[From Bill Powers (921009.0800)]
Greg Williams (921009) --
Do you disagree that the actions taken by a person at time t1 which
are necessary for successful control by that person at time t1
could result in unsuccessful control (and hence Bill-significance)
at time t2?
I don't disagree. In that case, the action DOES affect something else
that the person is controlling for, and creates loss of control. After
a person has done this for a few years, presumably some reorganization
will take place and the person will learn to control for predicted
perceptions. Of course if you prevent the person from doing the
improvident act, no learning will take place.
Do you disagree that the actions taken by a person at time t1 which
are necessary for successful control by that person at time t1
could result in improved control [at time t2]...?
Nope, don't disagree with that, either. This kind of control would
belong at about the program level, wouldn't it? Or perhaps principles
(saving for the future). If the improvement at time t2 were not
intended, however, it would just be a lucky accident, and could have
turned out oppositely.
In the first example above, if the ("kindly") storekeeper refused >to
take the person's credit card (t1), judging (quite >patronizingly)
that this particular assemblyline worker really >shouldn't buy a $400
pair of shoes, then that "external influence" >would be Bill-
significant at t1 and Greg-significant at t2, >wouldn't it?
If the customer doesn't reorganize, he/she will either protest loudly
and violently at the discrimination or find another store that will
sell the shoes. If that doesn't work, the customer still won't control
for saving for the future; evidently there's no reference level for
doing that, and a storekeeper or storekeeper conspiracy can't install
one. So the customer will blow the money on something else. It's very
unlikely that by the time t2 arrives, there will be any money left for
the trip, no matter what the storekeeper does. That's not what the
customer is controlling for.
To which I add: external variables capable of facilitating
successful control (allowing it where otherwise it would be
impossible) -- significant for a third quite different reason.
You can't facilitate a control system that doesn't exist. If the
control system does exist and isn't succeeding, the person is probably
reorganizing and will eventually succeed or change the goal. If the
control system is succeeding, then any direct help will be resisted.
If you facilitate control by making the environment more amenable to
control, then reorganization will stop and the struggling control
system will succeed. The next time the situation comes up, you will
probably get a phone call asking if you would kindly set up the
environment that way again, please. The third time, the phone message
will be "Ready whenever you are." The fourth time it will be "I'll be
needing you at 10:00, and don't be late." All you've done is to make
yourself part of someone else's control loop.
Are you saying that controlled variables do depend on the
environment in important ways? What ways, when control is
successful?
Historically. Environmental influences prior to time t1 can affect
the trajectory of controlled variables after time t1.
I disagree totally. The trajectories of controlled variables are the
trajectories of their reference signals, not of environmental
disturbances of those variables. The whole point of controlling a
variable is to render it independent of external influences and make
it depend only on the internal desire that it be in a particular
state. If you want to wash a cup, its previous history makes no
difference at all. It gets washed, even if it's clean, and whether you
pick it up or someone tosses it to you.
When the hierarchy is operating properly, all perceptions at all
levels remain close to their respective reference signals: they
remain under control.
But the hierarchy (or, more generally, control organization)
CHANGES over time. You think those changes aren't due in any way to
the environment; I think the environment plays an important role.
"Due in any way to the environment" is backing down quite a lot, isn't
it? Obviously the changes would be drastically affected by the
environment if its mean temperature dropped to -100 c. If that's the
kind of thing you want me to "admit", then OK, no problem. The
organism has to learn to control its perceptions in the world that
exists. It has to find one of the actions that will have the required
effects on a perception in order to control it. It even has to find
perceptions that are (a) controllable, and (b) relevant to maintaining
critical variables at their reference levels. All this requires that
the properties of and events in the environment be taken into account.
If the only eating implements available are a knife and a fork, and
the person thinks it impolite to eat with the fingers, the soup will
be eaten rather slowly or the person will say to hell with it and
slurp out of the soup-bowl. That's because of the environment. But the
environment didn't tell the person to eat the soup.
This is nothing that I haven't said dozens of times, but you seem to
pay no attention and you put words in my mouth to the effect that the
environment doesn't play any role at all. It does play a role: it's
there and it behaves according to its properties, according to what
you do to it. It does things all by itself, which is how disturbances
arise. If the environment changes in some basic way, the organism has
to reorganize so as to maintain control of it, and the new
organization will one of those that compensates for the change.
My whole point is that none of the changes in the organism would occur
if the organism itself didn't actively make those changes, for its own
purposes.
I wonder if you practice automotive "preventive maintenance"?
Yes, some of my perceptions actually encompass an imagined future,
which I control. Of course after I've faithfully changed the oil 20
times, the water pump fails. The imagined future is only imagined. Do
you do preventive maintenance on your water pump?
Suffice it to say that there are many people who (patronizingly or
not) believe in keeping drunks from behind wheels, and so forth, to
attempt to PREVENT the disruption of control by people.
Fine, they're controlling their own perceptions, aren't they? You may
have quite a struggle in getting the car keys away from the drunk, but
being sober and I hope bigger, you will win the conflict and have it
your way. The next time, the drunk will have a spare set of keys in
his sock.
Now you're getting into what is OBJECTIVELY good for people. This is
how most major conflicts between people are created.
