Counter-Control, Moderation and Grand Guignol (was Re: What is CSG for?)

CSGnet Archive CSG2004
2

Michelle Ivers (2004.03.24. 0900 EST)

[From Rick Marken (2004.03.23.0900)]

I don't believe it is possible to design a control system that controls

but

cannot be controlled.

I actually said that we are designed to control our perceptions rather than
BE controlled by others. I think what happened on this discussion forum is
a great example of what happens when people try to control others. Someone
stated that when a control system is pushed and doesn't like it, it pushes
back with equal force. Wouldn't you agree then that as control systems we
don't like being controlled?

Controlling what other people say (and do) seems like exactly the kind of
thing control systems would do.

I'm not sure how you think you can control what others say and do. In my
opinion, you can attempt to affect another control system, you can create a
disturbance for it, you can be part of its feedback function, but I don't
see how you can CONTROL someone else. Please correct me if I'm wrong.

Does this make his "comments" OK with you? If I said that it is my
perception that Michelle Ivers is an ignorant slut (you're probably too
young to remember the great SNL skit from which that phrase is taken) would
I be immune from criticism for having said it? I think not.

Rick, I made no comment as the the "OK-ness" of Bill Williams' comments. My
only comment was that he stated clearly that "it seems to me". (his
perception). If your perception of me is that I'm an 'ignorant slut' for
disagreeing with you then that's ok. (After all, its only your perception
Rick, and that doesn't really matter a whole lot to me.) Whether it comes
from a SNL skit or not (the fact that you didn't think I'd get the reference
anyway), says to me that I've probably caused you a great deal of
disturbance.

Of course I do. And you do as well or you would not have written this post
to me. I want (have a reference) to see substantive, intelligent posts on
topics related to PCT. You want (have a reference) to see me change my
references (so that I want people say to say nasty things about me) so as
not to set up a perfect opportunity for counter-control.

I think you've got me wrong here Rick. I don't have any references about how
other people behave. As I said to Bill Powers, I was merely offering a
suggestion to you about what you could do if you didn't like what was being
said. It doesn't affect me either way, and I don't care how you behave or
what you say or do in this forum. I cannot control what you say or do, (I'm
not sitting there next to you and stopping your fingers from hitting those
keys and then clicking on the send button) nor do I have any desire to
control you or anyone else.

Controlling per se sets up the perfect opportunity for "counter-control"
because counter-control is simply control that takes advantage of the fact
that another system is controlling. Counter-control occurs when the object
of control is a control system rather than a non-control system. The only
thing that distinguishes counter-control from any other kind of control is
that it is based on knowledge of a variable that a control system is
controlling. In counter-control you control the actions of a control system
by disturbing a known controlled variable. For example, I can
counter-control the actions of a thermostat by blowing on an ice cube near
the sensor. The cold air disturbs the temperature that is sensed and
controlled by the thermostat. The thermostat acts to counter this
disturbance by turning on the furnace. I can, thus control the thermostat's
actions, making the furnace go on and off as I desire; I am
counter-controlling the furnace. My ability to exercise this control exists
because I know that the thermostat is controlling temperature. If the
thermostat "died" (stopped controlling temperature) I would lose my ability
to control the thermostat's actions by applying disturbances to the
controlled variable.

I disagree with you here Rick. Counter-control occurs when you (I) have a
reference about another control system doing something. Eg. You appear to
want people to act a certain way when posting to this forum. You have made
your references for this very clear to all who read. It only takes one or
two people to post in a manner that is contrary to your references and you
are the one being counter-controlled. Over the last few months, it has been
really easy to create disturbances for you. (I would suggest that Bill
Williams
picked up on that quite quickly.)
Your example is of controlling a non-living control system (ie. a
thermostat). I'm not sure that this is a good example of what you wanted to
demonstrate, seeing as we are discussing interacting with other living
control systems.

The only way to avoid counter-control, as far as I
know, is to not control at all.

See above, the way to avoid counter-control is to stop having references
about other people doing things.

According to PCT, what living systems do is control.

Yes, living systems control - their own perceptions.

For example, you are counter-controlling when you buy a
$1.50 item and hand the salesperson a $5 bill. You are getting the
salesperson to hand you $3.50 by taking advantage of the fact that the
salesperson is almost certainly controlling for taking only the amount that
the item costs. I don't think either you or the salesperson would find this
counter-control something to be avoided.

I would probably be more inclined to say that I was controlling for owning
the $1.50 item, and that I gave over my $5.00 bill so that I could control
the perception of me owning the $1.50 item. When I go into the shop, I'm
not controlling for getting X amount of change. I'm controlling for
purchasing the item, window shopping etc.

There are situations where people can
take advantage of your controlling to get you to do things you would rather
not be doing, such as yelling or calling names. In this case, what one

wants

to avoid is not the counter-control itself but the undesirable (from one's
own point of view) actions one uses to counter the counter-controller's
disturbances to what one is controlling.

So it was Bill Williams who 'made' you type nasty comments and send them to
this forum? I still don't see how someone else can 'get you to do things
you would rather not be doing'. Surely you have the final choice in whether
you say things or not?

I was suggesting moderation as a means of raising
the general level of discussion on CSGNet so that CSGNet would be be a

place

where I could point intelligent people, like my acquaintance from grad
school, who would like to learn about PCT and how it is applied.

So your friend from grad school would like to see you refering to me as an
'ignorant slut' on a supposedly professional forum? Given that I'm certain
you don't even know who I am or where I'm from, I fail to see how this is
raising the general level of discussion on CSGNet.

Bill Powers seems to accept this mix with equanimity. I
would prefer to try to filter out the National Enquirer component so that
CSGNet becomes a respect source of information about PCT.

Did it occur to you that maybe Bill Powers is controlling his own
perceptions and attempting to match them with his references about how this
list operates?

Maintaining quality is a control process; maintaining the quality of the

discussions on

CSGNet requires a willingness to control what gets onto CSGNet. I think

Bill

Powers' reluctance to control access to CSGNet has made CSGNet

approximately

useless as a venue for teaching and promoting PCT to a scientifically
competent audience. Only those who already understand PCT quite well are
willing to wade through the sewage on CSGNet in order to get to the
occasional nugget.

Again Rick, that's only your perception of things. For example, I wonder
how many readers would agree with the language you used in this email as
being on a par with 'maintaining the quality' on here.

I would never think of pointing anyone I know to CSGNet.

That's a shame Rick.

If, however, CSGNet were
properly moderated by a qualified moderator -- a person with demonstrated
competence in PCT modeling, research and applications -- then I would be
happy to refer several people to PCT.

And would I be correct in guessing that you would consider yourself as such
a qualified moderator? If a moderator rejected one of your emails Rick,
would you continue to feel the same way?

These people might not end up becoming
"converts" to PCT but they would certainly be able to provide intelligent
criticism of it.

Yet again, having references about other people will only cause pain. These
other people may not want to provide criticism or become converts. Would
that be 'intelligent' by your standard? I'm guessing that each of us here
would have a different idea about what is considered intelligent.

And I will enjoy watching and playing in it
occasionally. But my real work will be carried on, as in the past, alone,
with a little help from my friends.

That's great to hear Rick.

Cheers
Michelle.

3

[Martin Taylor 2004.03.23.23.03]

Michelle Ivers (2004.03.24. 0900 EST)

[From Rick Marken (2004.03.23.0900)]

I don't believe it is possible to design a control system that controls

but

cannot be controlled.

I actually said that we are designed to control our perceptions rather than
BE controlled by others. I think what happened on this discussion forum is
a great example of what happens when people try to control others. Someone
stated that when a control system is pushed and doesn't like it, it pushes
back with equal force. Wouldn't you agree then that as control systems we
don't like being controlled?

Not necessarily. I think that quite a lot of the time we control a
percepion of how we are viewed by others, with a reference to
perceive that others view us positively. To control such a perception
with such a reference value almost explicitly says we are open to
control by others, because if the person we are trying to satisfy
shows displeasure, that is a disturbance to our perception that must
be countered. How to counter it? Do something the other wants us to
do--in other words, be controlled by the other.

Less socially, many workers don't really enjoy the job, but do it
because of the disturbance to a perception of available money (and
food and housing) that would occur if they didn't do the job. In
doing the job, they are offering to be controlled by someone else--a
boss or foreman, for example.

When you are allowing yourself to be controlled, what it really means
is that you are finding a way to control a perception of another
person's view of you at a level of perceiving the other to be
satisfied with you. If you like to think of it, there's a looping
feedback path that goes through your actions disturbing the other's
perceptions and their actions disturbing your perceptions. Maybe it's
counter-control, technically, but the result is cooperation, not
conflict.

Martin

4

Michelle Ivers (2004.03.24 1445 EST)

[Martin Taylor 2004.03.23.23.03]

Not necessarily. I think that quite a lot of the time we control a
percepion of how we are viewed by others, with a reference to
perceive that others view us positively.

