Different Worlds

[R_Marken]

[From Rick Marken (980422.1200)]

Me:

OK. Let me see if I understand your point. Are you saying that
"seeing things differently" and "having different references
for the same perceptual variable" are equivalent ways of describing
the PCT model of the conflict between Woody and Diane?

Bruce Gregory (980422.1040 EDT):

Yes.

Ok. Now, when I hear that two people "see things differently" I get
an image in my mind of two people who perceive the same objective
situation differently. In terms of the PCT model I imagine you are
saying that the perceptual function that produces the representation
of this objective situaiton is different for each person involved in
the conflict. Is this what you mean by "see things differently"?

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[Bill_Powers1]

[From Bill Powers (980422.1344 MDT)]

Bruce Nevin (980421.2242 EDT)--

I had in mind just sitting and attending to perceptions that are present.
It is not clear that I am controlling those perceptions; and as I just
attend to them, whatever they may be, there can be no conflict respecting
them. There are no preferences as to how these perceptions should be, there
are no reference levels for them.

You can't tell if a variable is under control, even by yourself, by just
observing it. You find out whether it's controlled by seeing what happens
when it's disturbed. Just watching a sunset isn't enough to rule it out as
a controlled variable. You would have to watch the sunset at a time when
something happened to make it two hours late. I think you would become
extremely upset if that were to happen, increasingly so as the sun hung
just above the horizon for five minutes, fifteen minutes, an hour, or two
hours. By the end of the two hours, your whole hierarchy of control systems
would be in an uproar and your whole understanding of your world would be
upside down.

If a variable stays in the state you prefer without any action being
needed, you never experience any error and in fact you don't even need to
learn what to do in case there is an error. I think we accept the world
that we experience, quite quickly, as the world we expect to experience,
meaning that we make our perception of it into a reference level for it.
There's no need to complete building the control system if there's never
any deviation of what we perceive from what we expect/want to perceive.

Further development of this thought is left to the student. Hint: if there
is a conflict between passively observed phenomena, is this conflict due to
the natural world being in conflict with itself, or to an "incorrect" (or
simply incorrect) way of perceiving the natural world?

Best,

Bill P.

[R_Marken]

[From Rick Marken (980422.1230)]

Me:

Ok. Now, when I hear that two people "see things differently" I get
an image in my mind of two people who perceive the same objective
situation differently. In terms of the PCT model I imagine you are
saying that the perceptual function that produces the representation
of this objective situation is different for each person involved in
the conflict. Is this what you mean by "see things differently"?

Bruce Gregory 9980422.1503 EDT):

Yes.

Ok. So does the PCT model say that there would be no conflict
between two people if they could perceive things in the same way?
For example, would there be no conflict between Diane and Woody if
they could perceive frequency of sex in the same way?

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[Bruce_Gregory]

[From Bruce Gregory (980422.1713 EDT)]

Bill Powers (980422.1344 MDT)

If a variable stays in the state you prefer without any action being
needed, you never experience any error and in fact you don't even need to
learn what to do in case there is an error. I think we accept the world
that we experience, quite quickly, as the world we expect to experience,
meaning that we make our perception of it into a reference level for it.

Yes. I've been thinking this for a while. It's nice to learn that it is not
a heresy!

Best Offer

[Bruce_Gregory]

[From Bruce Gregory (980422.1715 EDT)]

Bill Powers (980422.1329 MDT)

The reasoning in favor of conflict is like
that of the person who kept hitting himself in the head with a hammer
because it felt so good when he stopped.

Is this a veiled reference to me or to Rick?

Best Offer

[Fred_Nickols3]

[From Fred Nickols (980422.1755 EDT)]

Rick Marken (900421.1220)

Bruce Gregory (980421.1155 EDT)--

I understand and appreciate the point you make. I want only to
point out the utility of the different worlds metaphor. It avoids
fruitless arguments over who is right, Diane of Woody. In Diane's
world three times a week is too much. In Woody's world it is too
little. This is functionally equivalent to saying that each of
them has a different reference value for the perception "the right
amount of sex", but it _may_ avoid arguments about who has the
"correct" reference value.

Rick:

I think the "different worlds" metaphor is an attempt to make believe
that conflicts will go away ("avoid fruitless arguments") if we
can respect the fact that other people see things differently. But
conflicts would also go away, given my "same world" _model_, if
people could all just respect the fact that other people _want_
things differently.

Allow me to toss in a concrete example and see what y'all make of it.

Many years ago, while still working as an internal OD consultant in the
Navy's program, the Human Resources Management Center (HRMC) where I was
stationed hosted a conference of race relations officers from around the
Navy. The occasion for their conference was to prepare a statement that
would define the future directions of the Navy's race relations program.
Their intended audience was the Chief of Naval Personnel and, ultimately,
the Chief of Naval Operations (CNO).

The five-day conference was at its mid-point on Wednesday noon when I
was summoned to my commanding officer's office. "Nick," he said, the
guys in the race relations conference are getting nowhere and have asked
for some help. Will you go up and see what you can do?" Naturally, I
"volunteered."

I went up to the large seminar room where the conference was being held
and encountered a room full of Lieutenant Commanders and Commanders (and
a couple of Captains as I recall). A few minutes' of general conversation
with this group confirmed the skipper's charge: they were getting nowhere.
Every time a statement or section would be proposed for inclusion in the
end product (the report to BuPers), it would immediately come under attack
and the proposer would withdraw it. At the mid-point of their five-day
conference, they had not one item for inclusion in the report.

