Correlations and politics

[rsmarken]

[From Rick Marken (2007.08.12.1420)]

Bill Powers (2007.08.12.1130 MDT)--

I think there is a step that has to precede this level of analysis; namely, the step of
modeling the economy from first principles to see how it actually works...

By first principles I mean the simplest facts of control: consumers have
reference conditions for goods, labor, leisure, savings, and other things;
producers have their own reference signals as managers for the conduct of
business, and are also consumers with consumer-like goals.

I prefer to start with data and then try to develop models to explain
the data. What you are describing is what I see as the same back
asswards approach that economists currently take, which is to start
with first principles (theory) and then, maybe, see if the data
conforms. You are just suggesting different first principles.

I've made a number of observations of macro economic behavior that are
interesting because they are inconsistent with what economists,
working from first principles, think should happen. For example, I
found that economic growth precedes rather than follows growth in
capital investment. That's the data. If economists had started with
that observation, rather than first principles, then maybe their
models would explain this phenomenon. Instead, their models, starting
from first principles, say that investment precedes (and causes)
growth. The observation is thus considered some kind of mystery to be
explained by a patch to the first principle (that _must_ be right). In
this way, economist actually proceed exactly as talmudic scholars, who
already know the correct explanation; they just want to figure out how
to fit it to what it doesn't seem to explain.

I would prefer to develop a macro economics that starts with data and
where a model is developed to _account_ for the data (like it's done
in other sciences) and then new are collected to test the model. Since
macroeconomic phenomena are the result of human behavior, I would bet
that the models that works best to explain the economic data will be a
control model And, of course, that model -- control or not -- will
also have to take into account the known environmental constraints on
how people can do their economic thing. But with economics -- as with
other models of more mundane behavior -- I say "phenomena phirst".

Best

Rick

···

--
Richard S. Marken PhD
Lecturer in Psychology
UCLA
rsmarken@gmail.com

[Dr_Schenk]

[Martin Taylor 2007.08.12.14.54]

[From Bill Powers (2007.08.12.1130 MDT)]

<Schnitt>

<Schnitt>

···

Am Sun, 12 Aug 2007 15:30:05 -0400 hat Martin Taylor <mmt-csg@ROGERS.COM> geschrieben:

I think there is a step that has to precede this level of analysis; namely, the step of modeling the economy from first principles to see how it actually works.

We can build models on what we know, and argue about the results, and improve the models to deal with each others' objections, until we arrive at a model that we all agree is acceptable. Then we can watch it work and have some basis for consensus. Short of that, all we can do is exchange preferences and prejudices.

I'm not sure whether to call this defeatist or idealist. Either way, it's a good way to pre-empt other approaches, and, I think unhelpful. The equivalent in physics would be to argue that one should not study thermodynamics until one can model the quantum interactions among all the particles in a system.

"First principles" in PCT includes a lot of general statements that follow from the basic claim: "All behaviour is the control of perception", and from straightforward physical arguments.
Example: It is not possible to control exactly N independent perceptions through an environment with M degrees of freedom where M < N. (The degrees of freedom are, of course, measured in units of inverse time).
Example: If two control systems control their perceptions as the same function of environmental variables, they can do so provided each has a tolerance zone and the two tolerance zones overlap.
Example: The side effects of any action are highly likely to be nearly orthogonal to any specific perception, but it is highly likely that they will disturb some other perception. And so forth.

Statements like these require no models. They are, if correctly stated, simply consequences of normal science outside of PCT, and can be used to develop consequent statements that MUST be true of any PCT model that respects the underlying assumptions (which it must do if it is to represent observable data), in the same way that complicated theorems of Euclidean geometry follow from the interrelations of a few simple axioms. And as with Euclid, it is important that the assumptions of the axioms be respected, because failure in one basic proposition to represent observable reality can result in falsity of the derived propositions. That's where reality checks come in: does the real world actually behave that way?

Most PCT simulations violate obervable reality in that they ignore the necessary sources and sinks of energy. Usually, this doesn't matter, but when we get talking about resource limitations and conflict (resources often being food supply and fossil energy supply) even a PCT model probably would have to take into account the energy supply to the individual control units modelled.

It is always better, of course, if one can produce a model that uses only elementary control units linked by defined signal pathways, and that model accurately reproduces observed data. However, such success never demonstrates that the model is a true representation of the thing modelled. Nor does failure of the model to reproduce data accurately indicate that it is a false model; it may have inaccurate parameter values, or it may be correct but incomplete in that it fails to account for some influences that turn out to be important.

What is best is when different lines of enquiry support each other: models are constructed only of elements for which evidence can be found in the structure of the thing modelled, are based on physical (including chemical and physiological) understanding, and produce results that match data. But none of these lines of enquiry should be used rhetorically or practically to turn off the other lines.

-----------And Now for Something Completely Different-------------------

Going back some years, to your paper with Tom Bourbon in the PCT special issue of International Journal of Human-Computer Studies, I called it "propaganda", a term to which you objected. I did not then understand your objection, but recently I read somewhere a comment about the "American prejudicial connotation of 'propaganda', a connotation that does not exist in English usage" (or words to that effect). You must have thought that my description of your paper as "propaganda" suggested that it was in some way false or unfair, whereas to me it simply meant that the paper was intended to support a position or point of view. Since I was unaware of the difference in usage, as, I presume, were you, the misunderstanding was, in retrospect, almost inevitable.

Martin

--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/

[Bill_Powers1]

“First principles” in
PCT includes a lot of general statements that follow from the basic
claim: “All behaviour is the control of perception”, and from
straightforward physical arguments.

Example: It is not possible to control exactly N

independent perceptions through an environment with M degrees of freedom
where M < N. (The degrees of freedom are, of course, measured in units
of inverse time).
[From Bill Powers (2007.08.12.1530 MDT)]

Who is “Dr. Schenk?” He schnitted pieces from Martin’s post but
doesn’t seem to have written anything himself. I almost deleted it as
spam until I saw the subject. Did I overlook what he wrote?

Martin Taylor 2007.08.12.14.54 –

All right then, second principles.

What I have in mind is simply to treat consumers and producers (i.e.,
managers) as control systems rather than maximizers. The reason goods and
services are bought is that people want them. The reason wages are paid
is that managers need to hire workers to produce the goods. The reason
prices are manipulated is that other managers have to make sure the goods
are sold for as much money as possible but without a surplus building up
in inventory. The reason capital income is needed is that machinery
deteriorates and raw materials (and finished manufacturing equipment)
must be purchased. More income is needed because owners/stockholders
demand returns on their investments, and owners want to receive profit
just because they are owners.

All these basic control processes can be modeled in as much detail as we
like – we can have one good or many, one producer or many, one consumer
or many, one good perf consumer or many, one manager or many. We can
model the big chunks first, then the smaller ones. The model then can be
tested against data, as Rick recomments. Just taking data without a model
doesn’t do much good because you can’t anticipate what data will be
needed. This is an iterative process.

Martin’s first principles will of course show up in any realistic model.
The first one about degrees of freedom will show up as soon as too few
variables in a model are controlled by too many systems. The others will
also show up in any simulation. Since we already know about those first
principles they can be very useful in troubleshooting the model.

Is anyone actually interested in doing this?

Best,

Bill P.

···

Example: If two control systems control their perceptions as the same
function of environmental variables, they can do so provided each has a
tolerance zone and the two tolerance zones overlap.

Example: The side effects of any action are highly

likely to be nearly orthogonal to any specific perception, but it is
highly likely that they will disturb some other perception. And so
forth.

Statements like these require no models. They are, if correctly stated,
simply consequences of normal science outside of PCT, and can be used to
develop consequent statements that MUST be true of any PCT model that
respects the underlying assumptions (which it must do if it is to
represent observable data), in the same way that complicated theorems of
Euclidean geometry follow from the interrelations of a few simple axioms.
And as with Euclid, it is important that the assumptions of the axioms be
respected, because failure in one basic proposition to represent
observable reality can result in falsity of the derived propositions.
That’s where reality checks come in: does the real world actually behave
that way?

Most PCT simulations violate obervable reality in that they ignore the
necessary sources and sinks of energy. Usually, this doesn’t matter, but
when we get talking about resource limitations and conflict (resources
often being food supply and fossil energy supply) even a PCT model
probably would have to take into account the energy supply to the
individual control units modelled.

It is always better, of course, if one can produce a model that uses only
elementary control units linked by defined signal pathways, and that
model accurately reproduces observed data. However, such success never
demonstrates that the model is a true representation of the thing
modelled. Nor does failure of the model to reproduce data accurately
indicate that it is a false model; it may have inaccurate parameter
values, or it may be correct but incomplete in that it fails to account
for some influences that turn out to be important.

What is best is when different lines of enquiry support each other:
models are constructed only of elements for which evidence can be
found in the structure of the thing modelled, are based on physical
(including chemical and physiological) understanding, and produce results
that match data. But none of these lines of enquiry should be used
rhetorically or practically to turn off the other lines.

-----------And Now for Something Completely
Different-------------------

Going back some years, to your paper with Tom Bourbon in the PCT special
issue of International Journal of Human-Computer Studies, I called it
“propaganda”, a term to which you objected. I did not then
understand your objection, but recently I read somewhere a comment about
the “American prejudicial connotation of ‘propaganda’, a connotation
that does not exist in English usage” (or words to that effect). You
must have thought that my description of your paper as
“propaganda” suggested that it was in some way false or unfair,
whereas to me it simply meant that the paper was intended to support a
position or point of view. Since I was unaware of the difference in
usage, as, I presume, were you, the misunderstanding was, in retrospect,
almost inevitable.

Martin

–

No virus found in this incoming message.

Checked by AVG Free Edition. Version: 7.5.476 / Virus Database:
269.11.8/941 - Release Date: 8/7/2007 4:06 PM

–

No virus found in this incoming message.

Checked by AVG Free Edition. Version: 7.5.476 / Virus Database:
269.11.8/941 - Release Date: 8/7/2007 4:06 PM

[Richard_Kennaway]

[From Richard Kennaway (2007.08.13.1023 BST)]

[From Bill Powers (2007.08.12.0140 MDT)]
So the theory of consumer demand is based on the "scarcity" concept of economics. It also fails to include the concept of a reference level, because the reference level defines the amount of a good that is considered "enough", at which level the person will cease any effort to get more of it regardless of its price.