I would say that if someone held this reference point and wanted to be
perceived positively by others, then they would really be setting themselves
up for a lot of disappointment and counter-control. It doesn't make sense
to me to have a reference about other people doing things. I cannot make
other people do anything.

Less socially, many workers don't really enjoy the job, but do it
because of the disturbance to a perception of available money (and
food and housing) that would occur if they didn't do the job. In
doing the job, they are offering to be controlled by someone else--a
boss or foreman, for example.

I would say that in doing the job they are controlling their perception of
available money. I would suggest that it comes down to a conflict of hating
the job, having no money and then controlling the perception which was more
important to them. If working in a job that they enjoyed was more
important to them than a lack of funds then they would be controlling for
the perception of being happy in a job.

This is all only my opinion, but I still see it as each individual person
controlling their own perceptions. Lets say that its really important to me
that Bob likes me (I hope there's no Bob's out there reading). If I
perceived that Bob doesn't like me then I control that perception by doing
whatever to get the desired result. So when Bob likes me, I have no
perceptual error and everything is happy. Nowhere in that situation is Bob
controlling me. I see it as Bob is chugging along, controlling his own
perceptions, and I'm controlling mine. (it's just really unfortunate that my
reference is for Bob to do something.) If for some reason I go up a level
and control for something else that's more important to me than Bob liking
me, then Bob's opinion no longer affects me. Bob hasn't changed anything
he's done. He is still just chugging along.

If I'm wrong about this then please show me where.

Thanks for your thoughts Martin

Cheers
Michelle

5

[Martin Taylor 2004.03.24.00.17]

Michelle Ivers (2004.03.24 1445 EST)

[Martin Taylor 2004.03.23.23.03]

Not necessarily. I think that quite a lot of the time we control a
percepion of how we are viewed by others, with a reference to
perceive that others view us positively.

I would say that if someone held this reference point and wanted to be
perceived positively by others, then they would really be setting themselves
up for a lot of disappointment and counter-control. It doesn't make sense
to me to have a reference about other people doing things. I cannot make
other people do anything.

As Rick pointed out, you can, if you know what perceptions they are
controlling and are able to disturb those perceptions in ways that
make the actions you desire the most likely way they will counter
that disturbance.

I think you have a reference for other people doing things almost
every time you go about your everyday social interactions. At the
bank you have a reference for the teller to give you money (or take
your money). At a shop you have a reference for the seller to give
you an object you want. And so on.

>Less socially, many workers don't really enjoy the job, but do it

because of the disturbance to a perception of available money (and
food and housing) that would occur if they didn't do the job. In
doing the job, they are offering to be controlled by someone else--a
boss or foreman, for example.

I would say that in doing the job they are controlling their perception of
available money.

True.

This is all only my opinion, but I still see it as each individual person
controlling their own perceptions.

That's all anyone ever does, or can do.

Lets say that its really important to me
that Bob likes me (I hope there's no Bob's out there reading). If I
perceived that Bob doesn't like me then I control that perception by doing
whatever to get the desired result. So when Bob likes me, I have no
perceptual error and everything is happy.

Exactly what I was saying!

  Nowhere in that situation is Bob
controlling me.

I think that we may have a little difference in our understanding of
what "controlling" means. If you really want Bob to like you, and "If
I
perceived that Bob doesn't like me then I control that perception by doing
whatever to get the desired result", then Bob is controlling you,
because he gets you to act in a way that brings his perception of you
to a level near its reference value. Likewise, you are controlling
him, getting him into a state where your perception of him is near
its reference value.

I see it as Bob is chugging along, controlling his own
perceptions, and I'm controlling mine.

Yes.

(it's just really unfortunate that my
reference is for Bob to do something.) If for some reason I go up a level
and control for something else that's more important to me than Bob liking
me, then Bob's opinion no longer affects me. Bob hasn't changed anything
he's done. He is still just chugging along.

Yes.

If I'm wrong about this then please show me where.

I don't think you are wrong about what actually happens in the
control systems. Any disagreement seems to be in what we think of as
"controlling" another person. I think of it as getting another person
to act as I wish them to act. I'm not sure what you think of it as.

Counter-control is ordinarily used when the person you are trying to
control has a reference value of "Not" for perceiving him/herself
being controlled. Then your disturbing actions are likely to be
perceived as giving a value of "Yes" to his/her perception of being
controlled. That disturbance can be corrected only if the other
person acts in a way that they perceive you not to want, which
disturbs a perception in you. That way lies conflict. But most
control we do of other people isn't like that. It happens because the
other person wants to do whatever it is that you want them to
do--perhaps open a window, pass the salt, or sell you an item.

Martin

6

[Peter Small (2004.03.24)

Michelle Ivers (2004.03.24 1445 EST) wrote:

This is all only my opinion, but I still see it as each individual person
controlling their own perceptions. Lets say that its really important to me
that Bob likes me (I hope there's no Bob's out there reading). If I
perceived that Bob doesn't like me then I control that perception by doing
whatever to get the desired result. So when Bob likes me, I have no
perceptual error and everything is happy. Nowhere in that situation is Bob
controlling me. I see it as Bob is chugging along, controlling his own
perceptions, and I'm controlling mine. (it's just really unfortunate that my
reference is for Bob to do something.) If for some reason I go up a level
and control for something else that's more important to me than Bob liking
me, then Bob's opinion no longer affects me. Bob hasn't changed anything
he's done. He is still just chugging along.

If I'm wrong about this then please show me where.

Michelle, reading through your last two posts, I find myself totally
identifying with your take on PCT. But, I don't think this is the way
most of the others are seeing it. They would see the fact that your
wanting to be liked by Bob makes you vulnerable to counter control
because this is influencing your behavior (because you do things to
make Bob like you).

But, as you say, you can easily stop the counter control dead in its
tracks by moving to a different perspective where it isn't important
as to whether or not Bob likes you.

The problem I'm having with PCT is that many people are seeing PCT as
being at a higher level of organization than I feel is warranted. I
don't see it as an overall concept into which other concepts have to
fit. I see it more as a neat little concept that is useful to
incorporate into other concepts higher up the hierarchy.

This idea of concepts fitting into the hierarchy of other concepts is
brought out nicely by Herbert Gintis in an article he wrote in
February 2003 entitled "Towards a Unity of the Human Behavioral
Sciences" (http://www-unix.oit.umass.edu/~gintis/unity_abst.html)

In this paper, Ginis put a strong case for putting Game Theory at the
top of an organizing conceptual framework. This would put PCT into a
proper perspective. Not only this, it gives a new perspective to the
current discussion between Bill Powers and Bill Williams on using PCT
in economics.

Peter Small

Author of: Lingo Sorcery, Magical A-Life Avatars, The Entrepreneurial
Web, The Ultimate Game of Strategy and Web Presence
http://www.stigmergicsystems.com

···

--

7

From[Bill Williams 24 March 2004 6:50 AM CST]

[Peter Small (2004.03.24) in commenting on

Michelle Ivers (2004.03.24 1445 EST) wrote:

The problem I'm having with PCT is that many people > are

seeing PCT as being at a higher level organization > than I feel

is warranted.

This is Peter's perception, but it is a mistaken perception.

[Peter doesn't see control theory as] an overall

concept into which other concepts have to fit. I see it

more as a neat little concept that is useful to
incorporate into other concepts higher up the

hierarchy.

This idea of concepts fitting into the hierarchy of other > concepts is

brought

out nicely by Herbert Gintis in an > article he wrote in February 2003

entitled "Towards a > Unity of the Human Behavioral

Sciences" (http://www-unix.oit.umass.edu/~gintis/unity_abst.html)

In this paper, Ginis put a strong case

Don't believe this for a moment. Gintis's argument is fatally defective.

And, it is easy to demonstrate the deficiency. As far as I know, the

Santa Fe people have yet to make a fundamental contribution to

economic theory. The Gintis paper is based upon a fundametnally

mistaken premise-- maximization.

for putting Game Theory at the top of an organizing

conceptual framework.

Not at all. Not even close.

This would put PCT into proper perspective.

Peter's a bright guy, but he has got this all wrong. All wrong.

Not only this, it gives a new

perspective to the current discussion between Bill >Powers and Bill

Williams on using PCT in economics.

Peter is almost as fun to play with as Rick.

First the discussion between Bill Powers and myself has had nothing at all
to do with applying control theory to economics. So far at anyrate the only
subject of discussion has been a review of an area of agreement regarding
time and modeling, and the start of a discussion concerning macro economic
principles.

In the Gentis paper to which Peter provides a link there appears an
assertion in the section 3: "The Universality of the Rational Actor Model"
their appears an assertion:

In the first sentence in this section Gentis says,

"The rational actor model asserts that agents have _preferences_ reflecting
their utilities and the trade-offs among these wants, and that agents
maximize their utility...."

