Power in PCT (was PCT and Politics)

CSGnet Archive CSG2008
1

Bill Powers wrote:

[From Bill Powers (2008.08.23.0740 MDT)]

Fred Nickols (2008.08.22.0636 MDT) --

I grew up with a pretty commonplace notion of power; namely, the ability to control others. In the course of being trained as an OD specialist in the Navy, I came upon and adopted a very different definition of power: the range of options at one's disposal. The more options you have the more power you have. I also came to believe that the best way to acquire power (i.e., to increase your range of options) is to study, learn, experiment and find out what works and what doesn't work. Another way of increasing your own power is to increase the power of others (i.e., increase their range of options, even if that boils down to simply helping them see alternatives). In PCT terms, you help them do a better job of obtaining/maintaining reference conditions for variables they wish to control. They tend to return that favor.

Pretty much my history, and my conclusions, too.

What this says is simply that "power" means nothing more nor less than the ability to control: to act on one's world in such a way as to experience it the way you want to experience it. Starting with the idea of controlling other people, we both came to realize that this is an illusory goal: the more you try to control others, the more they resist -- and there are a lot more of Them than there are of You.

My notion of "power" in PCT is close to what you two agree on, but not identical (at least I think not).

Bill says ""power" means nothing more nor less than the ability to control: to act on one's world in such a way as to experience it the way you want to experience it." So far, so good. But this can be broken down into two parts (at least): that environmental feedback paths exist that a person could use to influence the controlled perception toward its reference value, and that the person has reorganised so that the output of the control process influences the useful paths more than it influences environmental feedback paths that would oppose the desired change in perception.

It's the second part of Bill's response with which I disagree (as I have done in respect to similar statements many times in the past). It simply isn't true that the more you try to control others, the more they resist. That occurs ONLY if your methods of controlling the other induces a conflict, as, for example, if the other perceives that the action you desire would please you and the other has a reference to displease you.

Since we are social animals, MOST of our power comes from getting other people to do what we want -- in other words, controlling them. This almost always is done without the slightest resistance on the part of the helper, whether the helper is aware of being controlled or not. If I ask you if you would mind opening the door because my hands are full, I do not expect you to say "No, because I resent being controlled". I expect you to say "Sure", and to open the door. Why? Because most people control for a perception of themselves as being civil and helpful rather than being uncivil and obstructive (and perhaps the person is controlling for seeing you to be pleased). Almost all of our real power comes from controlling other people in this kind of way.

Powerful politicians exert their power by controlling lots of people to act in concert, not by force or threat of force (except for military dictators), but by disturbing some perception many people control so that their control actions influence the politician's controlled perception appropriately (e.g. getting them to vote for him/her). Business managers control their employees in the same way. There may be internal conflict in the employee between acting to perceive the manager to be pleased (or equivalently to perceive the paycheck as reliable and growing) and acting to control some other perception, but usually the employee finds a reorganization that reduces or eliminates the conflict.

The main difference between powerful people and those who are not so powerful is that the powerful either have more means of influencing their (social and physical) environment or that they have reorganized better to use the means at their disposal. In our society, "means" often means "money", but money isn't everything. Friendship, honour, and the like can also be perceptions for which people have reference values, and for which disturbances can induce actions. A powerful person can control others by disturbing those perceptions or by providing environmental feedback paths that allow others to control them (offering the possibility of a knighthood used to be quite effective, and cost the monarch no money). Consider the power influenced by Ghandi -- greater than that of the British Empire -- and how little money he needed to exercise that power over people. He had the means, and he used it. He controlled many, many, people, not all of whom agreed with his objectives.

Power, to me, is essentially what Bill says: "the ability to control: to act on one's world in such a way as to experience it the way you want to experience it". A great deal of that ability is the ability to get other people to act so as to bring your experience of the world closer to the way you want to see it. Rick would like to get people to act one way (vote for Obama) so he could experience a world in which poor people have a little more power than they do, or than he thinks they would if people were to act in another way (vote for McCain). Rick doesn't have much power to influence that perception, but to control people toward voting for Obama is one of the possible environmental feedback paths that give him what power he has in that respect.

Looked at this way, "power" is largely about getting people to act in a way that brings your perceptions closer to their references, and that simply means controlling other people. Controlling another person won't work well if the method induces a conflict either in the other person or between their perceptual control systems and your own, but it works very well most of the time, in any "civil" society. We and our cultures have evolved so that it does.