In standard economic terms, that is the level at which a person's marginal utility for the good drops to zero. The marginal utility is the degree to which you want more than you have. That looks to me very like an error level.

My point is not to say that economics is already PCT-based -- it clearly is not. But my suspicion is that building it on PCT lines is going to be like putting Newtonian physics in under Galileo's observations of falling objects and pendulums. Objects will continue to fall and pendulums continue to keep time.

Note also that the control-system model also shows the Giffen Effect, in which raising the price of one good causes a person to buy more of it (this happens when there is a budget constraint and the two goods provide different kinds of benefit at different costs). Bill Williams thought this was an important contribution of PCT to economics.

I'm not sure why, because economists already have their own explanation of the Giffen effect. According to the Wikipedia entry, it was first named and explained in Alfred Marshall's "Principles of Economics" in 1895:

"As Mr. Giffen has pointed out, a rise in the price of bread makes so large a drain on the resources of the poorer labouring families and raises so much the marginal utility of money to them, that they are forced to curtail their consumption of meat and the more expensive farinaceous foods: and, bread being still the cheapest food which they can get and will take, they consume more, and not less of it."

True or false, that's the standard economic explanation. There appears to still be debate about whether the Giffen effect has ever been observed.

···

--
Richard Kennaway, jrk@cmp.uea.ac.uk, Richard Kennaway
School of Computing Sciences,
University of East Anglia, Norwich NR4 7TJ, U.K.

[Richard_Kennaway]

[From Richard Kennaway (2007.08.13.1031 BST)]

Martin Taylor wrote:

I'm just back from a week away, and find a weird series of interchanges between you two, based entirely, so far as I can see, on your individual ideas about morality. I have seen no evidence of scientific enquiry from either of you on this matter.

I saw the purpose as elucidating our respective system concepts and references.

Richard (and Rick): What are the side-effects of a transaction freely entered into? Do they ever disturb a controlled perception of anyone not party to the transaction? Do they ever affect another person's environmental feedback paths in such a way as to make control easier or more difficult?

They do indeed. In economics, this concept is called "externality".

What are "rights", and why are "rights" considered to be outside of the realm of discussion? Why is something a "right" in one culture not a "right" in another? Why do "rights" evolve and change over time within a culture?

I don't think I talked about "rights", only about my personal preferences -- high level references, if you like -- for the way I would like society to be organised.

"Rights" are either (1) what people talk about when they want to pass off their own preferences as objective truths, or (2) the powers that people in a particular society are guaranteed the use of by the institutions of that society (governmental or otherwise). Rights in sense (2) are matters of objective fact and are whatever they happen to be at any particular place and time.

Richard said:

"My belief is this:
    "People should be paid whatever anyone wants to pay them. If that means that work that I personally value highly is paid a fraction of the pay of work I value less, I don't care."

Consider this situation: X is a strong guy with a known-to-be-nasty group of friends (think Cosa Nostra). X conducts a transaction with Y, by simple discussion: X "I think you should pay me $100"; Y "Why would I want to do this"; X "Because you like me, DON'T YOU?"; Y "Here's $100". That's certainly a free transaction, and both parties benefit, Y by getting $100, X by being (temporarily) assured of not having his shop trashed or himself beaten up.

This is certainly not a free transaction. X is threatening violence against Y.

Now Z, who sweeps floors for X, comes along and says "I've swept your floors all month, and now I'd like the $100 we agreed you would pay me". X: "I haven't got the $100, as I gave it to Y".

I think you have your X's and Y's mixed up. Did you mean that Z sweeps Y's floor, and Y can't pay because X took his money? That's the only interpretation that makes sense to me.

Questions (for Richard, but anyone can answer): (1) Has Z been affected by the transaction, freely entered into by X and Y?

The question is moot, because no transaction was freely entered into by X and Y.

If Y had $100 and simply decided not to pay Z, that would be a breach of contract.

A fundamental precondition for people to be able to conduct free transactions is protection against force, and enforcement of contracts (more generally, protection against fraud). To what extent this is provided by a government is not the point. Even libertarians differ about whether and how much of a government it requires, but all are agreed on this: free transactions and enforceable contracts absolutely require societal institutions that support them. And it is "good" to have free transactions and enforceable contracts because it relieves people of the waste of expending a major part of their resources on defending themselves against each other day to day.

(2) According to your moral theory, does Z have any "right" to influence the transaction between X and Y? (3) If Y was not backed by his friends but was X's favourtie nephew, would that change your answers?

Z, and anyone else who knows about it, might very well disapprove of X. They might go so far as to involve the police in the matter. I would like (rights(1)) such transactions as between X and Y to be prevented, through some organisation of society that prevents them (rights(2)).

I'm not sure what the point is of X extorting from his favorite nephew.

Richard said:

My belief is this:
    "People should be paid whatever anyone wants to pay them. If that means that work that I personally value highly is paid a fraction of the pay of work I value less, I don't care."

In the example, did X value the work of Y more highly that that of Z?

I don't understand this -- Y did no work for X in your scenario. Nor X for Y.

Could you discuss other side-effects of transactions freely entered, such as, for example, a transaction whereby X and Y agree that Y should cut all the trees on the land of X, paying X handsomely for them, thereby subjecting the land of neighbour Z to extensive erosion?

You are talking about externalities, which in this example can be handled by property rights (in sense (2)). Z owns his land; X damages it; X owes compensation to Z, and there are institutions through which Z can pursue his claim if X is recalcitrant.

Could these questions be answered without using the concept of "rights" imposed by some higher authority or by your own systems principles? Could they be answered by reference to what would be likely to happen in an environment of control systems interacting through a common environment rather than only with each other in isolated dyads acting in an abstract infinitely resourced environment?

That is a very strange paragraph, in which you're trying not to say whatever it is you're saying. I mean, where do you get this nonsense about isolated dyads and infinite resources? Economics, practically by definition, is about limited resources, and markets with only two participants are a tiny sliver of the interactions that are studied.

Richard and Rick have been kind enough to share their beliefs. Here's mine:

0: One can consider a continuum from "good" to "bad", which is a perception in the observer. In what follows, unless otherwise stated I am the observer.

1. If any action or event either increases the ablity of some control system to control its perception or brings that perception closer to its reference value, without affecting the ability of any other control system to control its perception, that's "good". (The action or event in question may or may not be part of the output of the control system whose perception is under consideration).

Sounds like a PCT version of Pareto optimality. "If any action or event increases a person's utility, without affecting the utility of anyone else, that's "good"."

4. My personal "good-bad" judgment says that no one person is privileged over any other, so if some action or event improves the ability of a small number of people to control while equally reducing the ability of a large number of people to control, that's bad.

How do you measure "ability to control"? Your axioms (1) and (2) don't apply to this case, where there is a conflict or trade-off.

Economists have an answer to how you measure utility. (It should be noted, however, that not all economists are utilitarians.) What measure of "amount of control" can you derive from PCT?

5. [Back to quasi-technical argument] The change in ability to control from the addition or subtraction of one dollar to the available capital is greater, the fewer dollars one has. Hence, transferring one dollar from a rich person to a poor person would be "good" if it were an isolated event. Whether in any specific case it would be "good" depends on the side effects of the transaction.

This is politics by imagination. The dollar will not be transferred unless someone has the power to take the dollar from the rich person and give it to the poor one. The libertarian point, expressed in your terms, is that the very existence of people and institutions with that power itself constitutes a great danger of a large loss of control for all of the people subject to it, both rich and poor.

Richard:
Right, it means taking money from the rich and giving it to government contractors.

And employees, and improving the abilities of both rich and poor to control, not to mention enhancing the abilities of those contractors and employees to control their perceptions.

That's the first time I've seen corporate welfare schemes praised for being just that.

Whether this is "good" or "bad" depends on your personal sense of ethics and morality. In mine, as stated above, it's "good".

Is that still true when the contractor is Halliburton and the work is in Iraq?

In Richard's, I infer from the tenor of the exchange that it's "bad".

Yes.

···

--
Richard Kennaway, jrk@cmp.uea.ac.uk, Richard Kennaway
School of Computing Sciences,
University of East Anglia, Norwich NR4 7TJ, U.K.

[Bill_Powers1]

[From Bill Powers
(2007.08.12.0140 MDT)]

So the theory of consumer demand is based on the “scarcity”
concept of economics. It also fails to include the concept of a reference
level, because the reference level defines the amount of a good that is
considered “enough”, at which level the person will cease any
effort to get more of it regardless of its price.

In standard economic terms, that is the level at which a person’s
marginal utility for the good drops to zero. The marginal utility
is the degree to which you want more than you have. That looks to
me very like an error level.
[From Bill Powers (2007.08.13.0609 MDT)]

Richard Kennaway (2007.08.13.1023 BST) –

What is the name for having more than you want? Utility is the inverse of
error. As utility increases, error decreases. Where error is zero there
is a singularity in utility, even though it is said that marginal utility
goes to zero. I don’t think economists believe there is any such thing as
zero error, or infinite utility.

As we pass through zero marginal utility, the error goes from negative to
positive and the effort, accordingly, reverses. If utility passes through
a maximum, it is still of the same sign, so the action must reverse to
achieve maximization, without anything to indicate which side of the
maximum you are on and hence which way the action should be aimed. Too
much of a good can have the same utility as too little of it. The utility
concept does not fit the facts. You have to start introducing new
concepts as patches to make the utility idea work, such as marginal
utility and disutility. Nobody ever says how you measure the margin when
you’re not at it. Where is, and what determines, the amount of a good
that has zero marginal utility? And how much utility is there when the
marginal utility goes to zero?

My point is not to
say that economics is already PCT-based – it clearly is not. But
my suspicion is that building it on PCT lines is going to be like putting
Newtonian physics in under Galileo’s observations of falling objects and
pendulums. Objects will continue to fall and pendulums continue to
keep time.