The recent shift of mainstream economics to a game theoretic stance is often
said, as Peter and Gentis are asserting that game theory provides the answer
to everything, or at least a better answer to the question of how to provide
a foundation for social theory. However, as my quotation from Gentis
indicates, and I thank Peter for bringing the paper to my attention, the
basis of the game theoretic analysis, and this goes back to Frank Knight in
the 1920's, remains a conception of behavior that is based upon the
principle of maximization. The difficulty presented by adopting
maximization as your foundational principle is that it is a conception that
can not provide reasonable answers to easily specified situations. Among
these is the Giffen Paradox. The principle of maximization assumes that
preferences always take the form such that applying the principle of
maximization will identify an optimum pattern of behavior-- and when applied
in economics it doesn't work. The maximization principle has been known not
to work since Alfred Marshall introduced this problem into the mainstream of
discussion in his Principles text in 1895.

When I was a graduate student as a thesis I attempted to develop an
alternative to the principle of maximization. I used the Giffen paradox as
an

anomalous case from which to make this attempt. But, I didn't get anywhere,
so I bluffed my way into a Ph.D. As some people will tell you I

am good at bluffing. Thirteen years later, and still trying to find a
method that would explain the Giffen behavior I encountered Bill Powers. He

was interested in economics, but didn't see a way to apply control theory to
economic questions. Together we solved the paradox over a long

weekend.

The difficulty that I had, knowing the economic side of the issue, and the
difficulty Powers had knowing modeling, was that to solve the problem

you needed to know more than anyone single person at the time knew.
Knowing modeling alone wasn't enough, and neither was knowing the

economics. However, one the problem was solved by combining what we knew, I
did a literature search and I found that the problem had

been solved. See item 4 in the bibliography below. But, Beckman for some
reason didn't go on to apply control theory to enough other

problems in economics to demonstrate its applicability. If you happen to
read the paper which I wrote and Marken edited for "American Behavioral

Scientist" be warned that some parts are all fucked up. Marken thought, as
usual, that he was smarter, knew more, and etc etc, and he

attempted to improve the paper I'd written with consulting me.

So, giving the principle of maximization a game theoretic gloss, ala Gentis
doesn't explain the

Giffen anomaly. However, control theory does explain the anomaly, and in
addition provides the

basis for asserting that no commensurable preference function can explain
Giffenness.

The bibliography is a bit out of date, however, as far as I know nothing of
any significance has happened in the

last year or two, or even threee.

Bill Williams

1 Sigler, George J. 1947 "Notes on the History of the Giffen Paradox."

       Journal of Political Economy Vol 55 April p. 152-6.

2 Prest, A. R. 1948 "Notes on the History of the Giffen Paradox:

      Comment." Journal of Political Economy vol 56 p. 58-6O.

3 Sigler, George J. 1948 "A Reply." Journal of Political

      Economy Vol 56 April p. 61-2.

4 Beckman, M. J. 1953. "Comparative Statics in Linear Programing

     and the Giffen Paradox." Review of Economic Studies

     vol. 22-23. # 61.

5 Alessi, De 1968 "A Methological Appraisal of Giffen's Paradox."

      Weltsirtschalftliches Archive vol 1O1 p. 287-96.

6 Gates, J. K. & Pendleton, T. H. 1968. "Giffen's Paradox: An

       Alternative Solution." Indian Economic Journal. Vol. 15. # 5.

  Other Indian Journal Giffen papers

7 Dholaki, Ravindra H. & Dholaki Bakul H. ??? "Giffen's

       A Comment on the Recent Controversy." Indian

         Journal. Vol. 26. # 45. ???

8 Debroy, B. 1982. "More on Giffen and Gates-Pendleton." Indian

       Indian Economic Journal. Vol. 29. # 4.

9 Gram, W. P. 197O. "Giffen's Paradox and the Marshallian Demand

       Curve." Manchester School Economics and Social Studies

        March p. 65-71.

1O Lipsey R and Rosenbrluth G. 1971 "A contribution to the new theory

      of demand: A rehabilitation of the Giffen Good"

           Canadian Journal economics p. 131-63.

11 Welty, G. A. 1971 "Giffen's Paradox and Falsifiablity."

      Weltwirtschalftliches Archivive vol 107 p 139-46.

12 Ng, Yew Kwang. 1972. "Step-Optimization, Secondary Costraints, and

        Giffen Goods." Canadian Economic Journal. Vol. 5. # 4.

13 Vandermeulen, D. C. 1972 "Upward Sloping Demand Curves without

        the Giffen paradox" American Economic Review 62 (June): 453-58.

14 Steindl, Frank G. 1973 "Money and Bonds as Giffen Goods."

      Manchester School of Economics and Social Studies Vol 41 # 4

       December p. 418-24.

15 Henderson, J. P. 1973. "William Shewell's Mathematical Statement of

        Price Flexiblity, Demand Elasticity and the Giffen's Paradox."

         Manchester School Journal of Economics.

16 Goldfarb, Robert S. 1977 "The Demand Curve for an 'Otherwise-Normal'

        Good can have an upward sloping Segment under Coupon or Time
Rationing."

         Atlantic Economic Journal December p. 68-7O.

17 Boland, L. A. 1977 "Giffen Goods, Market Prices and Testability."

        Australian Economic Papers p. 1O4-9.

18 Koenker, Roger. 1977 "Was Bread Giffen? The Demand for Food

      in England Circa 179O." Review of Economics and Statistics

        vol 59 # 2 May p. 225-229.

19 Rashid, SAlim 1979 "THe Beeke good: a note on the origins of the
"Giffen

        Good" History of Political Economy vol 11 pp 6O6-7.

2O Hughes, Joseph P. 1981 "Giffen Inputs and the Theory of

       Multiple Production [Substitution and Expansion Effects

       in Production Theory: The Case of Joint Production]

         Journal of Economic Theory Vol 25 # 2 October p. 287-3O1.

21 Masuda E. & Newman P. 1981. "Gray and Giffen Goods." Economic

        Journal. Vol. 91. # 364.

22 Hoy, M. and Robson, A. J. 1981 "Insurance as a Giffen Good."

        Economic Letters vol 8 p. 47-51.

23 Dougan, William R. 1982 "Giffen Goods and the Law of Demand."

        Journal of Political Economy" vol 9O # 4 p. 8O9-15.

25 Battalio, Raymond C., Kagel, John H. and Kogut, Carl. 1983

        "Giffen Goods: They Really Exist." unpublished Paper Texas A & M

           University

26 Ammer, Christine and Ammer, Dean S. 1977,84 _Dictionary of Business

      and Economics: Revised and Expanded Edition_ New York: Macmillian

         p. 196. and 229.

27 Dwyer, Gerald P. & Lindsay Cottan M. 1984. "Robert Giffen

        and the Irish Potatoe." AER Vol. 74. # 1. p. 188-92.

28 Silverberg, Eugene and Walker, Donald A. 1984 "A Modern

       Analysis of Giffen's Paradox." International Economic Review

         vol 25 October p. 687-94

29 Ulrich, Kohli. 1985 "Inverse Demand and Anti-Giffen Goods"

        European Economic Review April vol 27 p. 397-4O4.

3O Dooley, P. C. 1985 "Giffen's Hint? " Austrialian Economic Papers

        vol 28 p. 142-44.

31 Pearce, David W., 1986 _The MIT Dictionary of Modern Economics_

         Cambridge. Massachusetts: MIT Press p. 169.

32 Kohli, Ulrich 1986 "Robert Giffen and the Irish Potato: Note"

        American Economic Review Vol: 76 # 3 June p. 539-542

33 Hurwicz, Leonid., James, Jordan., Yakar, Kannai. 1987

      "On the Demand Generated by a Smooth Concavifiable Preference

        Ordering." Journal of Mathematical Economics; Amsterdam

34 Walker, Donald A. 1987 "Giffen's paradox." in _The New Palgrave

      A Dictionary of Economics_ edited by John Eatwell, Murray Milgate

       and Peter Newman. volume 2 E to J London, United Kingdom:

         MacMillian Press

35 Silberberg, A., F. F. Waren-Boulton, and T. Asana 1987.

        Inferior-good and Giffen-good effects in Monkey Choice

         Behavior. Journal of Experimental Psychology: Animal

          Behavior Processes 13 July: 292-3O1.

36 Berg, Morten. 1987 "Giffen's Paradox Revisited." Bulletin of

        Economic Research vol 39 January p. 79-85.

37 Dooley, Peter C. 1988 "Porter's Hint and Alternative Theories

        of the Giffen Paradox: A Reply." Australian Economic

          Papers Vol. 27. # 5O. June p. 142-44.

