Social organization

[From Bill Powers (930524.0800 MDT)]

Bob Clark (930523.1610 EDT) --

... isn't it likely that individuals will construct such groups
in ways that resemble those with which they are already
familiar?

This question recurs frequently on the net. It is probably true
that people often attempt to structure social organizations in a
way that resembles an individual's hierarchy of control. A person
is used to willing that an act shall occur (like clenching a
fist) and observing its immediate occurrance. What would be more
natural than to set up a social situation so that when one wills
that another person do an act, one immediately perceives that act
being carried out? The idea of managers and executives is
patterned this way: the person higher in the organization issues
a command, which the person lower down must take as a reference
signal and bring about in reality. Thus the manager or executive
uses verbal orders, memos, lists of rules on bulletin boards,
etc. as output, to achieve the desired perceptions.

There are, however, some fundamental differences between the way
levels of control are implemented within an individual and the
way they have to be implemented in a social hierarchy.

The communication between levels inside an individual takes place
through upgoing perceptual signals and downgoing reference
signals. A higher system's output is identically a lower system's
reference signal (or one contribution to it). The perceptions in
the higher system are derived from direct copies of perceptions
in the lower systems, in the form of neural signals without any
translation being necessary. A control system of one level
continuously adjusts reference signals for the immediately lower
level and continuously recieves information about what the lower
systems are in fact perceiving.

In a social hierarchy, all communication between levels has to
take place through the lowest level of perception in each person.
The executive must translate the desired perception into symbols
and emit them as marks on paper or sound waves in air. The
subordinate then reads or hears those physical effects, perceives
them at many levels, and translates them, probably via memory
associations, into equivalent reference perceptions (memory
aside, this is the basic organization of Martin Taylor's "Layered
Protocol" scheme). The reference perception that results may
exist in the subordinate at the same level as the original one in
the executive, or at a higher or a lower level. The executive may
have ordered her secretary to make coffee for a conference; the
secretary may reluctantly have acquiesced, thinking of it as
being forced once again to perform a menial task beneath his
capabilities in order to keep his job. A simple order to perform
a sequence of operations is perceived as an order to play a
social role.

The translation problems in a social hierarchy are immense. The
order that the executive perceives as it is given is not
necessarily the order that the subordinate, after translating the
words into perceptions, experiences. The means by which the
subordinate starts to implement the understood reference-
perception may itself violate the executive's intention. The
subordinate, having all the internal levels of organization that
the executive has, may disagree with the order and modify it to
make it practical, in the subordinate's opinion. The subordinate
may act to achieve one goal, but report to the executive that
something else was done. Language is so ambiguous that the
subordinate may honestly do what was interpreted as the
executive's intent, but actually do something quite different,
and provide an honest verbal report as to what was done that is
interpreted by the executive as indicating compliance.

Because executives do not have direct access to the subordinate's
perceptual signals, the executive must operate a great deal of
time in the imagination mode. When an order is issued, the
executive must imagine how it will be understood and implemented;
there is no continuous direct access to what the subordinate is
perceiving, a fact which allows great divergences between the
actual control process being carried out and the process as the
executive imagines it. So executives are forced to make their
plans as if their own understanding of what their directives
meant had in fact been achieved exactly as intended. This means
that elaborate plans are often drawn up and assumed to be
implemented, with little relationship to what the subordinates
are actually doing -- or what disturbances have arisen since the
order was issued, required unexpected actions to counteract them
with unforeseen repercussions at the planning level. Executives
tend to perceive what they intended to happen, and be out of
touch with what is actually happening.

Another great difference between the internal hierarchy and the
social one is that a given level of control in the internal
hierarchy has exclusive domain over a given class of control
process, while in a social organization an executive is issuing
orders at some highest level, with subordinates providing their
own reference signals for their own lower systems from similarly
high levels. The result is as though the subordinate, at some
level, were receiving and obeying orders not only from the
executives of his own organization, but from executives in other
organizations, too.

Each person in a social hierarchy has a complete complement of
levels. Some of the goals at these levels may have to do with the
goals of the organization, but there are many other goals active
at the same time. The subordinate may have religious convictions,
family and love interests, hobbies and intellectual pastimes,
investments, vices, personal ambitions, moral principles of
conduct, health concerns, and political opinions. In all of these
areas, the person's desires and intentions put constraints on how
the processes of control within the person can be used for
another's purposes without conflict. The executive drops into the
middle of this complex system of goals an arbitrary goal designed
to achieve not the subordinate's own goals, but the objectives of
the organization as the executive understands them. It is highly
likely that the result will be conflict.