I was a chief at the time, with about 17 years of service. I thought I
knew what was going on. So this is what I did and said:

I told the group that I figured that since they were from all over the
Navy (which is to say, not from a single unit), that I would bet that
they had come to the conference with some pretty clear charges to make
certain that this point or that point was included in the report. In
other words, they all came to the party with pet issues to protect.

Nods, smiles, and a little laughter confirmed my hunch. I then asked
them to write down on their notepads the specific charges they had been
tasked to prosecute.

I then took them by surprise when, after all had stopped writing, I asked
them to state their charges and I would write them on the easel sheets.
One commander said, "Just a darn minute, chief, you didn't say anything
about making these public." I replied, "That's the problem, commander,
this s--t is all hidden and it needs to get out on the table." With a
little grumbling, they acceded to my request.

Once I had everyone's "hidden agenda" up on the easel sheets, I looked
them over and announced, "I don't see a darn thing up there that conflicts
with anything else. Do any of you?" No one did.

So, I said, "Looks to me like you can move on from here," and I left the
conference room. This whole thing took about an hour and a half, from
noon to about 1330.

Along about 1700 the skipper summoned me to his office again. "Nick," he
exclaimed, "what the h-ll did you do up there?" Nonplussed, I asked,
"What's the matter?" "Nothing," he replied. They're finished. A couple
of them are going to hang around and put the finishing touches on the
draft report, but the rest of them are leaving early. "What the h-ll did
you do?" he repeated. I told him and he started laughing. "That explains
it," he said. "They're basing their report on all those items you teased
out of them."

Now I've always maintained that those guys weren't really in conflict, they
just thought they were and they didn't really know because their agendas
were hidden. (That's a common occurrence, I know.)

The point of this anecdote isn't to pile praise on me (although I certainly
got plenty of that afterward). It is to provide a specific instance of
what appears to be conflict but wasn't and to provide an instance that can
also be dissected in terms of PCT and the current conversation about
"different worlds."

Comments, anyone?

Regards,

Fred Nickols
The Distance Consulting Company
nickols@worldnet.att.net
http://home.att.net/~nickols/distance.htm

        "The Internet offers the best graduate-level education
         to be found anywhere."

[bnhpct]

[From Bruce Nevin (980422.2256 EDT)]

Bill Powers (980422.1329 MDT)--

By conflict and its resolution we identify relevant aspects of that shared
world more clearly. And we may develop new capacities for control. True of
internal conflict too. Conflict is not a bug, it's a feature.

I should have put a facetious grin next to that. But Isaac was allergic to
those faces for a while.

I wouldn't call that the best way to identify relevant aspects of the
shared world.

No, cooperative exploration & test is better. Individual exploration & test
is good but not quite as good. Are other things better?

While you're in conflict, you've lost the use of the system
in you that is conflicting with the corresponding system in the other
person -- both persons have lost control of the variable that is at the
focus of the conflict, or some variable closely related to it. If the
conflict is internal, you're lost two control systems. If conflicts were
good for us, why would we try to resolve them?

The kernel of truth is that resolving them is good for us. And not just by
resolving the conflict and re-enabling some degree of control.

Of course during the resolution of conflicts we do expand specific control
abilities and learn how to handle the problems without conflict. Our
capacity to reorganize is, I would agree, a good thing.

It could not be exercised without conflict or some other disturbance
overwhelming control.

But to conclude
from that that conflict itself is a positive contribution to our existence
is a non sequitur. It does not follow that because our bodies can mend a
broken leg, deliberately breaking our legs would be good for us.

Bad analogy, Bill. The capacity to heal broken bones is already
established. Breaking and healing a leg develops no new capacities.

The state
of conflict is unequivocally bad for us because it destroys control, and
there is no net gain in deliberately inducing conflict just because we can
most likely learn to resolve it. The time and effort required to recover
from the conflict would most likely be available for more productive use if
the conflict had never existed. The reasoning in favor of conflict is like
that of the person who kept hitting himself in the head with a hammer
because it felt so good when he stopped.

Or the smoker, yes.

Bateson argued that the peculiar form of conflict that he identified as the
double bind was at the root of creativity, of learning--and also of
evolution, where an unresolved conflict leads to extinction. Alice's
bread-and-butter fly, made of toast and butter, must drink warm tea to
live. But of course this melts the butter and it falls apart and
consequently dies. It can't drink, and it can't not drink. If it is not to
die (and moreover always to die, as the Knight I believe it was mournfully
observed, and therefore become extinct) then it must do something different.

An easy resolution of conflict (or of overwhelming disturbance of any sort)
is to cease controlling one of the pair of perceptions. Sour grapes, I
don't care about that after all. For internal conflict, perhaps one of the
control loops takes the form Rick recently described for me, where
increasing error results in reduced gain.

But what if you can't not care? Then the conflict takes the form of a
double bind. You can't control it, and you can't not control it.

Reorganization tries something different, anything. And again. And again.
And again until error is reduced.

Going up a level takes you to the perceptual control for which the
uncontrollable perception was to be the means (but as it turns out cannot
be), and from there other means might be evident, or if not then
reorganization at that level might be more effective than at the lower
level, or going up another level since it turns out the conflict is
effective on the first level up too.

Yes, conflict does appear to be a lemon. And yes, we do seem to make
lemonade. Or warm tea with toast and butter. If Bateson is right, it is the
goad of evolution in the species and of learning in the individual. He may
have been wrong, this is all open to discussion and test, but anyway that
is what I had in mind.

  Bruce Nevin

[bnhpct]

[From Bruce Nevin (980422.2318)]

Rick Marken (980422.1930) --

This (and _all_) conflicts have nothing to do with the fact that
people live in different perceptual worlds. It has to do with
the fact that they live in the _same_ real world and, thus,
cannot keep the _same_ or similar perceptual representations
of that world in different states at the same time.

