test before or/and after coercion, ask what they would
be doing if they could (very problematic), probably other strategies I
can't think of.

As regards the value of qi it doesn't matter what r' is (the intention of
the victim) since the victim is overpowered by the coercer. This is the
original meaning of the claim that the intentions of the victim are
irrelevant.

As regards modelling coercion, the intentions of the victim cannot be
determined by the usual PCT means because their effect on qi is countered
by coercion. But we cannot pretend that the victim is not there, or model
the victim's output with a null value. That is not modelling coercion. That
is modelling control with an irrelevant bystander.

Rick's diagram includes the victim's output as a null (without a control
loop including r' there can be no value assigned to o' as a control
output). Instead, it seems necessary to postulate some best guess as to the
value of r' and then let the coercer overwhelm o' among the other
disturbances to qi.

  Bruce Nevin