PCT aspirin

CSGnet Archive CSG2008
1

Of course, these correlations
(the actual values of which are not
reported in the paper) do
not imply causality; adopting a religious

system concept reference is not necessarily the cause of high teen

pregnancy rates, high abortion rates and high crime rates and low
life

expectancy. But the existence of a positive (rather than a negative

correlation) between religious references and these measures of poor

social quality certainly suggests that failure to adopt religious

references will not lead to a decline in the quality of a society,
as

many religious people in the US have suggested.

system concepts are
“not very likely” to be the cause of social dysfunction;

just that they are “not necessarily” the cause
of this dysfunction. In fact,

there is other data that suggests that there is a causal
relationship

between religious beliefs and abortion rate, for example.
Abortion

rates are higher in regions where counselors are not allowed –
for

strictly religious reasons – to discuss the abortion option
with

expectant mothers than in places where they are allowed to
discuss

this option.

The point of my discussion above was not to say that there is a
causal

relationship between religiosity and social
dysfunction.
My point was

that “correlation does not imply causality” does not mean
that

correlation means there is definitely not a causal relationship

between the variables. You seem to think that I am arguing that

correlation actually does imply causality, which I am not. You seem
to

be arguing that correlation implies no causality, which is what I
take

issue with.

The point I was trying to make above is simply that correlation –
an

observed relationship between two variables – does not imply that
one

variable causes the other but it also does not imply that one
variable

does not cause another. For example, there is a strong, negative

correlation between current flow and resistance in an electrical

circuit. This correlation does not imply causality nor does it imply

no causality. In the case of current and resistance there actually
is

a causal relationship between the variables. That’s what I meant
when

I said that correlated variables are “not necessarily”
causally

related rather than “not very likely” causally related.
The

correlation between variables like current and resistance or
religion

and social dysfunction means only that the variables are _not

necessarily_ causally related; it says nothing about the likelihood

that they actually are (or are not) causally related.

You seem to be interpreting the mantra "correlation does not
imply

causality" to mean that "correlation means that causation is
very

unlikely", a frequent misinterpretation of the mantra. In fact,
the

mantra simply means that a high correlation between variables says

nothing about whether the variables are or are not causally related.

It should also be quickly pointed out, however, that no correlation

between variables does imply no causal relationship between the

variables. That was all I was arguing based on the Paul data. Where

there is a very low correlation between religiosity and social

dysfunction I conclude that, contrary to religious zealots,

religiosity seems to have no causal relationship to social health.

Where there is a strong positive relationship between religiosity
and

social dysfunction, all I am saying is that this is not necessarily
a

causal relationship and, anyway, it is opposite to the relationship

predicted by the religious zealots.

What we can agree on is that there are no data supporting a
relationship

between increases in religiosity and increases in social
problems.

Actually, I can no longer agree on that, given the finding of a .94

correlation between proportion believing in God and rate of teen

abortions.

Best

Rick

Richard S. Marken PhD

rsmarken@gmail.com

–

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.19.17/1253 - Release Date:
1/31/2008 9:09 AM

–

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.19.17/1253 - Release Date:
1/31/2008 9:09 AM
[From Bill Powers (2008.01.31.1515 MST)]

I understand that you’re thoroughly pissed off with me and are not
pleased that I have cast aspersions on your profession. I will cease and
desist soon, if it turns out that nothing I have to say makes any
impression.

Consider the following which you wrote:

Since causality is not indicated at all, it also follows that there is no
suggestion . that adopting religious references will lead to a decline in
the quality of a society. The other possibility is also not supported:
that a change in social quality in either direction leads to or causes a
change in religious beliefs (though that could be offered as an
alternative causal hypothesis). An epidemic of abortions can lead to the
development of moral (if not necessrily religious) scruples, since we
really don’t know at what point a baby would feel the experience of being
aborted.

It doesn’t matter how high the correlation is: it still doesn’t explain
why the two variables are correlated. To determine causation you have to
do experimental manipulations (as you correctly pointed out to me when I
said the studies were experimental), or observe natural variations that
carry out the manipulations for you, That was not done in any of the
studies. There was no observation of a change in quality of a society
together with an observation of a change of religious beliefs in that
same society, or any attempt to change one variable and predict the
effect on the other in the same society.

In my opinion, the religious issue is a red herring. There is both good
and evil to be found in effects of religion, but the roots of human
problems lie elsewhere, in my opinion, and will not be much affected by
whatever we decide to conclude about religion. If we either decrease or
increase the emphasis on religion, some aspects of society will get
better and some will get worse – by small amounts. Conflict between
control systems will continue in either case to be the main source of
human difficulties, both internal and external.