Martin

2

[From Bill Powers (2008.08.23.1325 MDT)]

Bill says ""power" means nothing more nor less than the ability to control: to act on one's world in such a way as to experience it the way you want to experience it." So far, so good. But this can be broken down into two parts (at least): that environmental feedback paths exist that a person could use to influence the controlled perception toward its reference value, and that the person has reorganised so that the output of the control process influences the useful paths more than it influences environmental feedback paths that would oppose the desired change in perception.

It's the second part of Bill's response with which I disagree (as I have done in respect to similar statements many times in the past). It simply isn't true that the more you try to control others, the more they resist. That occurs ONLY if your methods of controlling the other induces a conflict, as, for example, if the other perceives that the action you desire would please you and the other has a reference to displease you.

That is why I say that the most likely result of trying to control others is
resistance. No person knows enough about another to specify reference conditions for another's behavior without (intentionally or inadvertently) causing conflict for the other person: the person thinks, "if I do as you ask, I will cause an error inside myself."

In many case we learn through experience, or explicitly agree, that certain requests for behavior on the part of other people are complied with by reasonable people. Pass the salt, open the door, lend me your wife. Oops, that last one is not included in most agreements, and we quickly learn that the range over which we can expect willing compliance is not very large in comparison with what we would find most convenient.

Even innocent requests that a person would normally be willing to consider can arouse conflict. Therapists run into this all the time -- in fact, more often than most of them I know realize. It's commonplace for a therapist of normal persuasions to offer advice or prescribe actions to clients, but as listening to recordings of sessions in which this occurs will quickly show, most of the suggestions are immediately rejected by the client. Of those that are apparently accepted, some fairly large fraction are not carried out: that's called "noncompliance", and is common enough to have a name (the rejections also have a name: "resistance").

"What you need is a hug", you say to a distressed person standing apart from the group, not realizing that this person has just been told she has terrible body odor. When you try to control someone else's behavior, you simply don't know what the existing behavior is accomplishing for that person, that would be interrupted or even contradicted by changing to another behavior.

So Martin, I am not saying that people resist control by others just because they don't like being controlled (though that does happen). They resist because what the other person wants them to do would cause errors inside them of which the other person knows nothing. The resistance is not directed at the person making the request; it's directed at preventing or reducing the error that would be caused by compliance. I think you may have been assuming that I attribute the resistance to the mere attempt at control itself. What I've been trying to say is that the resistance arises because complying with the request would cause other errors to arise. Though on a less dramatic scale, it's like one person requesting another to unscrew a light-bulb (no problem) and stick his finger into the socket (BIG problem).

Best,

Bill P.

···

At 02:33 PM 8/22/2008 -0400, Martin Taylor wrote:

3

[Martin Taylor 2008.08.22.17.11]

Sorry about missing the date stamp on my last....

[From Bill Powers (2008.08.23.1325 MDT)]

It's the second part of Bill's response with which I disagree (as I have done in respect to similar statements many times in the past). It simply isn't true that the more you try to control others, the more they resist. That occurs ONLY if your methods of controlling the other induces a conflict, as, for example, if the other perceives that the action you desire would please you and the other has a reference to displease you.

That is why I say that the most likely result of trying to control others is
resistance. No person knows enough about another to specify reference conditions for another's behavior without (intentionally or inadvertently) causing conflict for the other person: the person thinks, "if I do as you ask, I will cause an error inside myself."

No, but consider...we have evolved to be social animals, meaning that members of a group help each other more than they act against one another. As individuals, we have reorganized to act with respect to others so that they are more likely to act helpfully than to act against us. Evolution and reorganization change the odds that how you act when you want someone to do something will generate the kind of conflict that will prevent them from doing what you want. It's survival of the fittest means of control.

Whenever we want someone else to act in a certain way, we must disturb some perception they are controlling (or, perhaps, provide them a means of controlling that had not been available to them). We create (or take advantage of existing) error in that person. They act to reduce the error, and we hope that the action is the one we want. Cultural conventions (reorganizations in individuals akin to those involved in learning the communal language) make it probable that this will often be the case, just as when we make noises with our mouth, the other is more likely than not to understand what we want them to understand.

In many case we learn through experience, or explicitly agree, that certain requests for behavior on the part of other people are complied with by reasonable people. Pass the salt, open the door, lend me your wife. Oops, that last one is not included in most agreements, and we quickly learn that the range over which we can expect willing compliance is not very large in comparison with what we would find most convenient.

