AWOL; importance

[From Bill Powers (921020.1100)]

David Goldstein (921020) --

There seems to be a conflict at your school regarding keeping
residents from going AWOL. If a zero level of going AWOL were the only
goal, then a simple way to achieve it (other than Rick Marken's) would
be to lock the doors and bar the windows and have armed guards to
prevent escapes. This would achieve the desired rate of AWOL
incidents: zero.

Obviously, this is not acceptable in relation to other goals:
financial, humanitarian, and conceptual (this is not REALLY supposed
to be a prison). This unresolved conflict leads to using methods that
are ineffective, which result in perhaps a slight reduction in AWOL
incidents and leaves other goals not fully satisfied.

There is another problem, as I understand it: your institution
operates within the constraints of a coercive legal system. The
residents are assigned to you against their will, and are not free to
leave whenever they want to. If they escape they will either be
returned to your institution or sent to a different one. There are, no
doubt, sufficient reasons for doing this in the sense that many people
are convinced that it's necessary. We can't change that.

So it seems to me that your problem as people working within the
institution is to find the least conflicted way of operating within
the external constraints imposed by the legal system. The most
important conflicts are not those between staff and residents, but
those between and within staff members. To eliminate those conflicts,
you must first find out what they are: what goals are behind each
method that is used, each rule, each principle? Then you must ask how
effective the actions taken to achieve the goals are, and whether each
action tends to satisfy ALL goals or tends to satisfy SOME goals while
exerting effects away from other goals.

Consider the goals involved in the use of operant methods. If operant
methods work, why not use them in the most effective way? The most
effective way is to make sure that residents can gain access to
neither food nor water unless they behave as you want them to behave.
No matter how hungry or thirsty they get, you must not reinforce them
until they perform continually closer to the standards you have set.
You must of course make sure that they can't give each other any food
or water, or find a way to get them themselves, or escape entirely
from the conditioning situation. You can put any conditions you like
on getting reinforcements: the residents must speak politely to each
other and the staff, use good middle-class English, not use words or
phrases offensive to any religion, sexual custom, or place of origin,
study a certain number of hours per day, sit quietly when not usefully
engaged, and keep their quarters neat and clean. The alternative is to
die of starvation or thirst. This method will work as long as the
residents are in the institution.

If you reject using an effective method, then you must have reasons
for not doing the things that are required. These reasons contain or
are related to goals that are in conflict with the goal of achieving
the above list of ideal behaviors (or whatever your list is). By
examining each thing that is done in this way, simply asking what
would be require to make it work with maximum effectiveness, you can
uncover all the conflicting goals that prevent using any approach in
its most effective form.

When you have truly laid out the list of conflicting goals, you will
realize the true magnitude of the problem, and will begin to
reorganize. You may find that you must prioritize your goals -- it may
be more important to prevent physical violence, for example, than to
get residents to avoid offending the staff with their language. You
may decide to change your definitions of desirable behavior. You may
eliminate the offense called AWOL by making leaves freely available
(to those who are not absolutely and inescapably confined). You may
decide to work out a new system entirely, in conference with the
residents, that minimizes all kinds of conflicts. There are many
creative solutions that will occur to you once you have convinced
yourself that your whole approach to the residents is in a state of
severe internal conflict.


Greg Williams (921020) --

The only way for the action to become important to the influencee
would be for it to be perceived and compared with a reference
signal of its own.

I agree, but it can become important RETROSPECTIVELY: "Ooops! I was
wrong to do that!"

The only way this can happen is for the action to disturb some other
perception in an uncorrectable way. The action itself is not
perceived, normally, and can never have a reference level. The effect
on the other perception is the first thing you notice; it may take
some time to track down the action responsible, because normally we
pay no attention to our actions. This is simply a case of conflict, in
which the action that controls one perception makes control of another
impossible. The fact that this takes place over time means only that
the solution may have to be found in a higher level of control. But
only the perceptions are important to the person; the actions causing
the problem will be changed without hestitation, as required. The
actions are not important. Only perceptions are.

And it can be important TO THIRD-PARTY "OBSERVERS" (such as
sociologists and police officers).

So, they have their own perceptions, and try to control for them. So
what? This still doesn't make actions important to the person using
them for controlling perceptions.

And again I say that many people attach great importance to the
outcomes of social interactions involving persons who are
controlling their perceptions which are dependent on others'
actions, regardless of whether the interactions are "important" as
you define it (that is, involving reorganization of the control
structure of someone upon whose actions another's controlled
perception depends) or are "unimportant." I'm saying that your
definition is simply not relevant for people who study such
interactions or for people who are actually involved in such

My definition is precisely relevant to just such people. Their
perceptions are important to them, if there are reference levels for
them, but their actions are not. The person who studies interactions
considers interactions important to perceive. The actions through
which such perceptions are achieved -- going to the library,
interviewing people, taking pictures, taking notes -- are unimportant
and will be changed as circumstances require. People actually involved
in interactions consider their own perceptions (of themselves, the
other, and the relationships between them) important, but will take
whatever actions are needed to maintain the perceived interaction in
the state they want. There's no preference for any particular action;
all that matters is the state of the perception.

Please note:

... "important" as you define it (that is, involving >reorganization

of the control structure of someone upon whose >actions another's
controlled perception depends) ...

At least try to argue against me instead of a straw man. I gave this
definition yesterday:

"DEF: Something is important to a person if the person perceives it
and has a reference signal to compare it with. "

If you like, I will expand it slightly: " ... and tries to control
it." Although that is not always the case -- one may not know how to
try to control it. The definition above covers it, I think.


Bill P.