Social Control Theory

CSGnet Archive CSG2004
1

From[Bill Williams 27 May 2004 2:50 AM CST]

[From Bill Powers (2004.05.25.1352 MDT)]

Bill Williams 25 May 2004 11:40 AM CST --

>In anycase I starting with, as I understand it the implications of

control

>theory, I have a arrived at a notion of "social" that may stand up to
>criticism. Given that it has been derived in a context of control

theory, as

>I understand it, is a bit different than other conceptions of "social."

This a very clear exposition of the way control theory enters into social
phenomena, and after one pass I can't remember disagreeing with anything

in

it.

Much as I value Bruce Nevin's joining into this discussion, I would like to
preserve
this element of agreement that we apparently reached. ( Not that you can't
decide
after re-reading and further reflection that my exposition contains a falacy
or falacies
that weren't apparent on a first reading. ) However what, I would like to
preserve is a
common understanding, if that is what it is, regarding how agents ( which
is as good
a word as any for present purposes ) interact with each other. I will
bring Bruce back
later to help out my argument.

First: it is physiologically implausible that any organism would be
constructed so that
its functioning could be controlled by an external cause. Autonomy seems to
be a
bed rock principle of organic behavior.

Second: Organic autonomy is not an unconditional autonomy. Rather it is an
autonomy
that is exercised in the context of some specific enviornment. The
implication of this is
that while the enviornment does not determine, or even influence behavior,
the
autonomous behavior of an organism will be different in different
environments. This,
as I understand it, is entirely consistent with a proper understanding of
control theory.
The control theory model with out any contradiction asserts that organic
behavior is
completely autonomous, and also that the behavior of an organism can and
will be
different depending upon the particular features of the environment.

These two propositions may appear to be entirely elementary and
unexceptional. At least
I hope they are. However, they specify a relationship between organic
behavior and the
environment which is different than the one that is ordinarily assumed.

Consider an organism in two enviornments A and B. Suppose we switch the
organism
from enviorment A to environment B. And suppose as a result the organism
changes its
behavior. Are we jusified in saying the switch from A to B _caused_ the
organism to
change its behavior? This question appears to me to be crucial in
understanding how
adopting the principles of control theory will neccesarily generate a change
in the conception
of social phenomena.

Ordinarily I think most people would say that switching the enviorment from
A to B _caused_
a change in the organisms behavior. However, if my premise that an
organism is autonomous
then can it be said that the switch from A to B _caused_ a change in an
autonomous creature?

I think there is a lot of opportunity for equvocation here. And, I think
the reason for this is that the
context of such a discussion introduces some abiguities. The context
involves an organism and
an environment. This is a different context than one in which "All we can
know is what we perceive.

If the slogan, "All we can know is what we perceive." is taken literally,
then it would appear that
the very notion of an enviornment is forbidden. We may not, in some sense,
be in a position to
know the environment other than through our perception of it. But, our
control theory, it would
seem does tell us that there _is_ an environment. And, that is knowing
_something_ about the
enviornment. So, having adopted a control theory ontology I feel confident
that I can refute the
claim that "All we can know is what we percieve." All I have to do is open
my little book, and
there in the diagram is an organism and an environment.

Now at present I am not sure how I am going to prove that there are other
people. Not that I really
doubt that there really are other people. Maybe the answer is to write a
book _Social Control
Theory_ by Bruce and Bill. But, first I would have to convince myself that
there really is a Bruce.
But, having done that, we could get to work, and eventually we would have a
book. Then when
people got tired of looking in control theory books written by solipists,
they could read this book
by Bruce and Bill. And, they could look at the title page, and they would
see "By george they have
solved the riddle of solipsism-- there are two people right here-- Bruce and
Bill." And, the solipists
will knash their teeth and wonder, however did they do that?

Bill Williams

Clark McPhail went through a similar analysis, arriving at similar

conclusions but in the context of sociology, where the resistance to
"individualism" must be at least as stiff as in economics.

A useful term in this connection is "agent," unless that, too, has bad
connotations in economics. It's not quite as evocative of pins through
butterflies as "specimen", although Phil Runkel has taken away a lot of
that unwanted connotation. An agent is an entity capable of acting on the
world to produce effects, the implication being that the effects are

chosen

and intended. It might have an advantage in that "agent-based economics"
seems to be gaining ground, and may be approaching the principles of
control, if only tentatively. The term is just a nomination, not a
decision. No reason we can't use both.

The idea of a theory of culture is attractive. From where I stand, the
chief difficulty would be in deciding just what we would want such a

theory

to accomplish. In the theory of agents, control theory can be used to

model

and predict the actions of a single agent in a specified set of

conditions,

and it also serves as a source of guidelines for research into the

internal

organization of single specimens. Out of control theory we get some very
general rules of thumb, about things like loop gain, stability,
relationships between disturbances and actions, and hierarchical control.

So we can ask what sorts of general principles we would like to draw from

a

control-theoretic analysis of culture, and where we can get the data to
work from. There are, of course, tons of studies from a century or more of
social research, but unfortunately the quality of the data is such that
it's hard to draw conclusions from them with much predictive power. We may
be in a situation like that vis-a-vis psychology, where any value of past
research data for use in exploring control phenomena is strictly
accidental. The experimental designs simply didn't anticipate the needs of
research into control processes. This leaves us with little choice but to
do our own experiments, or get someone else to do them (which is a lot
harder, believe me).

The other approach is to experiment with models. The "crowd" program was
designed for that purpose, and it revealed phenomena that, unknown to me,
were already well-known in the study of crowds -- the "arcs and rings"
configuration being one of the first to be noticed in the setup called
"guru." This is a branch of modeling that tends to show up in fields where
experimenting with real systems is difficult, expensive, or illegal. You
can answer such questions as "What is the effect on an astronaut's ability
to control position in space if one leg has to be amputated?" -- without
actually doing this to a real astronaut.

