Positive feedback; responsibility

[From Bill Powers (950204.1145 MST)]

Lars Christian Smith (940204 15:25 CET) --

     The more people became familiar with a particular clockface, the
     more they demanded it, the more the clockmakers produced it, etc.

Nice example, especially since it involves a social interaction NOT
behaving like a control system. This is an example of a flip-flop
positive feedback system, which has natural limits going in either
direction. During the transition you have a slow-motion picture of what
happens in toggle switch or an electronic bistable gate: the more you go
in one direction the greater the forces carrying you in that direction
until you hit maximum or zero.

A similar process might be seen during reorganization, where there is
more than one alternate behavior that will improve the situation. Once
you start reorganizing in one of the directions, chances are that you
will go all the way to the final form of that behavior, because going
back to select the other alternative would cause an increase in error
until you got close to the other final form.


Bruce Buchanan (950204.1245 EST) --

     PCTers like everybody else live in society, or more to the point
     live in families, and in work and legal relationships, where
     responsibilies are not fuzzy, and not only a matter of analysis,
     but are important matters of agreement and mutual expectations.

My point is that responsibilities don't become effective until they are
accepted. When they are imposed from without, the attempt to impose them
can easily be unsuccessful because behind them is an attempt by one
person to control the intentions of another person. It doesn't matter
whether it is important or even vital for a person to accept proposed
responsibilities; if he or she doesn't accept them, they will not come
under that person's control and that person will not, in fact, become
responsible for them.

The "importance" of taking on certain responsibilities is in the eye of
the beholder and dependent on the beholder's structure of goals. All of
society might be of the opinion that an absent father is responsible for
paying child support, yet there are millions of absent fathers who are
not factually reponsible for paying it: that is, they have not set
paying child support as a goal of their own, and hence are not,
factually, causing child support payments to be made.

Your comment seems to be from the prescriptive point of view, not from
the factual point of view. Of course we can discuss, argue, and reach
agreement about what responsibilities people OUGHT to take. But if they
don't in fact take them, they are not factually reponsible for making
whatever happen that is supposed to happen. If it happens anyway,
someone else was responsible. Your way of putting it simply exemplifies
how people go about trying to get others to assume responsibilies, in
this case by trying to convince them that there's a sort of natural law
at work, or that all right-thinking people would assume the
responsibilities. None of that bears on the question of who is factually
responsible for what.

     We are all raised in families, where our survival and prosperity
     depends greatly on whether we live within the rules we are given
     there. As adults we have more freedom of choice about the
     responsibilities we accept and make our own. Why should it be a
     problem to PCT if my higher level reference criteria for certain
     sorts of responsibilities are set to accord with those of my wife,
     say, or others?

Again, this is how a person would argue who is trying to get someone
else to assume a responsibility. It's no problem for PCT to explain how
a person accepts a responsibility and carries it out, or how a person
can reject a responsibility and thus not carry it out. Real people do
both. PCT has nothing to say about what people become responsible for --
that's a matter for individuals to work out with their friends, family,
colleagues, governments, and so on.

. . .[If] you don't choose to assume that
responsibility as your own goal, then I must try to control you into
taking up the responsibility. This brings reward and punishment into
the >picture . . .

     Certainly this circumstance is common enough. But I do not see why
     it must be the case in principle. It seems to me that an assumption
     that the individual is opposed to the acceptance of social
     responsibility is neither logically required nor the most common

PCT isn't about what some people do, or what most people do; it's about
what ALL people do. So from the PCT standpoint, all that can be said
about assuming responsibility is that doing so is intentionally taking
control of something and bringing it to or maintaining it at some
specific reference level. Whether most people will choose to accept
responsibilities that others want them to take, or whether they will
reject them, is irrelevant. Again, most people accept some
responsibilities and reject others; the theory has to cover both
outcomes, as it does. If a person does not adopt a suggested goal and
start working to carry it out, then those who still want that person to
adopt that goal can only start trying to control that person, with the
automatic result of resistance (as all control systems resist
disturbances). If the person does accept the goal, then that person
becomes factually responsible for whatever perception is now being
controlled by that person.

     The fact is that most people are very keen to be accepted and often
     only want to know what is expected of them to include this among
     their own reference variables.

Fine, if that's true. Finding out if it's true isn't a theoretical
question: it is or it isn't true, an empirical question. A theory has to
handle not only the positive instances but the negative instances, too.
"Most people" isn't "everybody." PCT explains what is going on when a
person takes on a responsibility, and what is going on when he or she
doesn't, and what is going on when someone else tries unsuccessfully to
make another person adopt a responsibility.

