# Where are the sociologists?

**URL:** <http://discourse.iapct.org/t/where-are-the-sociologists/12407>\
**Category:** CSG1993\
**Created:** [June 18, 1993, 10:34pm UTC](http://discourse.iapct.org/t/where-are-the-sociologists/12407 "1993-06-18T22:34:40Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Marken\_Richard\_S1](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/m/977dab/32.png) [@Marken\_Richard\_S1](http://discourse.iapct.org/u/Marken_Richard_S1)\
**Post date:** [June 18, 1993, 10:34pm UTC](http://discourse.iapct.org/t/where-are-the-sociologists/12407/1 "1993-06-18T22:34:40Z")

</div>

[From Rick Marken (930618.1500)]

After reading the posts from Bill Powers (930618.1000 MDT) and  
Tom Bourbon (930618.0958) on social control, I have been re-  
considering my comments on this topic. I had been assuming that  
there is such a thing as a "collective controlled variable" which  
is a variable that is not controlled by any individual but is a  
result of control by the collective. My example of a collective  
controlled variable was the three line pattern made cooperatively  
by the subject's in Tom's cooperation experiment. After thinking  
about this for a bit I realize that I am wrong -- there is no such  
thing as a collective controlled variable . A controlled variable  
ONLY exists if there is a perceptual signal in some system that  
represents that variable, a reference signal that specifies the  
reference level of that signal and an output system that can affect the  
perceptual signal. In Tom's task, the three line pattern is a controlled  
variable ONLY IF one or the other or both subjects are perceiving and  
controlling that overall pattern. Any disturbance to the pattern  
would be resisted (if possible) by the person (or persons) controlling  
it. The three line pattern would APPEAR to be controlled, however,  
even if each person were controlling just a piece of it -- the relation-  
ship between left and middle line for one subject, the relationship between  
middle and right line for the other. It might seem like the three line pattern  
is controlled in this case but it is NOT -- which could be proved if the  
subjects were allowed to see and control only their two line patterns  
with the three line result projected on another screen. Disturbances  
to the three line pattern that did not affect each subject's two line  
controlled variable would NOT be resisted.

So Hans' and my idea that "flockness" is under control (a "collective  
controlled variable" by my terminology -- ie. a variable that is  
controlled by the collective but not by any individual) is incorrect unless  
there is a control system present that is perceiving and acting to influence  
this variable.

Where are the sociologists during this discussion of social control, by  
the way? This is where PCT meets sociology. One of the goals of  
sociological research, it seems to me, would be to provide formal means  
for discriminating emergent (uncontrolled) social variables (like the  
flocking of birds) from controlled social variables (like the three  
line arrangement in Tom's cooperation experiment--assuming that that  
IS a controlled social variable). The difference, I think, is that there is  
no perceptual representation of an emergent social variable in any  
control system that has an influence on that variable; there is a  
perceptual representation of a controlled social variable in at least  
one of the control systems that influences that variable.

Best

Rick

---

<div class="post-metadata">

**Author:** ![Tom\_BOURBON3](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/t/85e7bf/32.png) [@Tom\_BOURBON3](http://discourse.iapct.org/u/Tom_BOURBON3)\
**Post date:** [June 21, 1993, 6:11pm UTC](http://discourse.iapct.org/t/where-are-the-sociologists/12407/2 "1993-06-21T18:11:04Z")

</div>

Tom Bourbon (930621.1301)

> [From Rick Marken (930618.1500)]
> 
> After reading the posts from Bill Powers (930618.1000 MDT) and  
> Tom Bourbon (930618.0958) on social control, I have been re-  
> considering my comments on this topic. I had been assuming that  
> there is such a thing as a "collective controlled variable" which  
> is a variable that is not controlled by any individual but is a  
> result of control by the collective. My example of a collective  
> controlled variable was the three line pattern made cooperatively  
> by the subject's in Tom's cooperation experiment. After thinking  
> about this for a bit I realize that I am wrong -- there is no such  
> thing as a collective controlled variable. ...

Glad you have seen the light, brother! I thought you were going a little  
too far in your interpretation of the three-line pattern. It \*is\* a  
controlled variable (but not a collective one), some times, but not others.  
(More on that in a while.)