Conflict could involve very dissimilar perceptions. All that is needed is
that the environmental feedback for controlling one conflict with the
environmental feedback for controlling the other. If the points of conflict
in the environment were imperceptible to the participants they would not
recognize it as conflict, but it would still be conflict nonetheless
wouldn't it? Or if that is problematic, try the case where an observer can
see that the two people are in conflict but they cannot.

  Bruce Nevin

[R_Marken]

[From Rick Marken (980422.1930)]

Me:

Ok. So does the PCT model says that there would be no conflict
between two people if they could perceive things in the same way?

Bruce Gregory (980422.1543 EDT) --

Not necessarily.

But this implies that it IS a possibilty though, right? Doesn't
the "different worlds" view suggests that people can solve a
conflict between them if they can learn to "see" (perceive) things
the same way?

They might perceive that sex three times a week is reasonable,
but they may prefer different times.

Then the conflict is about the right time for sex, right? What does
the fact that they perceive the reasonability of sex the same have to
do with the fact that they are in conflict over the time for sex?
And wouldn't your "differet worlds" notion say that this conflict
results from the fact that these two people perceive the appropriate
time for sex differently?

(They might have different reference values for perceiving sex in
the morning and sex at night, for example.)

The PCT model would say that this conflict exists because they
have different references for the _same_ perceptual variable --
time to have sex.

As Bill recently noted, all that is necessary for conflict to
exist is for Diane's controlling for sex in the morning to
interfere with Woody's controlling for sex at night.

That "necessity" is taken care of by the fact that both people have
to be present at the same time and place to have sex; that's the
way physical reality apparently works. If Woody does it at night
while Diane does it in the morning then neither will be producing
the perception of having sex.

This (and _all_) conflicts have nothing to do with the fact that
people live in different perceptual worlds. It has to do with
the fact that they live in the _same_ real world and, thus,
cannot keep the _same_ or similar perceptual representations
of that world in different states at the same time.

Presumably this problem could be resolved with less difficulty
than if Woody's reference level for frequency of sex was once a
day and Diane's reference level was once a month.

You'll have to explain this one. I sure don't see how that falls
out of the PCT model. Why is it easier to solve the conflict over
preferred time of sex than the one over preferred frequency of sex?

Are you sure you have carefully read and studied my "Cost of
Conflict" demo at

http://home.earthlink.net/~rmarken/ControlDemo/Conflict.html

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

[bnhpct]

[From Bruce Nevin (980422.2230 EDT)]

Bill Powers (980422.1344 MDT)--

Me (980421.2242 EDT)--

I had in mind just sitting and attending to perceptions that are present.
It is not clear that I am controlling those perceptions; and as I just
attend to them, whatever they may be, there can be no conflict respecting
them. There are no preferences as to how these perceptions should be, there
are no reference levels for them.

You can't tell if a variable is under control, even by yourself, by just
observing it. You find out whether it's controlled by seeing what happens
when it's disturbed.

If I am just observing whatever sensations arise in the region of the
nostrils, those perceptions cannot be disturbed. My attention to them can
be disturbed by distracting me, by applying anaesthetic to my nostrils so I
can't feel anything, etc., but whatever is perceived there just is whatever
it is, with no preference whatsoever as to what it should be, and therefore
those perceptions cannot be disturbed. The only way I can control
perceptions of myself meditating in this fashion is by not controlling
those particular perceptions. Attending to them without preference and
without control is the means of controlling perceptions of myself
practicing vipassana meditation.

Just watching a sunset isn't enough to rule it out as
a controlled variable. You would have to watch the sunset at a time when
something happened to make it two hours late. I think you would become
extremely upset if that were to happen, increasingly so as the sun hung
just above the horizon for five minutes, fifteen minutes, an hour, or two
hours. By the end of the two hours, your whole hierarchy of control systems
would be in an uproar and your whole understanding of your world would be
upside down.

Yes. Irrelevant.

If a variable stays in the state you prefer without any action being
needed, you never experience any error and in fact you don't even need to
learn what to do in case there is an error. I think we accept the world
that we experience, quite quickly, as the world we expect to experience,
meaning that we make our perception of it into a reference level for it.
There's no need to complete building the control system if there's never
any deviation of what we perceive from what we expect/want to perceive.

Yes. Irrelevant. Interesting as regards the role of disturbance in
learning, and of conflict as a particular common type of disturbance to
control, but irrelevant to the present point.

Further development of this thought is left to the student. Hint: if there
is a conflict between passively observed phenomena, is this conflict due to
the natural world being in conflict with itself, or to an "incorrect" (or
simply incorrect) way of perceiving the natural world?

I don't know what you mean by "a conflict between passively observed
phenomena." If conflict is possible then whatever you mean by "passive
observation" is different from what I am describing.

BTW, speaking of conflict resolution, I just sent four chapters off to my
dissertation advisor, and have two more in rough draft. The end is in
sight, depending of course on the responses of my committee.

  Bruce Nevin

[Bill_Powers1]

[From Bill Powers (980422.2133 MDT)]

Bruce Nevin (980422.2256 EDT)--

The kernel of truth is that resolving them is good for us. And not just by
resolving the conflict and re-enabling some degree of control.

