defining coercion

X-UIDL: 406266f26f48e5cfd8cb36a7f07b848d

[From Bruce Nevin (980510.2341 EDT)]

Bill Powers (980507.0234 MDT)--

I think we've drifted away from the central idea of coercion as I have seen
and experienced it (and sometimes done it). Coercion is control of somebody
else's behavior, not just incidentally while controlling some random
perception of one's own, but specifically to control a perception of the
other's actions.

I had misread your early exchanges with Tim Carey, e.g. your (980501.0211 MDT) in reply to Tim Carey (980501.0730). My fault. I was trying to accommodate the idea that the coercer disregards the coerced; all you said was that the coercer disregards the intentions of the coerced.

Suppose we just say that coercion is setting a reference level for another
person's actions and doing whatever is necessary to control them. Does that
cover enough cases to be useful?

Sounds good. Let's check through some cases.

I. If B resists, and A lacks the capacity to overcome B's resistance, then A is not successful in "doing whatever is necessary." So, OK, the definition entails whatever use of force is necessary, and requires that this use of force be sufficient to overpower B.

II. If B is doing what A wants already, A does nothing, but stands ready; in that case we don't know whether it's coercion or not unless we know one of these:

o A can overwhelm B.

o B is complying because of prior coercion or the threat of coercion.

o B wants to comply for some independent reason. (I'll return to this.)

Under the above definition, A is coercing B because A has reference perceptions for B's actions, even though A doesn't have to do anything. This case suggests an addition to the definition: that for B to be coerced, B must have intentions other than those served by the compliant actions.

III. If A is controlling some perception that B's actions disturb, A's resistance to the disturbance is not coercion (under the above definition) unless A specifically perceives B's actions and controls them. If A's resistance to disturbance indirectly constrains B's actions, but A is not controlling a perception of B's actions, then A is not coercing B. However, the environmental contingencies by which B feels these effects may be coercive, or may constitute a coercive system. Reading back through this thread, there seems to be some disagreement about the extent of being obligated to recognize when one's ends are benefitted by a system of contingencies that are coercive to others. Reminds me of the Black Panther/Weatherman notion of white-skin privilege.

ยทยทยท

-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-

Tim Carey's "signing off" summary (980506.0618) was in part:

2. The coercer may, in some instances, be helping or assisting the coercee
but this is still coercion.

Intending that one's intervention be for the other's benefit does not necessarily constitute helping them, depending on what their intentions and level of effort are. So helping has this interesting relation to coercion that it is similar but the helper is controlling a perception of the other person's intentions--something the can either infer from the Test or imagine but cannot perceive in their observations of the other person. It seems to me that helping has to be wanted by the person being helped. If that is the case, then helping is like collusion or cooperation, except that the helped person (like the coercer) requires certain actions by the other person in order to control his own perceptions. In both cases (the coercer the helped person) they might have reference perceptions for specific actions ("Just hold that hose in place, would you, while I tighten this." "Hold that damned thing tight or I'll whop you up 'side of the head again."), or they might require the other person's expertise ("It just died on me! Can you fix it?" "Fix that carburetor or I'll shoot you!").

A rule of thumb about helping is that at least 50% of the effort comes from the person being helped; less than that and the person is being rescued (Eric Berne & Claude Steiner). So under these definitions of helping and of coercion, they are mutually exclusive, but a coercer may believe that she is helping.

3. To the coercer the coercee's references are irrelevant but they are not
irrelevant to an observer who is interested in modelling coercion.

To model coercion it's enough that the coerced system have intentions that conflict with the coercer's intentions. The modeller does not have to know the specific intentions unless modelling a specific coercive relationship.

4. An observer needs to either Test or observe intelligently an interaction
to determine whether coercion is going on.

You can't apply the Test to the coerced system because by definition control is swamped by actions of the coercer. (For this reason I don't think Rick will be able to pull out a projection of what the participant's actions would have accomplished if the coercive resistance of the demo control system were not present. All you get is the participant's futile resistance to its resistance. You can get the sign--you know that I want the numbers bigger/smaller--but you can't tell what my reference is.) You can apply the Test to show that the coercer is controlling a perception. If you can show that the coercer's effort, or capacity for effort, is so strong that any disturbing action by the coerced system would be futile. This is enough.

-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-

IV. Consider again the third bullet under (II) above. What if B independently intends the same state of the environment as A does? The possibilities include:

o A has a reference perception of the manner of B's role in effecting this. Some particular sequence of steps, say, where the different sequence of steps that B knows (or the substitution of some different steps) brings about the same effect. A may coerce B into doing it the expected way.

o A has a reference for the outcome rather than the particular actions involved. A lets B bring about the mutually desired result in some unexpected way.

o A does not even know the actions but only the outcome, and must trust B's expertise.