Of course, in a black-and-white sense, you are right. Where I would demur is in "not very large in comparison with". Most of the time, when we really want someone to do something that they are able to do, we are not being mischievous. We don't ask them to do things we expect they will not want to do, unless we quite deliberately set up a conflict in the other (e.g. "lend me your wife and I will give you $1 million; refuse and I will cut off your hand"). even "Please open the window" is problematic if we know the person has just closed it to stop a draft, or to block outside noise.

That kind of conflict, in less dramatic form, probably occurs in most employer-employee relationships, in which the employer believes the employee would not do the work if left alone. Who would choose to put wheels onto semi-finished cars if they could be fishing? So, the employer offers an improvement in the employee's environmental feedback paths for controlling other perceptions, by offering money if the employee accepts a level of error in the "what I am doing right now" kind of perception. Yes, in controlling other people, there is often an induced conflict. There certainly is induced error. Systems not being linear, most of those conflicts don't escalate. Some do, and some lead to therapy, some to revolt. But compared to those that don't, the problematic ones are very few.

So Martin, I am not saying that people resist control by others just because they don't like being controlled (though that does happen). They resist because what the other person wants them to do would cause errors inside them of which the other person knows nothing.

I agree that this can happen. That's not in dispute. What is in dispute is that it is the normal case and that the conflict is severe enough in most cases to impede the person from acting as the controller wishes.

The resistance is not directed at the person making the request; it's directed at preventing or reducing the error that would be caused by compliance. I think you may have been assuming that I attribute the resistance to the mere attempt at control itself. What I've been trying to say is that the resistance arises because complying with the request would cause other errors to arise.

I appreciate that, but I would also point out that if the controller has disturbed a controlled perception or the controllee, by definition error has been induced. The question is whether the global error is increased or reduced by the controlee's action (wanted by the controller) to reduce the error in the perception the controller disturbed. Of course, if the controller acted to give the controlee an extra environmental feedback path, there's no induced error and the controlee should not exhibit any induced conflict.

I go back to the fact that we are not lone tigers prowling the jungle and fighting any other tiger we may encounter. We have evolved so that we usually find overall error to be reduced by helping each other.

In the other thread [Bill Powers (20078.08.23.1203 MDT)], you say: "The adversary system is built into American life in education, sports, business, law, religion, and politics." That is true on a large scale (and I think it's true of all cultures that have an ethos built upon the old Mesopotamian -- meaning mainly Judaism, Zoroastrianism, Christianity, Islam), but think how few of our everyday interactions actually involve adversarial relations, compared to the many that don't. An adversarial interaction is one in which each party's intent is do create persistent error in the adversary -- to deny the adversary the power to control some perception. Such interactions are not the most common, by many orders of magnitude. They are just the ones you notice.

Power is the ability to control one's own perceptions. One can increase one's own power sometimes by giving power to others (and in doing so, controlling the others).

Martin

···

At 02:33 PM 8/22/2008 -0400, Martin Taylor wrote:

4

[From Bill Powers (2008.08.22.17234 MDT)]

[Finally got back that day I’ve been skipping. Tomorrow, the 23rd, is
Alice’s birthday.]

[Martin Taylor 2008.08.22.17.11]–

, but consider…we have evolved
to be social animals, meaning that members of a group help each other
more than they act against one another. As individuals, we have
reorganized to act with respect to others so that they are more likely to
act helpfully than to act against us. Evolution and reorganization change
the odds that how you act when you want someone to do something will
generate the kind of conflict that will prevent them from doing what you
want. It’s survival of the fittest means of control.

OK, I’ll grant the plausibility of that premise, but I think the process
of optimization has barely got started. Perhaps at the family level, or
maybe the level of small tribes, this kind of relationship works. Some of
the time. But for larger groups it’s a long way from asymptote.

Whenever we want someone else to
act in a certain way, we must disturb some perception they are
controlling (or, perhaps, provide them a means of controlling that had
not been available to them).

I think we very seldom want to control someone else’s behavior; what we
want is certain consequences that affect us to occur. I say “Would
you hand me the salt, John?” and John says “Sure.” He
hands the salt to Sally, sitting between him and me, and Sally hands it
to me. Did John carry out the behavior I asked him to perform? No. He
inferred the consequence I wanted, as per your layered protocols, and
performed a behavior other than handing me the salt that would give me
what I want with the least total error for him. I controlled the
consequence (for me) of his behavior, but not his behavior.