I think I’ve said enough on this subject.

Best,

Bill P.

Rick Marken (2008.01.31.1140) –

So it’s not really correct to say, based on the evidence, that
religious

Then why did you say the following? “In fact,there is
other data that suggests that there is a causal relationship between
religious beliefs and abortion rate, for example. Abortion rates are
higher in regions where counselors are not allowed – for strictly
religious reasons – to discuss the abortion option with expectant
mothers than in places where they are allowed to discuss this
option”.

2

(Gavin Ritz 2008.02.01.12.71NZT)

I think this is actually great because we
are doing what humans do all the time, mostly in conflict, after all this behaviour got us here. Power,
knowledge, influence, conflict, relationship, accountability, tension is my
real interest.

One trying to influence the other through
non vested authority (probably authority earned), by means of knowledge, must
be regarded as an interesting conflict and power struggle.

And after all PCT is a conflict discipline
amongst others.

···

-----Original Message-----
From: Control Systems Group
Network (CSGnet) [mailto:CSGNET@LISTSERV.UIUC.EDU] On Behalf Of Bill Powers
Sent: Friday, 1 February 2008
12:02 p.m.
To: CSGNET@LISTSERV.UIUC.EDU
Subject: Re: PCT aspirin

[From Bill Powers
(2008.01.31.1515 MST)]

I understand that you’re thoroughly pissed off with me and are not pleased that
I have cast aspersions on your profession. I will cease and desist soon, if it
turns out that nothing I have to say makes any impression.

Consider the following which you wrote:

Of course, these
correlations (the actual values of which are not

reported in the paper) do
not imply causality; adopting a religious

system concept reference is not necessarily the cause of high teen

pregnancy rates, high abortion rates and high crime rates and low life

expectancy. But the existence of a positive (rather than a negative

correlation) between religious references and these measures of poor

social quality certainly suggests that failure to adopt religious

references will not lead to a decline in the quality of a society, as

many religious people in the US have suggested.

Since causality is not indicated at all, it also follows that there is no
suggestion . that adopting religious references will lead to a decline in the
quality of a society. The other possibility is also not supported: that a
change in social quality in either direction leads to or causes a change in
religious beliefs (though that could be offered as an alternative causal
hypothesis). An epidemic of abortions can lead to the development of moral (if
not necessrily religious) scruples, since we really don’t know at what point a
baby would feel the experience of being aborted.

It doesn’t matter how high the correlation is: it still doesn’t explain why the
two variables are correlated. To determine causation you have to do
experimental manipulations (as you correctly pointed out to me when I said the
studies were experimental), or observe natural variations that carry out the
manipulations for you, That was not done in any of the studies. There was no
observation of a change in quality of a society together with an observation of
a change of religious beliefs in that same society, or any attempt to change
one variable and predict the effect on the other in the same society.

In my opinion, the religious issue is a red herring. There is both good and
evil to be found in effects of religion, but the roots of human problems lie
elsewhere, in my opinion, and will not be much affected by whatever we decide
to conclude about religion. If we either decrease or increase the emphasis on
religion, some aspects of society will get better and some will get worse – by
small amounts. Conflict between control systems will continue in either case to
be the main source of human difficulties, both internal and external.

I think I’ve said enough on this subject.

Best,

Bill P.

Rick Marken (2008.01.31.1140) –

So it’s not really correct to say, based on the evidence, that
religious

system concepts
are “not very likely” to be the cause of social dysfunction;

just that they are “not necessarily” the cause of this
dysfunction. In fact,

there is other data that suggests that there is a causal relationship

between religious beliefs and abortion rate, for example. Abortion

rates are higher in regions where counselors are not allowed – for

strictly religious reasons – to discuss the abortion option with

expectant mothers than in places where they are allowed to discuss

this option.

The point of my discussion above was not to say that there is a causal

relationship between religiosity and social dysfunction.

Then why did you say the following? “In fact,there is other data that
suggests that there is a causal relationship between religious beliefs and
abortion rate, for example. Abortion rates are higher in regions where
counselors are not allowed – for strictly religious reasons – to discuss the
abortion option with expectant mothers than in places where they are allowed to
discuss this option”.

