[Martin Taylor 950821 11:00]
Various August messages from Bill Powers, Rick Marken, Hans Blom and
probably others to whom I apologize.
There have been quite a few messages over the last couple of weeks about
the place of "rules" in human-human interaction. Unless I missed something,
all of the discussion has dealt with a "rule" as some codifiable set of
instructions directing "correct" behaviour in particular contexts. That's
indeed what laws are, but I don't think much of our social interaction
works with such "rules." What we should be discussing is, I think, a
two-fold progression: the development of "conventions" and the codification
of conventions into "rules."
By conventions I mean just the same sort of thing as arises from reorganization
in a non-social (e.g., passive physical) environment, whereby one learns
to push or pull according to whether one wants a particular perception to
increase or decrease. There is no necessary perception of "This is the
situation--therefore I must push" (a "rule"). There is simply a perceptual
signal that is greater or less than a reference value, a consequently an output
of the appropriate sign to bring the perception nearer its reference value.
The actor need not perceive the connection explicitly; the appropriate
output "just happens."
But to an external observer (which we all are, in respect of any other person),
the output (push or pull) seems to happen regularly when we perceive the
other person to want something and not have it as we perceive them to desire.
The external observer can codify the reorganized control structure's effects
as an explicit "rule": when X wants A and the situation provides him with B,
then he often does Q.
If the external observer (not being an S-R psychologist who has omitted
the "wants A" part of the "rule") notes this kind of regularity and notes
also that the actions Q seldom have "bad" effects on other people (i.e.
generate little conflict), the external observer may be tempted to rephrase
the observation as a law: "When a person want A and the situation provides B,
the person SHOULD do Q." It is this kind of a rule, a codified and
enforced convention, against which Bill Powers was arguing ((950811.1150 MDT)
If you, personally, want human beings to live
and work in close proximity, then you, personally, will advocate that
there be certain ground rules, and you, personally, individually, will
do what you can to define what you think they should be and to persuade
or force others to accept them....
Now move on, from actions relating to effects on the non-social world to
actions with social effects--effects that often serve as disturbances to
the controlled perceptions of other people. Start again with ordinary
reorganization, with no "rules." What is likely to happen?
Let us suppose that in respect of some perception of a non-social CEV,
many people have reorganized so that "when the person wants A and the
situation provides B, then the person often does Q." Now imagine a person
X who wants a person Y to perform act Q and perceives Y to want A. It may
often happen that X eventually reorganizes in such a way that under these
conditions X will act to provide Y with situation B. Just as with the
"push-pull" example I started with, X may have no awareness that this is
happening.
But a social convention is beginning to develop: when person Y wants A,
Y acts so as to find a person X who wants Q and can provide situation B.
When a person X wants Q, X acts so as to find a person Y who wants A and
for whom X can provide situation B.
As an example, substitute in the above: A = money, B = employment, Q = some
skilled behaviour. But one can substitute many other equivalences.
There's no rule here, but a third party, Z, can observe the behaviour of
many X's and Y's and note that often when an X (e.g. a person with money)
wants Q (some skilled action to be performed), and a Y (a person without
money) is available who can act according to Q, X will provide for Y and
Y will accept from X a situation B (which we may call employment).
Having observed this, Z may codify a "rule" that says that under these
conditions Y SHOULD provide Q, or a rule that says X SHOULD provide B,
or both. The convention that is often beneficial to both has become a
"rule" satisfactory to Z, but not necessarily always to X and Y.
But if you tell them they should do all these things because there are
social forces that make them necessary or right, then you're lying.
Now one may ask what might be meant by the term "social force." In
simple mechanics, a "force" is what makes things move. It's an inferred
concept based on the observations of things moving or not moving under
a variety of concepts. Until Newton (Galileo, Aristotle??) there was
no such concept, but that didn't stop people from controlling their
perceptions by applying what we now call "force." Force is what makes
situations change if certain actions are done.
That notion of "force" can be extended, and must be extended, if we are to
apply to it the adjective "social." A "social force" must be what makes
situations one perceive change if one performs certain actions. It doesn't
matter for physical force that the perceived effects occur through a vast
complex of mainly electromagnetic and gravitational interactions, and it
doesn't matter for "social force" that the perceived effects occur through
the output actions of other control systems disturbed by one's own actions.
And it doesn't matter for either that specific actions don't always have
the same effects.
If we accept this notion of "social force," does "social force" exist?
If there is a convention developed through reorganization in a multitude
of interacting people, then certain actions will often change situations
in reliable ways. Not always, but often, which is true also of physical
systems. If there are "rules" or "laws" the reliability is increased,
but not guaranteed. One has to recognize--perceive--a law, which is not
the case for a convention. One may well act according to a convention
without recognizing that one is doing so; it is just a result of
reorganization.