This problem does not exist inside a properly functioning
individual, for one level of control system has exclusive say as
to how that level is going to achieve its goals. One level never
questions a reference signal given to it from a higher level; it
can't even perceive the world in the same terms as the higher
system. It has neither higher nor lower level purposes. The level
in a human being that is concerned with syntax knows nothing
about words. One level specializes totally in one kind of
perception, accepts reference signals from higher systems as
givens, and issues reference signals to lower systems with no
need to consider that the lower systems may provide their own
goals as well. This allows the whole internal human hierarchy to
function as a single coordinated unit with no conflict between
levels.

So while it may be quite natural for human beings to try to deal
with each other as if a social hierarchy were similar to an
individual's internal hierarchy, this is basically not a
practical mode of social operation, at least not as envisioned in
traditional hierarchical terms. Over the last few decades,
philosophers of management have begun to realize this, and the
traditional "command" structure is being recognized as the cause
of more problems than it solves. Hierarchical organizations do
not actually work nearly as well as their managers like to
pretend. The concepts of HPCT, whether or not they are completely
correct about an individual's internal organization, give us
strong hints as to why.

Restructuring social organizations in the light of HPCT is not
going to be easy. Human beings like to control. They like to
control everything they can, including other people. The delusion
that this is possible has led people to believe in things like
the divine rights of kings, dictators, and priests, the natural
superiority of the rich and the unprincipled, the acceptance of
class divisions and one's position in society -- the right, in
short, of some people to tell others what to want. it is not only
those who wield the power who promote this sort of concept; even
those over whom the power is wielded have been persuaded to
accept this hierarchical system as the natural order of things.

Perhaps it is the natural order of things, among people who do
not yet understand that each person is an autonomous control
hierarchy just like every other person. In that case, one primary
use of HPCT can be to change the accepted natural order of
things.

···

------------------------------------------------------------
Best,

Bill P.

[From Dag Forssell (930524 11.15)] Bill Powers (930524.0800 MDT)

A wonderful post. Thanks, Bill.

These are important issues, that speak to the application of PCT
and HPCT to our daily struggles in the real world.

For others interested, I want to call attention to the largest
paper (37 pages) in LCSII: CT psychology and Social Organizations.
Here, Bill spells out the theme of today's post in much more
detail.

Best, Dag

From Tom Bourbon (930524.1208)

[From Bill Powers (930524.0800 MDT)]

Bob Clark (930523.1610 EDT) --

... isn't it likely that individuals will construct such groups
in ways that resemble those with which they are already
familiar?

This question recurs frequently on the net. It is probably true
that people often attempt to structure social organizations in a
way that resembles an individual's hierarchy of control. A person
is used to willing that an act shall occur (like clenching a
fist) and observing its immediate occurrance. What would be more
natural than to set up a social situation so that when one wills
that another person do an act, one immediately perceives that act
being carried out? The idea of managers and executives is
patterned this way: the person higher in the organization issues
a command, which the person lower down must take as a reference
signal and bring about in reality. Thus the manager or executive
uses verbal orders, memos, lists of rules on bulletin boards,
etc. as output, to achieve the desired perceptions.

There are, however, some fundamental differences between the way
levels of control are implemented within an individual and the
way they have to be implemented in a social hierarchy.

The communication between levels inside an individual takes place
through upgoing perceptual signals and downgoing reference
signals. A higher system's output is identically a lower system's
reference signal (or one contribution to it). The perceptions in
the higher system are derived from direct copies of perceptions
in the lower systems, in the form of neural signals without any
translation being necessary. A control system of one level
continuously adjusts reference signals for the immediately lower
level and continuously recieves information about what the lower
systems are in fact perceiving.

In a social hierarchy, all communication between levels has to
take place through the lowest level of perception in each person.
The executive must translate the desired perception into symbols
and emit them as marks on paper or sound waves in air. The
subordinate then reads or hears those physical effects, perceives
them at many levels, and translates them, probably via memory
associations, into equivalent reference perceptions (memory
aside, this is the basic organization of Martin Taylor's "Layered
Protocol" scheme). The reference perception that results may
exist in the subordinate at the same level as the original one in
the executive, or at a higher or a lower level. The executive may
have ordered her secretary to make coffee for a conference; the
secretary may reluctantly have acquiesced, thinking of it as
being forced once again to perform a menial task beneath his
capabilities in order to keep his job. A simple order to perform
a sequence of operations is perceived as an order to play a
social role.