Experimenting with models reveals emergent phenomena not obvious in the
initial design of the model. Consider a simple design consisting of two
control systems which are independent except that the actions of each one
symmetrically affect the controlled variable of the other (as well as

their

own). It's not obvious that this elementary relationship creates positive
feedback, but an experiment with a model will quickly show the instability
that appears when the strength of the mutual effects exceeds a certain
threshold. This sort of exploration may provide a different route to a
theory of culture, by suggesting non-obvious phenomena that can emerge

from

social interactions. Of course the underlying models of agents have to be
realistic, and the interactions tested have to be in the realm of the
possible, but this method is being used increasingly in other sciences,
such as astrophysics and materials sciences. We have used it in PCT to
explore the properties of systems in conflict (Kent McClelland) without
actually putting real people into conflict with each other (well, not on
purpose).

The two approaches, experiments with real systems and experiments with
models, would, in the best of all possible worlds, be highly interactive,
with real-system experiments being used to develop and test models, and
experiments with models being used to point out new phenomena which can be
investigated in the real systems. It seems to me that this is bound to be
how we investigate social phenomena in the future.

Finally, we can now do something that nobody could have done (on any
feasible budget) only ten or a dozen years ago: model large numbers of
agents interacting with each other in a complex environment. I'm talking
about hundreds, even thousands, of agents, in models that manipulate many
megabytes of data. So we can have our cake and eat it too: our macro
models can be built out of large numbers of microeconomic interactions, to
show the relationship between the detailed and general views of the

system.

Charlotte Bruun wrote eloquently about this aspect of modeling in her
thesis and the paper more specifically about her model (she hasn't
communicated for some time, but I'm leaving it up to her -- a busy
professional, a wife, and a mother of several very small children).

The consumption-budget model we've been tossing around may be a place to
start, in that it involves a single agent, but is potentially part of a
larger system involving other sectors of the economy and other agents. If
we can get this one agent operating with acceptable realism in a sort of
amorphous fuzzy economy, we can then start giving form to other parts of
the economy with which this agent ineracts, as in econ004 and beyond. I
think we will discover things by experimenting with this model and its
successors, phenomena that we can hope to identify in the real economy. At
first, I'd be happy to "predict" things that everybody already knows --
that's how we make sure the model is working right. But I'm confident that
there are phenomena which no human being, however knowledgable or
insightful, could predict simply because the interactions are too complex
for anyone to comprehend without the aid of a model.

I haven't even mentioned reorganization and learning. I'm irrationally
daunted by the prospect of setting up a big model and letting
reorganization loose to work on it. I guess I don't have much faith that

it

will really work. But what the hey, we'll try it one of these days.

Today I had a big break-through in converting Little Man Version 2 into
Delphi. It was originally in C, which of course is not compatible with
Delphi which uses Pascal, The existing code in Pascal did not contain all
the original features, and while I did (with a lot of help from my

friends)

salvage the original Pascal code for Version One (from the late 1980s), I
could not find any trace of an equivalent form of Version 2 in Pascal from
1989 and the early 90s -- the one with real physical dynamics in it. So I
decided to try to merge the two versions with the old C code as a guide,
and today I managed to get the little man pointing while looking at the
finger and target with two independently-aimed eyes in a movable head,

with

binocular vision for control in the depth dimension -- in Delphi. This

part

still works in the imagination mode (no physical dynamics yet), but I've
already tested the interface with the physical dynamics part of the

program

and joining the two parts should not take much longer. Then all that will
remain will be to use the Delphi tools for editing parameters, and

building

a "live block diagram" of the Version 2 model -- everyone liked that
feature of Version 1, which was left out of version two for lack of space
on the screen, so I want to include it in the Version 2 model.

Then I think we will be able to turn to the economics modeling much more
seriously. I still have a lot of converting from Pascal to Delphi to do

(it

scares me to think that none of my old DOS-bsed programs will compile,

much

···

less run, on any working machine 10 years from now) but the Crowd
(converted) and the Little Man were the most complex programs and involved
learning a lot of new things about Delphi. The rest can go faster,
expecially if there is some volunteer help.

Best,

Bill P.

2

[From Bill Powers (2004.05.27.0956 MDT)]

Bill Williams 27 May 2004 2:50 AM CST--

Ordinarily I think most people would say that switching the enviorment from
A to B _caused_ a change in the organisms behavior. However, if my
premise that an organism is autonomous then can it be said that the switch
from A to B _caused_ a change in an autonomous creature?

The first thing to be said is that you are constructing a model here: in
this model there are one organism and two environments with different
properties. All these models exist in your perceptions, so you are still
not describing the World As It Is. We both agree that there very probably
is such a world, but we also agree, I hope, that our models of that world
probably are somewhat different from the actuality, and may be drastically
different in respects that we do not perceive. I think this is the key to
avoiding the ambiguities you discuss next.

I think there is a lot of opportunity for equvocation here. And, I think
the reason for this is that the context of such a discussion introduces
some abiguities. The context involves an organism and an
environment. This is a different context than one in which "All we can
know is what we perceive.

I agree that putting a model of an organism into a model of an environment
is critically different from just making a model of the organism. This is
why every one of my models and simulations includes a model of the (local)
envirnment of the control system. Indeed, as in the case of the Little Man,
modeling the environment was 90% of the problem; the actual control systems
are pretty simple.

If the slogan, "All we can know is what we perceive." is taken literally,
then it would appear that the very notion of an environment is
forbidden. We may not, in some sense, be in a position to know the
environment other than through our perception of it. But, our control
theory, it would seem does tell us that there _is_ an environment. And,
that is knowing _something_ about the enviornment. So, having adopted a
control theory ontology I feel confident that I can refute the claim that
"All we can know is what we percieve." All I have to do is open my
little book, and there in the diagram is an organism and an environment.