> ... A controlled variable  
> ONLY exists if there is a perceptual signal in some system that  
> represents that variable, a reference signal that specifies the  
> reference level of that signal and an output system that can affect the  
> perceptual signal. In Tom's task, the three line pattern is a controlled  
> variable ONLY IF one or the other or both subjects are perceiving and  
> controlling that overall pattern. Any disturbance to the pattern  
> would be resisted (if possible) by the person (or persons) controlling  
> it. The three line pattern would APPEAR to be controlled, however,  
> even if each person were controlling just a piece of it -- the relation-  
> ship between left and middle line for one subject, the relationship between  
> middle and right line for the other. It might seem like the three line pattern  
> is controlled in this case but it is NOT -- which could be proved if the  
> subjects were allowed to see and control only their two line patterns  
> with the three line result projected on another screen. Disturbances  
> to the three line pattern that did not affect each subject's two line  
> controlled variable would NOT be resisted.

For anyone who has not seen a demonstration of the task, or who missed my  
earlier detailed post, here is a diagram of the connections in Boss Reality  
(the innards and environs of a PC):

Position of-------------------------\>| Position of Left Cursor  
Left Handle |  
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\|/  
Magnitude of-------------\>ADD----------\>| Position of Middle Cursor  
Disturbance /|\  
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\>  
Position of--------------------------\>| Position of Right Cursor  
Right Handle

The program steps that determine where each of the three cursors appear  
on the screen during a particular update of screen information are:

Position of Left Cursor := Scaled Position of Left Handle;

Position of Right Cursor := Scaled Position of Left Handle;

Position of Middle Cursor := Sum of Momentary Value of Random Disturbance  
&nbsp;&nbsp;&nbsp;plus Scaled Position of Left Handle plus Scaled Position of Left Handle.

That is the set of connections in one small part of the world. Nothing in  
the connections specifies everything that will happen when one or two  
persons use the handles.

As part of a semi-private discussion off of the net, Rick had said:

> **···**
>
> ========================================
> 
> [Rick]
> 
> > I was trying to read Hans as saying that a controlled variable  
> > could be the cooperative result of controlling by 2 or more  
> > control systems. I think that's what is happening in Tom's  
> > experiment -- is it not? The pattern of the three lines is, indeed,  
> > a controlled variable, and control of that variable requires that  
> > the two parties involved control -- and control a particular  
> > variable each. So there is a controlled variable that is the result  
> > of collective action. Of course, the two people don't become a  
> > control system, but their individual controlling controls the  
> > position of the three lines.
> 
> [Tom]  
> Yes, and they can select another pattern of the three lines, in which case  
> each person will necessarily adopt a different reference perception for part  
> of the new intended pattern.
> 
> [Rick]
> 
> > I was thinking that, by my definition of flockess as the set of between  
> > individual distances, flockness is, indeed, controlled. The set of  
> > distances are controlled, and they are controlled at values which  
> > depend on the controlling done by each individual.
> 
> [Tom]  
> But, unlike my people, who KNOW they are "making a pattern of three lines,"  
> members of a flock need not have such an awareness or intention concerning  
> "being in a flock."
> 
> [Rick]
> 
> > This is where I might be wrong; maybe by that definition "flocking" is  
> > still emergent.
> 
> [Tom]  
> See above.
> 
> [Rick]
> 
> > ... Is the line pattern in Tom's experiment a controlled  
> > variable? Does control of that variable depend on the controlling done  
> > by a collective? Is flocking a controlled variable by my definition?
> 
> [Tom]  
> What if I do the task with my own two hands? Or what if I use a moused  
> version of the task and do it with one hand? Or a mouse-on-a-stick (I  
> intend to patent and franchise!) and do it with one joint (this is starting  
> to sound REALLY good!)? Is the line still controlled? Can we draw a line  
> (a division) between any of these permutations and the two-person  
> cooperative task?
> 
> In all of the one-person variations, there is no doubt the line is  
> controlled. In the two-person runs, my two people have always known the  
> "superordinate" goal. But what if one person knew the big picture and only  
> told the second person, "just keep this line even with that one." Would the  
> second person be participating in "keeping three lines in a row?"
> 
> The REAL social factors begin to pop out, for me, when I think of the  
> \*entire set\* of permutations. Some are social, some are not. In some, the  
> big picture is controlled, in others, not so for all participants.
> 
> =================================
> 
> Back to the present:
> 
> Whether three-in-a-row is a controlled variable, or an unintended  
> by-product of control of other variables, depends on the intentions and  
> perceptions of the control systems. This seemingly innocuous task is rich  
> with implications for several presumably independent areas of study,  
> ranging from social science on one extreme to neuroscience on another.
> 
> When one person performs (say the person is me -- TB) "I" control the three  
> lines and keep them in a row - - -, or in some other pattern of "my"  
> chosing: \_  
> &nbsp;&nbsp;&nbsp;\_ \_ \_ \_  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\_ or \_ \_ or \_ ...
> 
> My actions control two different relationships (middle re right; and middle  
> re left) that are necessary to the production of the "big" pattern. But  
> where do "I" fit in to the picture? Where am I, neurologically? "I" have a  
> clear understanding of what I intend, and of whether I see what I intend. I  
> know that I can achieve the intended result by moving the handles, or the  
> mouse, depending on the version of the program that is running. But by  
> conventional neurological theory, the brain areas responsible for producing  
> the appropriate movements differ across the various one-person  
> implementations mentioned in the post to Rick. Two completely  
> independent control systems can achieve the same results that "I" achieve --  
> two simple single-loop PCT models can do it. What, and where, "I" am becomes  
> less and less clear to me when I think about these demonstrations. (Hans, I  
> usually go out of my way to say that a person can be represented by two  
> independent PCT models, not that a person is two such systems. The same  
> style applies when there are two people, and two models. The people can be  
> modeled by, or as if they were, two exceedingly simple PCT systems.)
> 
> When two people perform the task, the results can be identical to those  
> produced by one person, but for different reasons. Each person controls  
> the relationship between two of the lines, and each might know and try to  
> control a perception of the three lines in a particular relationship. Or  
> each might control only the relationship between two of the lines and neither  
> know or care about the relationship among all three. Or one person might  
> know and control the relationship among the three lines, by directly  
> controlling the relationship between two, and by telling the other person to  
> control the relationship between two -- the second person remaining ignorant  
> of the first person's overall goal.
> 
> Only when a person who acts alone, or at least one of two people who act  
> together, perceives and adopts an intended value for  
> three-lines-in-a-particular-configuration does line-ness become a controlled  
> variable. In all other cases, it is an unintended, perhaps unperceived, side  
> effect of the control of other variables.
> 
> The same is true of flockness.
> 
> Of course, given an organism, or model, or artificial control device with  
> sufficiently complex perceptual functions, a previously uncontrolled  
> variable can be made the object of control. Similarly, given an assemblage  
> of such systems or organisms, a peviously uncontrolled (incidentally  
> affected) variable might become the object of collective control. (Hans, if  
> you want to study the individual, social, and evolutionary  
> development of control systems, I suggest this simple cooperative tracking  
> task to you. When you can explain all of its permutations and  
> implications, you should have gone a long way toward your intended end.  
> And along the way, non-engineer PCTers might be able to pick the brains of  
> a control engineer.)
> 
> I can offer an example of "emergent" collective control. A few years ago,  
> during a meeting of the American Society for Cybernetics, a group of CSTers  
> (the earlier name for PCTers) crowded together in an elevator, on the way to  
> dinner. Never before had so many CST-PCTers been assembled in any place,  
> much less one so small as the elevator. Someone (Mary Powers, I believe) in  
> the group (the gaggle? the flock?) remarked about how many of us there were.  
> Everyone else agreed -- never before had there been enough of us to fill an  
> elevator. Not long after that, we began to talk about having our own  
> meetings. We did, and CSG was born. And CSG begat CSG-L. And here we are.
> 
> Identifying our flockness turned it into a controlled variable, for each of  
> us. From something that had never been a controlled variable, we created  
> one. We continue to do so, in spite of difficulties or complications  
> that might otherwise disturb our individual perceptions of  
> CSG-all-in-one-place. But when it happens, it happens one of us at a  
> time. There is no CSG-in-the-sky that uses each of us to create the  
> perception of the CSG flock, the way any one of us can set the reference  
> signals for the characters in CROWD-GATHER and control our perceptions of  
> characters in a gaggle. (Not unless Bill has installed some pretty  
> interesting equipment on his hill in Colorado! Maybe he uses the net to  
> reset the references every year around this time.) In a few weeks, we will  
> make it happen again. Hans, why don't you come and see it happen?
> 
> [Rick]
> 
> > So Hans' and my idea that "flockness" is under control (a "collective  
> > controlled variable" by my terminology -- ie. a variable that is  
> > controlled by the collective but not by any individual) is incorrect unless  
> > there is a control system present that is perceiving and acting to influence  
> > this variable.
> 
> I second you on this. This is one of the points I made to Hans in my post  
> (939618.0958) when I said Hans, but not the heating system, controls "a  
> nice temperature" and that Hans, but not the characters in GATHER-CROWD,  
> controls a perception of "flockness." And now when individual CSG people  
> create their own perceptions of the meeting.
> 
> [Rick]
> 
> > Where are the sociologists during this discussion of social control, by  
> > the way? This is where PCT meets sociology. ...
> 
> I have been wondering the same thing. Has everyone gone on vacation?
> 
> Until later,  
> &nbsp;&nbsp;Tom Bourbon