This depends on whether in resolving a conflict we become better at
resolving conflicts. That would imply reorganization at a higher level, so
that we develop, perhaps, rituals or algorithms for quickly and
systematically resolving familiar kinds of conflicts instead of relying on
trial and error. The process of reorganization itself, however, I have
always assumed never changes or improves during our lifetimes, so
exercising it will not gain us anything (any more than breaking more bones
will enable us to heal faster). In the case of the kind of reorganizing
system I have visualised, there is no way to improve its output action, a
random process.

But to conclude
from that that conflict itself is a positive contribution to our existence
is a non sequitur. It does not follow that because our bodies can mend a
broken leg, deliberately breaking our legs would be good for us.

Bad analogy, Bill. The capacity to heal broken bones is already
established. Breaking and healing a leg develops no new capacities.

That is the analogy I intended. If reorganization doesn't become more
skillful with repeated use, then deliberately inducing it will not make it
work any better or faster. What we can hope for is to develop systematic
higher-level control systems that keep certain specific conflicts from
occurring. What we get out of this is that that kind of conflict no longer
occurs. But then we are no better off than we would have been if that kind
of conflict had been foreseen and avoided in the first place. There is
nothing beneficial about actually being caught in an active conflict, even
if the result is to learn how to avoid such conflicts in the future.

Bateson argued that the peculiar form of conflict that he identified as the
double bind was at the root of creativity, of learning--and also of
evolution, where an unresolved conflict leads to extinction.

This is confusing cause and effect. It is not the double bind or conflict
in general that is the root of creativity: creativity results from the
built-in organization that is capable of resolving that and other
conflicts, by means of generating new organizations. If you present a
conflict to an organism incapable of reorganizing above the level of the
conflict, the conflict will simply not be resolved. The conflict itself
doesn't contain any mechanism for resolving conflicts. You can't make an
organism creative by presenting it with conflicts unless it already
contains the means for creative reorganization.

Alice's
bread-and-butter fly, made of toast and butter, must drink warm tea to
live. But of course this melts the butter and it falls apart and
consequently dies. It can't drink, and it can't not drink. If it is not to
die (and moreover always to die, as the Knight I believe it was mournfully
observed, and therefore become extinct) then it must do something different.

Yes, and unless it already contains a mechanism for reorganizing before the
conflict appears, it will not be able to do anything different and it will
die. The conflict did not give it the ability to reorganize; it only made
the exercise of that ability necessary.

An easy resolution of conflict (or of overwhelming disturbance of any sort)
is to cease controlling one of the pair of perceptions. Sour grapes, I
don't care about that after all. For internal conflict, perhaps one of the
control loops takes the form Rick recently described for me, where
increasing error results in reduced gain.

But a system without any reorganizing function can't do anything new. If it
experienced the conflict once, it will continue to experience it forever.
If the conflict can't be resolved by exercising some preset procedure or
method that has already been learned, it can't be resolved at all. The
system can behave only as it is already organized to behave. If "sour
grapes" is your learned method of resolving a conflict, you will never have
to endure a conflict, because as soon as one occurs you will do the
sour-grapes routine and do away with it. That is, assuming that the only
conflicts you experience are those that can be fixed that way. I don't
count that as a real conflict, because it is instantly resolved. A real
conflict is one that persists because there is no known way of resolving
it. When there is no known way of resolving it, the only remaining way is
random reorganization, which does not work any better through practice.

But what if you can't not care? Then the conflict takes the form of a
double bind. You can't control it, and you can't not control it.

Reorganization tries something different, anything. And again. And again.
And again until error is reduced.

Going up a level takes you to the perceptual control for which the
uncontrollable perception was to be the means (but as it turns out cannot
be), and from there other means might be evident, or if not then
reorganization at that level might be more effective than at the lower
level, or going up another level since it turns out the conflict is
effective on the first level up too.

You're assuming that control systems of which you are not aware will not
operate until you become aware of them. That isn't true.

If there already exists a solution at any level, I don't count the result
as a real conflict. Any apparent conflict will last only as long as it
takes for the higher-level system to sense an error and go into action. I
would have a conflict every time I wanted to go through a closed door, if
it were not for the sequencing system that sees to it that I open the door
before trying to go through it. There was, perhaps, a conflict at one time
early in my life, before I understood how to control the order in which I
set reference levels, but that has not been the case for a long time.

Yes, conflict does appear to be a lemon. And yes, we do seem to make
lemonade. Or warm tea with toast and butter. If Bateson is right, it is the
goad of evolution in the species and of learning in the individual. He may
have been wrong, this is all open to discussion and test, but anyway that
is what I had in mind.

I think you are assuming that having a conflict is the only way to learn
how to control at higher levels. I'd agree that it is one way, but I think
it's far from the best way. Its cost can be very high in comparison with
that of other methods; it is frequently fatal. And you are assuming that
just because reorganization is _necessary_ in order to resolve a conflict,
it will automatically occur when a conflict occurs. What's hidden in this
picture is the reorganizing system, which must exist prior to the conflict
if there is to be any resolution. It is that reorganizing system that
creatively produces ever-greater competence in the hierarchy of control,
and does so whether or not conflicts occur. It does not need the "goad of
conflict." All it needs is a desire unattained.

Reorganization without conflict, I maintain, leads to a better result than
reorganization with conflict, and does so faster and with less damage.

Best,

Bill P.

[Bill_Powers1]

[From Bill Powers (980422.2229 MDT)]

Bruce Nevin (980422.2230 EDT)--

You can't tell if a variable is under control, even by yourself, by just
observing it. You find out whether it's controlled by seeing what happens
when it's disturbed.