A model of the organism and a model of the environment, drawn there on a
page of a book that you are perceiving. If you even momentarily overlook
the Observer, it is very easy to forget that you are perceiving, and to
mistake some aspect of your experience for Reality Itself. As soon as you
are reminded, you realize once again that perception is all we can ever
know about directly. All else is models and interpretations of models --
which are also perceptions. I have been around that block many times, and
almost as many times have forgotten the previous times. But if you keep
watching, pretty soon you realize that this, too, is a perception, and that
there is nothing in your experience, nothing whatsoever whether you call it
concrete or abstract, that is not a perception. Our evidence about a real
environment is just that, evidence, not absolute proof or final certainty.
I see no choice but to be satisfied with that.

Now at present I am not sure how I am going to prove that there are other
people. Not that I really doubt that there really are other
people. Maybe the answer is to write a book _Social Control Theory_ by
Bruce and Bill. But, first I would have to convince myself that there
really is a Bruce.
But, having done that, we could get to work, and eventually we would have a
book. Then when people got tired of looking in control theory books
written by solipists, they could read this book by Bruce and Bill. And,
they could look at the title page, and they would see "By george they have
solved the riddle of solipsism-- there are two people right here-- Bruce and
Bill." And, the solipists will knash their teeth and wonder, however
did they do that?

I think it was Samuel Johnson who was arguing against solipsism, maybe
against Bishop Berkeley, and kicked a rock, saying "Thus do I refute you!"
It is so easy to exempt part of your experience, like a rock or a book, or
a pain in your toe, from being called "a perception." If something is real,
it is "tangible", right? But what does tangible mean but touchable, and
what is touch if not a perception?

Our way out of all this, if we want to speak consistently in terms of PCT,
is to speak of models of reality, not of reality itself (other than the
reality of direct experience). If you put a hierarchical control system, an
organism, into an environment (both organism and environment being
recognized as perceptions or models), you will observe certain
interactions, including actions produced by the organism that we call its
behavior. If you put the same organism into an environment with different
properties, you will observe that that actions produced by the organism
change; they may change in amount, on in kind, or both. The "behavior" of
the control system has changed.

In neither case is the behavior, the action, what the organism is
controlling. What we call behavior is only what an outside observer can
see. Only the controlling system experiences the perceptions that are
actually being controlled.

So the first thing we must admit is that even through the behavior changes,
we do not know what what has changed with respect to controlled variables
or means of control.

The second thing we have to admit is that even if behavior is seen to
change, there is no justification for saying that anything about the
organization of the controlling system has changed. The feedback loop
always includes the environment, and in that environment there are both
variable disturbances and variations in the properties connecting actions
to perceived consequences. The environmental feedback function is variable.
As we can show with models, a control system with completely constant
parameters will automatically change its actions, in the face of such
environmental changes, to prevent any important changes in the perceived
quantity. So to an external observer, it may well seem that the change in
the environment has caused the behavior of the control system to change.
However, from the standpoint of the control system, the change in behavior
was exactly what was required to prevent the environmental change from
altering anything that mattered to the control system.

In the environment there are also, of course, other people. We can place
models of other people into our models of the environment, to correspond
with our perceptions of other people in our perceptions of an environment.
Working out the consequences, model-wise, gets harder to understand, but
doesn't introduce anything new into the situation that we didn't know about
before. Look at the Crowd: each agent avoids all obstacles including other
active agents. In principle, an interaction involving multiple active
agents simultaneously avoiding each other and randomly-placed passive
obstacles should become extremely complex, considering all the paths there
are for feedback loops both positive and negative to form. In practice the
complications work themselves out with no special effort; we don't even
have to consider them. The simulation just runs. Working out all the
ramifications would involve an enormous analytical effort and reams of
equations, but it turns out that to do that would be a waste of time. The
simulation shows us the solution. There is such a thing as knowing too much.

I hope this is furthering our social convergence. I hope others involved in
this discussion realize what is going on, get the hang of it, and join in
with the right spirit. Even if one specimen occasionally breaks training.

Best,

Bill P.

3

[From Rick Marken (2004.05.27.1130)]

Bill Powers (2004.05.27.0956 MDT)--

I hope this is furthering our social convergence.

Hope springs eternal.

I hope others involved in
this discussion realize what is going on, get the hang of it, and join in
with the right spirit.

I know exactly what is going on. I have seen you do this since we met.

I think it would be wisest for me to just leave this to you since I don't
happen to agree with your methods or your goal. I would rather see you
treat everyone the way you treat me: with complete intellectual integrity
and without fear of losing them when you point out that they are wrong. I
believe that the history of PCT shows that the only people you have lost
from taking this approach are people PCT didn't need in the first place.

Best regards

Rick

···

--
Richard S. Marken
MindReadings.com
Home: 310 474 0313
Cell: 310 729 1400

4

From[Bill Williams 27 May 2004 2:30 PM CST]

[From Bill Powers (2004.05.27.0956 MDT)]

I hope this is furthering our social convergence. I hope others involved

in

this discussion realize what is going on, get the hang of it, and join in
with the right spirit. Even if one specimen occasionally breaks training.

I just had a hard disk failure wipe out a fairly lengthy post replying in
detail to your today's post. Rather than try to recreate what I said right
now-- which I find an irritating task, let me say this.

It isn't something I haven't said before, but maybe it needs saying again.
I don't as I understand it, see that I have any disagreement with your
approach to modeling. I don't have much of an opinion about whether some
aspects of it are physiologically correct or not. I am not much excited
about whether there "really is" something like your idea of neuro-current or
not. However, from my standpoing the technique you use in modeling seems to
me to have an enormous field of applicability in economics.