My point was

that “correlation does not imply causality” does not mean that

correlation means there is definitely not a causal relationship

between the variables. You seem to think that I am arguing that

correlation actually does imply causality, which I am not. You seem to

be arguing that correlation implies no causality, which is what I take

issue with.

The point I was trying to make above is simply that correlation – an

observed relationship between two variables – does not imply that one

variable causes the other but it also does not imply that one variable

does not cause another. For example, there is a strong, negative

correlation between current flow and resistance in an electrical

circuit. This correlation does not imply causality nor does it imply

no causality. In the case of current and resistance there actually is

a causal relationship between the variables. That’s what I meant when

I said that correlated variables are “not necessarily” causally

related rather than “not very likely” causally related. The

correlation between variables like current and resistance or religion

and social dysfunction means only that the variables are _not

necessarily_ causally related; it says nothing about the likelihood

that they actually are (or are not) causally related.

You seem to be interpreting the mantra "correlation does not imply

causality" to mean that "correlation means that causation is very

unlikely", a frequent misinterpretation of the mantra. In fact, the

mantra simply means that a high correlation between variables says

nothing about whether the variables are or are not causally related.

It should also be quickly pointed out, however, that no correlation

between variables does imply no causal relationship between the

variables. That was all I was arguing based on the Paul data. Where

there is a very low correlation between religiosity and social

dysfunction I conclude that, contrary to religious zealots,

religiosity seems to have no causal relationship to social health.

Where there is a strong positive relationship between religiosity and

social dysfunction, all I am saying is that this is not necessarily a

causal relationship and, anyway, it is opposite to the relationship

predicted by the religious zealots.

What we can agree on is that there are no data supporting a
relationship

between increases in religiosity and increases in social problems.

Actually, I can no longer agree on that, given the finding of a .94

correlation between proportion believing in God and rate of teen

abortions.

Best

Rick

Richard S. Marken PhD

rsmarken@gmail.com

–

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.19.17/1253 - Release Date: 1/31/2008
9:09 AM

–

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.19.17/1253 - Release Date: 1/31/2008
9:09 AM

3

[From Rick Marken (2008.01.31.1710)]

Bill Powers (2008.01.31.1515 MST)--

I understand that you're thoroughly pissed off with me and are not pleased
that I have cast aspersions on your profession. I will cease and desist
soon, if it turns out that nothing I have to say makes any impression.

I _was_ thoroughly pissed off with you (I'm not anymore; who could
stay mad at you?), not because you cast aspersions on my profession
(I've been casting aspersions on my profession for the last 30 years
or so -- you taught me how -- so I certainly don't mind it when you
cast them) but because you cast aspersions on me. You said that I was
bashing religion by suggesting the possibility that the data indicate
a causal relationship between religion and social ills; you were also
saying that the data were not good for the very reasons that they were
good; because they showed no relationship between religion and social
ills.

Consider the following which you wrote:

>Of course, these correlations (the actual values of which are notreported in
>the paper) do not imply causality; adopting a religious
>system concept reference is not necessarily the cause of high teen
>pregnancy rates, high abortion rates and high crime rates and low life
>expectancy. But the existence of a positive (rather than a negative
>correlation) between religious references and these measures of poor
>social quality certainly suggests that failure to adopt religious
>references will not lead to a decline in the quality of a society, as
>many religious people in the US have suggested.

Since causality is not indicated at all, it also follows that there is no
suggestion . that adopting religious references will lead to a decline in
the quality of a society.

Of course.

It doesn't matter how high the correlation is: it still doesn't explain why
the two variables are correlated.

Of course. Maybe I should have said what you quote above slightly
differently (though I doubt that it would have made much difference if
you are looking for religion bashing in what I say). Maybe I should
have said: the positive correlation doesn't mean that religion causes
social ills but it is also inconsistent with the possibility that
religion decreases them. I think you were working very hard to see me
implying that religion is a cause of our problems. I was very
consciously (but, apparently, almost completely unsuccessfully) trying
to make it clear that the only thing these data suggest is that
religion is irrelevant.

I knew that the unfortunate positive relationships between religion
and some social ills could lead people to conclude that I was
presenting the data to show that religion is bad for society. So I
tried to make it clear that that was not by goal by repeating, over
and over, that the positive relationships between religion and some
social ills do _not_ imply that religion causes social ills; they are
simply inconsistent with the idea that religion makes things better.
Apparently, Martin caught this but nobody else seems to have. Tough
house;-)

Again, I was presenting these data with the idea of showing one
possible way to evaluate the potential social merits of system
concepts. The data seem to eliminate religion as a set of system
concepts that _could_ possibly make society better. And some societies
clearly are better, in terms of the social ills considered in the Paul
paper, than others. So I was thinking that the next step might be to
see if one could do a survey to find some system concept that does
correlate strongly and negatively with social ills. Then one could
start doing more detailed experimental studies to see whether that
system concept can be related to ways people control that makes life
in one society better than in others.

