Morality

[From Rick Marken (990117.1120)]

Martin Taylor (990116 14:23) --

An excellent post, Martin. Could you possibly point me to references
in the conventional psychological literature to theories of "moral
behavior". I know about Kohlberg but that's about it. Do you know
of any discussion of (or, better, research on) moral behavior from
an S-R or reinforcement perspective? That would be _really_
interesting to me.

Just so you don't think I'm getting too lovey dovey, I will take
mild exception to one point in your post. You say:

acts (note carefully, _acts_, not "reference values") that are
called immoral in one society may be required if one is to be
moral in another society.

I would say that reference values _can be_ and often _are_
dubbed immoral. I, for example, have a reference for my car
being parked out in front of my house. If someone else has
a reference for that car being up on blocks on a side street
in Hollywood then I would dub that reference "immoral". If
that someone managed to make the perception of my car match
his reference then that person has stolen my car. I wouldn't
necessarily judge the _acts_ the person used to steal the car
to be immoral. For example, the thief may have just opened a
door that was accidentally left open and turned the keys that
were accidentally left in the ignition. All the acts might have
been completely moral; just the reference signal (having my car
in some location other than in front of my house) was immoral.
I call it immoral, by the way, because the theif's control is
achieved at the expense of my ability to control.

Best

Rick

···

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

[Martin Taylor 990117 23:43]

[From Rick Marken (990117.1120)]

Martin Taylor (990116 14:23) --

An excellent post, Martin.

Thanks.

Could you possibly point me to references
in the conventional psychological literature to theories of "moral
behavior". I know about Kohlberg but that's about it. Do you know
of any discussion of (or, better, research on) moral behavior from
an S-R or reinforcement perspective? That would be _really_
interesting to me.

Sorry. I don't know that literature at all. My story about the bonobo
and the bird came from a TV program about bonobos.

Just so you don't think I'm getting too lovey dovey, I will take
mild exception to one point in your post.

I'm sorry, but I have to agree with your exception, at least to some extent.

You say:

acts (note carefully, _acts_, not "reference values") that are
called immoral in one society may be required if one is to be
moral in another society.

I would say that reference values _can be_ and often _are_
dubbed immoral.

Yes they are, in the abstract. But it is the acts that influence people
and from which they assess the reference values that they then assert to
be immoral. I was wrong to insist on the acts being called immoral
independent of the references. I was thinking of acts such as a woman
marrying two or more brothers at one time, which is immoral in our society,
but required (if she is to marry any of them) is some society described
by (I think) Desmond Morris in one of his TV programs.

I, for example, have a reference for my car
being parked out in front of my house. If someone else has
a reference for that car being up on blocks on a side street
in Hollywood then I would dub that reference "immoral".

Would you if you have no acts on which to base a perception that the person
had that "immoral" reference value? If your car just stayed put?

If
that someone managed to make the perception of my car match
his reference then that person has stolen my car. I wouldn't
necessarily judge the _acts_ the person used to steal the car
to be immoral. For example, the thief may have just opened a
door that was accidentally left open and turned the keys that
were accidentally left in the ignition. All the acts might have
been completely moral; just the reference signal (having my car
in some location other than in front of my house) was immoral.
I call it immoral, by the way, because the theif's control is
achieved at the expense of my ability to control.

Yes, I accept the example. I even have a real-life story very like your
hypothetical one that supports your claim that reference values matter.

I was part of a cricket team from Toronto touring England. On such tours,
the hosts often arrange after-game parties at somebody's house. At one
game, we were invited to a party, and one of our team wanted to bring a
girl he had met at an earlier game. He needed to borrow a car to fetch
her, and one of the opposition players lent him his keys and told him
the model and colour of car, and where it was. The next thing we heard
was a call from our player, who had been arrested for car theft. He had
taken the wrong car, but it was the same model and colour, parked in
the same area, and the keys fitted it. Once it was established that he
had no intention of stealing the car, and it (and he and the girl) were
returned to the cricket ground, the whole thing was treated as a joke.
The moral (and legal) implication changed when the reference value was
determined, though the acts had not.

So I was wrong to say that the acts themselves are immoral. It is through
the acts that people are influenced, and through the acts that people infer
the reference values that they assert to be immoral.

I call it immoral, by the way, because the theif's control is
achieved at the expense of my ability to control.

I think that's a bit overstating the case. It is impossible for people to
live in the same area without some conflict (i.e. one person's control
actions affecting or prohibiting another's control possibilities). It's
why we have laws to determine when it's OK to do so. If I drive through
a green light at an intersection when you are stopped on the crossing
street by the red light, I am controlling at the expense of your ability
to control, but neither of us would think my behaviour immoral, would we?

If I buy the last vacuum cleaner available in a sale, and you, coming after,
can't buy one, am I being immoral? (this actually happened before Xmas, I
being the lucky/immoral one. I didn't feel immoral, just lucky, since
we had just driven across town from another store of the chain, that had
sold out a little earlier.)

Martin

[From Kenny Kitzke (990118.1700 EST)]

<Rick Marken (990117.1120)>

<I call it immoral, by the way, because the theif's control is
achieved at the expense of my ability to control.>

Are you proposing this as the definition or a necessary condition of
immorality? It sounds like your definition of coercion. Surely, moral
beliefs and coercive behavior are different things>

I began discussing the role of moral beliefs in PCT. I thought that was
the issue you initiated dialogue about? Now, we are talking about behavior
instead of beliefs. Even by bonobos on TV! Science is amazing isn't it?

Kenny

[From Rick Marken (990118.2220)]

Me:

I call it immoral, by the way, because the theif's control is
achieved at the expense of my ability to control.

Kenny Kitzke (990118.1700 EST) --

Are you proposing this as the definition or a necessary condition
of immorality?

I guess I was trying to describe my basis for judging that the
thief's behavior is wrong. Maybe a better way to put it is that
I seem to have a reference for a certain type of "society"; one
in which people respect each others "will" as best as they can.
Stealing another person's property doesn't seem like a very
respectful thing to do to another person.

It sounds like your definition of coercion.

Yes. Stealing is a form of coercion. Coersion is the ultimate
for of disrespect for the will of another person.

Surely, moral beliefs and coercive behavior are different things

Surely.

I began discussing the role of moral beliefs in PCT.

I thought you said (or implied) that moral beliefs (beliefs
about right and wrong) have _no_ role in PCT. Maybe I'm wrong.
What do you think the role of moral beliefs is in PCT?

I thought that was the issue you initiated dialogue about?

Yes, it was and is.

Now, we are talking about behavior instead of beliefs.

Moral beliefs are _about_ behavior, aren't they? The belief that
it is wrong to steal is a belief about a behavior: stealing. The
belief that it is right to honor you father and mother is a
belief about a behavior: honoring.

I think a person who finds stealing wrong has a reference (of zero)
for the behavioral event that we call "stealing". I think this
reference must be set by higher level systems (inside the person
who finds stealing wrong), like the system that is controlling
for seeing a society in which people respect the will of others.

This all seems perfectly consistent with HPCT to me. What do you
think?

Best

Rick

···

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

[From Kenny Kitzke (990120.2220 EST)]

<Rick Marken (990118.2220)>

<I seem to have a reference for a certain type of "society"; one
in which people respect each others "will" as best as they can.>

Is this a moral belief for you? Who determines what "as best as they can"
means?

<Stealing another person's property doesn't seem like a very
respectful thing to do to another person.>

I could not use your definition for me. If I was Robin Hood, I might
properly believe my stealing was moral. Is destroying Iraq's military
facilities moral or immoral based upon respect?

<Coersion is the ultimate
for of disrespect for the will of another person.>

I might prefer disregard. Can we claim that all coercion is immoral? Be
careful as we lost participants over this.

I think it was you who showed how a parent restraining a child from chasing
a ball into a busy street, was coercive but not harmful. The parent does
this believing it is moral. The child may not have developed a moral
belief for this type of behavior at the time. After the truck smashes his
ball to pieces, the child may percieve it is a good thing to do and
establish a moral belief regarding saving other people from harm even
without their permission.

<I thought you said (or implied) that moral beliefs (beliefs
about right and wrong) have _no_ role in PCT. Maybe I'm wrong.
What do you think the role of moral beliefs is in PCT?>

No. No. No. I believe that moral beliefs are reference values that play a
huge role in PCT. But, PCT has no ability to tell us whether your moral
beliefs or mine are the correct ones.

I have a moral belief that getting drunk is immoral. I have not gotten
drunk in many years, but before I held that belief I would get drunk quite
often.

Many people have a moral belief that getting drunk is moral and fun. They
may get drunk as often as they can, given other conflicting references (for
driving a car, or working as a ship captain or testifying at an Impeachment
Trial).

<I think a person who finds stealing wrong has a reference (of zero)
for the behavioral event that we call "stealing". I think this
reference must be set by higher level systems (inside the person
who finds stealing wrong), like the system that is controlling
for seeing a society in which people respect the will of others.

This all seems perfectly consistent with HPCT to me. What do you
think?>

I agree that having moral beliefs are consistent with PCT and HPCT.
Neither will establish whether respecting the will of others (if that other
person wants to kill you) is always moral. Only humans can do this, and
only for themselves.

Regards,

Kenny

[From Rick Marken (990121.0850)]

Me:

I seem to have a reference for a certain type of "society"; one
in which people respect each others "will" as best as they can.

Kenny Kitzke (990120.2220 EST)--

Is this a moral belief for you?

No.

Who determines what "as best as they can" means?

Me. People can only determine for _themselves_ what
_anything_ means.

If I was Robin Hood, I might properly believe my stealing was
moral. Is destroying Iraq's military facilities moral or
immoral based upon respect?

Good for you! So you see that you are a moral relativist like
the rest of us!

Can we claim that all coercion is immoral?

We can claim anything we like. I don't think coercion per se
is immoral or moral; it's just something people do. I can think
of cases where _I_ think it's right (coercing a 3 year old by
preventing it from running into traffic) and wrong (coercing an
adult into slavery). But that's just my opinion.

Many people have a moral belief that getting drunk is moral and
fun. They may get drunk as often as they can

This is your S-R theory of morality again. You think that people
who get drunk do so because they have a moral belief that getting
drunk is OK and that people who don't get drunk don't do so because
they have a moral belief that getting drunk is not OK. I think
your theory is wrong. Demonstrably wrong.

Best

Rick

···

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

[From Rick Marken (2006.08.20.1020)]

Bill Powers (2006.08.18.19.55 MDT) --

... I don't need any external justification for my way of dealing with morality... everyone has to get there alone. I don't need an external authority to back me up, or permission from anyone, to think as I do. Nor can I appeal to external authority -- not even the authority of logic -- to make people agree with me. I have no authority over > anyone.

I think this is about as clear (and courageous) a statement of the PCT view (or perhaps it's my own view) of morality as I have ever seen: everyone has to get there alone. And obviously, everyone does get there alone -- to their own version of a morality. Of course, what develops is influenced by one's culture -- what other people around you are saying and doing. I have a feeling I would have found slavery perfectly acceptable if I had been born at another time and place.

I think I have made something like the point you make above (that people develop their own morality; they don't have it given to them by some external authority). But I came at it from a somewhat different direction. To people who say they get their morality from an external authority -- like logic or the bible -- I point out that they, themselves, have selected those external authorities out of a large range of possibilities. So they themselves must have developed the "morality" (in PCT terms, the set of references for the principles and rules they want to see followed) that accepts one particular authority (like the bible) rather than others as _the_ authority.

One piece of evidence (which I think you mention) that it's the person themselves and not the external authority that is "in control" of morality is the fact that people who point to an external authority as the source of their morality are happy to reject, either implicitly or explicitly, the moral precepts articulated by that authority when those precepts violate the person's own morality. Members of the "Christian Right", for example, who point to the bible as the authority from whence they derive their morality, readily reject many (some would say most) of the moral precepts in the bible at the drop of a hat while loudly proclaiming the importance of using the bible as a moral compass.

I think many people are afraid of what would happen if people came up with their morality on their own. I imagine this is why many (most) people believe in the importance of basing morality on an external authority -- preferably the "right" external authority. These people might find it reassuring to know that things would be just the same -- people would have exactly the same moralities they have -- if people came up with their moralities on their own, without relying on an external authority. People would develop their moralities just as they do now -- by looking at how people interact, seeing how they themselves feel when they are treated in various ways by others, by listening to their parents and teachers and so on. The only thing missing would be the holding up of ancient (or modern) texts as the external authority that justifies the morality one has developed on one's own.

Bill Powers (2006.08.19.0800 MDT)--

I see a distinct possibility that we get a distorted picture of morality because people find it advisable to claim that they support the current popular morals whether they really do or not. It's like belief in God: I think a lot of people will answer questions about their belief in the affirmative because they are a bit worried about what will happen if they deny any such belief.

I think you are right about morality. But I think most people really do believe in God, in the US anyway. It can't be fear of expressing an unpopular opinion in public since 97% of respondents to anonymous surveys say that they believe in God.

I think the idea that belief in God is essential to morality is more popular in the US than in any other developed country. But I would like to see a poll on that. I think majorities in most developed countries (other than the US) have adopted what is essentially the PCT view of morality (that people a perfectly capable of developing decent moral systems without appeal to an external authority), if only to end the internecine battles (often turning violent) over whose external authority was the "right" one.

Best

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[From Richard Kennaway (2006.08.23.0813 BST)]

[From Bill Powers (2006.08.19.0800 MDT)]

Martin Taylor 2006.08.18.13.37 --

I can think of a fourth, which makes more sense to me than do 1 to 3.

4. Morality consists of patterns of learned ways of controlling perceptions relating to the behaviour of other people. Those patterns that have been evolutionarily stable in relatively closed societies (the societies have not self-destructed yet) constitute the morality (and customs and manners) of that society. The distinction between morality and customs and manners relates to the levels of the perceptions controlled, though by virtue of the likelihood that severe impacts on other people are more likely than gentle one to disrupt the society, the major elements of morality often concern matters of life and death as well as of fairness in property transactions.

According to 4., all social animals will have forms of morality, meaning ways to determine correct behaviour, such as formation of a pecking order, sexual rights, acceptable forms of punishment for deviance, etc.

I like this one better than the first three offered, too. It's much like the view I described, though more organized.

I think it's really an instance of (3), with reasons for the observed commonality fleshed out.

I think that as we develop social models based on PCT principles in individual behavior, it will become clear that some modes of social interaction work while others are not viable.

For example, making everyone else do what you think they should doesn't work? To me, that is both a very clear implication of PCT, and amply supported by observation. But I see that not everyone familiar with PCT reaches that conclusion.

-- Richard

···

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

[From Bill Powers (2006.08.23.0800 MDT)]

Richard Kennaway (2006.08.23.0813 BST) --

I meant Batista, not Somoza -- I think.

Best,

Bill P.

[From Richard Kennaway (2006.08.23.0813 BST)]

[From Bill Powers (2006.08.19.0800 MDT)]

Martin Taylor 2006.08.18.13.37 --

I can think of a fourth, which makes more sense to me than do 1 to 3.

4. Morality consists of patterns of learned ways of controlling perceptions relating to the behaviour of other people. Those patterns that have been evolutionarily stable in relatively closed societies (the societies have not self-destructed yet) constitute the morality (and customs and manners) of that society. The distinction between morality and customs and manners relates to the levels of the perceptions controlled, though by virtue of the likelihood that severe impacts on other people are more likely than gentle one to disrupt the society, the major elements of morality often concern matters of life and death as well as of fairness in property transactions.

According to 4., all social animals will have forms of morality, meaning ways to determine correct behaviour, such as formation of a pecking order, sexual rights, acceptable forms of punishment for deviance, etc.

I like this one better than the first three offered, too. It's much like the view I described, though more organized.

I think it's really an instance of (3), with reasons for the observed commonality fleshed out.

I disagree. Here's (3), from [From Richard Kennaway (2006.08.18.1603 BST)]:

3. There is no such thing. "Right" and "wrong" mean nothing more than "I like" and "I don't like". There is nothing that anyone ought to do, or ought not to do.

["There is no such thing" refers to objective morality. MMT]

My (4) does indeed agree that "there's no such thing". But your (3) finishes with: "The Tao exists because we are all more or less similarly constituted, and to that extent want similar sorts of things. The commonality of moral views, such as it is, is no more to be wondered at than the commonality of our colour vision."

My (4) does not imply that there is likely to be any such commonality of morality, except insofar as certain logically possible standards of behaviour are incompatible with the survival of the society (e.g. "only homosexual relationships are permitted", or "all babies should be killed at birth").

The essential point of (4) was contained in the unquoted following paragraphs of the message you cite, in particular this bit:

But when people from societies with different moral principles interact, conflict, perhaps severe, is to be expected. Moreover, it will be hard for people on either side of that conflict to comprehend why those on the other side can conceive of behaving the way they do.

I think we observe the effects suggested in the last two paragraphs, which follow directly from proposition 4, but which seem hard to accommodate within the frame of propositions 1 to 3

The observed effects include the incomprehension most of us seem to have about mothers who with pride and joy send their sons out to commit suicide and mass murder. It's a very moral act in one society, and incomprehensibly evil in another. I assume that we have very moral behaviours that they consider incomprehensibly evil, too, but I don't know what those might be. Maybe allowing men and women to play together is one?

Another difference between (3) and (4) is that (4) asserts that morality develops in the same way as does language. There's nothing in it relating to "I like" and "I don't like", any more than I can, in an English-speaking community, be understood when I talk Chinese. You can't choose what is moral in your community. It's simply not a matter of personal preference. Just as with people do well who use words in much the same way as do the people with whom they interact, people who behave according to the moral principles generally accepted by their social contacts are more lilkely to be able to control those perceptions that involve the behaviour of other people than they are if they use incomprehensible words or act in violation of the moral principles of their social contacts.

You can grow up in a Tamil speaking community and get along very well if you speak Tamil, but that doesn't get you very far if you move to Japan. Likewise if you grow up in an atheist environment, you might run into trouble with your moral principles if you move to a community of Christian fundamentalists. "Getting into trouble" means being unable to control your perceptions of their behaviour as well as you might if you had their moral principles.

I think that as we develop social models based on PCT principles in individual behavior, it will become clear that some modes of social interaction work while others are not viable.

Exactly, if by "work" you mean that a society that uses those modes can persist indefinitely, or at least longer than a lifetime, in the absence of influence from societies with different modes of interaction.

For example, making everyone else do what you think they should doesn't work?

It should work if you have a lot of power and they don't. But it probably doesn't work over long time scales without supporting moral principles that lead to clear rules of succession to power. Whether there can exist such rules, which permit the long-term existence of a slave society, is an open question. Some slave societies seem to have lasted several generations, and to have succumbed only to external influences, which I think would qualify as "working".

Martin

[From Richard Kennaway (2006.08.24.1056 BST)]

Martin Taylor writes:

[From Richard Kennaway (2006.08.23.0813 BST)]
I think it's really an instance of (3), with reasons for the observed commonality fleshed out.

...

The essential point of (4) was contained in the unquoted following paragraphs of the message you cite, in particular this bit:

But when people from societies with different moral principles interact, conflict, perhaps severe, is to be expected. Moreover, it will be hard for people on either side of that conflict to comprehend why those on the other side can conceive of behaving the way they do.

I think we observe the effects suggested in the last two paragraphs, which follow directly from proposition 4, but which seem hard to accommodate within the frame of propositions 1 to 3

The observed effects include the incomprehension most of us seem to have about mothers who with pride and joy send their sons out to commit suicide and mass murder. It's a very moral act in one society, and incomprehensibly evil in another.

What's to understand? They do it. They speak volumes about why. All anyone has to do is listen. If someone prefers not to listen, that's just another preference, perhaps a preference for not understanding people whose goals are highly inimical to their own.

I assume that we have very moral behaviours that they consider incomprehensibly evil, too, but I don't know what those might be. Maybe allowing men and women to play together is one?

I don't see any lack of understanding, merely different attitudes.

Another difference between (3) and (4) is that (4) asserts that morality develops in the same way as does language. There's nothing in it relating to "I like" and "I don't like", any more than I can, in an English-speaking community, be understood when I talk Chinese. You can't choose what is moral in your community. It's simply not a matter of personal preference.

It is a matter of personal preferences: other people's. You can choose your own but not theirs.

For example, making everyone else do what you think they should doesn't work?

It should work if you have a lot of power and they don't.

What is this "should"? There is only "is" and "is not". And it's mostly "is not". One could hardly have a greater imbalance of power than that in a prison, yet prison governors can't even keep drugs and mobile phones away from the inmates.

-- Richard

···

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

[From Richard Kennaway (2006.08.24.1408 BST)]

[From Bill Powers (2006.08.24.0430 MDT)]
I'm not sure understanding comes that easily. It doesn't for me. I can see what people do, but what they say about why they do it often sounds like gibberish to me. I'm not sure they understand it themselves. People let themselves be pushed around by words a lot. They listen, they make up their own meanings for words, and they assume that that's what everyone else means. They think that if something has a name, it must be real.

"Should" is an example of that. I agree (now that I'm not channelling Friedman) that it actually refers to people's references, but people using the word often do not think that that is what they are talking about. They speak instead as if they think they are talking about something independent of themselves, as if "X should do Y" is true or false independent of anyone's preferences, even their own.

My category 3 of morality -- that there are no normative facts -- was intended to include all non-normative views of how normative behaviour might arise, rather than to exclude all of them and ascribe it to whims out of nowhere.

What is this "should"? There is only "is" and "is not". And it's mostly "is not". One could hardly have a greater imbalance of power than that in a prison, yet prison governors can't even keep drugs and mobile phones away from the inmates.

But don't the inmates think they should be able to get drugs and phones? If not, why try to get them?

They want drugs and phones. I don't know anyone in prison, but I would find it surprising if they claimed that they "should" be available.

To say "should" also hints at a conflict. When you say "I should do that," the implication is that you're not doing it.

Which is why, if I find myself about to use the word, it's a wake-up call to me that something I'm not currently attending to needs sorting out.

"Either do, or not do. There is no 'should' for a Jedi." :slight_smile:

-- Richard

···

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

[Martin Taylor 2006.08.24.10.28]

[From Richard Kennaway (2006.08.24.1056 BST)]

For example, making everyone else do what you think they should doesn't work?

It should work if you have a lot of power and they don't.

What is this "should"? There is only "is" and "is not".

Yes, but you don't know (other than through intuition, which can differ between equally capable analysts) whether "is" or "is not" more closely represents the facts. My use of "should" was intended to be equivalent to "probably would in many situations". It had nothing of the flavour of "should" in "you should (ought to) do this".

And it's mostly "is not".

This assertion says that your intuition differs from mine, although the word "mostly" suggests the difference is only one of degree.

One could hardly have a greater imbalance of power than that in a prison, yet prison governors can't even keep drugs and mobile phones away from the inmates.

I think that's not a good example. The governor's power is constrained. For sure, if the governor had the power to control exactly what passed the walls, and wanted not to have cell phones and drugs pass the walls, no cell phones or drugs would pass the walls. Governors in some prisons probably do have that power, but in most semi-civilized countries, they don't. The governor of a British or Canadian prison cannot normally prevent the inmates from contact with the outer world (or so we are led to believe).

However, I don't see the relevance of the example, even if it were internally valid. In PCT terms, we are talking about conflict, and your assertion seems to say that the conflict would not normally resolve to a situation in which the error value in the more powerful control system is less than that of the weaker system. I think you'd have to demonstrate that counter-intuitive (my intuition) claim by experiment, simulation, or mathematical argument.

If you want to soften the conditions to allow for "other people" to achieve their goals by means other than the ones being constrained by the dictator, then your claim is that the dictator doesn't have enough degrees of freedom to effectively oppose all of the different action outputs available to the oppressed. Since the dictator is human, that's almost certainly true. But then there's no dispute between us.

So "where's the beef?"

Martin

[From Richard Kennaway (2006.08.24.1712 BST)]

[Martin Taylor 2006.08.24.10.28]

[From Richard Kennaway (2006.08.24.1056 BST)]
One could hardly have a greater imbalance of power than that in a prison, yet prison governors can't even keep drugs and mobile phones away from the inmates.

I think that's not a good example. The governor's power is constrained. For sure, if the governor had the power to control exactly what passed the walls, and wanted not to have cell phones and drugs pass the walls, no cell phones or drugs would pass the walls.

That is a tautology: if he could, then he could.

However, I don't see the relevance of the example, even if it were internally valid. In PCT terms, we are talking about conflict, and your assertion seems to say that the conflict would not normally resolve to a situation in which the error value in the more powerful control system is less than that of the weaker system.

Not at all. The more powerful system generally is more successful in getting what it wants -- that is almost another tautology. I was talking about the difficulty, in practice, of forcing people to do what you want them to do.

-- Richard

···

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

[From Rick Marken (2006.08.24.1200)]

Bill Powers (2006.08.24.0930 MDT) to Richard Kennaway (2006.08.24.1408 BST)--

In case you hadn't noticed, while I was writing all that worthy prose I was not packing stuff into boxes. You can judge for yourself whether the goal of enlightening you was behind the writing, or simply that of putting off what I must now go and do.

The next time you feel like you _shouldn't_ be packing, could you please take a moment to help me out on my questions from my last post? In particular could you let me know how far off I am on the following:

Rick Marken (2006.08.23.1120)--

1. I thought one way conflict occurs is when two (or more) control systems control the same (or a similar) perceptual representation of the same environment. Conflict can occur when control systems control different perceptual representations of the same environment. But in that case the conflict does not occur _because_ the systems are controlling different perceptions.

2. The test for the controlled variable is a way to determine what perceptual variable a system is controlling. so it can be seen as a way for one person (the experimenter) to have a perception that is equivalent to what another (the subject) is perceiving and controlling.

Are these two points wrong in some fundamental way? If so, I know how I'm going to spend my retirement: learning PCT!

Thanks.

Best

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[From Bill Powers (2006.08.24.1315 MDT)]

Rick Marken (2006.08.24.1200)

  1. I thought one way conflict occurs is when two (or more) control
    systems control the same (or a similar) perceptual representation of the
    same environment. Conflict can occur when control systems control
    different perceptual representations of the same environment. But in that
    case the conflict does not occur because the systems are controlling
    different perceptions.

Yes, that is one way. But how do you determine that two control systems
are controlling the same perceptual variable? You have to define the
controlled variable in terms of your perceptions in order to test
it, and so all you can prove is that the variable under control by the
other systems is equivalent to the one you have defined.
You’re right in saying that conflict doesn’t arise because the
control systems are controlling different perceptions. Control systems
control different perceptions all the time without conflicting with each
other. What leads to conflict is when the different perceptions are
different functions of the same set of environmental.variables. Then,
except for rare coincidences, controlling one perception will require the
environmental variables to be in different states from when the other
perception is controlled. Then it’s impossible for both perceptions to be
controlled at the same time.

  1. The test for the controlled
    variable is a way to determine what perceptual variable a system is
    controlling. so it can be seen as a way for one person (the
    experimenter) to have a perception that is equivalent to what another
    (the subject) is perceiving and controlling.

What the Test shows is that the control system is controlling something
related to your perception of the controlled variable. When the variable
actually under control is constant, the variable you have defined is
constant. When you disturb the variable you have defined, the actual
variable is also disturbed, and the action that stabilizes the actual
variable also stabilizes the one you have defined.

That is as close as you can ever get to finding the true controlled
variable. A variable that is very different from the one actually under
control could still pass the test.

Best,

Bill P.

[From Rick Marken (2006.08.24.1320)]

Bill Powers (2006.08.24.1315 MDT)--

Thanks for replying. I feel like I _should_ be writing so this helps keep me away from that;-)

Rick Marken (2006.08.24.1200) --

1. I thought one way conflict occurs is when two (or more) control systems control the same (or a similar) perceptual representation of the same environment.� Conflict can occur when control systems control different perceptual representations of the same environment. But in that case the conflict does not occur _because_ the systems are controlling different perceptions.

Yes, that is one way. But how do you determine that two control systems are controlling the same perceptual variable? You have to define the controlled variable in terms of your perceptions in order to test it, and so all you can prove is that the variable under control by the other systems is equivalent to the one you have > defined.

Equivalent is good enough for me.

You're right in saying that conflict doesn't arise because the control systems are controlling different perceptions. Control systems control different perceptions all the time without conflicting with each other. What leads to conflict is when the different perceptions are different functions of the same set of environmental.variables.

Is this really the right way to say it? Aren't v1+ v2 and v1 - v2 different functions of the same set of environmental variables (v1, v2)? There would be no conflict if one system were controlling v1+ v2 and the other were controlling v1 - v2. Right? I think that when perceptions of the same environmental variables are different conflict will result when there are fewer output df then input df. Isn't that it?

2. The test for the controlled variable is a way to determine what perceptual variable a system is controlling.� so it can be seen as a way for one person (the experimenter) to have a perception that is equivalent to what another (the subject) is perceiving and controlling.

What the Test shows is that the control system is controlling something related to your perception of the controlled variable. When the variable actually under control is constant, the variable you have defined is constant. When you disturb the variable you have defined, the actual variable is also disturbed, and the action that stabilizes the actual variable also stabilizes the one you have defined.

That is as close as you can ever get to finding the true controlled variable. A variable that is very different from the one actually under control could still pass the test.

OK. That seems good enough to me ... for now, anyway. That is, until we' start going for absolute truth. Right now I'm content to go with relatively better truth than what we've currently got.

Best

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[From Bill Powers (2006.08.24.1525 <DT)]

Rick Marken (2006.08.24.1320) --

Is this really the right way to say it? Aren't v1+ v2 and v1 - v2 different functions of the same set of environmental variables (v1, v2)? There would be no conflict if one system were controlling v1+ v2 and the other were controlling v1 - v2. Right? I think that when perceptions of the same environmental variables are different conflict will result when there are fewer output df then input df. Isn't that it?

No, all you need is for p1 = v1 + v2 and p2 = v1+ 10*v2, or any two different enough functions. Yes, there is just one pair of functions that allows non-interacting control, but there is an infinity of functions that are not orthogonal and that therefore create interactions. Among those, some create interactions large enough to drive one or the other control system to its limits, and then there is conflict.

That is as close as you can ever get to finding the true controlled variable. A variable that is very different from the one actually under control could still pass the test.

OK. That seems good enough to me ... for now, anyway. That is, until we' start going for absolute truth. Right now I'm content to go with relatively better truth than what we've currently got.

OK, but I think you're whistling in the dark. Why not try to devise a control system that is controlling something completely different from what you might propose as a controlled variable, such that your definition will pass the test? As long as I'm arguing on one side and you're defending, you'll only think of positive instances of what you think is happening. If you try to find a counterexample, I can stop trying to persuade you. You're much more likely to persuade you than I am.

The Test is useful as a way of discovering what another control system might be controlling, as projected into your perceptual space. That enables you to make predictions in terms of your own perceptions. But the Test can't answer the basic epistemological question, as far as I can see.

Best.

Bill P.

[From Rick Marken (2006.08.24.1655)]

Bill Powers (2006.08.24.1525 <DT)--

Rick Marken (2006.08.24.1320) --

Is this really the right way to say it? Aren't v1+ v2 and v1 - v2 different functions of the same set of environmental variables (v1, v2)? There would be no conflict if one system were controlling v1+ v2 and the other were controlling v1 - v2. Right?

No, all you need is for p1 = v1 + v2 and p2 = v1+ 10*v2, or any two different enough functions.

What you had said was "What leads to conflict is when the different perceptions are different functions of the same set of environmental.variables." What I was questioning was whether this was the best way to express it. I was simply pointing out that it is not true that what leads to conflict is when different perceptions are different functions of the same set of environmental.variables. That's why I gave the example of v1+ v2 and v1 - v2: different functions of the same environmental variables that don't lead to conflict.

But it is true (I believe) that what always leads to conflict is when different perceptions are the same or similar (not completely orthogonal) functions of the same set of environmental variables. The functions v1 + v2 and v1+ 10*v2 are certainly different but they are highly correlated so they are also "similar" (non-orthogonal). The functions v1+ v2 and v1 - v2 are also different but they are not similar at all; they are orthogonal.

OK, but I think you're whistling in the dark. Why not try to devise a control system that is controlling something completely different from what you might propose as a controlled variable, such that your definition will pass the test? As long as I'm arguing on one side and you're defending, you'll only think of positive instances of what you think is happening. If you try to find a counterexample, I can stop trying to persuade you. You're much more likely to persuade you than I am.

I'm already persuaded. I agree that the Test can result in my concluding that the controlled variable is completely different than what it actually is. I can see, using your example, that it would be easy to conclude that it's v1+ v2 when it's actually v1 + 10* v2 that is controlled.

The Test is useful as a way of discovering what another control system might be controlling, as projected into your perceptual space. That enables you to make predictions in terms of your own perceptions. But the Test can't answer the basic epistemological question, as far as I can see.

I completely agree. I didn't mean to seem flip about it. But the fact is that I'm challenged enough by the difficulties of discovering what another control system might be controlling, as projected into my perceptual space. I'll leave the epistemological questions to people who are much smarter than me, which is nearly everyone over the age of 7.

Love

Rick

···

----
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400