If I am just observing whatever sensations arise in the region of the
nostrils, those perceptions cannot be disturbed. My attention to them can
be disturbed by distracting me, by applying anaesthetic to my nostrils so I
can't feel anything, etc., but whatever is perceived there just is whatever
it is, with no preference whatsoever as to what it should be, and therefore
those perceptions cannot be disturbed. The only way I can control
perceptions of myself meditating in this fashion is by not controlling
those particular perceptions. Attending to them without preference and
without control is the means of controlling perceptions of myself
practicing vipassana meditation.

Fine. And if disturbing those perceptions (by, for instance, my pressing a
finger against the region of your nostrils) doesn't result in any attempt
to restore them to their former state, you are not controlling them. But
the only way you can prove that you're not controlling them is to apply, or
have someone else apply, disturbances to them and see whether or not you
act to oppose the effects of the disturbances. If you have achieve a state
of purposeless observation, you will probably not act.

Just watching a sunset isn't enough to rule it out as
a controlled variable. You would have to watch the sunset at a time when
something happened to make it two hours late. I think you would become
extremely upset if that were to happen, increasingly so as the sun hung
just above the horizon for five minutes, fifteen minutes, an hour, or two
hours. By the end of the two hours, your whole hierarchy of control systems
would be in an uproar and your whole understanding of your world would be
upside down.

Yes. Irrelevant.

No, not irrelevant. In this case we find out that there is a control system
present but without a workable output function. In other words, you have a
preference for how a sunset should proceed, and lack only the means of
correcting an error should any error occur. The upsetness is the start of
reorganizing to become able to restore the sunset to its proper state. Of
course you can't actually do this, but you can control the perception in
other ways: don't look, or imagine the sunset proceeding properly.

If a variable stays in the state you prefer without any action being
needed, you never experience any error and in fact you don't even need to
learn what to do in case there is an error. I think we accept the world
that we experience, quite quickly, as the world we expect to experience,
meaning that we make our perception of it into a reference level for it.
There's no need to complete building the control system if there's never
any deviation of what we perceive from what we expect/want to perceive.

Yes. Irrelevant. Interesting as regards the role of disturbance in
learning, and of conflict as a particular common type of disturbance to
control, but irrelevant to the present point.

Well, then, maybe you'd better remind me what the present point is. Yours
seems to be different from mine.

Further development of this thought is left to the student. Hint: if there
is a conflict between passively observed phenomena, is this conflict due to
the natural world being in conflict with itself, or to an "incorrect" (or
simply incorrect) way of perceiving the natural world?

I don't know what you mean by "a conflict between passively observed
phenomena." If conflict is possible then whatever you mean by "passive
observation" is different from what I am describing.

If you see the sun rising and at the same time see it setting, this appears
to be a conflict between passively observed phenomena (passive meaning that
we can observe it but can't affect it). We reject this conflict because we
(normally) agree that conflicts can't occur in the physical world. That
leaves only our perceptions to blame; we must somehow be misperceiving, and
must find out just what we are perceiving wrongly.

BTW, speaking of conflict resolution, I just sent four chapters off to my
dissertation advisor, and have two more in rough draft. The end is in
sight, depending of course on the responses of my committee.

Wow, what a long haul. I really hope your persistence is rewarded.

Best,

Bill P.

[R_Marken]

[From Rick Marken (980422.2200)]

Bruce Nevin (980422.2318)

Conflict could involve very dissimilar perceptions.

Yes. But I believe conflict becomes less likely the more
dissimilar the perceptions (that is, the more dissimilar
the perceptual functions). If the perceptual functions
are linear (such as p = ax+by) then we can measure dissimilarity
as the orthogonality (linear independence) of the functions. I
forgot the equation that measures the orthogonality of two
linear functions (Richard Kennaway can surely help here) but
I think that the following two perceptual functions are as
different (orthogonal) as two linear functions of two variables
can be:

p1 = 1*x+1*y

p2 = 1*x -1*y

These perceptions are functions of the same environmental variables
(x and y). I can build control systems that simultaneously control
these very different perceptions of x an y with no conflict at all.
This is possible because the systems can act to produce values of x
and y that are solutions to _both_ equations simultaneously. The
less orthogonal the equations, the less good any simulataneous
solution.

Actually, I don't know how to hook these control systems to p1 and
p2 so that there _is_ conflict. It's possible (and possibly
provable) that systems that control orthogonal perceptual
variables _can't_ get into conflict, regardless of the nature
of the feedback connection between the systems and their
controlled perceptions.

But I will see if I can set up the world so that two systems
that can control orthogonal perceptions are in conflict (so
that they interfere with each other's efforts to control their
own perceptions).

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

[Oded_Maler6]

[From Oded Maler (980423-II)]

(Rick Marken (980422.1930))

> (They might have different reference values for perceiving sex in
> the morning and sex at night, for example.)

The PCT model would say that this conflict exists because they
have different references for the _same_ perceptual variable --
time to have sex.

Why not that at some time T one has a perception of "the right time
to do sex" and the other not, and they both control for "do the right thing"?

--Oded

[Richard_Kennaway2]

[From Richard Kennaway (980423.1608 BST)]

Rick Marken (980422.2200):

Yes. But I believe conflict becomes less likely the more
dissimilar the perceptions (that is, the more dissimilar
the perceptual functions). If the perceptual functions
are linear (such as p = ax+by) then we can measure dissimilarity
as the orthogonality (linear independence) of the functions. I
forgot the equation that measures the orthogonality of two
linear functions (Richard Kennaway can surely help here) but
I think that the following two perceptual functions are as
different (orthogonal) as two linear functions of two variables
can be:

p1 = 1*x+1*y

p2 = 1*x -1*y

Orthogonality is a bit of a red herring. As long as p1 and p2 are linearly
independent, it is possible to construct control systems to simultaneously
control them without conflict. For example, suppose p1 = x, p2 = x+y.
Suppose that the control loops produce outputs which affect x and y, in the
general way: output = (ea,eb), where e is the error and a and b are
constants, meaning that ea is added to x and eb is added to y.

Then if the control loop for p1 always produces equal and opposite outputs
affecting x and y respectively (i.e. a1 = -b1), then its actions will have
no effect on p2, and if the control loop for p2 acts only on y (i.e. a2 =
0), then it will have no effect on p1. Neither control loop is disturbed
by the other's existence. All that is necessary for this to be possible is
that the matrix defining (p1,p2) in terms of (x,y) be non-singular. If it
is singular, then p1 and p2 are essentially the same perception.

Actually, I don't know how to hook these control systems to p1 and
p2 so that there _is_ conflict. It's possible (and possibly
provable) that systems that control orthogonal perceptual
variables _can't_ get into conflict, regardless of the nature
of the feedback connection between the systems and their
controlled perceptions.

Suppose that in your example, b1 = a2 = 0, i.e. the controller for p1 only
acts on x, and the controller for p2 only acts on y. Then the output of
controller 1 will disturb p2, and that of controller 2 will disturb p1. Is
this an example of conflict, in the PCT sense? The controllers can still
succeed in controlling their perceptions, even though each is disturbing
the other.

An unresolvable conflict would result if b1 = b2 = 0, i.e. each of the
controllers tries to control its perception by acting only on x. Either
one can control its perception if the other is not operating, but together
they cannot.

-- Richard Kennaway, jrk@sys.uea.ac.uk, http://www.sys.uea.ac.uk/~jrk/
   School of Information Systems, Univ. of East Anglia, Norwich, U.K.

[Bill_Powers1]

[From Bill Powers (980423.0800 MDT)]

Rick Marken (980422.2200)--

I think that the following two perceptual functions are as
different (orthogonal) as two linear functions of two variables
can be:

p1 = 1*x+1*y

p2 = 1*x -1*y

These perceptions are functions of the same environmental variables
(x and y). I can build control systems that simultaneously control
these very different perceptions of x an y with no conflict at all.

Actually, I don't know how to hook these control systems to p1 and
p2 so that there _is_ conflict.

You pose an interesting question. The two equations above define two lines
in the x-y plane, y = p1 - x and y = x - p2. The solution is at the
intersection of these two lines. As long as there is a solution -- as long
as the lines intersect _somewhere_ -- there is technically no conflict.
However, if the values of x and y become extreme at the point of
intersection (it's way off the chart in some direction), we could say there
is an _effective_ conflict, because the outputs of the two systems, being
physically limited, can't become large enough for the systems to drive
their inputs to the solution point.

Now consider the equations

p1= x + y

p2 = a*x + a*y

These are the equations of two _parallel_ lines which either do not
intersect anywhere or else coincide. There is one and only one value of p1
that will work for a given value of p2 (p1 = a*p2), which means that p1 and
p2 can no longer be varied independently. This is the only case that
produces _absolute_ conflict, meaning that it's impossible for p1 to vary
independently of p2.

Conflict occurs, therefore, when the perceptual functions in the two
control systems approach being parallel in the coordinates of the next
lower perceptual space. Absolute conflict occurs if the two perceptual
functions are exactly parallel; that is, the one function is equal to the
other with the linear coefficients all multiplied by any constant (the
simplest being 1).

If the lines are nearly parallel, the intersection will occur only at large
values of p1 and p2, so there is _effective_ conflict which gets closer to
being absolute as the lines approach parallelism.

Note what this says about the subject of "different worlds." It says that
in fact conflict can occur only if the perceptual functions are identical
but for a constant multiplier -- that is, if they are parallel. Effective
conflict approaches absolute conflict as parallelism is approached. This is
the proof of Rick's earlier claim that conflict can occur only when the
different systems are experiencing the _same_ perceptual world (if the real
mathematicians here will permit me to use the word "proof").

I wonder whether the converse also holds -- that the occurrance of
conflicts between people could be a factor in a proof that an objective
world exists independently of our perceptions of it. That, of course, is a
critical question when we argue about whether one theory is as good as
another, or more right than another in some demonstrable way.

Best,

Bill P.

[Bill_Powers1]

[From Bill Power (980423.0937 MDT)]

Oded Maler (980423-II)--

Why not that at some time T one has a perception of "the right time
to do sex" and the other not, and they both control for "do the right thing"?

This doesn't work because "do the right thing" is too general. One person
says "have sex three times per week" is the right thing, and the other says
"Put three lumps of sugar in your tea" is the right thing. Both are "doing
the right thing," but if the second person changes to using 1 lump of
sugar, there is no conflict with the first person. To have a conflict you
must refer it to the same physical variable: twice a week versus three
times per week, for example. Then it truly becomes impossible to satisfy
both reference conditions at once, the basic requirement for a conflict.

The only way for people to have a conflict about one or both of them "doing
the right thing" is for them to agree on what doing the right thing would
be, and then for one of the people to decide to NOT do the right thing.

Best,

Bill P.

[bnhpct]

[From Bruce Nevin (980423.1152 EDT)]

First, please do not let this distract you from more important work.

Bill Powers (980422.2229 MDT)--

Well, then, maybe you'd better remind me what the present point is. Yours
seems to be different from mine.

Yes, it is.

Also I didn't realize that you limit the term "conflict" to cases that are
not resolved for an extended period while reorganization works. I was
supposing that reorganization might not always take a long time. Different
means for controlling the perceptual input may be close at hand, as it
were, involving just a few changes of connections or weights in the output
function or the perceptual input function. You seem to be saying that if it
does not take a relatively long time then the means for reducing error must
already have been available, and therefore it was not a conflict. I suppose
that if there is no conflict (and no overwhelming environmental
disturbance) we should expect control to resume in some small multiple of
the usual lag time for the highest level of control that is involved.
Between that relatively small interval and the kinds of protracted conflict
that you seem to have in mind I think there must be a wide range of
conflict and learning.

There may also be some misunderstanding reflected in your notion of
"passively observed phenomena (passive meaning that we can observe it but
can't affect it)." The perceptions observed in meditation are perceptions
that we can effect; we choose not to.

Bill Powers (980422.2133 MDT)

Bruce Nevin (980422.2256 EDT)--

The kernel of truth is that resolving [conflicts] is good for us. And not

just by

resolving the conflict and re-enabling some degree of control.

This depends on whether in resolving a conflict we become better at
resolving conflicts.

I am not talking about improvement in the process of reorganizing or of
resolving conflicts. I am talking about when new capacities are
developed--learning.

That is why I say that breaking and healing a leg is a bad analogy. I am
not supposing that the benefit of experiencing and resolving conflict is an
improvement in one's capacity to resolve conflicts. (Though I suppose, with
you, that improvements in algorithmic approaches to conflict are possible.)

I am supposing that conflict and its resolution is very common, and that
reorganization to reduce error due to conflict is a routine part of
learning (not all of it). But when we think of conflict we think of
occasions when it is not resolved and someone is stuck in it. That seems to
be what you have in mind.

From the point of view of each control system involved, conflict is just a

disturbance that overwhelms control. People stuck in internal conflict
commonly attribute their woes to factors in their environment, and they
commonly look for competing control systems to blame for environmental
interference, because from the control system's point of view there is no
difference.

Bateson argued that the peculiar form of conflict that he identified as the
double bind was at the root of creativity, of learning--and also of
evolution, where an unresolved conflict leads to extinction.

This is confusing cause and effect. It is not the double bind or conflict
in general that is the root of creativity: creativity results from the
built-in organization that is capable of resolving that and other
conflicts, by means of generating new organizations.

You left out that little word "at". "Conflict is at the root of
creativity," not "conflict is the root of creativity." I do not intend to
say (nor did Bateson) that the double bind is an agent that causes
creativity. I did say (as did Bateson) that the double bind is a
circumstance in which the organism must be creative to survive and thrive.
Your resistance to what you perceive as my misunderstanding has distorted
your perception of what I have said.

You can't make an
organism creative by presenting it with conflicts unless it already
contains the means for creative reorganization.

Can you see that what I have said does not contradict this?

I believe the reason why you imagined an unintended meaning (imagined it so
strongly as to ignore a word printed on the screen in front of you) was as
means of controlling a perception of the meaning conveyed by statements
like "factors in the environment control the behavior of individuals" with
a reference value of zero. You also control a perception "people, certainly
participants in CSG-net, should understand PCT." I think we need to
understand better the role that imagination plays. If sufficient inputs are
present for us to perceive a category then we imagine the other inputs are
present as well.

I
would have a conflict every time I wanted to go through a closed door, if
it were not for the sequencing system that sees to it that I open the door
before trying to go through it. There was, perhaps, a conflict at one time
early in my life, before I understood how to control the order in which I
set reference levels, but that has not been the case for a long time.

I suppose this is a conflict between "go through" and "no painful bumps"?
If it is a conflict, then, yes, this is an instance of learning.

I think you are assuming that having a conflict is the only way to learn
how to control at higher levels.

No, not the only way. I suppose this misunderstanding comes from the
misreading "conflict *is* the root of creativity." But I'll grant that I
invited this misunderstanding by saying (or attributing to Bateson)
"conflict is the goad of evolution and of learning". This certainly sounds
exclusive. Still, if the little boy unable to go through the door because
it isn't open yet is an example of conflict, then this mode of learning is
pretty general and commonplace.

I'd agree that it is one way, but I think
it's far from the best way. Its cost can be very high in comparison with
that of other methods; it is frequently fatal. And you are assuming that
just because reorganization is _necessary_ in order to resolve a conflict,
it will automatically occur when a conflict occurs.

I assume that reorganization occurs when error persists. To the control
systems involved conflict is a disturbance that overwhelms control. This
results in error that persists.

What's hidden in this
picture is the reorganizing system, which must exist prior to the conflict
if there is to be any resolution. It is that reorganizing system that
creatively produces ever-greater competence in the hierarchy of control,
and does so whether or not conflicts occur. It does not need the "goad of
conflict." All it needs is a desire unattained.

All it needs is disturbances that overwhelm control. From the control
system's point of view any disturbance that overwhelms control is like any
other. If it is due to internal conflict, then more than one control loop
is affected; if due to contention with another control system around their
effects in their shared environment, it may be easier to identify and
locate the source of disturbance. The typical demand is that the other stop
interfering. "Who left the lid to this jar on so loose? I just shook it and
look at the mess!" " Who screwed the lid to this jar on so tight? I can't
get it off!" If the disturbance is not due to conflict with another, it is
typical nonetheless to seek to assign blame. "Who the hell locked this
door! Oh. It's just stuck. Rain must have swelled it."

Reorganization without conflict, I maintain, leads to a better result than
reorganization with conflict, and does so faster and with less damage.

If you limit the word "conflict" to cases where it is unresolved for an
extended period, sure. Part of the misunderstanding is that I was using the
word more broadly.