But, given the reception to my little proposal I think I'll just go
back to the cursor control research and let someone else handle the
system concepts.

In my opinion, the religious issue is a red herring.

That was the kind of what I was trying to show. Some people advocate
adopting religious system concepts as a way to alleviate social ills.
The data suggest that religion is irrelevant to social ills; a dead
herring rather than a red one;-)

I now return control of the data-free discussion of system concepts
back to others on CSGNet.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

4

You said that I was bashing
religion by suggesting the possibility that the data indicate a causal
relationship between religion and social ills; you were also

saying that the data were not good for the very reasons that they
were

good; because they showed no relationship between religion and
social

ills.
[From Bill Powers (2008.02.01.0624 MST)]

Rick Marken (2008.01.31.1710) –

Since that is what I was saying, too, it doesn’t seem like a good reason
for an argument. I realize now, however, that there was a quite different
bee in my bonnet as well, the switch to statistical science, in which my
interest is usually about zero. I have never encountered a social system;
all I ever seem to find to interact with is other people who perceive
social systems in various ways.

Again, I was
presenting these data with the idea of showing one

possible way to evaluate the potential social merits of system

concepts. The data seem to eliminate religion as a set of system

concepts that could possibly make society
better.

OK, we can talk about that. I believe you’re referring to certain
fundamentalist Christian concepts having to do with magical relevations,
sexual morality, promises of heaven and threats of hell,
the-cross-and-the-flag superpatriot stuff, literal acceptance of the
Bible (particular selections from it, that is), and all that. I believe
that we could distill out of these general classifications the specific
contradictions of PCT principles and of science to which both you and I
would object. On the other hand, if we focused on other aspects of
religion to be found in the Bible (bind not the mouth of the kine that
treads the grain, do good to them that spitefully use you, the kingdom of
heaven is within you, judge not lest ye be judged, etc.) we could find
principles that are more in line with the ideas of PCT. There would be no
reason to reject those ideas.

You can tell a lot about a person from the parts of the Bible that the
person selects when trying to control the behavior of others and justify
the way he or she would behave anyway. For “Bible” you can
substitute “Koran”, “Bhagavad Gita,” “Popul
Vuh” or whatever is appropriate. That leaves us with the problem of
explaining why the person wants to control others (for the good or to
combat evil) and why the person wants to behave in certain ways, the same
problem we have whether or not religion is involved. These problems have
to do, in large part, with conflicts within and between people.

To find out about conflicts and do something about them, we have to deal
with one person at a time. One principle I sometimes remember to apply is
that of not ever doing science for the sake of advancing knowledge if the
process acts to the detriment of (or withholds help from) whomever or
whatever is being studied. First, do no harm. Thus I would prefer to
learn about system concepts by interacting with individuals while
exploring their internal organizations, and to learn about conflicts
while helping others to resolve theirs.

I think one reason that psychologists and others take such an indirect
statistical approach to studying behavior is that they don’t want to
influence the very thing they’re studying. By asking questions that don’t
seem to have anything to do with depression, for example, they hope to
detect and diagnose depression without letting the subject know they’re
doing so. This means, of course, that the information they get is going
to be inherently imprecise. “Do you prefer staying home and reading
to going out dancing?” is not diagnostic of any condition in
particular. Only when combined with answers to many other such indirect
culturally-biased questions would a state of sadness and discouragement
be indicated with some small degree of confidence. This absolutely forces
the psychologists to rely on statistics. This method necessarily leads to
low correlations because no one survey question is relevant excusively to
any particular problem.

The question “Are you feeling sad and discouraged?” might
reveal depression rather more reliably, but the answer might lead to more
extended discussion and, at worst, a therapy session that would
contaminate any further information to be obtained from this individual.
The person might actually stop being depressed, and damage the study. And
anyway, such direct questioning would frustrate one goal that many
psychologists seem to have: the goal of knowing more about people than
they know about themselves, and (if they’re clinicians) telling them
so.