However, I do have some disagreements with your approach to what we might
call sophistology, or the meta-psycho-physics that to me appears -- well
maybe this isn't the time, just now for more such as they say frank and
candid exchanges. But, I will say that, at least from
my view our difficulies may have more than we might think to your not
thinking that this aspect of your project is all that important. And, I
tend to attribute many of my objections to what you say to your idea that
the mode of expression in this regard is not as important as the modeling
itself. I might disagree, but I think that I can appreciate why you might
think this the way that you do.

Since it will be a bit before I feel inclined to retype a detailed reply to
your post of the 27th I have included a very rough draft of what may come
close to being a preface for the book I've been working at on applying
control theory to anomalies in economic theory. As I've said, the draft is
very rough, and I expect it will under go many changes, but I've had a great
deal of difficulty arriving at a mode of presentation that I think has a
chance of communicating at all effectively.

I've not yet written the section introducing your contribution. You've been
more generous to me than accurate in descibing our relative contributions to
arriving at a control theory explaination of the Giffen effect. So, I want
to exercise some care in how I introduce you.

IBM C:\VE\O.

     25 May 2OO4

                        Preface

   Science can only travel on a road that has been opened

by technique. Not everyone would accord scientific status

to contemporary economic theory, but whether a science or

not, the state-of-the-art in economics is limited by the

technique which it can employ. Unfortunately, thus far the

only technique that has been available for use in economic

analysis to represent an economic agent has the principle of

maximization. Veblen skewered orthodoxy's conception of

"Economic Man" as a "hedonistic globule of desire." Since

then the critics Hobson, Keynes, Joan Robinson, and Galbraith

among many others have subjected this conception of economic

behavior to a severe, and sustained criticism. However, in

the absence of any alternative means to carry out economic

analysis, the defenders of the orthodox neo-classical

tradition have been in a position to dismiss their critics

complaints by asking,

"What do you have that is any better to put in its place?"

Not, only was this a question to which heterodox economists

had no genuinely satisfactory answer, it was also a question

that occurred to at least some of the heterodox economists

themselves. Certainly the heterodox economists had long enough

to ponder the difficulties that the lack of a plausible

alternative to orthodoxy's principle of maximization created.

Among the unfortunate results was a failure of nerve on the

part of some influential heterodox economists, who in frustration

predicted that the task of creating a theoretical alternative

to the principle of maximization was an impossibility. They

recommended that heterodox economists accept the domination

of the orthodox scheme of theory and abandon efforts to find

an alternative basis for theoretical economics. Among the

chief methodological lessons that should have been learned

during the 2Oth century was not to trust old scientists who

tell us something can not be done. Young economists who tell

us this are no more trust worthy. Ironically the

pronouncements that creating a theoretical alternative to

the orthodoxy system have grown in volume at just the time

when a technique that is capable of generating a theoretical

alternative to Orthodoxy has become available.

   The alternative a technique, or body of techniques, known

as _Control Theory_ has a lengthy history tracing back into

the origins of the industrial revolution where it found use

in regulating the operation of machinery. Over time what

was initially a body of ad hoc contrivances created by the

tinkering of millwrights and other craftsmen, gradually

became a body of systematic knowledge with a theoretical

foundation. Following the extensive development of control

theory during World War Two, the application of control theory

proliferated through out science, industry, medicine, weapons

development and crucially found a connection with biology

and the behavioral sciences. However, the acceptance of this

body of technique in the biological and behavioral sciences

has been very gradual, and only quite recently has it become

acceptable to recognize their role as central themes in the

theoretical description of the life process. In psychology

the application of a control theoretic point of view remains

a distinctly minority position. Partly this has been due the

mathematical intractability of this technique, and the

enormous computationational demands which control theory

makes if the application is implemented in terms of a

computational algorithms. The arrival of micro-chip based

computers in the 198O's removed the computational

barrier to the widespread application of computer based

applications of control theory in industry, and also as a

side effect removed the barrier to the use of control

theory in fields such as heterodox economics that have not

enjoyed lavish financial support.

Insert Powers stuff ...................................

The personal computers available then were based upon 8 bit

chips running at less than a mega-hertz. Now ( 2OO4 ) the

computation speed has increased by more than a 1OOO fold

and 64 bit machines are becoming available while the price

of a typical machine has fallen by a factor of about 1O.

   Eventually this new computational technology can be

expected to have an effect upon the economists approach to

economic analysis. For me the arrival of cheap desk top

computers liberated me from a mathematical regime that was

intractable and allowed me to approach economic analysis

using the algorithmic method of simulation. The principle

result of this opportunity presented here is a demonstration

that control theory provides a superior replacement for

orthodoxy's principle of maximization. There are a number

of well known anomalies that have been emerged as a result

the inability of the principle of maximization to adequately

account for various aspects of economic behavior.

As Geoff Hodgson 199? described the situation,

     "... because no alternative theory human agency was

     proposed to replace the idea of orthodoxy's utility

     maximization the center of the neo-classical

     system remained unchallenged." up. 59.

Hodgson argued from a Darwinian that the system of orthodox

economics had remained unchallenged during the 2Oth century

because the biological sciences had, in Hodgson's terms

"failed to deliver the goods."

5

[From Bill Powers (2004.05.27.1636 MDT)]

Bill Williams 27 May 2004 2:30 PM CST --

However, I do have some disagreements with your approach to what we might
call sophistology, or the meta-psycho-physics that to me appears -- well
maybe this isn't the time, just now for more such as they say frank and
candid exchanges. But, I will say that, at least from
my view our difficulies may have more than we might think to your not
thinking that this aspect of your project is all that important.

Who says I don't think it's important? But I'm not sure what you're
alluding to here, so perhaps I'd better wait for a fuller exposition.

...your idea that the mode of expression in this regard is not as
important as the modeling itself.

I really can't figure out what I said that this pertains to. I'm not good
at figuring out veiled references; much better just to lay it out for me in
simple terms.

The introduction is well done, with a confident approach to introducing
control theory. It's good that there is some opinion already ranged against
the maximization idea. That could make it easier to introduce an alternative.

Best,

Bill P.

P. S. At my last annual eye exam last week I found out why I have to get so
close to the computer screen to see anything. Cataracts, which I have had
for a few years, but rapidly worsening. I get the right lens replaced June
9, and the left one on the 23rd. Then WATCH OUT, Nothing will escape me. I
can't wait to see what things look like in the telescope without all that
gunk in the way.

6

From[Bill Williams 27 May 2004 6:50 PM CST]

Heres a problem you might think about. In mainstream economic there has
been for past 70 years a growing awarness of a persistent anomoly in the
behavior of economy in the way that it adjusts to change. If you assume
that rationality consists of maximization then at least according to
orthodox theorists there is no reason why the adjustments to change should
be any more likely to take place through changes in prices as opposed to
changes in the quantities of commodities exchanged. However, it has been
noticed that rather consistently the shifts in the quantities exchanged
fluctuates 3 to 5 times more than the proportionate shift in the prices. I
had been vaguely aware of this problem, but hadn't given it much thought.

Then it occured to me, in a transaction between two economic agents defined
in terms maximization, the way the price is set is assumed to be costless.
The assumption is that both the buyer and seller are able to caluculate
precisely their gains and losses, and that there are no irrational ego
involvements, and so on and so forth.

The orthodox theory doesn't contain a description of the process of
bargining involved in establishing a price. But, people do bargining and
sometimes put some effort into bargining. In control theory terms setting
the price is can be a costly conflict. A commuter getting of the train in
Manhattan doesn't want to spend time haggling with a newstand paper boy over
the price that is going to be paid for a paper. And, companies and unions
would often rather avoid the struggle over a wage change and a possible
strike.

The result, at I least I think it makes sense, is shift the analysis of the
transaction into control theory terms. Then it is easy enough to think
about setting the price as a conflict. Then once a price has been set,
neither buyer nor seller may wish over a considerable range of situations to
re-engage in a conflict by attempting to change the price. Changes in the
quantity transacted however would appear to be a quite different matter.
Given that the price is for the time being fixed, either a sale and purchase
is agreed to or not. It is a go, no-go situation, and there is no conflict
involved.

So, for one of the most important anomolies in contemporary economic theory,
it would appear that control theory has a decisive advantage over
maximization as an explaination of what is going on.

I haven't been thinking about this very long. I'm not familiar except
superficially with the literature and there is a big literature on the
topic. Nor do I yet have any striking examples to illustrate the effect.
Still, it would appear to me that a control theory conception of economic
agents does explain the tendency to make adjustments in terms of quantity
rather than adjustments in terms of price.

What do you think?

Bill Williams

7

From[Bill Williams 27 May 2004 7:30 PM CST]

[From Bill Powers (2004.05.27.1636 MDT)]

I had when I was a child a moderate problem with astigmatism. When I was
old enough to be fitted with contacts the result was an amazing improvement
in my vision. The amazement lasted, as I remember it, all of about thirty
minutes.

Your problem may be more severe and your amazement at the improvement might
last longer.

But, then thirty minutes is quite a long time-- when you are amazed.

Bill Williams

8

[From Bill Powers (2004.05.27.1917 MDT)]

Bill Williams 27 May 2004 6:50 PM CST --

Heres a problem you might think about. In mainstream economic there has
been for past 70 years a growing awarness of a persistent anomoly in the
behavior of economy in the way that it adjusts to change. If you assume
that rationality consists of maximization then at least according to
orthodox theorists there is no reason why the adjustments to change should
be any more likely to take place through changes in prices as opposed to
changes in the quantities of commodities exchanged. However, it has been
noticed that rather consistently the shifts in the quantities exchanged
fluctuates 3 to 5 times more than the proportionate shift in the prices. I
had been vaguely aware of this problem, but hadn't given it much thought.

I'm not sure what the situation is that you're describing. Are you talking
about shifts in the quantity purchased of a particular good in relation to
shifts in prices of that same good? If that is the case, this implies that
there is a strong desire to get the goods at the lowest possible price, so
even a small increase in price will result in consumers looking for a
different source of the good, or a different good that can be used for the
same purpose. And even a relatively mild bargain would result in consumers
abandoning their other sources. That could be handled by minimax theory,
and also by control theory.

One thing about Charlotte Bruun's model surprised me. The only reason for
which a consumer buys a good in her model is that the consumer tries to
match his consumption to that of consumers near by (her model is divided
into geographic regions). There is nothing to indicate why anyone buys any
good, other than that other people buy it.

What is the reason given in orthodox economics for the fact that people buy
goods?

I think it makes sense [to] shift the analysis of the
transaction into control theory terms. Then it is easy enough to think
about setting the price as a conflict. Then once a price has been set,
neither buyer nor seller may wish over a considerable range of situations to
re-engage in a conflict by attempting to change the price.

This is a possibility, through I have heard that it's only in the United
States (and maybe a couple of other countries) that bargaining is
considered to be unpleasant. In Arab countries, I've read, it's a social
occasion, and simply offering a fixed price is taken as an insult -- you
don't want to interact with me, you think you're too good to enter into a
back-and-forth negotiation, you think you can dictate to me how much I will
spend.

In the commercial world with which I have always interacted, the place of
bargaining is taken by shopping around. The stores post their prices, and
you go looking for the lowest possible price that is easy to take advantage
of. The merchants read their competitors' ads and send spies around town to
see what other merchants are charging for goods, and offer bargains which
they calculate will bring shoppers to their stores. This isn't a one-on-one
negotiation about the price of a particular good, but more like a
many-to-many interaction. It's true that you can't usually beat the price
down in a given store at a given time, but many customers interacting with
many sellers can do about the same thing on a slower time-scale, just by
shifting their custom from one place to another.

Changes in the quantity transacted however would appear to be a quite
different matter. Given that the price is for the time being fixed, either
a sale and purchase is agreed to or not. It is a go, no-go situation, and
there is no conflict involved.

Again, I wonder what the underlying explanation has been for the fact that
people purchase goods at all. I can't really change the quantity of food
that I buy by a significant amount, and people with less money than I
have can't afford to lower the total calories by even a few per cent. Are
we talking about an affluent society in which people have no trouble
meeting their needs, or (as you have proposed) a scarcity society in which
needs are far from being met?

I'm sure you're right that at least in America we find bargaining to be an
unpleasant conflict situation and avoid it. But this doesn't prevent
merchants from raising and lowering prices at frequent intervals, or people
from shifting their purchases to a place that sells for less.

So, for one of the most important anomolies in contemporary economic theory,
it would appear that control theory has a decisive advantage over
maximization as an explaination of what is going on.

You probably have a point but I'm not prepared to declare victory for our
side just yet. I'd feel more secure with a working model to look at.

I haven't been thinking about this very long. I'm not familiar except
superficially with the literature and there is a big literature on the
topic.

It would be good to see some of the data behind the conclusions you
describe. I'm only guessing about how people really react, and for that
matter, about the "changes" you describe.

  Nor do I yet have any striking examples to illustrate the effect.
Still, it would appear to me that a control theory conception of economic
agents does explain the tendency to make adjustments in terms of quantity
rather than adjustments in terms of price.

I think we can try out some simple models to get at this question. First we
need to finish setting up the model of a single specimen repeatedly buying
a single good and paying for it out of income or cash reserves or
borrowing. Then we can go in several different directions. We might
introduce a second good at a different price, as you did in the Giffen
analysis, or a second source for the same good at a lower or higher price.

We might even think about modeling the process of bargaining. I did this
once, using a rather complicated model in which each party tried to
estimate the other party's real maximum offer or minimum acceptance price,
and changed the bid or asked price by a fraction of the "distance" toward
the estimated settling price on each round of negotiation. I didn't know if
it was realistic -- we'd have to compare it with real people bargaining and
see if we could adjust the parameters to make the model behave the same way.

What we could learn from such a model, if successful, would be whether
there are real disadvantages to the bargaining approach over the
price-setting approach with only gradual adjustments. I don't think we
could get away with just saying that bargaining is a conflict situation
without spelling out exactly what the conflict is. People differ so much in
every way that it's hard to say something is true of "people" without being
wrong about a large fraction of them. I'm sure you're tired of hearing Bill
Powers' Basic Law of Sociology, but here it is again: Some Do, and Some Don't.

Best,

Bill P.

9

[From Marc Abrams (2004.05.27.2242)]

[From Bill Powers (2004.05.27.1917 MDT)]

Bill Williams 27 May 2004 6:50 PM CST --

You guys might want to look at the System Dynamics literature. There is
a ton of work that has been done in economics in SD & negative feedback.
They have a very strong dislike for econometrics.

Marc

Considering how often throughout history even intelligent people have
been proved to be wrong, it is amazing that there are still people who
are convinced that the only reason anyone could possibly say something
different from what they believe is stupidity or dishonesty.

Being smart is what keeps some people from being intelligent.

Thomas Sowell

Don't argue with an idiot; people watching may not be able to tell the
difference.

Anon

I don't approve of political jokes. I've seen too many of them get
elected

Anon

10

From[Bill Williams 27 May 2004 11:00 PM CST]

[From Bill Powers (2004.05.27.1917 MDT)]

Bill Williams 27 May 2004 6:50 PM CST --

First, in the orthodox scheme, and by orthodox I mean the micro
neo-classical analysis of individual consumers and producers it is assumed
that a consumer is the principle of maximization. Then commodities have a
utility function such as Ux = f( C/ ( X 1 x X 1) ) or since this isn't
all that clearly expressed -- the Utility obtained from the consumption of
commodity X one is a function of some constant divided by X one squared.

So, suppose you have a mix of commodities and want to maximize your utility
given a limited budget. . You can maximize the utility obtained by
adjusting the consumption of commodities so that the marginal utility, or
dU/dX obtained from each commodity divided by the price of the commodity is
the same for each commodity.

When you assume this maximization of utility subject to a budget constraint
then if the price of one of commodities increases then the maximizing
consumer will purchase less of the commodity now that its price has
increased.

Given this model of a consumer, a consumer ought to make adjustments in the
rate of purchase of all commodities to every change in price of a single
commodity. That is there would be decrease in the purchase of the commodity
who's price has increased and an increase in the purchase of those
commodities who's price has remained constant. However, typically
consumers are not nearly as responsive to price changes as the neo-classical
economists would expect.

And, the Giffen case in which a consumer reacts to a price change in the
opposite way from which would be expected according to the assumption that
consumption is a process of maximization as I described above pulls the legs
out from under the neo-classical table.

But, no one, and I mean this in the sense that if anyone has had any idea
other than maximization subject to a budget constraint as a model of an
economic agent it hasn't appeared in the professional literature-- even as a
curiosity. I simplify don't know of a single case-- not ever, and most
people don't believe there is any economically meaningful to model economic
behavior other than a process of maximization-- probably I should put it
stronger and say then "know" that there is no other system, and they suspect
that it isn't possible to devise an alternative.

There is a reluctance to use maximization as the conception of agency, but
this reluctance is often tied to a revulsion againstscience and math and
modeling.

So, when orthodox theorists build models based upon maximization, the come
the conclusion that when such an economy is subject to random disturbances
it will adjust by dividing the reaction into price changes and changes in
the quantities of commodities sold and purchased. Further they expect that
price changes will be proportionate to changes in the quantities of
commodities transacted. But, that isn't what happens, at least in the
American economy. Rather than prices fluctuating very much adjustments
take place in the quantities transacted-- and by a big margin such as 3 to 5
times as much a change in the volume of transactions as in the price of the
transactions. Prices contrary to the orthodox economists expectation are
considered to be "sticky. " The everyday supermarket and department store
with fixed prices is an violation of the economists model of economic
behavior. So are the prices for many industrial commodities that change
less than do the volume of sales for these commodities. And, there is also
the experience that workers are far more resistant to a company reducing
their money wage than they are to a real wage reduction through inflation.
According to orthodox theorists it shouldn't make any difference to a worker
whether the workers wage is going to be reduce by the company or by
inflation. But the worker is far more resistant to the wage reduction
imposed by the company than to a reduction that takes place through
inflation.

What these anomalous departures with regard to what the orthodox theorist
expects and what is commonly observed amounts to is a massive disconnect
between what many people know to be true from their first hand experience
and what orthodox economic theory predicts will be the case. As a result
confidence even among many thoughtful orthodox theorists in the worth of
their work is not very high, but in the absence of some viable alternative,
they will stick to what they know-- which is maximization.

Whether or not I have a control theory model that explains in particular why
economists contrary to what they expect keep observing that prices are
sticky and adjustments take place more in the volume of transactions, these
folks know that they have a problem-- but they don't see any solution. Not
necessarily that you are going to find anyone who is going to be open to
trying any other method than the one they know, but then someone if they
knew that an alternative existed might sometime be willing to try some other
method than maximization.

If I don't find or think of some clear cut example of price stickiness I may
be reduced to telling a "just so story" of how if one discards maximization
a control theory perspective might explain why prices are, in an industrial
economy, sticky. There really aren't many cases in economics that provide
as decisive an opportunity for a demonstration of control theory as the
Giffen case.

I'm not sure if you will find this post helpful or not. In someways I think
I may just more-or-less repeated my self in only a slightly different way.
But, perhaps you will find the emphasis on how the orthodox economists
explain why a consumer buys something helpful. It really is, at least in my
view, a very strange way of explaining economic behavior. But, at present
the maximization explanation is the only systematic one available that can
explain economic behavior in response to a change in prices.

As I see it, what I need to do is generate a text that brings to the
attention of the heterodox side of the economics profession a notice that
there is now an a theoretical alternative to maximization that can serve as
the basis for the creation of an alternative to orthodoxy. To do this I need
some working examples of a control theory analysis, but mainly what is
needed is an accessible text that presents control theory so that it not
seen as just a technocratic version of National Socialism. I am sure there
are some reasonable people out there somewhere, but many of the people that
I am acquainted with think that Darwin was somehow responsible for Hitler,
and still think that behaviorism is the latest in scientific psychology.

Bill Williams

11

[From Bill Powers (2004.05.28.1259 MDT)]

Bill Williams 27 May 2004 11:00 PM CST --

First, in the orthodox scheme, and by orthodox I mean the micro
neo-classical analysis of individual consumers and producers it is assumed
that a consumer is the principle of maximization. Then commodities have a
utility function such as Ux = f( C/ ( X 1 x X 1) ) or since this isn't
all that clearly expressed -- the Utility obtained from the consumption of
commodity X one is a function of some constant divided by X one squared.

So, suppose you have a mix of commodities and want to maximize your utility
given a limited budget. . You can maximize the utility obtained by
adjusting the consumption of commodities so that the marginal utility, or
dU/dX obtained from each commodity divided by the price of the commodity is
the same for each commodity

Have you transcribed that equation accurately? The function 1/X^2 doesn't
have a proper maximum: it goes to infinity at X = 0. The maximum utility
would occur at zero quantity of X, and it would be infinite. Of course you
say that utility is some function f of this expression, but it would be
awkward to find a form for that function that would produce an ordinary
smooth maximum like a bell-shaped curve. [ X^2 is the old Fortran notation
for the square of X].

  I would think that the proper form would be more like

Ux = C/(X - Xm)^2

where Xm is the value of X that gives infinite utility. This still doesn't
seem right, though, because utility still goes to infinity, and so does
dU/dX. To get a simple bell-shaped curve, you would need something like

            C
Ux = ----------------,
        A + B*(X - Xm)^2

Now when X = Xm, the utility would be at a maximum of C/A, and would
decline smoothly for values of X greater or less than Xm, the peak getting
narrower as B increases. No infinities.

There is a relationship between utility maximization defined this way and
control. Suppose we define a state of zero error in a control system as
being the state of maximum utility of the associated perception. For either
positive or negative errors, the utility is less than maximum.

With that definition, we get an inverse relationship which could be of the
last form given above.

The exact form of the utility function is conjectural, of course -- I doubt
that utility can be measured. However, error can be measured in the context
of a control system. and using the third formula above, be converted to
utility. Lacking any independent measure of utility the definition wouldn't
be very useful, but at least there would be a way to convert from one
measure to the other.

In the control verson, X would be the actual magnitude of the perceptual
signal for a given commodity and Xm would be the reference magnitude, the
desired amount or intake rate of that commodity. If there is resistance to
the use of "perception", the argument can be converted to terms of
observable controlled quantities and reference levels, with no dependence
on a model of the controlling system.

Given this model of a consumer, a consumer ought to make adjustments in the
rate of purchase of all commodities to every change in price of a single
commodity. That is there would be decrease in the purchase of the commodity
who's price has increased and an increase in the purchase of those
commodities who's price has remained constant. However, typically
consumers are not nearly as responsive to price changes as the neo-classical
economists would expect.

If the problem is only that the proportion of quantity change to price
change -- is that ":elasticity"? -- is wrong, the solution could just be a
matter of changing the formula for computing utility. I'm still not clear
as to what is being talked about here, though the last sentence above does
make it seem as if elasticity is meant.

In a control model the solution would be to select the right controlled
variable and adjust the gain of the control loop to fit the data. A large
change in quantity purchased in response to a small change in price
indicates that the most important controlled variable has to do with price,
or cost, and that changing quantity is a means of controlling cost. When
there is a budget constraint, a person may have no choice but to decrease
purchases a lot when prices increase by even a small amount -- you can't
spend money you don't have and can't or don't want to borrow. If there are
alternative commodities, this doesn't necessarily mean giving up the good
-- it's the same situation as in your Giffen analysis, in that you can
still get the calories by switching from meat to bread, but the consumption
of meat is highly senstive to the price of meat (ignoring the Giffen effect
due to change in the price of bread). Look at your Giffen setup and what it
says about changing the price of the MORE expensive commodity. You can
probably get any elasticity you want by changing loop gains in the model.

And, the Giffen case in which a consumer reacts to a price change in the
opposite way from which would be expected according to the assumption that
consumption is a process of maximization as I described above pulls the legs
out from under the neo-classical table.

It may also say something about the sensitivity of quantity to price changes.

I'm not sure if you will find this post helpful or not. In someways I think
I may just more-or-less repeated my self in only a slightly different way.
But, perhaps you will find the emphasis on how the orthodox economists
explain why a consumer buys something helpful.

The thing is, they don't say that people buy things because they want or
need them. The idea of a reference level is missing, as far as I can see.
Keynes said they buy things because they have the money; Charlotte Bruun
proposed that they buy things at the same rate their neighbors buy them.
The orthodox picture says they buy things to maximize something called
utility. But nobody seems to get the idea that they buy things to bring
their inputs to a reference level set by each consumer internally, even
though we can allow other reasons like keeping up with the Joneses (which
is also an internally-set goal, come to think of it). The other reasons are
really quite unimportant when you get into a scarcity situation, What
matters when there is scarcity is staying alive, staying warm and dry,
staying healthy: the stuff of survival. Behavior under those circumstances
is set not by prices but by our basic biological and neurological
construction. If we can't afford to buy food, we beg for it, or grow it, or
steal it.

I'm wondering where it was in my previous exposition that you began to feel
that the argument was straying into metaphysics. Was it when I said that
the environment is a perception, too? Enlighten me.

Best,

Bill P.

12

[From Bill Powers 92004.05.28.0849 MDT)]

Bill Williams 27 May 2004 11:00 PM CST --

It occurs to me that nobody in this conversation has yet defined what
"maximize" means in economics. In my last post, I assumed that it means
what it means in mathematics, system analysis, physics, and so on. In these
fields it means that there is an output variable which is a function of an
input variable, the function being such that as the input variable
increases, the output variable rises at first, reaches a highest magnitude,
and then decreases again as the input continues to increase. So there is
actually a maximum point in the input-output relationship. "Maximizing"
then means finding the input magnitude that will bring the output just to
its maximum possible value. Either a larger or a smaller input would result
in a smaller output.

What some people mean by "maximizing" a variable, on the other hand, is
simply making it as large as possible, with no inherent limit on the output
except the limit on the amount of input you can generate when you try as
hard as you can to make more input. If consuming 1000 calories per day
provides some utility, 5000 would provide more, and 50,000 even more than
that. My discussion of yesterday would not apply to this kind of maximizing.

Best,

Bill P.

13

[Martin Taylor 2004.05.29.22.50]

[From Bill Powers (2004.05.27.1917 MDT)]
I have heard that it's only in the United
States (and maybe a couple of other countries) that bargaining is
considered to be unpleasant. In Arab countries, I've read, it's a social
occasion, and simply offering a fixed price is taken as an insult -- you
don't want to interact with me, you think you're too good to enter into a
back-and-forth negotiation, you think you can dictate to me how much I will
spend.

I took a guided bus tour in Turkey four years ago. Our guide
explained the cultural norms of haggling.

In a bazaar, you haggle. The seller quotes a price. You offer between
half and two-thirds. Less than that says you don't want the object,
but are too chicken to say so. More than that says that you don't
know the system, and therefore are an unpredictable customer. The
final price will be between two-thirds and thre-quarters of the
initial price (I may have the fractions wrong throughout this).

In a shop, you don't haggle. But you do ask if perhaps they might
throw in somwthing beyond the simple item. For example, if you are
buying a dishwasher, you ask if they might perhaps include an
extended warranty at the same price.

Bargaining does take effort, which means that the cost of doing the
bargaining ought to be taken into account in any analysis of the
"correct" price for a transaction. When the sticker price is known by
both sides to be non-negotiable, that additional cost is zero, and
all that matters is a potential conflict internal to either party as
to whether consummating the transaction will increase or decrease
their overall ability to control their perceptions (will I be better
off if I do this, or not?).

I think that this whole business is just one of those cultural
artifacts, like the structures and words of language. Perceptually,
they are as real as a rock, even if they exist only in the
interactions among people.

Incidentally, at some point in this, my 1993 presentation was
mentioned as supporting a convergence of reference levels. I think
that's misleading. What I did was to invoke the notion of
reorganization to argue that a baby will learn to use those
communication techniques that get the results that bring his/her
perceptions closer to their reference values. Usually, those happy
results occur when the baby's communications happen to be enough like
those used by others in the same social context. The baby need never
try to model its language after the language used by those around it,
but its needs will be best satisfied if its language comnes to
resemble the language of those around it. If mimicry makes that
happen quicker than random variation would, then perhaps mimicry
might be a tool. But there's no concept of deliberate "convergence".

Martin