V. A does knows only the outcome, not the tributary actions, and must rely upon the knowledge and skill of B to produce the outcome, yet A is coercive: "Fix that carburetor or I'll shoot you." "You're not going to be a musician, you're going to medical school to become a doctor, or I'll cut you off without a penny!" These are instances of coercion that constrain outcomes without touching on the particular actions that effect those outcomes.

-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-

Ordinarily, we social beings have some perceptions of the intentions of others, and we control to accommodate them. "I'm sorry, I didn't mean to get in your way." The coercer A does have some perception of the intentions of the coerced system B and sets the reference for accommodating B's intentions to zero.

How can we perceive another's intentions? We imagine them based on what we would do in their situation, or we infer them from the other person's resistance to disturbances that we perceive--informal variants of the Test. The clearest form of resistance to disturbance, real or anticipated (imagined) is when people declare their intentions. Butch Cassidy "Hey! I'm walking here!"

We ordinarily accommodate others. It is also true that we are impatient and bullying, but that is the distinction that makes for coercion.

So I think you cannot clearly understand a model of coercion without modelling accommodation also. When you can model how control system A includes among the perceptions that it is controlling a perception of the intention of another control system, then you can model what it is that a coercer is capable of controlling but does not.

A coercer sets a reference level for another person's actions and does whatever is necessary to control them. The coercer knows that the other person's actions effect their intentions, just as her own do. The coercer disregards the other's intentions as of no account.

If one control system overpowers another, but is unable to perceive the consequences that its actions has for the other (the influence through the environment is not apparent), it is not coercing. The overwhelmed control system might declare the violation, unless too frightened. If the declaration is recognized as such, but ignored, then the overpowering becomes coercive.

-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-

We started with "coercion is setting a reference level for another person's actions and doing whatever is necessary to control them."

On looking at some cases, two things seem to be missing. A coercer can have a reference perception of an outcome for which the coerced system's contribution is required, but without having a reference perception of the coerced system's specific actions beyond a kind of generic "engaged in bringing about the outcome". And the coercer disregards the intentions of the coerced system.

How about: Coercion is requiring the participation of another control system in order to control a perception, and doing so regardless of the other's intentions. (Controlling a perception already includes "doing whatever is necessary".)

The definitions of cooperation and collusion are similar except that the intentions of others are accommodated. Collusion seems to me to be cooperation in coercing others.

  Bruce Nevin

[From Bill Powers (980511.1057 MDT)]

Bruce Nevin (980510.2341 EDT)--

I'm not quite sure why we want to define coercion in such meticulous
detail. Even a perfect definition won't help us converse with others who
use the word, because they won't be defining it the way we do. For example,
Tim Carey wouldn't be assuming your definitions, because he says that the
coerced person should _feel_ coerced.

If we say that A has a reference level in mind for B's behavior ("Sit up
straight, Johnny"), we can make various assumptions about the parties
involved, to generate predictions of what they're likely to do next. Is
there something else we should be doing with this? If I ever knew, I forgot
what it was.

Best,

Bill P.

[From Bruce Nevin (980511.1357)]

Bill Powers (980511.1057 MDT)

I'm not quite sure why we want to define coercion in such meticulous
detail. [...] Is
there something else we should be doing with this? If I ever knew, I forgot
what it was.

If you don't know what you're talking about you can't model it.

An alternative, of far less value IMO, is to model something, say it is a
model of coercion, then insist that the technical definition of coercion
within PCT is what the model models. That *really* screws up communication.

  BN

[From Bill Powers (980512.0144 MDT)]

Bruce Nevin (980511.1357)--

An alternative, of far less value IMO, is to model something, say it is a
model of coercion, then insist that the technical definition of coercion
within PCT is what the model models. That *really* screws up communication.

Right. The first thing to do is not to produce a model, but to say what you
mean by coercion. Obviously there's some phenomenon or some experience to
which people refer by that word.

For me, coercion comes in two flavors: doing and experiencing. When I do
coercion it is my goal for someone else's behavior that counts, that
provides the defining element. When I experience coercion it is the force
or threat of force someone else uses on me if I deviate from a certain
behavior that counts. When I think of A coercing B, which one counts
depends on which point of view I take, A's or B's. As a modeler, I see
these phenomena and experiences as involving conflict between unequal parties.

The test of coercion comes when B deviates accidentally or on purpose from
the behavior A desires to see: A immediately applies persuasion, threats,
or force to bring B's behavior back to A's reference level. When B's
behavior is satisfying A, A does nothing to B, so the link is not readily
apparent. But this is as temporary a state as is the state of a car that
just happens to be following the path desired by the driver with no force
applied to the steering wheel. The driver is fully focused on the movement
of the car and will exert steering efforts at the first sign of deviation;
so it is for the person trying to control someone else's behavior. The
controller, even when not acting, is hovering over the controllee, watching
and listening with sharpened senses and hair-trigger corrective action at
the ready.

Of course maybe I'm using "coercion" to mean "a relationship with others
that I particularly dislike," so other people's concepts of coercion could
be entirely different.

Best,

Bill P.