     This description obviously is intended to reflect the PCT
     framework. However when it comes to a discussion of the moral and
     legal implications of responsibility, this framework is not very
     useful. For it seems to me that the term responsibility for most
     people includes notions of commitment and accountability, which
     puts its meaning in terms of higher levels of language and more
     abstract and selective criteria to be applied to lower level

I think it's very useful. People are always talking about
responsibilities that people "have" and become frustrated and confused
when it turns out that in fact, they don't have them even when they
"should." The confused language in which we talk about responsibility
reflects confused thinking about it. The frustration and confusion show
up in the way we try to set up rules and apply them as customs or laws.
When, despite all our persuasions and threats, large numbers of people
simply refuse to behave as if they have the responsibilities we claim
they have (meaning "should have"), we have no way of dealing with this
but to apply physical force. In effect, we are baffled. We end up
concluding that there are good people in the world, who just naturally
know what their responsibilities are and carry them out, and evil
antisocial people who don't know their responsibilities and fail to
carry them out.

PCT tells us what a responsibility IS, where it resides, and how it
works. Once we understand that responsibility is not a THING that exists
in the air or can be passed on or injected into someone else, we can
begin to talk about social responsibilities in a rational way. The real
problem lies in how we negotiate with other people to share the work of
maintaining a livable world for everyone.

     Unless a person is emotionally arrested at a stage of unreconciled
     adolescent rebellion he or she will not feel that everyone else is
     always trying to impose upon them!

Do you always volunteer to do every good work that everyone you meet
wants you to do? Do you always agree with others as to what your
responsibilities to them or someone else or the nation or the
environment are? Do you know anyone who has different concepts of
personal responsibility from those that you accept? Do you agree with
every government regulation that affects you? I think that a person
doesn't have to be "emotionally arrested" (whatever that means) to rebel
against some attempts to impose unwanted responsibilties. You have to
get out of the parochial frame of mind in which you take it for granted
that all right-thinking people will accept the same social goals as
personal goals. What's your attitude to Palestinians, who say that the
West has a responsibility to right all the wrongs the Jews have
committed under Western protection? How do you feel about the ayatollahs
who claim that you have a responsibility to keep your wife's face
concealed and to pray five times a day facing Mecca and to shun pork and

Responsibility is often spoken of as if it existed
independently of any person. And the reason for doing this,
paradoxically, is to let the speaker pretend that his or her own goals
are not his or her own reponsibility.

     I think this is a fair statement of a situation that is common.
     However I also think that most adults would see this in effect as a
     possible con, an argument from authority which requires one to be
     wary and look for explanations.

I think you mean "most adults who are reasonable intelligent people of
the kind I would like them to be." I don't think you're talking about
real people. In my experience, large numbers of people regularly deny
responsibility for their own actions and try to put them off onto other
people -- "You're making me angry!" You're right that when people are on
the receiving end of this sort of treatment they do see it as a con, but
then they turn right around and do it to other people with perfect self-

. . . we are responsible only for the intended
perceptions, and their external counterparts.

     While I would agree with this (although noting that intentions are
     not always necessarily conscious), I would also note that the
     consequence may well be something we have made our own by a promise
     or agreement to perform, perhaps for some other benefit we expect
     to receive. As I see it, this need not be precluded by PCT.

It's not precluded by PCT; PCT has to accept what we see happening and
explain it. But even making a promise or agreement to perform is no
guarantee that a person will in fact be responsible for performing the
act. That depends on whether the person has accepted keeping promises
and agreements as a personal responsibility. There is no natural law
that says people have to keep their promises, live up to contracts. The
mass of laws on contracts attests to that. If they accept keeping their
word as a personal goal, they will probably keep their word. If they
haven't, they probably won't, or will do so only when it's convenient.

There is simply no level at which we can claim that people HAVE
responsibilities. They accept some and reject others, with whatever
consequences result.

But when a student does break the rule, administrators
tend to say "Well, you knew the rule, so the consequences are your
responsibility, not mine." That is dishonesty ...

     Why is this neccesarily dishonest? The fact is that the student
     knows that acceptance of the rules is a condition for attending the

It is dishonest because it conceals the intentions of the person who
actually decides to enforce the rules. That is why I spoke about the
terrorist first, who said that if the airplane gets bombed it's the
responsibility of the people who refused to free the prisoners -- they
knew the conditions, didn't they? It's easy to see the dishonesty when
you're on the other side. When you are the person laying off
reponsibility, it's easy to point to all the good reasons there are for
the rules, and thus why enforcing them is necessary. But the honest way
to do this is to say "I believe in this rule and I intend to enforce it,
whether you agree or not." In other words, you take responsibility for
your own beliefs and the actions you use to maintain them in force. This
is much harder to do, because often people feel ashamed of the things
their own rules seem to require them to do to other people, and they
would rather that some group entity or abstract principle take the blame
for the part of the situation they aren't enjoying.