38 White, Michael. 1988 "Porter's Hint and Alternative Theories

        of the Giffen Paradox: A Rejoinder." Australian Economic

          Papers Vol. 27. # 5O. June p. 145-46.

39 # Corcoran, Elizabeth 1989 "Science and Business: An Economic Animal"

        Scientific American Vol: 260 # 3 March p: 73.

4O Mason, Roger S. 1989 _Robert Giffen and the Giffen Paradox." Oxford:

       Philip Allen

41 Briys. E. G. Dionne, and Eeckhoudt G., 1989 "More on Insurance as a

       Giffen Good. Journal of Risk and Uncertainty vol 2 ( December

         p. 415-2O.

42 Silberberg, JAlan, Hursh, Sstan R? 199O "Quinine Pellets as

      an Inferior Good and a Giffen Good is Rats." Journal of the

        Experimental Ananalysis of Behavior Vol. 53. # 2. March

43 Williams, William D. Giffen Effect Batson News Letter

44 Williams, William D. Giffen Effect Advances in Psychology Springer
Verlag

45 Williams, William D. 199O "The Giffen Effect" American Behavioral

        Scientist Vol: 34 Issue # 1 Sep 1990 p: 106-109.

46 White, M. 199O "Invention in the Face of Necessity: Marshallian

        Retoric and the Giffen Good." The Economic Record vol 66 # 192 March

         p. 1-11.

47 Van Marrewijk, C. and Bergeijk. 199O "Giffen Goods and the Subsistence

        Level." History of Political Economy vol 22 # 1 p. 145-48

48 Van Marrewijik, Charles; van Bergeijk, Peter A G 199O "Giffen goods

        and the subsistence level." History of Political Economy

              Vol: 22 Iss: 1 Date: Spring 1990 p: 145-148

49 Hebert, Robert F reviews: 199O _Robert Giffen and the Giffen

         Paradox_ by Roger S. Mason Journal of Economic History

              Vol: 5O Iss: 2 June 1990 p: 5O2-5O3

5O Battalio, Raymond C; Kagel, John H; Kogut, Carl A 1991

       "Experimental Confirmation of the Existence of a Giffen Good"

         American Economic Review Vol: 81 # 4 Sept p. 961-970

51 Gilley, Otis W. and Karels, Gordon V. 1991 "In Search of Giffen

        Behavior" Economic Inquiry vol 29 January p. 182-89.

52 Dooley, Peter C 1992 "Why did Marshall introduce the Giffen paradox?"

        History of Political Economy [PHPE] Vol: 24 # 3 Fall p. 749-752

53 Boland, Lawrence A. 1992 "Giffen Goods vs Market-Determined Prices" in

       _The Principles of Economics: Some Lies My Teachers told me_

           London: Routledge p. 196-233.

54 Rutherford, Donald., 1992 "Giffen good, Giffen paradox in

      _Dictionary of Economics_ London and New York: Routledge

        p. 19O-91.

55 Davies, John E. 1994 "Giffen Goods: The Survival Imperitive and

       the Irish Potatoe Culture." Journal of Political Economy

         p 547-65.

56 Wichers, Robert C. 1994 "In Search of Giffen Behavior: Comment."

       Economic Inquiry vol 32 January p. 166-7

57 Spiegel, Uriel. 1994 "The Case of a "Giffen Good" Journal of Economic

        Education Spring 1994 # 1 pp 137-47.

58 Yoram, Barzel, et. al. 1995 "The Demand Curves for Giffen Goods

       are Downward Sloping." in _Productivity, Change, Public

         Goods and Transaction Costs: Essays at the Boundaries

          of Microeconomics_ Economists of the Twenieth Century

             series Aldershot U.K.

59 Garratt, Rod 1997 Indivisibilities, inferior goods, and Giffen goods

      Canadian Journal of Economics Vol: 30 # 1 Feb p. 246-251.

61 McDonough, Terrence; Eisenhauer, Joseph 1995 "Sir Robert Giffen

       and the great potato famine: A Discussion of the role of a

        Legend in neoclassical Economics" Journal of Economic Issues

           Vol: 29 # 3 September p. 747-759

6O Weber, Christian E. 1997 "A Difficulty in the Search for Giffen

        Behavior." Unpublished Working Paper Seattle University

62 Weber, Christian E. 1997 "The Case of a Giffen Good: Comment"

       Journal of Economic Education Winter p. 37-44.

63 Spiegel, Uriel. 1997 "The Case of a Giffen Good: Reply"

        Journal of Economic Education Winter p. 45-7.

64 Nachbar, John H. 1998 "The Last Word on Giffen Goods?"

        Economic Theory vol 11 # 2 March p. 4O3-12.

65 Rosen, Sherwin. 1999 "Potato Paradoxes" The Journal of

        Political Economy Vol 1O7 # 6 p. 294-313.

8

[From Bill Powers (2004.03.24.0628 MST(]

Michelle Ivers (2004.03.24. 0900 EST) --

I actually said that we are designed to control our perceptions rather than
BE controlled by others.

Michelle, I think there is some degree of misunderstanding here. The basic
demonstration of counter-control was shown a couple of decades ago at some
CSG meeting, I forget which, using the rubber-band demo. I think you know
this but I'll describe it for anyone who doesn't.

In this demo, a "controller" takes hold of one end of a pair of rubber
bands knotted together, and an "experimenter" takes hold of the other end.
The two participants agree to keep the rubber bands stretched in a plane
just above a tabletop.

The controller picks a spot on the table and moves his or her end of the
rubber bands to bring the knot joining the rubber bands over that spot, and
to keep it there. The experimenter moves the other end of the rubber bands
around just over the tabletop so as to disturb the position of the knot
(not too fast, so the controller can easily keep control).

So far we have a demonstration of a controller keeping an environmental
variable (distance between knot and spot) at a constant value (zero)
despite unpredictable disturbances being generated by the experimenter.

As long as the controller maintains control of the same variable in any
constant reference condition, the experimenter can now control the position
of the controller's hand -- that is, pick another spot on the tabletop, and
move the controller's hand to that position. This is done by manipulating
the experimenter's end of the rubber bands. If you try it, you'll see it's
quite easy.

This is "counter-control." It does not involve any conflict. Both parties
can continue to control the variables each is interested in without
preventing the other from doing the same. What makes it a bit unusual is
that the variable one party is controlling is the other's means of control!
Incidentally, if both parties are aware of this, each can try to control
the other's means of control, but this will probably end in confusion since
nobody is controlling the knot any more.

So counter-control does not mean "getting even" or "controlling back in
retaliation" or anything like that. It just means using the fact that one
person is controlling for something as a way of controlling that person's
means of control.

It's true that if human beings are indeed hierarchies of negative feedback
control systems, they are interested in controlling what they perceive. But
this means only that they can, if they wish, control anything they can
perceive and also affect. It puts no other limits on what they can control.
Since they can perceive various aspects of other people's behavior, they
can control those aspects that are not being controlled by the other
people, if they can find a way to affect them. Of course they can also
control the aspects of other people's behavior that the other people ARE
controlling, if the controller is so much bigger and stronger that the
controlee's resistance makes no difference. People can control other
people, and they often do. They do it by applying, or somtimes just by
threatening to apply, overwhelming physical force.

That's not very nice, but PCT doesn't say anything about what's nice or not
nice. It just says how things work. If we want a society free of brutality,
killing, selfishness, greed, and other stupidities (I obviously do), we
have to find some grounds other than PCT for persuading others to want the
same thing. We have to construct attractive system concepts to which others
may be inspired to subscribe, from which we can develop workable sets of
principles with which to guide our reasoning and our procedures and our
social interactions. We can't control what goals other people adopt at
their highest levels; we might be able to control their actions, but that's
not what they care about. They care about the perceptions they control by
using actions, any actions that are required that don't conflict with the
other things they control..

Controlling the behavior of others can lead to difficulties when the result
is direct conflict. But this should not be taken to mean that PCT says we
shouldn't try to control other people's behavior. All it says is "If you
decide to control another person's behavior, expect difficulties." Who
knows, you may enjoy avoiding or overcoming the difficulties. Think of
people who play football, or chess.

Best,

Bill P.

9

Michelle Ivers (2004.03.25.1000 EST)

[From Bill Powers (2004.03.24.0628 MST(]

Thanks for this Bill.

Controlling the behavior of others can lead to difficulties when the result
is direct conflict. But this should not be taken to mean that PCT says we
shouldn't try to control other people's behavior. All it says is "If you
decide to control another person's behavior, expect difficulties." Who
knows, you may enjoy avoiding or overcoming the difficulties. Think of
people who play football, or chess.

I believe I made in error in that I was looking at things differently. I was
more referring to the fact that just because someone else is creating a
disturbance for us doesn't give us an excuse for our own behaviour. As in
the legal system, using the defence "he was doing something that disturbed
my perceptions" would not be acceptable for murdering someone.

