🚀 go-pugleaf

[Martin Taylor 990505 23:20]
>[From Tim Carey (990506.0550)]

>6. Finally, in the reading of PCT that I have done so far, I have seen the
>environment aspect of the model discussed in either feedback function or
>disturbance function terms. I've been trying to fit a coercer, therefore,
>into either the d or f part of the model. The idea of "controlling qo or qi"
>seems to suggest that they are neither part of f or part of d. Is there a
>"coercer function" that I'm unaware of or some other aspect of the
>environment that I've missed.

The clue is that for any one person, every other person is or can be part
of the environmental feedback path in the control loop of one or more
perceptual variables.

That's what society is. The coercion situation is no different.

For the coercee (EE), the coercer (ER) has made it known that a certain
action A (not behaviour) will result in certain effects on a CEV (I insist
on the "E" in CEV--it means "environmental" as opposed to "perceptual").
ER thus implements part of the environmental influence path for action A.
Now, in the normal course of events it would be unlikely for EE's action
to influence the CEV in question, which is ordinarily the CEV of a
_different_ control system, influenced by _different_ actions, but
ER has created an influence path by which (as EE sees it) the action
will influence the CEV in question. (In one way of looking at it, the
influence of A on that CEV is a side-effect of whatever control A was
used for, but EE must be able to perceive that this side-effect path
exists. This opens a whole 'nother can of worms about perception of
environmental feedback functions and about learning, which I want to
avoid contemplating in this message).

Call these two control loops A (the one in which the Action is the one
being controlled for by ER) and O (the one in which the perception
being controlled is of the CEV that ER will influence with Overwhelming
force if the action in A is not what ER wants). Because A's action
normally would not influence the CEV of O, the two loops ordinarily
do not conflict. But ER has installed a new connection in the environmental
influence path, through him/herself, whereby the action A will disturb
the CEV if it is not as ER desires. This sets up a _potential_ conflict
in EE. The conflict becomes real when ER acts on the CEV.

So long as EE's action A is as ER desires, EE's control loop O
works with no problem. But if EE disturbs ER's perception of A away from
ER's reference value for A, then the CEV of O will be disturbed,
possibly with such force that EE cannot control the O perception,
possibly only gently, "reminding" EE that overwhelming force is
available and might be used.

Now see what happened in that last sentence. Not only is ER in an
environmental feedback path of EE, but also EE is in the environmental
feedback loop of a controlled perception of ER. Each forms part of the
environmental feedback path of one or more perceptions controlled by the
other. ER has chosen to make it so, and nothing EE can do will make it
not so. It is in this that the "unconcern" for the wishes of EE enter
the coercion concept. ER enforces the entwining of the environmental
feedback paths, each through the other. EE does not, and has no control
over whether those "through-the-other" feedback paths exist.

Cooperation, it seems to me, is different. In a cooperative relationship,
EE also is controlling a perception of the state of ER (so the labels EE
and ER lose their mnemonic quality). Conversation is like that. In our
paper in the soon-to-appear PCT issue of IJHCS, we identify ten perceptions
of states in the recipient of a message that are controlled by the
originator, and nine in the originator that are controlled by the
recipient. When all nineteen perceptions are near their reference levels,
the message has been successfully passed. I don't see coercion happening
here, even though each party will continue to act to alter the states
perceived of the other party, and perhaps may get pretty snotty about
it--"You are wilfully trying not to understand, aren't you."

The difference in the cooperative relationship is that each party refrains
from controlling those internal states that the other is attempting to
control. I have a problem, therefore, in accepting the assertion that
coercion occurs _regardless of whether the coerced action is a controlled
perception of EE_. With Isaac, I think that the word "coercion" is better
restricted to those cases analyzed above, where the two loops in EE that
I called A and O are brought into conflict by ER's self-insertion into
the environmental feedback structure of EE. It matters (if ER is to
perceive the interaction as coercive) that ER perceive EE to be controlling
the CEV to which the overwhelming force is to be applied.

I still don't see the value in using the word "coercion" to describe
cases of direct manipulation of another individual's body or parts thereof.

Martin