We create (or take
advantage of existing) error in that person. They act to reduce the
error, and we hope that the action is the one we want. Cultural
conventions (reorganizations in individuals akin to those involved in
learning the communal language) make it probable that this will often be
the case, just as when we make noises with our mouth, the other is more
likely than not to understand what we want them to
understand.

Some of the problem here may be the difference between qualitative and
quantitative conceptions of error. We don’t have to “create”
error in anyone else. All real control systems are experiencing error all
of the time. Normally we can prevent error from becoming troublesome by
altering our actions as disturbances come and go; the amount of error
never gets very large for very long. The basic principle of
reorganization is that organization continues to change until a control
system is formed with (or altered to have) sufficient capability to keep
the error below the threshold where reorganization starts.
Controlling someone else’s behavior by applying disturbances relies on
keeping the disturbances small enough that they will not cause
significant reorganization. If the disturbance causes significant error
in the other person, reorganization will start, and will continue until
that error is kept small again. After that, the same disturbance may or
may not give rise to the same behavior as before. Probably not, since
it’s only the effect of the behavior that matters to the
controller.
I think successful social “intercontrol” can happen only when
the disturbances used to get other people to do as one wishes or needs
are small enough to avoid triggering reorganization in the other person.
When someone asks me to pass the salt, I can handle the disturbance
without significantly disrupting my control of anything I’m doing. I do
that without reorganizing, so I am quite reliable as a passer of salt,
provided I don’t have to get up, go to the kitchen, and refill the
salt-cellar first, every damned time.
So yes, I agree that through living and reorganizing in a society, we all
learn how to control our own experiences in such a way that small
disturbances generated by other people don’t materially affect the
quality of control, not enough to cause reorganization. Within those
limits, people can get us to behave in some ways just by applying subtle
and carefully chosen disturbances.
We say “John, would you mind passing the salt?” and not
“Hey, stupid, can’t you see that I need the salt?”
What I’m referring to when I talk about, for example, clients resisting
attempts by therapists to control their behavior, is the consequence of a
disturbance large enough to cause reorganization. If a control system
doesn’t already exist that is capable of opposing that degree of
disturbance, control systems will appear that are capable of doing
that. In any case, the person simply won’t allow that much of an error to
exist and will act or learn to act to prevent its having much effect.
Since reorganization is random, that may or may not end up giving the
producer of the disturbance the consequence that is wanted.

I think the bottom line, on which we can probably agree, is that control
of other people’s behavior is possible and perfectly routine as long as
the disturbances do not result in large enough errors to cause
reorganization. The rubber-band game illustrates this nicely. As long as
the experimenter does not use disturbances of the knot too large or fast
for the controller to resist, and still control accurately, the
experimenter can put the controller’s hand anywhere – anywhere, that is,
that does not hurt the hand.

Best,

Bill P.

5

[Martin Taylor 2008.08.23.13.59]

[From Bill Powers (2008.08.22.17234 MDT)]

[Finally got back that day I've been skipping. Tomorrow, the 23rd, is Alice's birthday.]

[Martin Taylor 2008.08.22.17.11]--

, but consider...we have evolved to be social animals, meaning that members of a group help each other more than they act against one another. As individuals, we have reorganized to act with respect to others so that they are more likely to act helpfully than to act against us. Evolution and reorganization change the odds that how you act when you want someone to do something will generate the kind of conflict that will prevent them from doing what you want. It's survival of the fittest means of control.

OK, I'll grant the plausibility of that premise, but I think the process of optimization has barely got started. Perhaps at the family level, or maybe the level of small tribes, this kind of relationship works. Some of the time. But for larger groups it's a long way from asymptote.

This is a very interesting response, since it implies (contrary to your usual assertion) that groups are control systems that may reorganize to increase their ability to control their (group) perceptions.

I don't want to follow that line of thought in this thread, and I don't think it is necessary. At least it is not necessary from my viewpoint, since I would argue that reorganization of individuals who grow up within communities of reasonably well structured culture would lead to the situation you describe. It's the "domains of spin glass" situation I described in my "Mutuality" paper <http://www.mmtaylor.net/PCT/Mutuality/index.html&gt;\.

An individual reorganizes so that his actions that disturb those with whom he most often interacts are more likely to bring his perceptions near their reference levels than to drive them further from their reference levels -- in other words, the application of social power within the closer group (family, church congregation, work colleagues) is likely to be as you describe: " Controlling someone else's behavior by applying disturbances relies on keeping the disturbances small enough that they will not cause significant reorganization.". If the applied disturbances did often induce reorganization in the other, the cultural milieu would change, and the person's ability to control the other would be reduced. Persistent failure to get what one wants by interactions with those closest is likely to lead to reorganization in oneself, not in the others.