Are you REALLY pleading for a completely "hands-off" approach to
regulating social interactions? Should the police wait until AFTER
the wreck to arrest the drunk?
I seem to recall saying that if my kid runs into traffic, I grab the
kid and get it out of the street, and to hell with avoiding conflict.
You do what you think is right. We all do, including Saddam Hussein.
If you strongly want people to behave in a certain way, you make laws
and establish a police FORCE. You do whatever it takes to get what you
want, as long as all your goals are satisfied, including your goals
for how people should relate to each other and what kind of society
you want to live in. I don't know where you got the idea that I'm
pleading for a completely hands-off approach to social interactions.
I'm only pointing out who is responsible for what. If you decide to
drag the drunk out of the car, that's your decision carried out to fit
your own goals. The environment didn't make you do that; the drunk
didn't make you do that. You did it because of what you believe, what
you predict, and what you want to happen or not happen; that's your
responsibility. You did it as part of making the world be the way you
want to perceive it. You can, of course, argue that the drunk and his
or her family is better off for what you did; that argument, too, is
your responsibility and reflects your convictions and goals. You can't
disguise your responsibility by saying "Yes, but he REALLY would have
killed himself and orphaned his children!" That's how you understand
it. It's your goal that he not be killed and that his children not be
orphaned and his wife widowed. Your goal, inside you. You can use
logic, statistics, persuasion, ridicule, hypnotism, violence, appeals
to right and wrong, quantum mechanics, or whatever tools you command
to achieve your goal and to get others to seek the same goal, and it
will still be your own goal and yours alone. You can't say "the
environment made me do it." The environment is just what it is.
Is there a problem in the mind of an observer who predicts that the
presently successfully controlling TV watcher is going to have
problems controlling in the future which are due in part to his/her
current ACTIONS and attempts to alter how the person ACTS, rather
than what the person is controlling for?
Not if this control process succeeds. But it's still the controller
arranging things to suit himself or herself on the basis of private
beliefs and goals. Unfortunately, it will be difficult to change how
the TV watcher ACTS without disrupting the perception under control,
namely, seeing the program on the TV. The result will probably be
conflict, which may create problems. The conflict is that the watcher
wants to be watching TV, and the controller wants the watcher NOT to
be watching TV. If conflict is to be avoided, the watcher has to
change the goal, not the ACTION of sitting down and turning on the TV
and directing the bloodshot eyes to it. The controller can't change
the action without disturbing the controlled variable. At best, the
controller can try to create a conflict and hope it will be resolved
as the controller wants: "OK, I'm driving into town to see the circus,
anybody want to come along?" Most likely, the dedicated controller
will walk up to the TV set, turn it off, and announce "That's all for
today." Implying, "and there's nothing you can do about it, kid."
Except go to a friend's house where the TV is on 24 hours a day.
And you certainly can't fix getting killed by a drunk
driver after you're already dead. So you attempt to alter the
drunk's ACTIONS beforehand -- you don't try to alter his/her belief
than drunk-driving is a God-given right! -- and deliberately
conflict his/her control system by throwing him/her in jail. Then
you give him/her the choice of alcoholism treatment or no driver's
license.
Right. Overwhelming physical force always works as a way of making
other people satisfy your goals, if you have the necessary resources
and are willing to watch your back from then on.
...if you ... want [benefits] when it's too late to get
them, perhaps because a con artist has successfully controlled for
certain of his/her perceptions dependent on certain of your
actions, or perhaps because your teacher was a lousy communicator,
or perhaps because your parents believed that you should do
whatever you want to do and not what they think is best, too bad!
Yep, too bad. You can't control what you're not controlling for.
That's how it works. If you don't learn from experience you're going
to get conned over and over. If you had a bad education it's going to
stay bad until you decide to continue it. If your parents let you do
anything you wanted, you're going to want a lot a weird things unless
you learn from experience that this doesn't get you what you want. If
your parents make you do what they think is best, and it's not best,
you're still screwed. That's life.
Do you prefer telling the widow of the drunk driver's victim that
she's just going to have to regain control of her perceptions now
that he husband is gone to telling the drunk to get into the police
car OR ELSE?
Of course not. I'd try to keep the drunk out of the car and see that
he got safely home. I might try to steer him to AA on the way. There
are lots of things I would try to do because of my own goals for the
kind of world I want and the kind of person I want to be in it. If the
drunk did kill himself anyway, which is likely, I'd tell the widow
that he is really dead and she must adjust her perceptions to fit that
fact, and make sure her kids know how he killed himself.
···
------------------------------------------------------------------
Control theory fits the world as it is, and people the way they are.
But an understanding of control theory leads to assigning causality
differently, and as a result will change the way people do things and
the reasons for which they do them. A resort to arbitrary control of
others by means that rest ultimately on violence will be seen as a
defeat, not an accomplishment. A viable steady-state society clearly
can't be organized on those principles, even though they work in the
short term and are all we have available now. Even if you use them and
think they're working just fine. When everyone realizes this, a
different organization of society will result. If it doesn't, I'll
kill everybody, and THAT will teach them.
------------------------------------------------------------------
Best,
Bill P.