The above seems to me to establish that it is valid to talk about a concept
of "social force." There's nothing "right" or "wrong" about a social force,
any more than there is anything right or wrong about the wind that may
disturb the tracking of your car down the centre of its lane. But if you
want the car to stay in the centre of the lane, you apply force in the
opposite sense. And so it is with social force. If social forces change
situations in ways you do not want, you apply social force in the opposite
sense, if you can. "If you can" means either according to conventions
embedded in your reorganized structure, or by actions at the program level(?)
using perceptions of rules (e.g. you hire a lawyer). Social forces make
certain actions "necessary" in exactly the same sense that physical
forces make certain actions necessary--no matter what the mechanism, the
"necessary" action serves to move the perception in such a way as to
reduce the error.
Social systems don't define "responsible." People do. You are behaving
responsibly if you're doing everything I think you ought to be doing,
and not doing anything I think you shouldn't be doing.
"Responsibility" is another ill-defined term. One might define it as acting
so as to control one's own perceptions with minimal side effects. Since
side-effects are not observable (by definition) to the system doing the
controlling, this definition ALWAYS requires an outside observer, though
that observer might be within the same brain as the controlling system
(conscience?).
One might define "responsbility" as controlling in such a way as to minimize
conflict (within the same body or with other people). But conflict is again
not perceived directly. This definition also requires an external observer,
which may be in the same body or outside it.
As you say, social systems don't define "responsible." People do. Or, I
would add, observers internal to one's own brain do. "Responsibility"
may be asserted by busybodies, parents, policemen... or by one's own
conscience--the perception of the effects of one's actions in comparison
with the reference perception for those effects.
Social systems don't define resonsibility under any of these definitions,
but they do determine it. The "social forces" generated by convention and
codified convention determine the levels of side-effect and to a large
extent the degree of conflict consequent on any action in any particular
context. There is nothing necessarily right or wrong about generating
conflict, but reorganization tends to alter people in such a way as to
reduce conflict, leading to conventions. When such conventions have
developed, then people tend to SAY that acting according to convention
defines "right" (= moral, not legal) actions. When these conventions
are codified, they become laws, which determine "right" (= legal, not moral)
or "wrong" actions.
This view of convention as the semi-stable result of multitudes of individual
reorganization is contrary to what Rick Marken (950815.1010) seemed to imply
to Hans Blom:
the rules are arbitrary; people make them up or learn
them and then decide to follow them (control for following them) or not all
on there own.
The codified rules may be arbitrary, but the conventions by which we live
are not. There may be an infinite number of different sets of conventions
that would allow for stable social systems, and we seem to see many such
sets in the world, but within any one set of conventions, those conventions
exist because over time they tend to reduce conflict. If they didn't,
people would often reorganize in such a way as not to act according to those
conventions, and the conventions would disappear.
PCT shows that people will follow the rules only if they want to (have a
reference for following them); and adults will have a reference for following
the rules only if doing so satisfies one of their higher level goals.
Yes, exactly so. But this "wanting to" need in no way be conscious. And
for rules that are codified conventions rather than arbitrary choices of
some dictator, most of the time, most people will "want to" follow them,
no matter how bizarre they may seem to someone from a differen social system.
Hans Blom (950821) brings up the point that one needs a more refined "model"
of a familiar person than of a "person-in-general." Clearly so, but this
"model" need not be perceived any more than a "rule" need be codified.
Reorganization occurs within an individual, and if two persons interact
particularly frequently, then each has a more profound effect on the
reorganization of the other than does any particular person from the general
public. Interpersonal conventions develop in exactly the same way as do
gross social conventions; and sometimes those interpersonal conventions
get codified into "rules." (My wife used to be quite fond of saying "let's
make a rule" but she does it much less often now. Is it because I tend
to follow her rules more than I did, or because she has unconsciously found
that I don't? I don't know, and I bet she hasn't noticed, either. But we
get along fine, either way).
Societies consist of individuals, all different. Yet,
some individual behaviors (stimulus-response patterns or elementary
control systems) are remarkably alike in a given society, culture or
subculture. Therefore a model of a _society_ may be helpful, even if
it applies to not a single individual of that society. Such a model
may be so general, however, that it easily leads to stereotypes.
Explicit models of a society are for sociologists, and for lawyers. But
if you accept, say, that neural networks "model" in their interconnections
the patterns that they attempt to classify, whereas expert systems do the
same "modelling" in their rule structures, then I can accept that people
"model" their society in the structure of their reorganized control
hierarchy. Either way, reorganization or explicit model, what you say
is correct.
Bottom line: in social interactions, HPCT with random reorganization will
normally lead to quasi-stable sets of conventions for action. These will
tend to represent local minima in the sense that changing them increases
the degree of conflict within the society. If these conventions are made
explicit, in the form of laws or "rules" that MUST be obeyed, it is quite
possible that conflict will increase (along with the employment of lawyers).
Social force is as respectable a concept as is physical force, and one
effect of codifying convention is to increase the forces involved in
control of perceptions that involve the actions of other people.
Final parenthetical comment: All the above applies to the development and
use of language.
Martin