Bill Powers (980422.2229 MDT)--

Fine. And if disturbing those perceptions (by, for instance, my pressing a
finger against the region of your nostrils) doesn't result in any attempt
to restore them to their former state, you are not controlling them. But
the only way you can prove that you're not controlling them is to apply, or
have someone else apply, disturbances to them and see whether or not you
act to oppose the effects of the disturbances. If you have achieve a state
of purposeless observation, you will probably not act.

I probably would act to control the perception of myself meditating and
would wonder why you were interrupting me. This does indeed show that I am
a neophyte meditator (as I freely own). The perceptions that are the object
of meditation are not controlled in themselves. Perhaps this is the reason
this technique works as it does. Those perceptual signals go up the
hierarchy as inputs to perceptions which, to the extent that I am a poor
meditator, are controlled, and therefore the low-level perceptions of heat,
cold, pressure, etc. are controlled as among the means of controlling the
higher ones. As I practice over time there is less control of perceptions
like who I think I am (look at me, I'm medititating, see me meditate).

There could be ways to tease these levels of control apart even for a
typically self-absorbed person like me. If I understood that you would be
doing this at unpredictable times, the pressure would be just another
sensation to be observed, yes. And before that, the change in temperature,
disturbance of air currents, perhaps electrostatic effects. At this scale
of perception, pressing with a finger is pretty heavy-handed (you will
pardon the expression).

Watching a sunset is not an instance of vipassana meditation practice.
However, I am told that an advanced meditator would take in your delayed
sunset with perfect equanimity.

If a variable stays in the state you prefer without any action being
needed, you never experience any error and in fact you don't even need to
learn what to do in case there is an error. I think we accept the world
that we experience, quite quickly, as the world we expect to experience,
meaning that we make our perception of it into a reference level for it.
There's no need to complete building the control system if there's never
any deviation of what we perceive from what we expect/want to perceive.

Sounds like you're saying that recognition is the match of perceptual input
with reference with no error and can be accomplished by the input half of
control systems with no output function. I suppose there is a comparator
and error output, such that if there ever were an error signal
reorganization would develop some kind of output function. You probably
said that in B:CP and I didn't take it in at the time.

The signal from a "recognition system" (no output function) could be passed
up to higher levels but it could never be controlled. This suggests that
either higher level systems do not influence the reference value, or they
rely upon it's being always the same (static, predictable change, etc.)

This has been a puzzlement for me for some time. Reference signals come
from memory. Yet they come from error output of higher levels. How is this
resolved? Just a pointer to the relevant part of B:CP would help.

  Be well,

  Bruce Nevin

[Bruce_Gregory]

[From Bruce Gregory (980423.1322 EDT)]

Rick Marken (980422.1930)

Me:

> Ok. So does the PCT model says that there would be no conflict
> between two people if they could perceive things in the same way?

Bruce Gregory (980422.1543 EDT) --

> Not necessarily.

But this implies that it IS a possibility though, right? Doesn't
the "different worlds" view suggests that people can solve a
conflict between them if they can learn to "see" (perceive) things
the same way?

No.

The different worlds interpretation predicts exactly the same things that
PCT predicts. No more. No less.

> They might perceive that sex three times a week is reasonable,
> but they may prefer different times.

Then the conflict is about the right time for sex, right? What does
the fact that they perceive the reasonability of sex the same have to
do with the fact that they are in conflict over the time for sex?
And wouldn't your "different worlds" notion say that this conflict
results from the fact that these two people perceive the appropriate
time for sex differently?

> (They might have different reference values for perceiving sex in
> the morning and sex at night, for example.)

The PCT model would say that this conflict exists because they
have different references for the _same_ perceptual variable --
time to have sex.

Not necessarily. To say that "time to have sex" is a perceptual variable, we
would have to demonstrate that it is perceivable and that both woody and
Diane are perceiving it and protecting different reference values. A
non-trivial hurdle before it is prudent to say "The PCT model would say..."

> As Bill recently noted, all that is necessary for conflict to
> exist is for Diane's controlling for sex in the morning to
> interfere with Woody's controlling for sex at night.

That "necessity" is taken care of by the fact that both people have
to be present at the same time and place to have sex; that's the
way physical reality apparently works. If Woody does it at night
while Diane does it in the morning then neither will be producing
the perception of having sex.

At least not with each other.

This (and _all_) conflicts have nothing to do with the fact that
people live in different perceptual worlds. It has to do with
the fact that they live in the _same_ real world and, thus,
cannot keep the _same_ or similar perceptual representations
of that world in different states at the same time.

> Presumably this problem could be resolved with less difficulty
> than if Woody's reference level for frequency of sex was once a
> day and Diane's reference level was once a month.

You'll have to explain this one. I sure don't see how that falls
out of the PCT model. Why is it easier to solve the conflict over
preferred time of sex than the one over preferred frequency of sex?

Maybe you should put your wife on at this point.

Are you sure you have carefully read and studied my "Cost of
Conflict" demo at

http://home.earthlink.net/~rmarken/ControlDemo/Conflict.html

I'll look at it again.

Best Offer

[Bruce_Gregory]

[From Bruce Gregory (980423.1355 EDT)]

Rick Marken (980422.2200)

Bruce Nevin (980422.2318)

> Conflict could involve very dissimilar perceptions.

Yes. But I believe conflict becomes less likely the more
dissimilar the perceptions (that is, the more dissimilar
the perceptual functions).

Presumably you and a flu virus have very different perceptual functions. As
you define conflict, you apparently cannot be in conflict with the virus.
How exactly would you define your interaction?

Best Offer