     A theoretical position which serves to rationalize the opposition
     of students to administrative rules seems to me mistaken, as well
     as sending all the wrong messages to anyone who wishes to exploit
     such misunderstandings.

I disagree. If the students have not accepted the rules for themselves,
then the situation is not at all what the administrators think it is.
The administrators think that the problem is "disobedience", when in
fact it is the failure to take into account that students are also
control systems and can't help acting like control systems. If the
administrators take a punitive, stern, implacable, controlling attitude
toward the students, then of course the students are going to resist. If
they didn't, there would be something wrong with them.

A theoretical position can'be mistaken just because it leads to
conclusions that are inconvenient for someone. The fact is that students
do often oppose administrative rules, and this probably indicates
something wrong with the adminstration, or the rules. The students are
being required to do something that is contrary to human nature. But
this is the last thing administrators want to hear -- that they are
creating the problem themselves. What they want is for the students to

Ed Ford's associates have produced some startling effects by using PCT
in the classroom. There are definitely rules, but certain rules have
simply disappeared. As one of his associates put it, they have redefined
what is meant by a "discipline problem." A disipline problem is behavior
that infringes on the rights of the teacher to teach and the other
students to learn. Coming late to class, forgetting pencils and paper,
not doing homework, falling asleep in class, and other such problem
behaviors are not discipline problems. They may be academic problems, or
social problems, or emotional problems, but they are not discipline
problems and they do not call for displinary measures. They call for

And even a discipline problem is handled by offering choices within the
rules. The disruptive student can remain in class by ceasing to disrupt.
Or the student can go to what used to be called "detention" but has been
changed into a social skills class -- but even in that class the student
has the option of just sitting there, as well as the option of working
with an adult to figure out how to solve the problem. And the student
can get back into class at any time by negotiating with the teacher to
try some different kind of behavior to get what is wanted. Sure, all
this takes place within a framework of clear rules, but the discipline
that takes place has become self-discipline, not punishment. The
students are treated with respect at all times, even in the "social
skills class" and even after they return to class. They are treated as
human beings, as we understand them under PCT.

My point is that it's easy to blame the victims, but far more effective
to understand what is going on and act accordingly. PCT doesn't
prescribe any ways of dealing with children or anyone else. But knowing
PCT, people like Ed and his associates can start trying out different
approaches, gradually changing the relationships between rule-givers and
rule-livers in the light of our new understanding of how people work.
And this seems to solve some rather severe problems.

Interestingly, the teachers, students, bus-drivers, cafeteria workers,
and advisers are all quite happy with this program. But the
administrators, apparently, hate it.

      ... it seems irresponsible to argue that one might be justified in
     ignoring rules at any time because one has not given one's personal
     assent. This would not really be defensible in any organization,
     not least one devoted to education.

But you never obey a rule unless you have given your personal assent to
it, and adopted it as your own rule. It would be impossible to act
according to a rule that has not become your own goal -- who would
operate your arms and legs? You might adopt it because you really agree
with it, or because you fear the consequences of not adopting it, but if
you obey the rule, adopt it you did. I am open to a theoretical
discussion of how it could be otherwise.

The real question is how much conflict has been created by your adoption
of the rule. If you agreed with the rule, there was probably none. But
if you disagreed, you disagreed because following the rule would cause
errors relative to other goals you have. Being threatened with even
greater errors if you do not follow the rule may persuade you to adopt
it, but the other goals have not been dealt with and the errors are
still there. Unless some way is found to reduce the internal conflict
that results from accepting an unwanted rule, the person accepting it is
going to be under great internal stress and will continually be looking
for a way to lessen the other errors. So you'll end up with kids running
in the hall when no teacher is looking, or popping pills in the restroom
stalls, or sneaking out behind the gym to show off a new gun, or waiting
until school is out and then knocking up a cheerleader, or getting
knocked up. When you see shrieking kids exploding out of a school every
day at 3:15, you know there is something wrong. What are they

This is what Hugh Gibbons calls "respect for the will of others."

     Is it really? As described, it seems more like a mindless
     _acceptance_ of constraints imposed upon the situation by others.
     This is what I understand the Mafia mean by respect! This might be
     necessary for the very young and inexperienced but is scarcely
     desirable in principle.

The first thing you have to do to get along with other people is to
learn what they want and how their behavior enables them to get it. This
is just as true when you're dealing with the Mafia as when you're
dealing with a saint. Other people will always act to control their own
worlds in the states they prefer. And as long as that doesn't harm you
or anything or anyone you care about, more power to them. Sometimes,
however, you don't accept their goals; to respect the will of another is
not necessarily to agree with it. Before you can act rationally toward
another person, you have to understand how that person works. If you
decide to use force on the other person, you should understand what the
likely result is. If you're caught in an irrational system, such as Nazi
Germany 60 years ago, you need to understand that there are people whose
wills are causing the situation, and you need to respect that fact
before you start thinking of ways out. It would probably be to your
advantage to understand what the administrators of that system will do
if you break the rules they have set, and plan accordingly. Wishing that
they would change the rules, or hoping that you could persuade them to
make an exception, would not be realistic and would decrease your
control over the situation.

     ... perhaps the PCT framework would also accommodate the view that
     responsibility can be assigned after it is understood and accepted,
     i.e. incorporated in one's own reference criteria?

Well, I would be very interested in hearing how you would propose to
assign responsibility unilaterally. Isn't the whole problem the step of
it's being "understood and accepted"? AFTER it's accepted, there's no
problem with assigning it; it's already been taken. All the problems
come BEFORE it's accepted. I may understand completely what
responsibility you would like to assign to me, but that may only
increase my resolve to reject it. What, I'm supposed to turn on the gas
every morning in the "shower room"? No, but thanks for thinking of me.
------ (950204.1245 EST)

     Well, it still seems to me that "Feeling free", as a basic if
     subjective aspect of experience, in which we are aware of possible
     choices and the consequences of alternative commitments, is still a
     legitimate way to describe one perspective. Admittedly it is not
     the whole story. But then neither is any particular theoretical
     formulation of the situation, including PCT.

That sounds pretty harmless, a "legitimate way to describe one
perspective." I don't think it would be very satisfying to anyone who
wants to know whether the feeling of freedom is an illusion, and if so,
in what respects.

     To choose to ignore the "highest" goal implies, I think, an
     overriding goal or value. It may be a choice to end life in order
     to end pain and suffering. The lives of the saints show that
     people may die willingly for what they believe to be a higher
     value, perhaps represented by their god, and also believe this is a
     true assertion of freedom.

But that just pushes the problem up a level. I don't think we can go on
doing that forever; our brains are finite and when we try to go up one
level too many we find ourselves looking at bone.
----------- (950204.1245 EST)
RE: positive feedback

     I understand that this might be the point of view of the control
     systems engineer. From the point of view of the biologist and
     living systems, however, this perspective is incomplete.

     The S-shaped curve of natural growth involves processes of positive
     feedback in the early stages, when the active elements are few and
     the resources in surplus supply. At that stage growth may be
     exponential. Sooner or later, however, space and material resources
     - nutrients, blood supply - become relatively sparse, and further
     growth brings about negative feedback.

Sorry, but you wouldn't even be talking about feedback, positive OR
negative, if the engineers hadn't got there first. What we have here is
a TECHNICAL term that has been grooved upon by people who were free-
associating, not understanding what was being talked about.

Feedback is the effect of a variable on itself via a closed loop. If the
effect is in the same direction as an initial perturbation, the feedback
is positive; if opposed, negative. Positive feedback is not simply some
force that causes an effect that eventually comes up against a limit.
All your examples are _open-loop_ examples, and don't involve positive
feedback at all. An explosion may or may not involve positive feedback;
the main effect is simply rapid propagation of a burning front through
the explosive.

The only reason I object to loose usage of positive feedback is that the
resulting misinformation leads people away from understanding what the
phenomenon of positive feedback actually is, how it works, and why it is
the opposite of control. You're talking about different phenomena -- why
use the wrong word for them?

     It is not positive feedback that is the hazard as much as pride,
     ignorance and overweening ambition - a cybernetic interpretation of
     an old story!

That simply confirms my opinion that using the term loosely leads to
missing an important phenomenon. Positive feedback in a behavioral
system is always a hazard, simply because of the way it works.
Behavioral systems with positive feedback become unstable and destroy
themselves, for reasons which are easy to explain mathematically. There
are harmless and even useful forms of positive feedback, but not inside
a living control system except perhaps in LOCAL circuits. Positive
feedback in your example of the Sun would destroy the Solar System. The
stellar dynamics inside the Sun contains strong _negative_ feedback
which keeps the size of the Sun just right for getting rid of its
internal energy. When the star reaches the end of its life, the helium
is all used up (or is it the hydrogen?) and positive feedback sets in.
The star may expand into a red giant, or it may collapse in on itself
and then blow up in one gigantic oscillatory cycle -- a nova. Either
way, positive feedback will be the end of us.

Bill P.