As anyone knows who has witnessed an MOL demonstration, the method of
levels involves extremely direct questioning. One Cognitive Behavior
Therapist who is now an MOL practitioner said, on observing his first MOL
session, “You just get straight to the schema, don’t you?” An
MOL practitioner might say to the explorer, “You seem to be sad or
discouraged right now. Is that what’s going on?” Of course the
explorer might reply, “No, I just realized that I’m a little sorry
for that jerk” and the guide would accept that without arguing that
the explorer is, too, sad and discouraged and needs to stop denying
it.

Maybe, taking all religious concepts together and summing up their net
effects on social ills, religion does not improve social conditions. But
we can also separate helpful from unhelpful religious concepts, to find
those that might be candidates for system concepts that people might
consider when looking for something better in their own lives. I suspect
that doing this would pretty much eliminate religion itself as a factor,
since most people seem to think that you either accept the whole package
or none of it. But what would be wrong with accepting the specific
aspects that seem useful just because they make sense and lead where we
would like to go? That’s what we do with all ideas, isn’t it? Why reject
an idea just because some religious person said it was a good
idea?

If I’ve managed to say what I mean, you can see that there is essentially
no role for population statistics in any of this. If some population
measure seems to be valid, I immediately want to know what makes it
valid. I don’t just apply it as if it amounted to knowledge, the way an
insurance company would. I really don’t think I know of any system
concepts that would make any particular person’s life better. All I know
is what would make my life better if I and others adopted it, or what I
predict might make it better. And I don’t think that the way to find
better-for-me system concepts is to send out questionnaires and adopt the
most common answers as my own (or avoid them).

Best,

Bill P.

5

[From Rick Marken (2008.02.02.1035)]

Bill Powers (2008.02.01.0624 MST) --

Rick Marken (2008.01.31.1710) --

you were also saying that the data were not good for the very reasons
that they were good; because they showed no relationship between
religion and social ills.

Since that is what I was saying, too, it doesn't seem like a good reason
for an argument.

I agree. It was an unnecessary argument.

I realize now, however, that there was a quite different
bee in my bonnet as well, the switch to statistical science, in which my
interest is usually about zero. I have never encountered a social system;
all I ever seem to find to interact with is other people who perceive social
systems in various ways.

I agree. I think this is possibly the only major intellectual
disagreement we have. It surfaced in the discussion of economics. In
that context, you didn't like studies of measures of aggregate
behavior and I did. I agree with you that group data is emergent from
properties of the individuals in the groups. I also agree that using
group data as a basis for studying properties of individuals is wrong
(this is the mistake made by conventional psychologists who study the
average performance of groups of individuals to determine properties
of individuals in these groups). But I think it's OK to study groups
by looking at summary (statistical) characteristics of those groups,
and I think you don't. Which is fine with me.

Thus I would prefer to learn
about system concepts by interacting with individuals while exploring their
internal organizations, and to learn about conflicts while helping others to
resolve theirs.

Yes, you want to take a "bottom up" approach, like what you want to do
in economics. I am willing to try to study and model group level
phenomena at the group level. For example, at the group level there
are clear differences between societies in terms of things like
homicide rate, teen abortion rate, etc. I am willing to try analyze
these things at the group level, in terms of modeling and determining
relationships between group level variables.

I think we have learned a lot about electricity but studying it at the
group level. Ohm's law, for example, is worth knowing even though it
is emergent from properties of individual atoms and electrons. I was
just at a concert yesterday and I realized that the music from the
orchestra could be valuated at the group level (I could compare the LA
Phil's performance to the NY Phil's) even though I know that the group
result is emergent from the properties of the individual members of
the orchestra.

I think one reason that psychologists and others take such an indirect
statistical approach to studying behavior is that they don't want to
influence the very thing they're studying.

I agree. I think they also do it for other reasons that you have also
mentioned. For example, psychologists use groups in research because
they think they are "averaging out" all the apparent random variation
in behavior when they do this. Psychologists are trying to understand
the nature of individuals by studying the behavior of groups, and this
is a mistkae. But I think it is not a mistake to try to understand the
nature of groups by studying the behavior of groups. I think we
disagree about this.

Maybe, taking all religious concepts together and summing up their net
effects on social ills, religion does not improve social conditions. But we
can also separate helpful from unhelpful religious concepts, to find those
that might be candidates for system concepts that people might consider when
looking for something better in their own lives.

I agree. And that is basically what I would suggest as the next step
in the group level research: see if one can come up with possible
system concepts -- possibly derived from looking at particular
religious tenets -- that might be strongly associated with differences
in various social measures across countries.