I think it will be more like substituting Newtoniam physics for the
concepts of mechanism that preceded it. It used to be thought that when
you throw an object, it is imbued with “impetus”. The object
continues to move until it runs out of impetus, at which point it falls
to the ground. It is hard to judge the angle at which a cannonball fired
into the distance falls, so people really thought cannonballs fall
straight down. Air resistance was unknown, and the steepening of the
downward leg of the parabola or ellipse could not be accounted for.

Aristotle believed that the forward motion was maintained by air curling
in behind the moving object and pushing it forward.

I think economics is still in its pre-Newtonian, or Aristotelian,
phase.

economists already
have their own explanation of the Giffen effect. According to the
Wikipedia entry, it was first named and explained in Alfred Marshall’s
“Principles of Economics” in 1895:

“As Mr. Giffen has pointed out, a rise in the price of bread makes
so large a drain on the resources of the poorer labouring families and
raises so much the marginal utility of money to them, that they are
forced to curtail their consumption of meat and the more expensive
farinaceous foods: and, bread being still the cheapest food which they
can get and will take, they consume more, and not less of
it.”

This is a pretty confused explanation – do you really think that the
marginal utility of money explains the Giffen effect? This was known as
the Giffen Paradox before Williams and I worked out the control-system
explanation.

I think what really explains it is the need for calories, provided both
by meat and by bread, coupled with a preference for meat over bread.
Nothing at all need be said about reference levels for money or marginal
utility of money, because the constraint is simply that the money supply
is too small to let all the calories be supplied by meat. People subject
to the Giffen effect are not free simply to spend more money on bread
when the price of bread goes up. They have to buy more from WalMart and
less from Lord and Taylor if WalMart’s prices go up. Or as Bill Williams
liked to put it, the jet set has to travel more by commercial airlines
and less by private jets when commercial airlines raise their
fares.

The Giffen effect is about conflict.

True or false,
that’s the standard economic explanation. There appears to still be
debate about whether the Giffen effect has ever been
observed.

Yes, Bill Williams ran into that. Economists are well-known for
preferring theory to observation. A PCT model will do much better at
fitting the facts. However, nobody seems to wants such a thing. Economics
is another religion.

Best,

Bill P.

[Kenneth_Kitzke_Value]

[From Kenny Kitzke (2007.08.13)]

<Bill Powers (2007.08.12.0140 MDT)>

<He spoke of people “maximizing,” which is what most economists I know about believe is behind economic decisions. In businesses, I have always heard, the objective is to maximize return on investment. I don’t necessarily believe that this is a universal objective among actual businessmen (as opposed to theoretical ones). In fact, Newell (or was it Simon?) got a Nobel Prize for proposing that managers “satisfice” instead of “maximize” – that they set objectives for profit or production and try to satisfy them.>

I’ll let Richard speak for himself as to what he meant. I find his insights into PCT, mathematics and life goals quite “satisfying.”

I see nothing unusual or inherently wrong about humans desiring to “maximize” their income; the fruit of their work. This applies to any form of work such as being an employee, business owner or even an investor. Do you? Of course, what one does to “maximize” that income can be illegal, immoral or unethical. Hopefully that would be the exception and not the norm.

Return on Investment is not a common measure of income for small businesses. It applies more to large capital intensive businesses. No one knows in advance what the maximum is. Large businesses do like to set goals or objectives in their “profit” plans because of this. I do not believe that is a good approach. I simply look to do better than last year. It is amazing how continual steady growth, even at 8%, produces dramatic results over a few years (like compound interest).

Barbers, insurance salesmen and consultants like me probably never use it. I surely don’t. But, we do seek income and profit; and often try to maximize it…but not always. I, for example, have been in the situation of paying over 50% of my “net income” in taxes. The effort it takes to use less than 50 cents of every dollar I earn has left me with a choice: more free time or more income? I have often chosen the former and, on purpose, have turned down income producing activities.

I see a major disconnect between some of your feelings about businesses when you are talking about Walmart or General Electric or Enron and the local real estate agent (an independent contractor under the law) or a baker or furnace installer, folks who operate businesses often as a sole-proprietor, small partnership (like a legal firm) or a small business, even a family business, with say less than 10 employees. Such business owners are unlikely to think of becoming “super rich.” They work to earn a living to some standard (reference) that is perceived as “right” or adequate for them.

There are rotten apples in every category of workers and there are some golden delicious ones too. I see things in big business that sickens me, but I can find such things in labor unions, government bodies, public schools and churches. There is a distribution and while that may tell us something about the makeup of the categories on the average, it will not say anything about any specific specimen as you so rightly have observed.

Every human being is unique with their own reference perceptions that not only escape our precise understanding, but are often not easily identified even by the person. With practically infinite varieties of personal reference perceptions, it is little wonder that human behavior is also best understood one specimen at a time.

···

---

Get a sneak peek of the all-new AOL.com.

[Martin_Taylor2]

[Martin Taylor 2007.08.13.17.50]

[From Richard Kennaway (2007.08.13.1031 BST)]

Martin Taylor wrote:

I'm just back from a week away, and find a weird series of interchanges between you two, based entirely, so far as I can see, on your individual ideas about morality. I have seen no evidence of scientific enquiry from either of you on this matter.

I saw the purpose as elucidating our respective system concepts and references.

It may well have done that, but those are matters of private interest. I understood the interchange to have been intended to advance in some way our understanding of PCT in relation to political theory. I intervened in an apparently misguded attempt to steer the discussion more nearly in that direction.

Richard (and Rick): What are the side-effects of a transaction freely entered into? Do they ever disturb a controlled perception of anyone not party to the transaction? Do they ever affect another person's environmental feedback paths in such a way as to make control easier or more difficult?

They do indeed. In economics, this concept is called "externality".

Good. Though I'd quibble with identifying PCT side-effects with "externalities". The meanings and implications certainly overlap, but they aren't the same; or do you think they are?

Now let's pursue the question of how this agreed fact relates to the questions of how PCT relates to politico-economic theory. I asked the several questions, because it seemed clear to me that your whole approach is based on transactions freely agreed between two parties, with absolutely zero consideration of the effects on the controlled perceptions of other people not involved in the transaction. I'd hoped Rick would answer them from his standpoint, too.

What are "rights", and why are "rights" considered to be outside of the realm of discussion? Why is something a "right" in one culture not a "right" in another? Why do "rights" evolve and change over time within a culture?

I don't think I talked about "rights", only about my personal preferences -- high level references, if you like -- for the way I would like society to be organised.

Fair enough. Then I surmise that you understand "rights" in much the same way as I do.

"Rights" are either (1) what people talk about when they want to pass off their own preferences as objective truths, or (2) the powers that people in a particular society are guaranteed the use of by the institutions of that society (governmental or otherwise). Rights in sense (2) are matters of objective fact and are whatever they happen to be at any particular place and time.

Good. The possibility of "rights" (2) seems to imply some means of enforcing these rights, by acting in some way against anyone who would violate them.

Richard said:

"My belief is this:
    "People should be paid whatever anyone wants to pay them. If that means that work that I personally value highly is paid a fraction of the pay of work I value less, I don't care."

Consider this situation: X is a strong guy with a known-to-be-nasty group of friends (think Cosa Nostra). X conducts a transaction with Y, by simple discussion: X "I think you should pay me $100"; Y "Why would I want to do this"; X "Because you like me, DON'T YOU?"; Y "Here's $100". That's certainly a free transaction, and both parties benefit, Y by getting $100, X by being (temporarily) assured of not having his shop trashed or himself beaten up.

This is certainly not a free transaction. X is threatening violence against Y.

Let's pursue the assertion that this is "certainly not a free transaction." I'd like you to demonstrate, using only PCT concepts, that this transaction is any less free than any other that might have been conducted between these two people.

Some points: (1) X never threatened any violence in the overt dialogue. The threat, if any, is contextual, based on the society and culture in which both X and Y live. That society and culture is a fact of life for both of them. That being so, in your view, could ANY transaction between these two be free, X always having the same set of friends?

If there could be a free thransaction between these two, how? If not, the implication is that there are societies in which Libertarianism could not take hold, beacuse there exist classes of people between whom free transactions are impossible. You may well agree that this is unfortunately so, but if you do, you must also agree that Libertarianism is a Utopian system, that could not evolve in the real world, although it might be evolutionarily stable were it ever to exist.

(2) Assuming that X is indeed perceived by Y to be threatening violence, how does this make the transaction less free? By agreeing to it, Y brings his perception for the future state of his shop and himself nearer their reference values than they would be if he rejected the transaction, and X brings his perception of the amont of money he holds nearer its reference value. These seem to me to be precisely the conditions for a free transaction; both parties bring some controlled perception nearer to its reference value than it would have been without the transaction. Or are you saying that some controlled perceptions SHOULD not enter into consideration in conducting a free transaction? If you are, on what basis are the permissible controlled perceptions to be selected?

Now Z, who sweeps floors for X, comes along and says "I've swept your floors all month, and now I'd like the $100 we agreed you would pay me". X: "I haven't got the $100, as I gave it to Y".

I think you have your X's and Y's mixed up. Did you mean that Z sweeps Y's floor, and Y can't pay because X took his money? That's the only interpretation that makes sense to me.

Yes, I got them mixed up, as you correctly divined. I would have wished you had answered the rest of the questions on that basis, but since you didn't, I'll try rephrasing some of them.

Questions (for Richard, but anyone can answer): (1) Has Z been affected by the transaction, freely entered into by X and Y?

The question is moot, because no transaction was freely entered into by X and Y.

It's still a question. The same issue would arise if the transaction X and Y had entered had been one you would have been happy to call free, so long as it gave X $100.

If Y had $100 and simply decided not to pay Z, that would be a breach of contract.

Yes, that's an appropriate label. Does it tell us anything in the language of PCT?

A fundamental precondition for people to be able to conduct free transactions is protection against force, and enforcement of contracts (more generally, protection against fraud).

So, in this case, what is it that protects Z against the effects of the transaction between X and Y? Even more simply, if Y simply chooses not to honour the contract with Z, what can Z do about it -- in other words, what constitutes "enforcement" without force or the expectation of force?

To what extent this is provided by a government is not the point. Even libertarians differ about whether and how much of a government it requires, but all are agreed on this: free transactions and enforceable contracts absolutely require societal institutions that support them.

"That support them" how? Surely "societal institutions" must do something to influence a controlled perception on the part of the fraudster or the one contemplating the use of force? In the non-libertarian world, that influence is in the form of an intent to set up a conflict: the "societal institution" has the ability to control some perception controlled by the criminal with more force than the criminal can muster. In the libertarian world, what is done instead to "support" enforceable contracts and the non-use of force?

And it is "good" to have free transactions and enforceable contracts because it relieves people of the waste of expending a major part of their resources on defending themselves against each other day to day.

Yep. That is the intent of the _enforceable_ rule of law in present society. It sounds more and more as though your preferred society is the one we (are supposed to) live in.

(2) According to your moral theory, does Z have any "right" to influence the transaction between X and Y? (3) If Y was not backed by his friends but was X's favourtie nephew, would that change your answers?

Z, and anyone else who knows about it, might very well disapprove of X. They might go so far as to involve the police in the matter. I would like (rights(1)) such transactions as between X and Y to be prevented, through some organisation of society that prevents them (rights(2)).

I'm not sure what the point is of X extorting from his favorite nephew.

I hope that wasn't a deliberate side-step away from an answer to (3). To clarify, you were correct in thinking that X and Y got switched right after they were initially introduced, and they stayed switched. Y can't pay Z because out of the goodness of his heart he gave his favourite nephew (now X) the $100 because he agreed with "Because you like me, DON'T YOU?" not from fear of reprisal. My question was whether your answers would be changed by this change in the relation of X and Y. The question stands.

Richard said:

My belief is this:
    "People should be paid whatever anyone wants to pay them. If that means that work that I personally value highly is paid a fraction of the pay of work I value less, I don't care."

In the example, did X value the work of Y more highly that that of Z?

I don't understand this -- Y did no work for X in your scenario. Nor X for Y.

Precisely the point of the question. The question stands.

Could you discuss other side-effects of transactions freely entered, such as, for example, a transaction whereby X and Y agree that Y should cut all the trees on the land of X, paying X handsomely for them, thereby subjecting the land of neighbour Z to extensive erosion?

You are talking about externalities, which in this example can be handled by property rights (in sense (2)). Z owns his land; X damages it; X owes compensation to Z, and there are institutions through which Z can pursue his claim if X is recalcitrant.

Government (or private police) use of force, in other words. But remember, Y did the cutting and the damage. X agreed to it, for money. Which one should, in your view, be subject to enforcement? You say the owner of the land neighbouring the land damaged, not the person who did the damage. Why?

"Externalities" is a word. It is not the same as side-effects in the PCT sense, though the implications do overlap.

Could these questions be answered without using the concept of "rights" imposed by some higher authority or by your own systems principles? Could they be answered by reference to what would be likely to happen in an environment of control systems interacting through a common environment rather than only with each other in isolated dyads acting in an abstract infinitely resourced environment?

That is a very strange paragraph, in which you're trying not to say whatever it is you're saying. I mean, where do you get this nonsense about isolated dyads and infinite resources?

From your talk about free transactions freely agreed between the partners in the transaction. In nothing that I read did you mention the interactions with others affected by the transaction. Nor did you indicate that the possibility had occurred to you that the transaction might involve depriving someone else of resources they might have otherwise been able to use.

Actually, the question is independent of whether in the back of your mind you were actually thinking of a regulated society in which the side-effects on others are taken into account in enforceable laws that regulate permissible transactions. The question is whether it is _possible_ to treat what would happen in a system of interacting control systems, without invoking any concept of "rights".

  Economics, practically by definition, is about limited resources, and markets with only two participants are a tiny sliver of the interactions that are studied.

Agreed, but I thought you were discussing libertarian principles and PCT, not classical economic theory or even "Economics" more generally.

Richard and Rick have been kind enough to share their beliefs. Here's mine:

0: One can consider a continuum from "good" to "bad", which is a perception in the observer. In what follows, unless otherwise stated I am the observer.

1. If any action or event either increases the ablity of some control system to control its perception or brings that perception closer to its reference value, without affecting the ability of any other control system to control its perception, that's "good". (The action or event in question may or may not be part of the output of the control system whose perception is under consideration).

Sounds like a PCT version of Pareto optimality. "If any action or event increases a person's utility, without affecting the utility of anyone else, that's "good"."

4. My personal "good-bad" judgment says that no one person is privileged over any other, so if some action or event improves the ability of a small number of people to control while equally reducing the ability of a large number of people to control, that's bad.

How do you measure "ability to control"? Your axioms (1) and (2) don't apply to this case, where there is a conflict or trade-off.

That's why it's a separate statement.

As for how one measures "ability to control", that's an issue that has long bothered me. It's not clear at all. But I am leaning toward the idea that it's only at the top level (control of systems perceptions, in Bill's hypothetical hierarchy) that the measurement should be made.

Why is it a difficult issue? Partly because of the importance of "tolerance" in both the colloquial and the engineering sense. Tolerance means that you can't simply use as a measure the deviation of the perception from its reference value. Tolerance means that sufficiently small deviations are functionally equivalent to zero error. Tolerance allows systems to control in circumstances that would otherwise lead to conflict.

Economists have an answer to how you measure utility.

PCT would say that it's based on conflict. One thing has more utility than another if it is preferred when you can have either one or the other, but not both. I don't know whether most economists treat utility as a time-varying function, though. Maybe you do.

What measure of "amount of control" can you derive from PCT?

You don't "derive" it from PCT. PCT recognizes that control is what is going on. Any measure must be based on observables (which might be self-reported).

It's quite possible that in the development of PCT science we might some day find that a reasonable proxy measure is something as simple as "happiness", and that the US principles of "life, liberty, and the pursuit of happiness" are PCT-correct, whereas the Canadian principles of "peace, order, and good government" are simply enablers for the US set.

5. [Back to quasi-technical argument] The change in ability to control from the addition or subtraction of one dollar to the available capital is greater, the fewer dollars one has. Hence, transferring one dollar from a rich person to a poor person would be "good" if it were an isolated event. Whether in any specific case it would be "good" depends on the side effects of the transaction.

This is politics by imagination. The dollar will not be transferred unless someone has the power to take the dollar from the rich person and give it to the poor one.

Yep. But it's not imagination, really. There may be occasions on which one dollar makes a big difference to the ability of a rich person to control some perception, but they are likely to be few compared to the times this is true for a poor person. So, using the quasi-axioms I presented as my own moral beliefs, transferring the dollar from the rich person to the poor one is more often likely to be "good" than "bad".

  The libertarian point, expressed in your terms, is that the very existence of people and institutions with that power itself constitutes a great danger of a large loss of control for all of the people subject to it, both rich and poor.

Now THAT is politics by imagination! If you can effectively answer the questions you omitted to answer, which I restated above, without entering into self-contradiction, you may have made a start. But at present, it's a simple assertion that I think has very little chance of being demonstrable.

Richard:
Right, it means taking money from the rich and giving it to government contractors.

And employees, and improving the abilities of both rich and poor to control, not to mention enhancing the abilities of those contractors and employees to control their perceptions.

That's the first time I've seen corporate welfare schemes praised for being just that.

I see. To employ someone to produce something is "corporate welfare". Interesting use of language!

Whether this is "good" or "bad" depends on your personal sense of ethics and morality. In mine, as stated above, it's "good".

Is that still true when the contractor is Halliburton and the work is in Iraq?

Relevance?

If we are to have a "clean" discussion of the reasonableness or self-consistency of the Libertarian concept based on the principles of PCT, then the use of emotionally charged language is not helpful. (I would prefer a discussion of the evolutionary stability of the Libertarian society, but that's probably too much to ask).

Remember, in PCT terms, "force" is essentially physical. The threat of force is different. It involves placing someone in a conflict situation by making it likely that some controlled perception will increase its error (e.g. not feeling hurt) unless some other perception increases its error instead (giving money to the extortionist means having less when you wnat more, but not being hurt). People may have multiple ways of getting out of a conflict situation (think MOL), but they have no way of getting out of a situation in which they are subject to force (other than by persuading the source of the force to stop applying it -- possible when that source is human, not when it's something like a mountain landslide).

So, you are quite correct in saying that "the very existence of people and institutions with that power itself constitutes a great danger of a large loss of control for all of the people subject to it, both rich and poor", but only when you apply it to specific individual transactions. It is also true that excessive use of institutional power can lead more generally to loss of control for most (not all -- there are those who wield the levers of power) (witness the Nazis or Soviet Russia).

But just as one can die from eating excessive salt and also from eating no salt, so I believe a society would be likely to die from both excessive government regulation and from insufficient government regulation. I mean "die" in the sense that in either case, most would have little ability to control much of anything, whereas a very few would have the ability to control a great deal.

Now that last paragraph IS politics by imagination.

Martin

[rsmarken]

[From Rick Marken (2007.08.13.2035)]

Bill Powers (2007.08.13.0609 MDT)--

True or false, that's the standard economic explanation. There appears to
still be debate about whether the Giffen effect has ever been observed.
Yes, Bill Williams ran into that. Economists are well-known for preferring
theory to observation. A PCT model will do much better at fitting the facts.
However, nobody seems to wants such a thing. Economics is another religion.

Bill, your posts on economics have been clear, incisive and
informative (thanks for the Wiki pieces). In an earlier post you asked
whether anyone would be interested in joining in on development of a
PCT based economics. I would but I've found this economics thing to be
quite unsatisfying, as a social enterprise, anyway. Economics is,
indeed, a religion and I just don't have a very good time dealing with
religious people.

Best

Rick

···

--
Richard S. Marken PhD
Lecturer in Psychology
UCLA
rsmarken@gmail.com

[Bill_Powers1]

Now let’s pursue the question of
how this agreed fact relates to the questions of how PCT relates to
politico-economic theory. I asked the several questions, because it
seemed clear to me that your whole approach is based on transactions
freely agreed between two parties, with absolutely zero consideration of
the effects on the controlled perceptions of other people not involved in
the transaction.
[From Bill Powers (2007.08.14.0855 MDT)]

Martin Taylor 2007.08.13.17.50 –

If I hold a gun to your head and say “Your money or your life,”
you are entirely free to decide to give up either your money or your
life. The same is true if I say “Your money or your child’s
life.” In neither case have I physically damaged either you or your
child. I have neither increased nor decreased your ability to control
because I have not, as yet, either taken your money or killed anyone. All
I have done is utter a threat, which I may or may not carry out.

It does not seem to me that the question at hand can be resolved in this
way, because this argument takes into account only actual physical damage
or restraint used as a means of controlling someone else. This argument
recognizes only accomplished events, which is too low a level of
perception to be useful in discussing a system concept. In B:CP I made
the same mistake, saying that only overwhelming physical force can be
used as a way of controlling another person. It’s not the force that
matters, but the higher purpose for which it is applied.

If the above arguments are accepted, then truly there is no way for one
person to override another’s control except by the actual use of superior
physical resources. The problem with this conclusion is that it leaves no
way to organize a society except to forbid nothing and force nothing, and
leave the settlement of all conflicts up to the disputants. Since some
people are physically weaker than others, it will be necessary for the
weaker to find equalizers, like the Colt .45 six-gun of the Old West,
called even then “The Equalizer,” or the longbow of England.
Weapons of stealth, too, like poison and fire, which minimize the
identification of the user, will proliferate. It will still be true that
the intelligent will prey on the less intelligent, tricking people of
slow or damaged wit into acts that appear beneficial on the surface but
prove in subtle and baffling ways to be detrimental to them, but of
course beneficial to the perpetrator. The ruthless will prey on people of
conscience, the avaricious on the generous, the untrustworthy on the
trusting, the glib on the gullible. There is a reason why we have words
so readily at hand for all these things.

We will get, in short, pretty much the world as it is. The argument for
total freedom says we must leave everything alone, because everyone is
already exerting as much control as possible. When you analyze to the
bitter end, this even includes all the people who are – of their own
free choice – trying to control other people and choosing to experience
the consequences.

On the other hand, any proposals for how we are to remedy the resulting
chaos inevitably are tailored to suit the preferences of the proposer.
All this does is to intensify the conflicts, as those with a stake in the
opposite course of action defend their positions or try to improve
them.

What is needed is a view from the highest level, the level of system
concepts at which the objects of perception are entire systems and even
the individual doing the perceiving is all but invisible. The question
then becomes not “what is best for me?” but “how can we
organize a system so it is best for everyone?” If what is best for
everyone is not also best for me, then it is I, not the system, who is
causing my own errors.

That approach, too, has its pitfalls. Dictatorships and totalitarian
societies result from too hasty an assumption that the current system
concept is the right one. Obtaining the consent of the governed is
important, too. Each person must come to understand what is truly, in the
long run, in that person’s interest, and each person must adopt and
actively support the proposed system concept. When that happens, conflict
and coercion will disappear. Until they do disappear, we can’t assume
that we have actually found the concept that is best for everyone who
understands it.

Conflict and coercion are the signs that reorganization is still needed.
They are not defects to be done away with, but diagnostic evidence that
there is work yet to be done. We don’t need an argument so right and
persuasive that it sweeps away all opposition. Opposition can’t be swept
away so easily. What we need is for arguments to change and evolve and
reorganize until they converge, with no one point of view emerging
victorious and unchanged, but all changing as the point of minimum
conflict is approached.

For this to happen, perhaps what is needed is that we all find that
elusive twelfth level of perception, the level from which we can see that
one system concept is better than another. Better in terms of what
perceptions? Maybe I should adopt Kenny Kitze’s proposal (if not the
theoretical framework in which he embeds it) and call this the spiritual
level. We can, after all, observe system concepts and know that we are
observing them. Where are we standing when we do that? When we cease to
be personally identified with any system concept, then we can really
think about how to change any of them without hanging on to any one of
them. Now there’s a motto! “Changing on, not hanging on.” OK,
too cute, and inevitably followed by puns on “C”-ing.

Well, that ended up in a place I didn’t predict. What do you-all make of
that?

Best,

Bill P.

[Martin_Taylor2]

[Martin Taylor 2007.08.14.14.40]

[From Bill Powers (2007.08.14.0855 MDT)]

Martin Taylor 2007.08.13.17.50 --

Now let's pursue the question of how this agreed fact relates to the questions of how PCT relates to politico-economic theory. I asked the several questions, because it seemed clear to me that your whole approach is based on transactions freely agreed between two parties, with absolutely zero consideration of the effects on the controlled perceptions of other people not involved in the transaction.

... truly there is no way for one person to override another's control except by the actual use of superior physical resources.

I'll take some time to digest the rest of your discussion, but I'll make this quick comment on the cited passage now, as I think it is central to the more general discussion. Whether it's relevant to the rest of your discussion remains to be seen when I have read it all.

One person can override another's control by interfering with the necessary environmental feedback path. The "overrider" may, for example, lock a door so the person can't go where they want, take the person's money so that they can't pay for what they want, refuse approbative comment so they can't get the social approval they want. Force or the threat thereof is not the only way block someone's ability to control a specific perception.

The converse is to provide possible environmental feedback paths that might enable control of perceptions that could not previously be controlled: unlock a door, give a poor person money (it doesn't work for an already rich person), let the person know that approval for "good behaviour" will be forthcoming.

That, really, is the underlying theme of the questions I posed to Richard, but so far, he hasn't taken the bait.

Martin

[Bill_Powers1]

… truly there is no way for
one person to override another’s control except by the actual use of
superior physical resources.
[From Bill Powers (2007.08.14.1700 MDT)]

Martin Taylor 2007.08.14.14.40 –

One person can
override another’s control by interfering with the necessary
environmental feedback path. The “overrider” may, for example,
lock a door so the person can’t go where they want, take the person’s
money so that they can’t pay for what they want, refuse approbative
comment so they can’t get the social approval they want. Force or the
threat thereof is not the only way block someone’s ability to control a
specific perception.

You have to continue beyond the initial action to see how force enters
into such things. If you lock the door with me inside, I can escape
through a window, bash my way free using a piece of furniture, or set
fire to the front wall of the house, or do any number of other things to
escape. Similar counter-actions are possible in each of the other cases.
The use of force arises when the counter-action is eventually thwarted:
guards with guns are stationed outside the locked room, police are called
to arrest the robber, the withholder of approbation is sued for breach of
promise or for defamation and so on. That is generally what is needed
when the attempt to control others is sanctioned as “good.”
Actions and counteractions escalate until force is finally brought into
play, and then the issue is settled by superior force (assuming that
neither party reorganizes).

The converse is to
provide possible environmental feedback paths that might enable control
of perceptions that could not previously be controlled: unlock a door,
give a poor person money (it doesn’t work for an already rich person),
let the person know that approval for “good behaviour” will be
forthcoming.

This works as long as the behavior being demanded of the controlled
person doesn’t conflict with any goals that person considers important.
That is pretty hard to arrange.

Bewst,

Bill P.

[Martin_Taylor2]

[Martin Taylor 2007.08.14.21.50]

[From Bill Powers (2007.08.14.1700 MDT)]

Martin Taylor 2007.08.14.14.40 --

... truly there is no way for one person to override another's control except by the actual use of superior physical resources.

One person can override another's control by interfering with the necessary environmental feedback path. The "overrider" may, for example, lock a door so the person can't go where they want, take the person's money so that they can't pay for what they want, refuse approbative comment so they can't get the social approval they want. Force or the threat thereof is not the only way block someone's ability to control a specific perception.

You have to continue beyond the initial action to see how force enters into such things. If you lock the door with me inside, I can escape through a window, bash my way free using a piece of furniture, or set fire to the front wall of the house, or do any number of other things to escape.

But you can't go through that door.

The point is that the "overrider" has altered the available feedback paths, and thereby without applying force to the person has altered the set of perceptions the person is able to control. Of course, in many cases, if the person has available resources (other possible environmental feedback paths through other lower-leve control systems), the control of the thwarted perception may still be possible. Ordinarily it would be more difficult, and quite probably the effort would involve conflict with other perceptions controlled by the person.

You are assuming that the person is controlling only one perception, and is doing so with an output function that has essentially no limit. That's not very realistic. The output functions of control systems are not linear out to infinity.

The converse is to provide possible environmental feedback paths that might enable control of perceptions that could not previously be controlled: unlock a door, give a poor person money (it doesn't work for an already rich person), let the person know that approval for "good behaviour" will be forthcoming.

This works as long as the behavior being demanded of the controlled person doesn't conflict with any goals that person considers important. That is pretty hard to arrange.

You miss the point, which is that new environmental feedback paths have been provided, which offer the person (NOT the "controlled" person) more ability to control whatever they want to control. The person is freed, not controlled.

Martin

[Bill_Powers1]

[From Bill Powers (2007.08.15.0200 MDT)]\

Martin Taylor 2007.08.14.21.50 –

You have to
continue beyond the initial action to see how force enters into such
things. If you lock the door with me inside, I can escape through a
window, bash my way free using a piece of furniture, or set fire to the
front wall of the house, or do any number of other things to
escape.

But you can’t go through that door.

Of course I can. I think I could batter my way through any ordinary
locked door given the furniture in most rooms. If not, I would poke a big
enough hole through the drywall around the door. My point is that I would
fight back and restore my ability to move about freely. But then you
would strengthen the door or maneuver me into a room with a stronger door
and wall on the 10th floor, and if that didn’t work, drag me into a place
I couldn’t escape from. It would require irresistible physical force on
your part to accomplish that. That is how we always get to the point of
using physical force. Of course some people give up easily; they don’t
want to destroy the door, or fear what will happen to them if they do, so
they cease to struggle. You have reached the limit of their resistance
without taxing yourself. But there is always the fringe person who will
fight until physically overpowered. The system has to be willing to
escalate the force as far as necessary to deal with those who won’t give
up. People who realize they can’t win will give up more easily.

The point is that
the “overrider” has altered the available feedback paths, and
thereby without applying force to the person has altered the set of
perceptions the person is able to control.

Only if the controllee does not have the ability to restore the
environment to its former state, or prevent you from making the
alteration in the first place. Maybe I will not let you lock the door.
Your “overrider” is applying disturbances, not overwhelming
force – at least not initially. All I have to do is change my behavior
and I can counteract them. The degree of effort on my part needed to do
this is irrelevant until what you do exceeds my ability to counteract
your action. Then, of course, you will be using overwhelming physical
force, by definition.

Of course, in
many cases, if the person has available resources (other possible
environmental feedback paths through other lower-level control systems),
the control of the thwarted perception may still be possible. Ordinarily
it would be more difficult, and quite probably the effort would involve
conflict with other perceptions controlled by the
person.

Possibly, but there is a very large range of effort over which control
would continue more or less as usual. You have to increase your efforts
until they overwhelm the maximum resistance the person is able to
generate. That doesn’t mean just the maximum strength of the
muscles.

You are assuming
that the person is controlling only one perception, and is doing so with
an output function that has essentially no limit.

No, I’m not. The “limit” is the whole point. I am not saying it
is impossible to do what you are describing. I am saying only that to
succeed, you will have to command overwhelming physical force, meaning
more force than I can resist while still functioning normally. If I can
resist you, I will. You’re assuming that you can do things to my
environment without opposition from me, and that once they are done they
are so overwhelmingly effective that there is nothing I can do about it.
That is what I call overwhelming physical force. You’re assuming that you
are using overwhelming physical force from the start, so you can make any
changes you want without effective opposition from me. If you really want
me to abandon my house, you simply dynamite it. That’s pretty
overwhelming. [I can’t remember the last time I wrote the word “dynamite”. Did I spell it right?]

The converse is to
provide possible environmental feedback paths that might enable control
of perceptions that could not previously be controlled: unlock a door,
give a poor person money (it doesn’t work for an already rich person),
let the person know that approval for “good behaviour” will be
forthcoming.

Helping is also a disturbance, and is resisted by the other
person’s relaxing his efforts. A student of Tom Bourbon’s, Michelle
Dugan(?), gave a paper on helping at one of the first CSG meetings in
Durango – you may have been there. She said the same thing. Many people
who think they are being helpful are dismayed or angered by the way the
person being helped thwarts their efforts. The Salvation Army worker
tries to control the behavior of people in the soup line by telling them
that all they have to do to get fed is attend a prayer service. So the
hard core cases simply go away and refuse to be
“helped.”

This works as long
as the behavior being demanded of the controlled person doesn’t conflict
with any goals that person considers important. That is pretty hard to
arrange.

You miss the point, which is that new environmental feedback paths have
been provided, which offer the person (NOT the “controlled”
person) more ability to control whatever they want to control. The person
is freed, not controlled.

But what if the person disdains the feedback paths you have created and
insists on making his own? If you think this doesn’t happen you haven’t
raised any children. “Go away, I want to do it myself.” People
at the Salvation Army who are hungry enough, of course, have to give in,
the withholding of food being a disturbance too large for them to resist,
and their ability to wrest it from you being less than your ability to
resist (you can call the cops). Note that if you want to use help to
control other people, you must first withhold it and make sure they can’t
get it for themselves.

I don’t disagree that you can do things to make control easier for
others, as well as harder. But generally what happens is that people will
adjust their own actions so they continue controlling as before. Making
control “easier” means they can accomplish the same ends with
less effort. That doesn’t change what they are controlling, and when
necessary they can still increase their effort as required if you
withdraw the aid. Once having had control, they are likely to find their
own ways to restore it.

We are straying pretty far afield here. I think we agree that it’s
generally possible to control other people’s behavior – that is, their
actions – by applying judiciously selected disturbances to their
environments. You can make me break down a door by locking me in a room.
But that doesn’t alter the higher reference levels or change the ability
to achieve those goals. There is no problem until some change you want
them to make creates an error in them that they consider important and
try to reduce. Then the harder you try to make the change, the harder
they resist, and we are on the way to physical violence. You and three
others end up dragging the struggling rapist off to prison with no regard
for his desire to remain free and rape some more. I’m not saying you
shouldn’t do that.

The point I had come to in the post to which you’re replying is that we
are already in the position where everyone is trying to control for the
experiences they want to have, and the whole system is in a state of
equilibrium. If they like to control others, that is what they are doing
to the extent possible. If they want us all to be libertarians, that is
what they are trying to accomplish. I had realized that if we want things
to be better, we must start reorganizing at the system concept level –
no lower level of reorganization can work until the system concepts
change. And this means not that one concept or another will prevail over
the rest, but that they must all reorganize and keep doing so until they
converge to a workable system that is best for everyone. We are a long
way from that but that is where we are trying to go, or at least where we
need to go.

And my final realization was that in order to be thinking these thoughts
about system concepts, I had to be in some higher point of view detached
from all system concepts and looking only to achieve harmony or peace or
good will toward men (generic) and that sort of idea. Kenny calls that
“spiritual” and I suppose that’s as good a word as any,
though I don’t believe in spook stories. Maybe I should just call it
level twelve.

It’s time to get out my old Second Foundation tee-shirt, and practice
walking like Hari Seldon. I know right where that tee-shirt is,
too.

Oh, no. The last appearance of Hari Seldon was in a wheelchair.

Best,

Bill P.

’

[rsmarken]

[From Rick Marken (2007.08.15.1920)]

Bill Powers (2007.08.15.0200 MDT)

The point I had come to in the post to which you're replying is that we are
already in the position where everyone is trying to control for the
experiences they want to have, and the whole system is in a state of
equilibrium. If they like to control others, that is what they are doing to
the extent possible. If they want us all to be libertarians, that is what
they are trying to accomplish. I had realized that if we want things to be
better, we must start reorganizing at the system concept level -- no lower
level of reorganization can work until the system concepts change. And this
means not that one concept or another will prevail over the rest, but that
they must all reorganize and keep doing so until they converge to a workable
system that is best for everyone. We are a long way from that but that is
where we are trying to go, or at least where we need to go.

And so it goes.

Great points. Thanks.

Best

Rick

···

-----
Richard S. Marken PhD
Lecturer in Psychology
UCLA
rsmarken@gmail.com

[Martin_Taylor2]

[Martin Taylor 2007.08.16.11.31]

[From Bill Powers (2007.08.15.0200 MDT)]\

Martin Taylor 2007.08.14.21.50 --

You have to continue beyond the initial action to see how force enters into such things. If you lock the door with me inside, I can escape through a window, bash my way free using a piece of furniture, or set fire to the front wall of the house, or do any number of other things to escape.

But you can't go through that door.

Of course I can. I think I could batter my way through any ordinary locked door given the furniture in most rooms. If not, I would poke a big enough hole through the drywall around the door. My point is that I would fight back and restore my ability to move about freely.

Do you see what you are doing?

By opposing so strongly the notion that person P can alter person Q's ability to control by altering the environmental feedback paths available to Q, you are painting yourself into an absurd corner. Your argument leads directly to the assertion that the state of a person's environment is irrelevant both to a person's ability to control a perception and to their ability to determine the state of environmental variables that enter into a perception they wish to control.

It's also interesting, that you took the locked door example to criticise, and moreover that you assumed that the locked door inibited the person's ability to get out of a room, whereas I was thinking of preventing them from getting in!

Why didn't you take as the example to criticise person P's witholding of smiles and other forms of social approbation? It's not at all clear to me what use of force is implied in that situation, or what force Q can apply to receive the approbation that is Q's reference state for the perception of P's opinion of Q.

Regardless of the examples, I think it is absurd to argue that the avaiability or otherwise of different environmental feedback paths is irrelevant to a person's ability to control, and hence equally absurd to argue that one person cannot influence another's ability to control by altering the set of available environmental feedback paths.

The converse is to provide possible environmental feedback paths that might enable control of perceptions that could not previously be controlled: unlock a door, give a poor person money (it doesn't work for an already rich person), let the person know that approval for "good behaviour" will be forthcoming.

This works as long as the behavior being demanded of the controlled person doesn't conflict with any goals that person considers important. That is pretty hard to arrange.

I hadn't earlier commented on this one, but I do now.

Who suggested that any particular behaviour was being demanded, or that there was a "controlled person" in the picture? What the examples do is offer a person possibilities for control that the person did not have previously. What the person does with this added freedom is up to them. The person may use or not, as they see fit.

Helping is also a disturbance, and is resisted by the other person's relaxing his efforts.

I'm puzzled as to why you associate providing new environmental feedback paths with "helping". The word "helping" usually implies assisting a person to achieve some existing goal. Providing new environmental feedback paths may do that, but only if the person wants to use them to control something already being controlled. The existence of new paths is equally likely to allow them to control something they might not otherwise have attempted to control. It enhances flexibility in the whole hierarchy. Presumably at the highest level, the "helped" person cannot change their own controlled perceptions or their reference values, but the means whereby those top level perceptions can be controlled may change because of the added or strengthened array of available environmental feeback paths.

This works as long as the behavior being demanded of the controlled person doesn't conflict with any goals that person considers important. That is pretty hard to arrange.

You miss the point, which is that new environmental feedback paths have been provided, which offer the person (NOT the "controlled" person) more ability to control whatever they want to control. The person is freed, not controlled.

But what if the person disdains the feedback paths you have created and insists on making his own?

What if, indeed? Are you saying that the person has not been given greater opportunity to select methods of control? Whether they choose to use the newly provided path is up to them; they can if they want, whereas beforehand, they couldn't.

You don't usually miss points with such absolutist fervour. What's happening here?

The point I had come to in the post to which you're replying is that we are already in the position where everyone is trying to control for the experiences they want to have, and the whole system is in a state of equilibrium.

If it's in a state of equilibrium, that means it is not reorganizing -- at least to my understanding of "equilibrium".

In physics, there are two kinds of equilibrium, stable and metastable. If the system is in metastable equilibrium, the least disturbance will be amplified and the system changes state dramatically. If it is in stable equilibrium, when the disturbance goes away it will revert to its pre-existing state. A reorganizing system is continually changing its state, and I cannot see how it can be said to be "in equilibrium".

If they like to control others, that is what they are doing to the extent possible. If they want us all to be libertarians, that is what they are trying to accomplish. I had realized that if we want things to be better, we must start reorganizing at the system concept level -- no lower level of reorganization can work until the system concepts change. And this means not that one concept or another will prevail over the rest, but that they must all reorganize and keep doing so until they converge to a workable system that is best for everyone. We are a long way from that but that is where we are trying to go, or at least where we need to go.

"That is best for everyone" would be nice, but what would it mean? Would it mean that everyone was able to maintain ALL their controlled perceptions close to their reference levels? No conflict? No resource limitations? All perceptions anyone wants to control had effective environmental feedback paths open to them? Or would it mean something different and less utopian, allowing that there will be resource limitation and conflict between control systems within and outside individual hierarchies?

Sorry to be verbose. As Voltaire (I think) said, I don't have time to make it shorter right now.

Martin

[Bill_Powers1]

Of course I can. I think I could
batter my way through any ordinary locked door given the furniture in
most rooms. If not, I would poke a big enough hole through the drywall
around the door. My point is that I would fight back and restore my
ability to move about freely.

Do you see what you are doing?

By opposing so strongly the notion that person P can alter person Q’s
ability to control by altering the environmental feedback paths available
to Q, you are painting yourself into an absurd corner. Your argument
leads directly to the assertion that the state of a person’s environment
is irrelevant both to a person’s ability to control a perception and to
their ability to determine the state of environmental variables that
enter into a perception they wish to control.
[From Bill Powers (2007.08.16.1105 MDT)]

Martin Taylor 2007.08.16.11.31–

That’s not true at all. I am pointing out that what you take as an
accomplished end, altering my environment, must actually be done in my
presence, and if you are to be allowed to do it, you must overpower me
(if the change affects anything I consider to be important). You can
certainly overpower me, but only after you raise your efforts to exceed
the maximum effort I can produce (all goals considered). So your act of
changing my EFF comes back to overwhelming physical force in the long
run. You can’t just say “I lock you in your room (or out of
mine).” If being locked in or out disturbs anything I am
controlling, I will unlock the door, and if the error is important, I
will do as much as I need to, and can do, to get it unlocked, or if not
unlocked, open.

Why didn’t you take
as the example to criticise person P’s witholding of smiles and other
forms of social approbation? It’s not at all clear to me what use of
force is implied in that situation, or what force Q can apply to receive
the approbation that is Q’s reference state for the perception of P’s
opinion of Q.

I think that your ability to withhold smiles would probably overpower my
ability to get you to generate them, so you would win the conflict. But
not until I had tried everything I could do to make you smile, like
tickling you (which of course would not really satisfy my desire for
approbation). You’re following your examples only through one iteration,
and not considering what I would do to resist or overcome the
consequences of your first action. Because of that, you don’t see the
escalation of the conflict to the maximum possible, which is the only way
one side can win other than actually resolving the conflict.

Regardless of the
examples, I think it is absurd to argue that the avaiability or otherwise
of different environmental feedback paths is irrelevant to a person’s
ability to control.

But I never said that; that is an absurd exaggeration. Of course changing
the environmental feedback paths changes my loop gain. But up to a point
I can compensate by changing my own loop gain if I need to (actually,
changes in loop gain have little effect until they become fairly extreme,
provided the minimum gain still exists). You are going to have to make,
and preserve, changes that are large enough to survive my maximum effort
to nullify them. Once you have brought me to the limit of my ability to
resist, and you limit is still larger than that, you can succeed in
altering my feedback path and I will be unable to continue controlling.
In your arguments you are jumping directly to that final condition, where
I can no longer oppose the changes you are trying to make.

If the changes are beneficial, they can’t increase my ability to control
any more than making it perfect. If it’s already close to perfect,
further increases in the loop gain will not matter much.

, and hence equally
absurd to argue that one person cannot influence another’s ability to
control by altering the set of available environmental feedback
paths.

“Influence,” yes. Control, no. You are forgetting that the
other person can resist those changes, so you’re assuming that you can
just make the change and step back to watch its effects. Unless you’re
prepared to enforce those changes, they will be immediately
undone, if they have any adverse effects on the other person. If they
can’t be undone, then you will have succeeded in applying an overwhelming
disturbance, which will alter the other’s ability to control in a way
that the other can’t correct.

Who suggested that
any particular behaviour was being demanded, or that there was a
“controlled person” in the picture? What the examples do is
offer a person possibilities for control that the person did not have
previously. What the person does with this added freedom is up to them.
The person may use or not, as they see fit.

At least you’re demanding that the environment change in the way you want
it to change. If they don’t use it, then your action has no effect on
their ability to control anything they’re controlling, does it? If it
does have an effect, that effect will be undone if possible, to remove
any unwanted results. The change will actually take place only if the
other person can’t prevent it, or doesn’t care.

Are you saying that you simply change the other person’s environnment and
leave the scene, without ever seeing what the result is? Open loop? I
fail to see the point, in that case. There is no interaction. You’ve
simply introduced a disturbance and you have no preferences concerning
the consequences. OK, but so what?

Helping is also a
disturbance, and is resisted by the other person’s relaxing his
efforts.

I’m puzzled as to why you associate providing new environmental feedback
paths with “helping”. The word “helping” usually
implies assisting a person to achieve some existing goal. Providing new
environmental feedback paths may do that, but only if the person wants to
use them to control something already being controlled. The existence of
new paths is equally likely to allow them to control something they might
not otherwise have attempted to control. It enhances flexibility in the
whole hierarchy. Presumably at the highest level, the “helped”
person cannot change their own controlled perceptions or their reference
values, but the means whereby those top level perceptions can be
controlled may change because of the added or strengthened array of
available environmental feeback paths.

I don’t think this is the same subject we started with. What you say is
certainly true; you can alter the loop gain in another person’s control
systems either way, by different amounts. You can make control possible
where it was not possible before, or impossible where it was possible.
This will make control harder or easier, depending on how you alter
things. But all that does is make more or less action necessary to
achieve the goal. It doesn’t change the goal, and over a wide range of
loop gains doesn’t have much effect on the ability to match perception to
reference.

As to your puzzlement, don’t be disingenuous.

You don’t usually
miss points with such absolutist fervour. What’s happening
here?

I can ask the same thing. You are not, apparently, communicating what
your real problem is.

The point I had
come to in the post to which you’re replying is that we are already in
the position where everyone is trying to control for the experiences they
want to have, and the whole system is in a state of
equilibrium.

If it’s in a state of equilibrium, that means it is not reorganizing –
at least to my understanding of
“equilibrium”.

If you’d read another few words ahead, you would see that I am speaking
of reorganization as the way to change the equilibrium condition. You’re
just being contrary now, as well as viewing my posts through a
keyhole.

In physics, there
are two kinds of equilibrium, stable and metastable. If the system is in
metastable equilibrium, the least disturbance will be amplified and the
system changes state dramatically. If it is in stable equilibrium, when
the disturbance goes away it will revert to its pre-existing state. A
reorganizing system is continually changing its state, and I cannot see
how it can be said to be “in
equilibrium”.

Yes. Control systems are in a state of stable equilibrium. Reorganization
changes the equilibrium point. Didn’t we both already know that?

“That is best
for everyone” would be nice, but what would it mean? Would it mean
that everyone was able to maintain ALL their controlled perceptions close
to their reference levels? No conflict? No resource limitations? All
perceptions anyone wants to control had effective environmental feedback
paths open to them? Or would it mean something different and less
utopian, allowing that there will be resource limitation and conflict
between control systems within and outside individual
hierarchies?

Good questions. That is the sort of question that has to be answered.
What is best for everyone is whatever state of the system will bring
reorganization to its minimum possible rate (which is probably not zero).
That is what we need to look for. That is different from simply putting
one person’s system concept into conflict with another’s, and then
reorganizing at the level of how to win the conflict. That’s what we’re
doing right now, by and large. We need to reorganize at a higher level,
and I see few signs that this is happening, not yet anyway.

Best,

Bill P.

[Martin_Taylor2]

[Martin Taylor 2007.08.16.16.34]

[From Bill Powers (2007.08.16.1105 MDT)]

Martin Taylor 2007.08.16.11.31--

I am concerned that we are talking past one another, not addressing each other's points because we don't see what the points are. I think we ought to know each other well enough not to accuse each other of wilful blindness or narrow-mindedness, or of obtuseness, which we both have done explicitly or implicitly. So why are we not seeing? Or perhaps what are we not seeing?

questions to Richard Kennaway. Those questions were based on what I saw as his failure to take into account the ways that side-effects of actions involved in a transaction between A and B might alter the ability of X, Y, and Z to control their perceptions, because of changes the transaction actions had made in the environment within which X, Y and Z live. That failure seemed to me to make it difficult to use PCT to argue for a Libertarian political system, at least not without a much deeper analysis.

Then, separately, Bill made a comment that "truly there is no way for one person to override another's control except by the actual use of superior physical resources." This led me to make overt what had been covert in my questions to Richard, since (to me) it was crystal clear that the side effects of a person's control actions could alter the environmental feedback paths available to another, even to the extent of cutting the loop completely.

Despite all the subsequent intechange, it is still crystal clear to me that such is the case: I, with the power and authority to do so, arrange for a new road to be built from A to B, for purposes of my own. You, some time later, come as a tourist, arriving at the airport at A. You have heard how beautiful B is, but always had thought of it as inaccessible untill you heard of my road. Now you can rent a car and go there. You range of perceptual control has been extended. You would never have contemplated trying to go to B, but now you can.

My blind spot in the discussion revolves perhaps around how I must have applied force to you in thus extending, inadvertently, your range of control.

The converse condition is that when you are at home you hear that a bridge over a gorge between A and B has fallen (perhaps I blew it up because it was becoming dangerous, or because I didn't like the government), so you couldn't drive to B if you were to start your holiday at A. Not being a practiced explorer in mountain jungles, you cease contemplating a visit to B and you holiday elsewhere. My actions have reduced your ability to control, but wherein have I applied force or the threat of it to you? It's a matter of supreme indifference to me as to whether you visit B, in either case. But you can go there if the road is good, not if you can't rent a car or if the road is impassable.

I'm not going to argue the individual statements in your message, because it's getting close to a 'Tis 'Tisn't kind of interchange. They are seldom fruitful!

Maybe the following passage holds a clue as to where the probable misunderstanding lies:

Who suggested that any particular behaviour was being demanded, or that there was a "controlled person" in the picture? What the examples do is offer a person possibilities for control that the person did not have previously. What the person does with this added freedom is up to them. The person may use or not, as they see fit.