Thank you for explaining things to me.

Cheers
Michelle.

10

[Peter Small 25 March 2004]

It isn't worth countering Bill Williams' comments because it is clear
to me he hasn't even bothered to read beyond the first few lines of
Gintis's article. He rejects the paper on the basis of a complete
misunderstanding of the way Gintis defines maximization.

To have the gall to tell us that he realized Gintis's "error" when he
was a student is just beyond the pale.

Have a look at Gintis's Web site and decide for yourself Who's
opinion you would rather trust: Bill's or Herber Ginis - Professor of
Economics at the University of Massachusetts

http://www-unix.oit.umass.edu/~gintis/

While your on this web site, it may be worth looking at his articles:
The Origins of Human Cooperation
and
The Evolution of Cooperation in Heterogeneous Populations

It might provide some insight into the flames that occur in this forum.

Peter Small

Author of: Lingo Sorcery, Magical A-Life Avatars, The Entrepreneurial
Web, The Ultimate Game of Strategy and Web Presence
http://www.stigmergicsystems.com

···

From[Bill Williams 24 March 2004 6:50 AM CST]

[Peter Small (2004.03.24) in commenting on

Michelle Ivers (2004.03.24 1445 EST) wrote:

The problem I'm having with PCT is that many people > are

seeing PCT as being at a higher level organization > than I feel

is warranted.

This is Peter's perception, but it is a mistaken perception.

[Peter doesn't see control theory as] an overall

concept into which other concepts have to fit. I see it

more as a neat little concept that is useful to
incorporate into other concepts higher up the

hierarchy.

This idea of concepts fitting into the hierarchy of other > concepts is

brought

out nicely by Herbert Gintis in an > article he wrote in February 2003

entitled "Towards a > Unity of the Human Behavioral

Sciences" (http://www-unix.oit.umass.edu/~gintis/unity_abst.html)

In this paper, Ginis put a strong case

Don't believe this for a moment. Gintis's argument is fatally defective.

And, it is easy to demonstrate the deficiency. As far as I know, the

Santa Fe people have yet to make a fundamental contribution to

economic theory. The Gintis paper is based upon a fundametnally

mistaken premise-- maximization.

for putting Game Theory at the top of an organizing

conceptual framework.

Not at all. Not even close.

This would put PCT into proper perspective.

Peter's a bright guy, but he has got this all wrong. All wrong.

Not only this, it gives a new

perspective to the current discussion between Bill >Powers and Bill

Williams on using PCT in economics.

Peter is almost as fun to play with as Rick.

First the discussion between Bill Powers and myself has had nothing at all
to do with applying control theory to economics. So far at anyrate the only
subject of discussion has been a review of an area of agreement regarding
time and modeling, and the start of a discussion concerning macro economic
principles.

In the Gentis paper to which Peter provides a link there appears an
assertion in the section 3: "The Universality of the Rational Actor Model"
their appears an assertion:

In the first sentence in this section Gentis says,

"The rational actor model asserts that agents have _preferences_ reflecting
their utilities and the trade-offs among these wants, and that agents
maximize their utility...."

The recent shift of mainstream economics to a game theoretic stance is often
said, as Peter and Gentis are asserting that game theory provides the answer
to everything, or at least a better answer to the question of how to provide
a foundation for social theory. However, as my quotation from Gentis
indicates, and I thank Peter for bringing the paper to my attention, the
basis of the game theoretic analysis, and this goes back to Frank Knight in
the 1920's, remains a conception of behavior that is based upon the
principle of maximization. The difficulty presented by adopting
maximization as your foundational principle is that it is a conception that
can not provide reasonable answers to easily specified situations. Among
these is the Giffen Paradox. The principle of maximization assumes that
preferences always take the form such that applying the principle of
maximization will identify an optimum pattern of behavior-- and when applied
in economics it doesn't work. The maximization principle has been known not
to work since Alfred Marshall introduced this problem into the mainstream of
discussion in his Principles text in 1895.

When I was a graduate student as a thesis I attempted to develop an
alternative to the principle of maximization. I used the Giffen paradox as
an

anomalous case from which to make this attempt. But, I didn't get anywhere,
so I bluffed my way into a Ph.D. As some people will tell you I

am good at bluffing. Thirteen years later, and still trying to find a
method that would explain the Giffen behavior I encountered Bill Powers. He

was interested in economics, but didn't see a way to apply control theory to
economic questions. Together we solved the paradox over a long

weekend.

The difficulty that I had, knowing the economic side of the issue, and the
difficulty Powers had knowing modeling, was that to solve the problem

you needed to know more than anyone single person at the time knew.
Knowing modeling alone wasn't enough, and neither was knowing the

economics. However, one the problem was solved by combining what we knew, I
did a literature search and I found that the problem had

been solved. See item 4 in the bibliography below. But, Beckman for some
reason didn't go on to apply control theory to enough other

problems in economics to demonstrate its applicability. If you happen to
read the paper which I wrote and Marken edited for "American Behavioral

Scientist" be warned that some parts are all fucked up. Marken thought, as
usual, that he was smarter, knew more, and etc etc, and he

attempted to improve the paper I'd written with consulting me.

So, giving the principle of maximization a game theoretic gloss, ala Gentis
doesn't explain the

Giffen anomaly. However, control theory does explain the anomaly, and in
addition provides the

basis for asserting that no commensurable preference function can explain
Giffenness.

The bibliography is a bit out of date, however, as far as I know nothing of
any significance has happened in the

last year or two, or even threee.

Bill Williams

1 Sigler, George J. 1947 "Notes on the History of the Giffen Paradox."

       Journal of Political Economy Vol 55 April p. 152-6.

2 Prest, A. R. 1948 "Notes on the History of the Giffen Paradox:

      Comment." Journal of Political Economy vol 56 p. 58-6O.

3 Sigler, George J. 1948 "A Reply." Journal of Political

      Economy Vol 56 April p. 61-2.

4 Beckman, M. J. 1953. "Comparative Statics in Linear Programing

     and the Giffen Paradox." Review of Economic Studies

     vol. 22-23. # 61.

5 Alessi, De 1968 "A Methological Appraisal of Giffen's Paradox."

      Weltsirtschalftliches Archive vol 1O1 p. 287-96.

6 Gates, J. K. & Pendleton, T. H. 1968. "Giffen's Paradox: An

       Alternative Solution." Indian Economic Journal. Vol. 15. # 5.

  Other Indian Journal Giffen papers

7 Dholaki, Ravindra H. & Dholaki Bakul H. ??? "Giffen's

       A Comment on the Recent Controversy." Indian

         Journal. Vol. 26. # 45. ???

8 Debroy, B. 1982. "More on Giffen and Gates-Pendleton." Indian

       Indian Economic Journal. Vol. 29. # 4.

9 Gram, W. P. 197O. "Giffen's Paradox and the Marshallian Demand

       Curve." Manchester School Economics and Social Studies

        March p. 65-71.

1O Lipsey R and Rosenbrluth G. 1971 "A contribution to the new theory

      of demand: A rehabilitation of the Giffen Good"

           Canadian Journal economics p. 131-63.

11 Welty, G. A. 1971 "Giffen's Paradox and Falsifiablity."

      Weltwirtschalftliches Archivive vol 107 p 139-46.

12 Ng, Yew Kwang. 1972. "Step-Optimization, Secondary Costraints, and

        Giffen Goods." Canadian Economic Journal. Vol. 5. # 4.

13 Vandermeulen, D. C. 1972 "Upward Sloping Demand Curves without

        the Giffen paradox" American Economic Review 62 (June): 453-58.

14 Steindl, Frank G. 1973 "Money and Bonds as Giffen Goods."

      Manchester School of Economics and Social Studies Vol 41 # 4

       December p. 418-24.

15 Henderson, J. P. 1973. "William Shewell's Mathematical Statement of

        Price Flexiblity, Demand Elasticity and the Giffen's Paradox."

         Manchester School Journal of Economics.

16 Goldfarb, Robert S. 1977 "The Demand Curve for an 'Otherwise-Normal'

        Good can have an upward sloping Segment under Coupon or Time
Rationing."

         Atlantic Economic Journal December p. 68-7O.

17 Boland, L. A. 1977 "Giffen Goods, Market Prices and Testability."

        Australian Economic Papers p. 1O4-9.

18 Koenker, Roger. 1977 "Was Bread Giffen? The Demand for Food

      in England Circa 179O." Review of Economics and Statistics

        vol 59 # 2 May p. 225-229.

19 Rashid, SAlim 1979 "THe Beeke good: a note on the origins of the
"Giffen

        Good" History of Political Economy vol 11 pp 6O6-7.

2O Hughes, Joseph P. 1981 "Giffen Inputs and the Theory of

       Multiple Production [Substitution and Expansion Effects

       in Production Theory: The Case of Joint Production]

         Journal of Economic Theory Vol 25 # 2 October p. 287-3O1.

21 Masuda E. & Newman P. 1981. "Gray and Giffen Goods." Economic

        Journal. Vol. 91. # 364.

22 Hoy, M. and Robson, A. J. 1981 "Insurance as a Giffen Good."

        Economic Letters vol 8 p. 47-51.

23 Dougan, William R. 1982 "Giffen Goods and the Law of Demand."

        Journal of Political Economy" vol 9O # 4 p. 8O9-15.

25 Battalio, Raymond C., Kagel, John H. and Kogut, Carl. 1983

        "Giffen Goods: They Really Exist." unpublished Paper Texas A & M

           University

26 Ammer, Christine and Ammer, Dean S. 1977,84 _Dictionary of Business

      and Economics: Revised and Expanded Edition_ New York: Macmillian

         p. 196. and 229.

27 Dwyer, Gerald P. & Lindsay Cottan M. 1984. "Robert Giffen

        and the Irish Potatoe." AER Vol. 74. # 1. p. 188-92.

28 Silverberg, Eugene and Walker, Donald A. 1984 "A Modern

       Analysis of Giffen's Paradox." International Economic Review

         vol 25 October p. 687-94

29 Ulrich, Kohli. 1985 "Inverse Demand and Anti-Giffen Goods"

        European Economic Review April vol 27 p. 397-4O4.

3O Dooley, P. C. 1985 "Giffen's Hint? " Austrialian Economic Papers

        vol 28 p. 142-44.

31 Pearce, David W., 1986 _The MIT Dictionary of Modern Economics_

         Cambridge. Massachusetts: MIT Press p. 169.

32 Kohli, Ulrich 1986 "Robert Giffen and the Irish Potato: Note"

        American Economic Review Vol: 76 # 3 June p. 539-542

33 Hurwicz, Leonid., James, Jordan., Yakar, Kannai. 1987

      "On the Demand Generated by a Smooth Concavifiable Preference

        Ordering." Journal of Mathematical Economics; Amsterdam

34 Walker, Donald A. 1987 "Giffen's paradox." in _The New Palgrave

      A Dictionary of Economics_ edited by John Eatwell, Murray Milgate

       and Peter Newman. volume 2 E to J London, United Kingdom:

         MacMillian Press

35 Silberberg, A., F. F. Waren-Boulton, and T. Asana 1987.

        Inferior-good and Giffen-good effects in Monkey Choice

         Behavior. Journal of Experimental Psychology: Animal

          Behavior Processes 13 July: 292-3O1.

36 Berg, Morten. 1987 "Giffen's Paradox Revisited." Bulletin of

        Economic Research vol 39 January p. 79-85.

37 Dooley, Peter C. 1988 "Porter's Hint and Alternative Theories

        of the Giffen Paradox: A Reply." Australian Economic

          Papers Vol. 27. # 5O. June p. 142-44.

38 White, Michael. 1988 "Porter's Hint and Alternative Theories

        of the Giffen Paradox: A Rejoinder." Australian Economic

          Papers Vol. 27. # 5O. June p. 145-46.

39 # Corcoran, Elizabeth 1989 "Science and Business: An Economic Animal"

        Scientific American Vol: 260 # 3 March p: 73.

4O Mason, Roger S. 1989 _Robert Giffen and the Giffen Paradox." Oxford:

       Philip Allen

41 Briys. E. G. Dionne, and Eeckhoudt G., 1989 "More on Insurance as a

       Giffen Good. Journal of Risk and Uncertainty vol 2 ( December

         p. 415-2O.

42 Silberberg, JAlan, Hursh, Sstan R? 199O "Quinine Pellets as

      an Inferior Good and a Giffen Good is Rats." Journal of the

        Experimental Ananalysis of Behavior Vol. 53. # 2. March

43 Williams, William D. Giffen Effect Batson News Letter

44 Williams, William D. Giffen Effect Advances in Psychology Springer
Verlag

45 Williams, William D. 199O "The Giffen Effect" American Behavioral

        Scientist Vol: 34 Issue # 1 Sep 1990 p: 106-109.

46 White, M. 199O "Invention in the Face of Necessity: Marshallian

        Retoric and the Giffen Good." The Economic Record vol 66 # 192 March

         p. 1-11.

47 Van Marrewijk, C. and Bergeijk. 199O "Giffen Goods and the Subsistence

        Level." History of Political Economy vol 22 # 1 p. 145-48

48 Van Marrewijik, Charles; van Bergeijk, Peter A G 199O "Giffen goods

        and the subsistence level." History of Political Economy

              Vol: 22 Iss: 1 Date: Spring 1990 p: 145-148

49 Hebert, Robert F reviews: 199O _Robert Giffen and the Giffen

         Paradox_ by Roger S. Mason Journal of Economic History

              Vol: 5O Iss: 2 June 1990 p: 5O2-5O3

5O Battalio, Raymond C; Kagel, John H; Kogut, Carl A 1991

       "Experimental Confirmation of the Existence of a Giffen Good"

         American Economic Review Vol: 81 # 4 Sept p. 961-970

51 Gilley, Otis W. and Karels, Gordon V. 1991 "In Search of Giffen

        Behavior" Economic Inquiry vol 29 January p. 182-89.

52 Dooley, Peter C 1992 "Why did Marshall introduce the Giffen paradox?"

        History of Political Economy [PHPE] Vol: 24 # 3 Fall p. 749-752

53 Boland, Lawrence A. 1992 "Giffen Goods vs Market-Determined Prices" in

       _The Principles of Economics: Some Lies My Teachers told me_

           London: Routledge p. 196-233.

54 Rutherford, Donald., 1992 "Giffen good, Giffen paradox in

      _Dictionary of Economics_ London and New York: Routledge

        p. 19O-91.

55 Davies, John E. 1994 "Giffen Goods: The Survival Imperitive and

       the Irish Potatoe Culture." Journal of Political Economy

         p 547-65.

56 Wichers, Robert C. 1994 "In Search of Giffen Behavior: Comment."

       Economic Inquiry vol 32 January p. 166-7

57 Spiegel, Uriel. 1994 "The Case of a "Giffen Good" Journal of Economic

        Education Spring 1994 # 1 pp 137-47.

58 Yoram, Barzel, et. al. 1995 "The Demand Curves for Giffen Goods

       are Downward Sloping." in _Productivity, Change, Public

         Goods and Transaction Costs: Essays at the Boundaries

          of Microeconomics_ Economists of the Twenieth Century

             series Aldershot U.K.

59 Garratt, Rod 1997 Indivisibilities, inferior goods, and Giffen goods

      Canadian Journal of Economics Vol: 30 # 1 Feb p. 246-251.

61 McDonough, Terrence; Eisenhauer, Joseph 1995 "Sir Robert Giffen

       and the great potato famine: A Discussion of the role of a

        Legend in neoclassical Economics" Journal of Economic Issues

           Vol: 29 # 3 September p. 747-759

6O Weber, Christian E. 1997 "A Difficulty in the Search for Giffen

        Behavior." Unpublished Working Paper Seattle University

62 Weber, Christian E. 1997 "The Case of a Giffen Good: Comment"

       Journal of Economic Education Winter p. 37-44.

63 Spiegel, Uriel. 1997 "The Case of a Giffen Good: Reply"

        Journal of Economic Education Winter p. 45-7.

64 Nachbar, John H. 1998 "The Last Word on Giffen Goods?"

        Economic Theory vol 11 # 2 March p. 4O3-12.

65 Rosen, Sherwin. 1999 "Potato Paradoxes" The Journal of

        Political Economy Vol 1O7 # 6 p. 294-313.

--

11

[Martin Taylor 2004.03.25.10.44]

[Peter Small 25 March 2004]

It isn't worth countering Bill Williams' comments because it is clear
to me he hasn't even bothered to read beyond the first few lines of
Gintis's article.

I have read the paper, and was intending to provide a critique, but
don't have time yet to do it properly. My main issue is in the notion
of rational decision-making as being important in people's behaviour.
Of course, it does happen, but I don't think it is central. I do
appreciate the game-theoretic result--it fits with what happens in
many ecological interactions with non-human species. But it seems to
be an evolved rather than a thought-out pattern of behaviour based on
decision-making of any kind more complex than is available to a bird
or a bee.

As an aside...

Have a look at Gintis's Web site and decide for yourself Who's
opinion you would rather trust: Bill's or Herber Ginis - Professor of
Economics at the University of Massachusetts

One Professor of Economics against another? Horrors. Who would have thought it?