...

Bill is speaking (far more generally and eloquently) of the phenomenon I
recently discussed while describing the interactions and communications that
can occur when people perform the cooperative tracking task. If one person
sits down and performs the task with both hands, it is easy. Any "social"
and "communicative" processes occur in the person, whom we model with HPCT.

As soon as two people try to learn the task -- to learn the perceptions to
control -- the situation changes. Even if, as I described in one post, an
experienced instructor tries to "talk them through" the task, the situation
changes again. The instructor operates as a fully-constituted and
fully-functioning HPCT system, as do both of the participants. I could try
to isolate the various functions in a PCT system into each person, but the
results would be disastrous, so far as the new, expanded, team achieving the
same results as one person alone, or two people in the usual collaboration.

Trying to isolate functions would mean that one person would watch one of
the subordinate relationships (say, middle cursor-left cursor),
another person, the other relationship (middle cursor-right cursor).
Each of these "input-function-people" will be a complete person, modelable
as an HPCT system, but each would be expected to "transmit" a unidimensional
"signal" to his or her associated "comparator-person," also modelable as a
full HPCT system, but treated as though merely a simple device. And so it
goes, around the modeled control loop. If I step my way around ONE SIMPLE
PCT LOOP, I quickly realize the impression that social organizations
literally, or in any meaningful approximation, duplicate the organization
of a PCT system, is illusory.

If the analogy fails for a single loop, at one level of control, I do not
think it likely to hold for a complete HPCT system. Bill nicely describes
some of the problems that would arise in the hierarchical case.

Until later,
  Tom Bourbon

[from Ray Jackson (930525.1900 MST)]

Bill Powers (930524.0800 MDT) said:

Restructuring social organizations in the light of HPCT is not
going to be easy. Human beings like to control. They like to
control everything they can, including other people. The delusion
that this is possible has led people to believe in things like
the divine rights of kings, dictators, and priests, the natural
superiority of the rich and the unprincipled, the acceptance of
class divisions and one's position in society -- the right, in
short, of some people to tell others what to want. it is not only
those who wield the power who promote this sort of concept; even
those over whom the power is wielded have been persuaded to
accept this hierarchical system as the natural order of things.

Bill, thanks for a wonderfully stated post. As I read through it, I
was reminded of the many "enlightened" minds currently influencing
organizational science: Deming, Senge, Peters, Waterman, and so on.
A very popular figure on the "empowerment" front, Ralph Stayer,
states:

        Before, I didnUt have power because I just had people
        wandering around not giving a damn. Real power is getting
        people committed. Real power comes from giving it up to
        others who are in a better position to do things than you
        are. Control is an illusion. The only control you can
        possibly have comes when people are controlling themselves.

Nice thought. These types of comments are all well and good when you
want to put together a presentation full of anecdotes and timeless
pearls of wisdom. All this hoopla over change and empowerment
provides an awareness of the need for change, but a closer look
reveals that it is only a superficial understanding of the
phenomenon. Fortunately, HPCT provides us with the explanation of
WHY these anecdotes are so powerful, and how to duplicate them.
That's why I'm so thankful to you, Ed and Dag.

Perhaps it is the natural order of things, among people who do
not yet understand that each person is an autonomous control
hierarchy just like every other person. In that case, one primary
use of HPCT can be to change the accepted natural order of
things.

You're absolutely right, but I see the natural order of things
changing for a variety of factors: a slow realization that control
hierarchies cause more problems that they solve (as you mentioned),
the fact that some places are successfully developing hierachies
which are oriented more towards communication than control (what I'm
pursuing at Motorola), and a growing consciousness of the needs of
the individual (people can no longer be a commodity). Of course,
none of these things change quickly, and any change will likely be
unsuccessful without a clear understanding of exactly what this
empowerment phenomenon is -- so, the real value of HPCT is that it
DEFINES empowerment as nothing else can.

Best Regards,
Ray

···

********************************************************************
Ray L. Jackson, M.Ed. rljackson@attmail.com
Manufacturing Training
Motorola Computer Group Work: 602-438-3894
2900 S. Diablo Way Pager: 602-244-3252 #2545
Tempe, AZ 85282
********************************************************************