    I think we very seldom want to control someone else's behavior; what
    we want is certain consequences that affect us to occur. I say
    "Would you hand me the salt, John?" and John says "Sure." He hands
    the salt to Sally, sitting between him and me, and Sally hands it to
    me. Did John carry out the behavior I asked him to perform? No.

I would say "Yes, he did", based on the PCT principle that what I wanted was not for John to make specific muscular movements, but to act so that I received the salt. What I disturbed was John's perception that I was satisfied with a state of the world that was in his power to change. I presumed him to be controlling a perception of my satisfaction with such states of the world. For some of them, to correct them might involve substantial conflict in him, but not in this case. He corrected the error in his perception of my satisfaction by getting the salt to me. The particular means he used are irrelevant. "You can't see what someone is doing by looking at what he is doing."

The cultural norms that allow a person to disturb another's perceptions (or to provide means for the other to improve control) evolve through repetitive interactions among people as one matures in one's own context of contacts. Those contacts can be (fuzzily) categorized as belonging to groups such as "family", "colleagues", "neighbours", "co-nationals" ... Many of these labels reflect the likelihood that a person belonging to the group will act moderately predictably when a particular kind of disturbance is applied, and the closer the group the more probable it is that the reactions to disturbance will be predictable (though not the muscle movements, as with John, Sally, and the salt).

None of which says there won't be conflicts within the group, when the resources are insufficient to allow all in the group to bring their perceptions to their reference levels simultaneously. But it does say that conflicts are less likely to be inadvertently induced by one's actions when dealing with a member of one's own group than when dealing with a member of a different group. Consider the damaging effect Adlai Stevenson caused by sitting with the worn sole of his shoe visible to his Arab interlocutors. He didn't have any intention of insulting them, but they perceived insult.

When many people of a group (say "nation", say "reformist", say "conservative", say "terrorist cell") agree that certain states of the world are good and work toward achieving those states, while members of different groups take incompatible states as good, inter-group conflict is assured. But that's not the only kind of conflict that can occur between groups of different culture (family tradition, religious belief, regional conversational practices). Consider the conflict over the cartoons of Mohammed, innocuous to the eyes of people accustomed to vicious political cartooning of authority figures, intolerable to some Muslims. Different cultures lead to different controlled perceptions, and to different reference values of similar controlled perceptions, simply because controlling those perceptions at those values are what allow members of the same culture to be able to control each other with minimal conflict.

So, using no concepts of groups as control systems, it seems more likely for large groups to be in conflict in a coordinated way than for individuals within small groups, or for coordinated small groups to be in conflict with each other.

We can, of course, resume consideration of the possibility that some groups are actually control systems. But I don't think it is necessary in this particular thread.

Martin

6

[Martin Taylor 2008.08.23.13.59]

WTP: OK, I'll grant the plausibility of that premise, but I think the process of optimization has barely got started. Perhaps at the family level, or maybe the level of small tribes, this kind of relationship works. Some of the time. But for larger groups it's a long way from asymptote.

MMT:

This is a very interesting response, since it implies (contrary to your usual assertion) that groups are control systems that may reorganize to increase their ability to control their (group) perceptions.

That's a pretty long reach. I meant nothing of the sort. I was speaking of the same kind of reorganizing processes you speak of in talking about language acquisition. You inferred, but I did not imply. Most of your post simply echoes what I meant. But not the "spin glass" part. You know that's not my cup of tea.

WTP: I think we very seldom want to control someone else's behavior; what

   we want is certain consequences that affect us to occur. I say
   "Would you hand me the salt, John?" and John says "Sure." He hands
   the salt to Sally, sitting between him and me, and Sally hands it to
   me. Did John carry out the behavior I asked him to perform? No.

I would say "Yes, he did", based on the PCT principle that what I wanted was not for John to make specific muscular movements, but to act so that I received the salt.

Isn't that exactly what I said? " ... what we want is certain consequences that affect us to occur." The consequence I wanted was receiving the salt. I didn't care what behavior John used to generate that consequence. I asked him to "hand me the salt," but he handed it to Sally, not to me. That didn't matter, because it wasn't his behavior I was trying to affect.

Best,

Bill P.

···

At 02:45 PM 8/23/2008 -0400, you wrote: