is anybody out there, out there, ...

CSGnet Archive CSG1998
1

ground control to major tom..i cannot believe that everyone is tongue-tied or
of the "no-comment" variety. are we still reeling from the coersive can o'
worms?
does anyone really think that the "universal error curve" has even one
experiemental confirmation? did it not seem to come right on the heels of a
questionable position on coersion. am i sniffing too much glue? will rick
and abbott ever meet on MTV's celebrity death matches?
when the cats are away, do the meeces play?

lolligaging to a heat wave

i.

2

[From Bruce Nevin (980602)]

12:31 PM 6/2/98 EDT, Marc Kurtzer--

I'm in a class, paying the price of great Perl, Today through Thursday.
Friday, I'm off to San Jose through Wednesday. Thursday, up to Gloucester
for my daughter's wedding on Saturday.

does anyone really think that the "universal error curve" has even one
experiemental confirmation?

It's a generalization on observations. Is it a phenomenon to be modelled?
is it an explanatory principle? Not always easy to distinguish between a
phenomenon, a generalization, and an explanatory principle. A
generalization has exceptions, difficult to disconfirm. An explanatory
principle (e.g. gravity) has apparent exceptions that can be explained
(lighter-than-air craft). A phenomenon is something incontrovertibly
requiring explanation. What's sloppy?

What was the questionable position on coercion?

  Bruce Nevin

3

i.kurtzer (980602.1700)
[From Bruce Nevin (980602)]

  12:31 PM 6/2/98 EDT, isaac Kurtzer--

I'm in a class, paying the price of great Perl, Today through Thursday.

Fritz Perl?

>does anyone really think that the "universal error curve" has even one
>experiemental confirmation?

It's a generalization on observations. Is it a phenomenon to be modelled?
is it an explanatory principle? Not always easy to distinguish between a
phenomenon, a generalization, and an explanatory principle. A
generalization has exceptions, difficult to disconfirm. An explanatory
principle (e.g. gravity) has apparent exceptions that can be explained
(lighter-than-air craft). A phenomenon is something incontrovertibly
requiring explanation. What's sloppy?

Heck, to me it is none of these yet. It has no empircal data. Therefore, the
phrasing as "universal" seems rather strange. Also, it is couched with
"error" which is a very specific term, whose connection remains unexplicated
in theory and unexamined in fact. If it is any of the above options i would
say "explanatory principle", but thats not saying much. Right now, it is
less tenable than "reinforcement" .

What was the questionable position on coercion?

that coersion is a interaction that only requires classification based on only
one participant's intention. As ALL other herethough defined interactions
appealed to all participants' intentions, at the least, strong reasons must be
given for this breach of taxonomic logic. I remain unconvinced.

i.

4

Marc Kurtzer wrote:...

[From Dick Robertson 980604.1041CDT]
Well, I'm still here until I leave for Vancouver next month. Hope to
see you there.
But, I can't think of anything intelligent to say about the universal
error curve. Is it really important--have I missed something?
Best, Dick R.

5

i.kurtzer (980604.1330)

[From Dick Robertson 980604.1041CDT]

Well, I'm still here until I leave for Vancouver next month. Hope to
see you there.

you bet.

But, I can't think of anything intelligent to say about the universal
error curve. Is it really important--have I missed something?

i think that you haven't missed anything at all. Not saying anything till
data comes is the most intelligent thing to say.

i.

···
6

ground control to major tom..i cannot believe that everyone is tongue-tied or
of the "no-comment" variety. are we still reeling from the coersive can o'
worms?

Well, 14 of us have just been at the European CSGNET Workshop, where a good
time was had by all. I found the presentations and discussions
fascinating. I expect the other participants will have a whole lot to say
when they get back online.

-- Richard Kennaway, jrk@sys.uea.ac.uk, http://www.sys.uea.ac.uk/~jrk/
   School of Information Systems, Univ. of East Anglia, Norwich, U.K.

7

[From Rick Marken (980609.1420)]

i.kurtzer (980602) --

does anyone really think that the "universal error curve" has even
one experiemental confirmation?

No. But there is a lot of anecdotal evidence for it. For example,
I find that I stop controlling for things that I would otherwise
control for when it becomes difficult or impossible to control them.
When I am on a business trip, for example, I rarely snack after
dinner because there is no easily accessable food in the room (at
least, not in the places I stay;-)); when I am at home I rarely
_stop_ eating snacks until I go to bed;-)

I know there are other possible explanations for this phenomenon
but one is certainly the "universal error curve" (I think "universal"
just means that this error curve applies to all control systems in
the hierarchy).

isaac asks:

did it not seem to come right on the heels of a questionable
position on coersion.

Bruce Nevin (980602) asks in reply:

What was the questionable position on coercion?

i.kurtzer (980602.1700) replies:

that coersion is a interaction that only requires classification
based on only one participant's intention...I remain unconvinced.

Coercion refers to control of behavior. The only relevant intentions
are those of the controller (coercer); the coercer's intention to
control some aspect of another person's behavior is what we are
calling _coercion_. If there is no visible resistance to coercion
(and we know that this happens; witness the non-resistance of most
Jews to being placed on boxcars shipped to death camps) there is
still coercion if it is the coercer's intention to control some
aspect of a person's behavior.

Why do you resist this definition of coercion? Do you really want
to say there is no coercion if there is no resistance (which is what
would happen if the intentions of the coercee were consistent with
those of the coercer)? Doesn't that seem to open up some scary
possibilities? For example, there was no widespread resistance
to the Soviet regime that I know of; does that mean that Lenin,
Stalin, et al were not coercive?

i.kurtzer (980604.1330) --

Not saying anything [about the universal error curve] until data
comes is the most intelligent thing to say.

Bruce Abbott (980604.1525 EST) --

How would you go about collecting data bearing on this issue?

I'd test 30 subjects in a three-level between subjects tracking task
with 10 subjects in each condition. The maximum size of the
disturbance would be varied across conditions: low, medium and high.
The dependent variable is the amount of time the subject stays in
the tracking task before quitting voluntarily (up to a maximum of
10 days;-)). Subjects would be carefully instructed about how they
should try to stay at the task as long as possible, for the good of
science. The universal error curve theory predicts that there will
be a significant effect of maximum disturbance on quitting; the
greater the maximum, the less time until quitting. ;-))

Richard Kennaway (980609) --

14 of us have just been at the European CSGNET Workshop, where a
good time was had by all.

Wonderful. Could you (or someone) give a brief report on who was
there and what was said?

I am very sorry I missed the conference. It's the first PCT relevant
meeting I've missed since 1983. I hope this is not the start of
my slide down the slippery slope. But it sounds like you folks did
just fine without me;-)

I hope some of you can make it to Vancouver.

Best

Rick

···

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

8

i.kurtzer (980610.1430)

[From Rick Marken (980609.1420)]

i.kurtzer (980602) --

> does anyone really think that the "universal error curve" has even
> one experiemental confirmation?

No. But there is a lot of anecdotal evidence for it.

Rick, this is the same type of argument that supports psi research. I will
not accept it.

For an example why:

For example,
find that I stop controlling for things that I would otherwise
control for when it becomes difficult or impossible to control them.

I'm sure there are other persons that could describe the manner in which they
pursued their day according to a different theory. In science its not enough
to evaluate a theory in the first person.

Again:

When I am on a business trip, for example, I rarely snack after

dinner because there is no easily accessable food in the room (at
least, not in the places I stay;-)); when I am at home I rarely
_stop_ eating snacks until I go to bed;-)

this is not the stuff that i want a science of behavior to be _based on_.
Substantiating a theory on first person evaluations gives so much freedom to
the researcher that it cannot be accepted as evidence. Lots of questionable
evidence does not make it any better.

I know there are other possible explanations for this phenomenon
but one is certainly the "universal error curve" (I think "universal"
just means that this error curve applies to all control systems in
the hierarchy).

WHAT PHENOMENA?! this now has the staus of something real ot be explained?!
RIck, back up for one second and see how quickly this just introduced WORD has
now become a phenomena. You would chew Abbott's hide for weeks if he ever did
something like that.

isaac asks:

> did it not seem to come right on the heels of a questionable
> position on coersion.

which i still think seems right.

Bruce Nevin (980602) asks in reply:

> What was the questionable position on coercion?

i.kurtzer (980602.1700) replies:

> that coersion is a interaction that only requires classification
> based on only one participant's intention...I remain unconvinced.

rick replies:

Coercion refers to control of behavior.

would you agree that there are at least two persons that are involved? That
coersion is one type of social interaction?
If so then you must explain why it is defined differently from all other
social interactions, preferably with data showing how that definition is more
empirically fruitful. Since that data does not exist, i remain unconvinced.

The only relevant intentions
are those of the controller (coercer); the coercer's intention to
control some aspect of another person's behavior is what we are
calling _coercion_.

We does not include myself, Tom, and Tim. You my friend are skirting the
argument from authority.

If there is no visible resistance to coercion

you have already mandated coersion's existance according to your criteria.
That is putting the cart before the horse.

(and we know that this happens; witness the non-resistance of most
Jews to being placed on boxcars shipped to death camps) there is
still coercion if it is the coercer's intention to control some
aspect of a person's behavior.

could we use an example that is not so provoking?

Why do you resist this definition of coercion?

see above

Do you really want
to say there is no coercion if there is no resistance (which is what
would happen if the intentions of the coercee were consistent with
those of the coercer)?

Let me think about this. But i would appreciate you also deciding whether you
will give up your position if that (right above) is what all this (definition
of coersion) hinges on. I will not pull out this leg if you will not allow
that to topple your idea.

i.kurtzer (980604.1330) --

> Not saying anything [about the universal error curve] until data
> comes is the most intelligent thing to say.

Bruce Abbott (980604.1525 EST) --

> How would you go about collecting data bearing on this issue?

rick:

I'd test 30 subjects in a three-level between subjects tracking task
with 10 subjects in each condition. The maximum size of the
disturbance would be varied across conditions: low, medium and high.
The dependent variable is the amount of time the subject stays in
the tracking task before quitting voluntarily (up to a maximum of
10 days;-)). Subjects would be carefully instructed about how they
should try to stay at the task as long as possible, for the good of
science. The universal error curve theory predicts that there will
be a significant effect of maximum disturbance on quitting; the
greater the maximum, the less time until quitting. ;-))

was that a joke?

to rick my friend without daggers and wishing for no daggers

i.

9

[From Rick Marken (980611.1600)]

Me:

I know there are other possible explanations for this phenomenon
but one is certainly the "universal error curve"

i.kurtzer (980610.1430) --

WHAT PHENOMENA?!

The phenomenon of "giving up". When snacks become difficult to
obtain I give up on controlling for having them. I have also
seen other organisms "give up" trying to control for things
when those things become inaccessible.

RIck, back up for one second and see how quickly this just
introduced WORD has now become a phenomena. You would chew
Abbott's hide for weeks if he ever did something like that.

I don't think so. I've observed "giving up" in myself and others.
This is a phenomenon. Whether an error curve is needed to
explain this phenomenon is another matter that can only be
evaluated using experimental test.

Me:

Coercion refers to control of behavior.

isaac

would you agree that there are at least two persons that are
involved? That coersion is one type of social interaction?

Yes.

If so then you must explain why it is defined differently from
all other social interactions

I don't think I understand the question. Coercion is a particular
type of social interaction where one party (the coercer) controls
some aspect of the behavior of another (the coercee). Coercion
is control where the controlled variable happens to be an aspect
of the behavior of another living system. This definition is
similar to the definition of all other social interactions
because both parties to the interaction are included in the
definition; the coercer appears as the control system; the
coercee appears as the controlled variable. What's the problem?

Me:

The only relevant intentions are those of the controller (coercer);
the coercer's intention to control some aspect of another person's
behavior is what we are calling _coercion_.

isaac:

We does not include myself, Tom, and Tim.

Why not? If coercion is control of behavior then how are the
intentions of the coercee relevant to this definition of coercion?

Me:

Do you really want to say there is no coercion if there is no
resistance

isaac:

Let me think about this.

Feel free. No coercion;-)

But i would appreciate you also deciding whether you will give up
your position if that (right above) is what all this (definition
of coersion) hinges on. I will not pull out this leg if you will
not allow that to topple your idea.

If you want to use the word "coercion" only when coercer and
coercee want the same variable in different states then that's
fine with me. What I will not change is my conclusion that the
RTP teacher is trained to control at least one aspect of the
behavior of students in the class.

Me:

I'd test 30 subjects in a three-level between subjects tracking
task ...

isaac:

was that a joke?

Sort of. I'm trying to re-adopt my old "conventional psychologist"
mental set regarding the way to test psychological theories. I was
hoping this would be a way to encourage Bruce Abbott and others to
contribute to a discussion of ways to test the PCT model. All my
yapping about the "proper" way to do PCT research has gotten me
nowhere. So I'm going with Bruce A.'s approach and see where we
get by applying conventional psychological research principles to
the study of a control phenomenon.

I'm starting with the assumption that the conventional approach
is just fine and that all my blather about testing for controlled
variables and comparing model behavior to actual behavior is
irrelevant because it's already part of the conventional approach.

The universal error curve seems to provide a nice opportunity for
us to develop, online, experiments that will test a possible
modification to the basic control model. So I was hoping that
we could just go about developing such an experiment, starting
from a conventional perspective, and revising (if necessary)
until there is some common agreement that the experiment (or
experiments) will test the theory. So far we have only had
one iteration so either everyone is satisfied with the design
I proposed or no one want to work on it anymore. Whatever.

Best

Rick

···

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

10

i.kurtzer (980612.0000)

[From Rick Marken (980611.1600)]

Rick:

I know there are other possible explanations for this phenomenon
but one is certainly the "universal error curve"

i.kurtzer (980610.1430) --

WHAT PHENOMENA?!

rick:

The phenomenon of "giving up".

aughgh!! Dr. Marken, I think it it premature to assume that any _single_ form
underlies those scenarios that you have---without explication--demarcated as
"giving up" . I am not saying people do not use this word for themselves as
well as apply it to others. But that in itself does not secure phenomenal
status. It remains an empirical question whether a _phenomena_ (as opposed to
semantically isolated instances) exists.

When snacks become difficult to
obtain I give up on controlling for having them.

Anecdotal evidence in the first person that support the author's position will
not convince me, clear experimental evidence will. That goes for anyone else
out there that believes this "universal error curve" now turned
"theory"!!!aurhgh!!!

I have also
seen other organisms "give up" trying to control for things
when those things become inaccessible.

B.F. Skinner also saw alot of things. Your seeing is not my believing, my
Doctor friend.

RIck, back up for one second and see how quickly this just
introduced WORD has now become a phenomena. You would chew
Abbott's hide for weeks if he ever did something like that.

I don't think so. I've observed "giving up" in myself and others.
This is a phenomenon. Whether an error curve is needed to
explain this phenomenon is another matter that can only be
evaluated using experimental test.

again B.F.

Rick:

Coercion refers to control of behavior.

isaac

> would you agree that there are at least two persons that are
> involved? That coersion is one type of social interaction?

Yes.

great

If so then you must explain why it is defined differently from

> all other social interactions

I don't think I understand the question.

i'll risk repition for clarity. The problem I have is that coersion is within
the class of scenarios we have deemed "social interaction" , but as you have
defined coersion it violates one of the defining criteria shared by all the
other herethough noted interactions. This includes a whole list of social
interactions Bourbon has presented at two conferences so i know you are
familiar with them. In _all_ other types of interaction we have appealed to
the intentions of both parties, so by knowing those intentions we can
determine whether it is a social interction at all and if so which one:
helping, cooperating, conflict, etc . Now with coersion you say that only one
person's intention is sufficient in classifying the social interaction. This
strikes me at many levels as yucky. First, this comes very close to lineal
causality. Secondly, since this is a breach of tradition it seems we should
have good reasons for doing so. The question is what are your good reasons.

Coersion is a type of social interaction where one party (the coercer)

controls

some aspect of the behavior of another (the coercee).

you have not referred to the OTHER party's intention. That's my problem.

Coercion
is control where the controlled variable happens to be an aspect
of the behavior of another living system. This definition is
similar to the definition of all other social interactions
because both parties to the interaction are included in the
definition;

no, it is not similiar. The coersee is included passively, as a mere
bystander in how someone else can completely usurp the abilty to define that
interaction.
That
A) smacks of lineal causality and
B) is contrary to protocal

the coercer appears as the control system; the
coercee appears as the controlled variable. What's the problem?

that

RIck:

The only relevant intentions are those of the controller (coercer);
the coercer's intention to control some aspect of another person's
behavior is what we are calling _coercion_.

isaac:

We does not include myself, Tom, and Tim.

Why not?

lineal causality and breach of protocal without good reason

If coercion is control of behavior then how are the

>intentions of the coercee relevant to this definition of coercion?

They are not. It's a bad defintion.

Rick:

Do you really want to say there is no coercion if there is no
resistance

isaac:

Let me think about this.

Feel free. No coercion;-)

i know Rebbe.

But i would appreciate you also deciding whether you will give up
your position if that (right above) is what all this (definition
of coersion) hinges on. I will not pull out this leg if you will
not allow that to topple your idea.

If you want to use the word "coercion" only when coercer and
coercee want the same variable in different states then that's
fine with me.

how will you use the word?

What I will not change is my conclusion that the
RTP teacher is trained to control at least one aspect of the
behavior of students in the class.

does that make it coersion?

Rick:

I'd test 30 subjects in a three-level between subjects tracking
task ...

isaac:

was that a joke?

Sort of. I'm trying to re-adopt my old "conventional psychologist"
mental set regarding the way to test psychological theories. I was
hoping this would be a way to encourage Bruce Abbott and others to
contribute to a discussion of ways to test the PCT model. All my
yapping about the "proper" way to do PCT research has gotten me
nowhere. So I'm going with Bruce A.'s approach and see where we
get by applying conventional psychological research principles to
the study of a control phenomenon.

i suspect you'll piss him off; he hates insincerity.

I'm starting with the assumption that the conventional approach
is just fine and that all my blather about testing for controlled
variables and comparing model behavior to actual behavior is
irrelevant because it's already part of the conventional approach.

you are a meshugina! aren't you?

your boobala and greatest hawker of Mind Readings
hey if anyone is jealous come to the conference and you can be my friend too!!

i.

11

[From Rick Marken (980612.0820)]

i.kurtzer (980612.0000)

aughgh!! Dr. Marken, I think it it premature to assume that
any _single_ form underlies those scenarios that you have---
without explication--demarcated as "giving up" .

I know. We have to do experiments. I've suggested one. What
do you think? Any suggestions?

Now with coersion you say that only one person's intention is
sufficient in classifying the social interaction. This strikes
me at many levels as yucky. First, this comes very close to
lineal causality.

Why? Coercion is closed loop control of behavior.

Secondly, since this is a breach of tradition it seems we should
have good reasons for doing so. The question is what are your
good reasons.

Here is a diagram of what I think of as coercion:

             r
                         >
        p --> C -->

···

        > Coercer

       >s> >o>
----------------------------
        > qo
        > >
        qi<-------- Environment
        >
        qo' (d)
----------------------------
       >o'| Coercee

------------------------------

qi is a variable controlled by both the Coercer and
the Coercee. The Coercee's effect on qi (via qo') is
just a disturbance to qi from the point of view of
the Coercer. The same is true for the Coercee; the
Coercee's effect on qi is just a disturbance to qi from
the point of view of the Coercee. Since the Coercer is
much stronger than the Coercee, the Coercer keeps
qi under control; the deprives the Coercee of any
control over qi. Note that this analysis requires no
knowledge of the intentions (r') of the Coercee. All
we have to know is that _both_ Coercer and Coercee have
effects on qi _and_ that the Coercer is much stronger
(higher output capability) that the Coercee.

A concrete example: qi is the Coercee's location in space.
The Coercer wants this variable in the state "in the
social skills room". The Coercer is able to produce this
perception of qi becuase he is so much stronger than the
Coercee. The Coercee's intentions only influence the kind
of effects (disturbances) the Coercee will have on qi.
If the Coercee intends to go to the social skills room
then the disturbance to qi (from the Coercer's perspective)
will be 0; if the Coercee intends to be somewhere other than
in the social skills rooom than the disturbance to qi will
be substantially greater than 0. But this doesn't matter
to the Coercer, who is able to produce the output required
to get qi (actually, the perception of qi, p) to the reference
state (in the social skills room).

The coersee is included passively, as a mere bystander in how
someone else can completely usurp the abilty to define that
interaction.
That
A) smacks of lineal causality and
B) is contrary to protocal

Are you feeling better about it now?

Best

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

12

i.kurtzer (980615.0000)

rick's most recent detailing of his proposal for coersion:

< qi is a variable controlled by both the Coercer and
< the Coercee. The Coercee's effect on qi (via qo') is
< just a disturbance to qi from the point of view of
< the Coercer. The same is true for the Coercee; the
< Coercee's effect on qi is just a disturbance to qi from
< the point of view of the Coercee. Since the Coercer is
< much stronger than the Coercee, the Coercer keeps
< qi under control; the deprives the Coercee of any
< control over qi. ***[this contradicts the first sentence--i]***
< Note that this analysis requires no
< knowledge of the intentions (r') of the Coercee. All
< we have to know is that _both_ Coercer and Coercee have
< effects on qi _and_ that the Coercer is much stronger
< (higher output capability) that the Coercee.
  >>

Therefore, ALL interactions about a common CV will fall under coersion given a
suitable gain difference. Does that not seem strange? Does that make for a
very strange taxonomy with helping as a subclass of coersion?

i.