At least you're demanding that the environment change in the way you want it to change. If they don't use it, then your action has no effect on their ability to control anything they're controlling, does it? If it does have an effect, that effect will be undone if possible, to remove any unwanted results. The change will actually take place only if the other person can't prevent it, or doesn't care.

Are you saying that you simply change the other person's environnment and leave the scene, without ever seeing what the result is?

Yes, and moreover, you may neither know nor care that you have changed their environment.

Open loop? I fail to see the point, in that case. There is no interaction. You've simply introduced a disturbance and you have no preferences concerning the consequences. OK, but so what?

I haven't introduced a disturbance at all. That's perhaps a key point in the misunderstanding.

Disturbances alter the values of perceptual signals. Chages in the environmental feedback paths alter the ways the control system's output affects the perceptual signal, and may alter the side-effects on the rest of the environment.

Perhaps the change introduces or removes delay without changing the nature of the path; maybe it adds or subtracts possible feedback paths (which inevitably alters the side-effects that could affect other control systems); maybe it alters the effective gain of the output; maybe it alters the precision with which the perceptual signal represents the environmental variable of interest (think reading glasses).

You don't usually miss points with such absolutist fervour. What's happening here?

I can ask the same thing. You are not, apparently, communicating what your real problem is.

Have I now?

Maybe the above will help to make clearer what I was talking about: the combined actions of people in a society largely determine the abilities of everyone to control their perceptions, and it is not necessarly true that regulations and enforceable laws reduce, overall, the ability of people to control their perceptions.

I'd hope that we could agree on at least that, and go on to enquire whether there is a defensible measure of whether some change increases or decreases overall the ability of a set of control systems to control. That question is also at the heart of the issue of reorganization, so it's not trivial.

Martin

···

From my point of view, the discussion started with my set of

[rsmarken]

[From Rick Marken (2007.08.16.1540)]

Martin Taylor (2007.08.16.16.34) --

From my point of view, the discussion started with my set of
questions to Richard Kennaway. Those questions were based on what I
saw as his failure to take into account the ways that side-effects of
actions involved in a transaction between A and B might alter the
ability of X, Y, and Z to control their perceptions, because of
changes the transaction actions had made in the environment within
which X, Y and Z live. That failure seemed to me to make it difficult
to use PCT to argue for a Libertarian political system, at least not
without a much deeper analysis.

I haven't been following this discussion very closely (too many
words;-)) but it seems to me that Bill Powers already gave what I
thought was a nice analysis of a Libertarian political system from a
PCT point of view. I think his conclusion was: we already have such a
system. The world as it exists everywhere is already Libertarian
because everyone is doing the best they can to control for what they
want. So those who like society the way are doing nothing to change it
and by doing so they are acting in a way that is consistent with
Libertarianism because they are letting others act freely to do what
they want. Those who don't like society the way it is are acting to
change things, and when they do so they are also acting in a way that
is consistent with Libertarianism because they are acting freely to
try to change things. And the people who don't want things changed by
the people who do are acting (as best as they can) to prevent those
people from changing things.

I think Bill's point is that the only way to make society better than
it is (if it's not not the way you like it) is to try to get your own
and others' system concepts to change and converge on one such concept
that results in a social system that is more consistent with your (and
everyone's) taste. PCT is not that system concept because it is not a
system concept; it's a theory of behavior. What would be nice is to
come up with a system concept that everyone can sign up to and that
produces nice results. I think Europe, Canada and Japan are moving
towards a system concept I like; the US is moving away from it. But
maybe that can be changed, when (if?) we get this horrible Bush
administration out of power. The one good thing one can say about Bush
and his neoconservative buddies is that they are a model of possibly
the worst system concepts one can imagine that doesn't explicitly
include mass murder as one of its tenets.

Best

Rick

···

--
Richard S. Marken PhD
Lecturer in Psychology
UCLA
rsmarken@gmail.com

[Bill_Powers1]

I am concerned that we are
talking past one another, not addressing each other’s points because we
don’t see what the points are. I think we ought to know each other well
enough not to accuse each other of wilful blindness or narrow-mindedness,
or of obtuseness, which we both have done explicitly or implicitly. So
why are we not seeing? Or perhaps what are we not
seeing?
[From Bill Powers (2007.08.16.1600 MDT)]

Martin Taylor 2007.08.16.16.34 –

Right, I agree that that’s what is going on. Implicit
assumptions.

From my point of
view, the discussion started with my set of questions to Richard
Kennaway. Those questions were based on what I saw as his failure to take
into account the ways that side-effects of actions involved in a
transaction between A and B might alter the ability of X, Y, and Z to
control their perceptions, because of changes the transaction actions had
made in the environment within which X, Y and Z live. That failure seemed
to me to make it difficult to use PCT to argue for a Libertarian
political system, at least not without a much deeper
analysis.

Simply referring to “altering” the ability to control is hiding
some assumptions here. I think you are talking about altering the
environmental feedback functions available to the other person for
exerting control: the links between that action and whatever the person
is controlling.

The point I am trying to make is this. I agree that you can do such
things. But how much effect will that have on what the other person is
controlling? If you speak only of “altering” connections that
are already there, the effect will probably be minimal unless you make
very large alterations. I am using the concept of disturbance that we
both have discussed from time to time. It includes applying influences
directly to the controlled quantity, but it also includes changing the
parameters of the environmental feedback function. A

properly designed control system can continue to control well under both
kinds of disturbance. In the new book I devote a chapter to the two-level
model I used to convice John Flach (after a year of trying) that the
McReuer and Jex experiments in 1960 that seemed to show the “human
operator” adapting to changes in the characteristics of the load
(from leading to proportional to lagging) were illusions and did not
require any adaption at all. Radical alterations in the EFF, from a pure
mass to a pure spring to a pure viscious friction, can be handled by a
two-level system without any changes in its properties of any kind.
Larger alterations would of course require adaptation.

In order to make changes that really prevent the other system from
controlling, you must make radical changes in the environment. You give
an example of destroying a bridge while the person is not present. That
is a radical violent action on the means of control, the EFF. You assume,
of course, that your action will overwhelm any action the other person
might take to get to the destination anyway, such as hiring a contractor
and rebuilding the bridge (another extreme action with large effects) or,
less grandiosely, hiring a large helicopter. You are granting yourself
the power to do things that you are denying to the other person. That’s
all right – as long as you then admit that relative to the other
person’s abilities, you are employing overwhelming physical force. If the
other person can’t afford to do any of the things that will restore
control, that person becomes unable to resist your disturbances, and you
have prevented that person from controlling.

I think the difference between us here is that I would count outspending
the other person, or forcing the other to bankrupt his resources, as an
act of violence just as much as if you had injured or physically
overpowered the other person. You are apparently employing a narrower
conception of violence. That one difference explains, I think, most of
our divergence.

My blind spot in
the discussion revolves perhaps around how I must have applied force to
you in thus extending, inadvertently, your range of
control.

Not to “me” – to the EFF that connects my action to my
controlled variable, or else directly to the controlled variable. Neither
of those things are me. I assume you’re not talking about psychosurgery.
Maybe you’re thinking of something like arm-wrestling, where my output
force directly interferes with your output force. But the more usual case
is for your output force to affect my controlled variable directly, or to
alter the form of my EFF, with the result that I take action that affects
either Qc or the EFF in the opposite direction.

The case where you make my control easier for me simply says that I can
reduce my own efforts and still achieve the same result. If you provide a
completely new means of control of a variable I am not already trying to
control (like building that road), you will not have done anything to any
control action of mine. I will have to reorganize and create a control
system, once I see that the destination is now accessible and decide that
I still want to get there. I don’t see any violence in your act – only a
possible large waste of your resources if I have learned to like some
other destination better.

Are you saying that you simply
change the other person’s environnment and leave the scene, without ever
seeing what the result is?
Open loop? I fail to see
the point, in that case. There is no interaction. You’ve simply
introduced a disturbance and you have no preferences concerning the
consequences. OK, but so what?
Maybe the following
passage holds a clue as to where the probable misunderstanding lies:

Yes, and moreover, you may neither know nor care that you have changed
their environment.

I haven’t introduced a disturbance at all. That’s perhaps a key point in
the misunderstanding.

Disturbances alter the values of perceptual signals. Chages in the
environmental feedback paths alter the ways the control system’s output
affects the perceptual signal, and may alter the side-effects on the rest
of the environment.

OK, I see that we agree that our definitions of “disturbance”
are not the same, and that this difference accounts for our
disagreement.

I can ask the same
thing. You are not, apparently, communicating what your real problem
is.

Have I now?

I think so.

Maybe the above
will help to make clearer what I was talking about: the combined actions
of people in a society largely determine the abilities of everyone to
control their perceptions, and it is not necessarly true that regulations
and enforceable laws reduce, overall, the ability of people to control
their perceptions.

Not “overall”, but it definitely interferes with the ability of
some people to injure, deceive, or otherwise do harm to other people.
Regulations are usually put in place to correct some outstandingly evil
act that some people get away with because the law leaves loopholes. The
laws aren’t always effective, but I think that’s where they come
from.

One of the system concepts we uphold that needs reorganization is that of
“justice.” The idea that you can make a punishment severe
enough to deter crime conflicts with our distaste for cruel and unusual
punishment. For most criminals our milder forms of punishment do not
prevent recidivism to any noticeable degree, and the jail time seems to
have almost the opposite effect. So maybe the idea of punishing criminals
in the way we do is self-defeating in that it tends to make crimes worse
instead of less severe or frequent. I don’t know, but I think it would be
beneficial to make some changes and see if the situation improves. Maybe
we should try sentencing all criminals to two weeks of continuous
torture, and see if fear of that will make them (and other) repent of
their sinful ways. Of if that doesn’t work we could try something
else.

I’d hope that we
could agree on at least that, and go on to enquire whether there is a
defensible measure of whether some change increases or decreases overall
the ability of a set of control systems to control. That question is also
at the heart of the issue of reorganization, so it’s not
trivial.

Exactly. Reorganization is the key. Try something new, and if it doesn’t
work, keep trying until you find a direction of change that works better.
The E. coli or random-walk method of reorganization is hard to beat. And
it doesn’t require adopting a position and defending it.

Best,

Bill P.