While your on this web site, it may be worth looking at his articles:
The Origins of Human Cooperation
and
The Evolution of Cooperation in Heterogeneous Populations

Here's my second issue: To me it's clear that cooperation has been
the ignored complement of competition ever since the first two cells
found that the behaviour of each enhanced the survival of the other
when they stuck together.

The right question is to determine the environmental conditions under
which cooperation gives way to competition, not to ask why
cooperation occurs. Slime molds provide a beautiful example of how
that balance can change over time and according to resource
availability. And before you ask, I haven't looked at Gintis's work
beyond the referenced paper, so maybe he already says all this. If
so, it would differ from what he seems to say in his paper you
referenced.

Martin

12

[From Bill Powers (2004.03.25.1134 MST)]

Peter Small 25 March 2004 --

Have a look at Gintis's Web site and decide for yourself Who's
opinion you would rather trust: Bill's or Herber Ginis - Professor of
Economics at the University of Massachusetts

I don't base my decisions about whom to trust on their affiliations or
degrees, but on what they write or say. I would say that Gintis is a very
powerful thinker, and I greatly admired some parts of his paper on the
Unity of Sciences. The way he describes Rational Actor theory makes it seem
similar to PCT, except that he never says what a goal is or how a person
can act in such a way that the result is to satisfy a goal. I consider that
a gap in his thinking. Also, "behavior" in his way of thinking seems to
consist only of reaching a decision as to what action to take, but there is
no explanation of how that decision gets converted into actually producing
the action, or more to the point, assuring that the necessary consequence
of the action takes place.

The most interesting part of the paper was the series of studies on
"self-interested actors," a basic assumption of economics which was refuted
in every small culture studied. The actual behavior in the Ultimatum Game,
moreover, conformed to patterns of behavior common in each culture. This is
very much in line with HPCT, in which there is a "principles" level of
organization that governs how detailed procedures and strategies are
organized. It would make sense in the HPCT model that this level would
determine behavior of all sorts, whether in games or in social interactions
-- and of course these studies strongly support that idea. It is not
specific behavior that is the hallmark of the culture, but general
principles that apply to all sorts of behaviors of different kinds.

If I have any overall fault to find in Gintis' approach, it is that
abstract explanations are predominant, with the result that (in my opinion)
the explanations of behavior offered are too complex. I think that much of
the complexity of behavior interpreted in terms of game theory and
evolutionary theory and other such notions may be in the eye of the
beholder. One can always read complex reasoning and statistical
relationships into behavior patterns even if those patterns are produced by
much simpler processes. The result is what I call "overinterpretation"
(borrowed from someone else), meaning that the descriptions and
explanations are not at the simplest level that will actually provide a
workable explanation. An example is a rule for describing how a raindrop
falls: if the ground has not yet been encountered, keep the probability of
descent greater than 0.5; if the ground has been encountered, set the
probability of further descent to zero. Of course that rule would in fact
produce the observed behavior of the raindrop if there were some agency in
the raindrop that could understand and implement it, but I think it far
more likely that the raindrop falls because of gravity, and stops falling
because it hits the ground. At least I would try to verify that explanation
before spending a lot of time on the other one.

Best,

Bill P.

13

From[Bill Williams 25 March 2004 11:20 PM CST]

[Peter Small 25 March 2004]

It isn't worth countering Bill Williams' comments because it is clear
to me he hasn't even bothered to read beyond the first few lines of
Gintis's article.

I guess I missed Peter's magical apperance when he was watching
over my shoulder while I read Gintis's article. Unfortunately there
seem to be lots of things that are clear to Peter that unfortunately
are not so.

He rejects the paper on the basis of a complete
misunderstanding of the way Gintis defines maximization.

I think I understand maximization well enough. Maximization involves
finding a maximum value in for a function or functions.. There are
various techniques for doing so. But, find a maximium the numbers
genrated by the functions or functions have to add up to a maximum.
To do this the numbers have to be comensurate. If Gintis had genuinely
found a new way to "define maximization" then he would not only be
one of the really great economist. And, he would also be a very
outstanding mathematician. And, Gintis is neither.

Not only is the section I objected to defective, the whole paper takes
the form of a promise for the future rather than any thing that Gintis
delivers today.

Let us consider just how, Gintis defines his conception of behavior.

Gintis says,

"The true power of each disciplines contribution to knowledge will
only appear when suitably qualified and deepened by the
contributions of the others." p. 20.

Is there any theory here? Not really. What Gintis proposes is an
eclectic fusion of various findings of social theory-- including
economics. Are there any new insights? None that Gintis reveals,
not yet at least. However, you can make a career out of this sort of
stuff. This is the sort of thing that many people think of as
making a contribution to social theory. I say it is puffed up version
of jam tomorrow, but no jam today.

To have the gall to tell us that he realized Gintis's "error" when he
was a student is just beyond the pale.

But, does this neccesarily mean that what I said is not true? In fact
I did know of the problem that Gintis ignores in the paper that Peter
suggests. My problem seems to be that I read the appendixes to
books.

The error Gintis makes is assuming the adaquacy of a conception
of economic behavior that was adopted late in the 19th century-- the
"neo-classical" tradition, a tradition that is based upon maximization.
Control theory provides a better basis for a theory of economic
behavior than does maximization. The Giffen paradox provides a
test point. Orthodox economic theory based upon maximization does
not explain the Giffen effect. Control theory does. Since I published
the Giffen stuff the first time experiments have been conducted that
have been reported as "confirming" the Gffen effect.

Have a look at Gintis's Web site and decide for yourself Who's
opinion you would rather trust: Bill's or Herber Ginis - Professor of
Economics at the University of Massachusetts

OK. Why not look up the citations to B. F. Skinner and then look up the
citations to W. T. Powers. Then decide for youself who's opinion you
would rather trust-- Fredie Skinner or Bill Powers?

I must say that Peter has a rather excentric approach to argumentation.

Maybe Rick could explain to Peter how serial killers make good police
officers. And, in turn Peter could explain to Rick how Gintis has
generated a new method of maximization and that orthodox economics
is really quite OK.

Bill Williams

14

From[Peter Small 26 March 2004]

Bill William's writes:

The error Gintis makes is assuming the adaquacy of a conception
of economic behavior that was adopted late in the 19th century-- the
"neo-classical" tradition, a tradition that is based upon maximization.
Control theory provides a better basis for a theory of economic
behavior than does maximization. The Giffen paradox provides a
test point. Orthodox economic theory based upon maximization does
not explain the Giffen effect. Control theory does. Since I published
the Giffen stuff the first time experiments have been conducted that
have been reported as "confirming" the Gffen effect.

Bill, I think you are an accountant at heart, not an economist.
Otherwise, I just don't know how you can equate Gintis's concept of
maximization with numbers.

When Gintis talks about maximization, he is talking about maximizing
the chances of achieving a goal. He is not talking about maximizing
the value of a goal.

The Gintis concept of maximization would see Irishmen in the Irish
Famine maximizing their chances of survival by buying more potatoes,
even though the price increases. In other words, there is no Giffen
paradox in terms of Game Theory maximization.

I suggest you read up a little on Game Theory, before you make more
of a fool of yourself.

Peter Small

Author of: Lingo Sorcery, Magical A-Life Avatars, The Entrepreneurial
Web, The Ultimate Game of Strategy and Web Presence
http://www.stigmergicsystems.com

···

--

15

From[Bill Williams 26 March 2004 2:20 PM CST]

I think I've said it before, but Peter really is a lot of fun. And,
one of, at least for me, fun-ist things about Peter is the way
he says things that do such great things for my self-esteem.
Not you understand that my self-esteem necessarily needs
augmenting, but then every little bit can sometimes be handy-
especially when ignorant sluts pop up and are mean spirited
enough to remind me that I am virtually at the very end of my
life. Like they say, life is a bitch and then you die. Welll the
first part is certainly true enough. And, besides who does
she think she is? Goddam hate filled monster!

So, it is such a relief to receive to deal with Peter's post which,
rather than remind me that death is on the way, restricts himself
to saying,

Bill, I think you are an accountant at heart,

Peter, for a guy (me that is) who has had difficulty keeping his
checkbook somewhere short of commercial felonies, I wish even
a little bit of what you say were true.

Peter goes on to say, that Bill Williams,

not an economist. Otherwise, I just don't know how you can
equate Gintis's concept of maximization with numbers.

Well, you ought not to try to create the misleading impression that
I stand alone in equating the concept of "maximization" with a
mathematical quantitative content.

From my 50 cent Standard College Dictionary _c 1950 which says,

p. 737. that,

"Maximum 1: The greatest quantity, amount, degree, or magnitude
that can be assigned. 2: A value of a varying quantity that is greater
than any neighboring value. 3: The highest possible of all the
values which a variable of a function can express; the point
at a varying quantity ceases to increase and begins to decrease.
4 The moment of greatest brightness in a variable star, or its
magnitude at such time.

However, Peter goes on to say, that

When Gintis talks about maximization, he is talking about maximizing
the chances of achieving a goal. He is not talking about maximizing
the value of a goal.

Whatever. Makes no difference at all. If Gntis is going to change the
meaning of a term, which I doubt very much that he actually did, he
would have an intellectual obligation to provide at least some notice
that he is going to use a word in a way that is very different than the
way the world ordinarily uses the word. For your information, Gentis
is basically a muddle headed old Marxist, who in his declining years
had the astonishing insight that the orthodox folks actually had it right.
But, he never actually got to the point of ever understanding how the
principle of maximization works in orthodox economic theory. He as
the paper you provided the link to indicates, now he thinks that he
can, or rather that someone in the future can, meld all sorts of stuff
into a future comprehensive system. As a matter of logic, neither he
nor anyone else, is every actually going to complete the program the
way he describes it.

Peter goes on to say tthat,

The Gintis concept of maximization would see Irishmen in the Irish
Famine maximizing their chances of survival by buying more potatoes,
even though the price increases. In other words, there is no Giffen
paradox in terms of Game Theory maximization.

Does Gintis actually have a running model of how maximization would
work in regard to the Giffen case. ( And, by the way the Giffen case
doesn't have anything more to do with the Irish Famine than saying that
Rick Marken called Michelle an "ignorant slut" has to do with slander.
The potatoes story seems to have been a slur on the Irish thought up
by Jews. ) In the references I supplied papers 27, 32, 40, 55, and 61
bear directly upon the misleading association of the Irish Potato
Famine and the Giffen effect. I would in particular suggest that the
McDonough and Eisenhauer paper ( # 61) provides a good but still
relatively concise treatment of why the common but mistaken idea
that there is any reality to the conjectural attribution of a Giffen effect
to the case of the potatoes and the Irish Famine is entirely bogus. But
McDonough and Eisenhauer make a mistake of their own by a weak
argument that entire Giffen effect is a monstrous myth.

Peter goes on to say,

> I suggest you read up a little on Game Theory, before you make more

of a fool of yourself.

Peter thinks he is being sly, by the way he poses his argument. But, I have
seen this trick before. What Peter is saying, is "Stop before you fuck
another pig." the not too subtle premise of the statement is that I have
already fucked a pig. Peter, when Rick gets back from his apparently
much needed vacation, why don't you see if Rick falls for this argument?

16

From[Bill Williams 26 March 2004 2:20 PM CST]

In replying to Peter just now, I neglected to mention that in my opinion
Rick did a good job of explaining why Peter is mistaken about what he
claims. See,

[From Rick Marken (2004.03.26.1120)]

Peter, you have managed to create a very rare phenomena on CSGnet. It
seems that Rick Marken, Bill Powers, Martin Taylor, and myself are all in
agreement, whatever other differences divide us, that you do not have a
sufficient understanding of control theory to provide a plausible basis for
the arguments you are making.

If we thought you actually understood control theory, some of us might be
more inclined to take what you claim more seriously.

Bill Williams

17

[Martin Taylor 2004.02.26.1646]

> From[Bill Williams 26 March 2004 2:20 PM CST]

Peter, you have managed to create a very rare phenomena on CSGnet. It
seems that Rick Marken, Bill Powers, Martin Taylor, and myself are all in
agreement, whatever other differences divide us, that you do not have a
sufficient understanding of control theory to provide a plausible basis for
the arguments you are making.

If we thought you actually understood control theory, some of us might be
more inclined to take what you claim more seriously.

I think this is quite likely to be unfair to Peter. What I think he
lacks is not an understanding of "control theory" but an
understanding that "perceptual control theory" is different. The two
concepts are not the same. A lot of conventional control theory is
based on the control of output to match some planned pattern in time
and space. The _design_ of this output is supposed to counter
whatever might be unfortunately influencing the thing to be
stabilized. And Peter is quite right in saying that this won't work
under the chaotic conditions that often prevail in the real world.

Why I think Peter may have "standard" control theory in mind is that
he started off by using the word "design" quite a lot, and that in
that context, his comments do make sense. I prefer to think that
someone of obvious intelligence and a good technical background is
making sense in the context of his own perceptions. And I think Peter
is such a person.

A long time ago, I mentioned on CSG-L soem control theorist from (I
think) Sweden (name long forgotten) who had written a strong critique
of "standard" control theory, and favoured what appeared to be a
rudimentary form of PCT. I tried to get him to join CSG-L, but I
don't think he ever did.

Anyway, if Peter is indeed thinking of "control theory" the way much
of the world does, his criticisms are well taken. They just don't
apply to "perceptual control theory".

Martin

18

Froom[Bill Williams 26 March 2004 4:15 PM CST]

[Martin Taylor 2004.02.26.1646]

I stand corrected. I think what Martin says that perhaps I have been unfair
to Peter might be true. I will rephrase what I said, and to make it simple
I will say it this way-- if Peter understood control theory, or PCT, HPCT
the way that Martin does, I don't think that he would be saying the things
that Peter does.

I don't insist that Peter agree with me. I would be satisfied if Peter
agreed with Martin.

Martin is, as everyone knows, much more of an irenist than-- well, I think
you could say most of the rest of us. And, for that reason it is good to
have Martin around.

Bill Williams

19

From[Peter Small 26 March 2004]

From[Bill Williams 26 March 2004 2:20 PM CST]
Peter thinks he is being sly, by the way he poses his argument. But, I have
seen this trick before. What Peter is saying, is "Stop before you fuck
another pig." the not too subtle premise of the statement is that I have
already fucked a pig. Peter, when Rick gets back from his apparently
much needed vacation, why don't you see if Rick falls for this argument?

I know you have a compelling desire to win at competitive games,
Bill, but do you really feel the need to out-gross Nick on metaphors?

Actually, I was quite sympathetic with Nick after he explained how he
came to use that unfortunate phrase. I'm sure even Michelle will be
inclined to forgive him?

Peter Small

Author of: Lingo Sorcery, Magical A-Life Avatars, The Entrepreneurial
Web, The Ultimate Game of Strategy and Web Presence
http://www.stigmergicsystems.com

···

--

20

[Peter Small 2004.02.26]

Martin wrote:

The _design_ of this output is supposed to counter
whatever might be unfortunately influencing the thing to be
stabilized. And Peter is quite right in saying that this won't work
under the chaotic conditions that often prevail in the real world.

Anyway, if Peter is indeed thinking of "control theory" the way much
of the world does, his criticisms are well taken. They just don't
apply to "perceptual control theory".

I think you may be misunderstanding my meaning of chaos. I am not
applying chaos to the outside world. I'm applying chaos to the neural
units that create the internal representations of the outside world
that appear in the mind: the appraisals, the perceptions and the
reference. To my understanding, these manifest as attractors.

As I understand PCT, it is about trying to eliminate any errors that
occur between the perceived world and a reference world in the mind
that represents the world as you would like to perceive it.

As I see the internal worlds as being created as attractor states,
which are very much influenced by emotional states, I cannot see how
you can eliminate emotions from the theory.

In my way of looking at things, the error signal takes the form of
emotions. These alert the cognitive system to the presence of an
error. The cognitive side then assesses the situation (guided by the
emotions) to discover the source of the error. This information is
then used to promote corrective activity.

As I'm understanding how this works in the brain, different solutions
are tried out in the brain until emotions tell you that you may have
found a way to correct for the error. Emotions then prompt you to
take the actions to try out the solution.

As an alternative, if a solution is not forthcoming, the brain
restructures the internal model to change priorities until the
emotional error signal subsides and is replaced with an emotion of
contentment.

Before reading Marc Lewis's paper, I was completely in the dark as to
how this could be arranged neurologically. His paper provided
sufficient information to explain *exactly* how these processes are
carried out. It is as clear and as concise as any description one
might provide to explain a computer system.

If I am wrong here, perhaps you could explain to me how references
are constructed and in which way the presence of an error is detected
and then sent to the cognitive area of the brain to prompt corrective
action? Explanations that include black boxes are not acceptable.

Peter Small

Author of: Lingo Sorcery, Magical A-Life Avatars, The Entrepreneurial
Web, The Ultimate Game of Strategy and Web Presence
http://www.stigmergicsystems.com

···

--

21

[From Bill Powers (2004.03.26.1650 MST)]

Bill Williams 26 March 2004 4:15 PM CST --

Martin is, as everyone knows, much more of an irenist than-- well, I think
you could say most of the rest of us. And, for that reason it is good to
have Martin around.

I don't know how many of you others here have been sneaking off to look in
a dictionary, but I confess that I thought Bill was making a muddled
reference to irony until I did my own sneaking. Irenic is an adjective
relating to peacemaking. Of course I never let on that I didn't know the
word, so don't tell him.

Best,

Bill P.