If I've managed to say what I mean, you can see that there is essentially
no role for population statistics in any of this.

Right. There is no role for it given your goals. But there is a role
for it, given my interest in understanding the behavior of the
populations themselves.

If some population measure seems to be valid, I immediately want to know
what makes it valid.

What makes voltage, resistance and current valid population measures
of electricity? At the group (population) level the question doesn't
seem to me to be one of "validity". That's a question conventional
psychologist's would ask because they are using the population measure
to study individuals. I'm interested in the population measures in
themselves, as measures of characteristics of the population. The
"goodness" (not validity) of the measures would be evaluated in terms
of how strongly the measures relate to one another. And once we (well,
I) have a good group level model -- good in the sense that it fits the
group level data extremely well -- then the goodness of the measures
is simply the fact that they are the measures that are used in the
model.

I really don't think I know of any system concepts that would make
any particular person's life better. All I know is what would make my life
better if I and others adopted it, or what I predict might make it better.
And I don't think that the way to find better-for-me system concepts is to
send out questionnaires and adopt the most common answers as my own (or
avoid them).

I agree!

Best regards

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

6

[From Fred Nickols (2008.02.02.1144 MT)]

[From Rick Marken (2008.02.02.1035)]

<snip>

But I think it is not a mistake to try to understand the
nature of groups by studying the behavior of groups. I think we
disagree about this.

Rick: Would you please give an example of group behavior?

···

--
Fred Nickols
Toolmaker to Knowledge Workers
www.skullworks.com
nickols@att.net

7

[From Rick Marken (2008.02.02.1140)]

Fred Nickols (2008.02.02.1144 MT)--

Would you please give an example of group behavior?

Actually, I already gave one: Music created by a group of individuals.
A symphony results from the combined efforts of several individuals.
No group, no symphony. This is an example of group behavior where the
result (symphony) is probably intended by each individual and can only
occur if everyone acts in concert. So the symphony -- the combination
of all the orchestral parts -- is a controlled result of the group.
Other examples of this kind of group behavior, which I would call
"cooperative control", are buildings, roads, universities, square
dances, etc.

Another kind of group behavior occurs when the group result is not
intended but is directly related to the fact that individuals are
controlling particular variables. I call this "collective control". An
example of this kind of group behavior is GDP, which is a result of
individuals or smaller groups of individuals controlling for producing
the stuff they want. Unlike the case of an orchestra, where the
symphony is the intended result, GDP is not an intended result but is,
nevertheless, a direct result of individuals producing particular
intended results -- the production of goods and services for
themselves and or in order to get money to purchase the goods and
services of others.

A third kind of group behavior occurs when the group result is a side
effect of the controlling done by a group of individuals Perhaps this
kind of group behavior could be called "side effect of collective
control". An example is carbon dioxide emission levels, which is a
side effect of people controlling, not for the production of C02, but
for keeping warm or running the stuff they are controlling.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

8

[Martin Taylor 2008.02.02]

[From Bill Powers (2008.02.01.0624 MST)] replying to

Rick Marken (2008.01.31.1710) --

I realize now, however, that there was a quite different bee in my bonnet as well, the switch to statistical science, in which my interest is usually about zero. I have never encountered a social system; all I ever seem to find to interact with is other people who perceive social systems in various ways.

Rick has given his own repsonse to this [From Rick Marken (2008.02.02.1035)]. I agree with much of what Rick said, but I also have a different slant on the issue.

"I have never encountered a social system; all I ever seem to find to interact with is other people who perceive social systems in various ways."

Likewise, I have never encountered a feedback loop; all I have ever encountered is a bunch of stimulus-response functions that influence each other in various ways.

Isn't a control system a social system of stimulus-response functions? Don't feedback loops sometimes pass through the control systems in two or more people? Aren't such feedback loops "social systems"? When the side-effects of one person's control affect the behaviour of another's environmental feedback function, and vice-versa, making each control more effectively, isn't that a different kind of feedback loop that can occur only as part of a social system?

When a child learns the language and culture of those who live and play with it, by reorganizing so that its verbalizations, gestures, and behaviour are effective in controlling its own perceptions, is it not part of a social system?

Your comment reminds me of the autistic child who sees no lawn, but sees only a myriad blades of grass.

As for the use of statistics, we know well that it doesn't interest you, but that does not mean that it is useless or that its use is counterproductive. The interpretation of statistics can indeed be misleading, but then so can the interpretation of any statement. One just has to be careful in judging what claims are warranted.

Martin