standards as conventions

CSGnet Archive CSG1992
1

[From Rick Marken 920612 11:00)

Martin Taylor (920607 1710) suggests "standards" should be viewed
as conventions that make it easier to cooperate.

It seems to me that standards
allow you to pre-empt a possibly painful random reorganization by permitting
you to set references that are appropriate if the other behaves in a
conventionalized way--according to standards.

I agree that there is much to be gained from conventionalized behavior.
This is particularly true in the technological world were it helps
enormosly to design systems that have a standard response to actions.
Thus we can be pretty confident that a clockwise turn will result
in the screw going "in" or the power going "on" or "increasing".
What we tend to conventionalize is the feedback function, g(o), that
relates out outputs to our inputs.

If there are any "absolute" standards, they will be those that have allowed
the social groups using them to survive and prosper.

Conventional standards (like the clockwise turn standards) can be "absolute"
to the extent that we can get all objects to abide by this convention.
This can be done in principle -- though it's difficult (and sometimes not
desired) in practice; some people may have a need for a counter-clockwise
"in" device. But the goal of absolute standards (conventions) is at least
feasible for inanimate objects because these objects have no purposes of
their own that might conflict with the convention. Such is not the case with
living systems.

We can't do without standards in a time-limited social world.

Indeed, people adopt (and require) conventional ways acting for the
same reason that they produce artifacts that have conventional ways of
acting; it makes control simpler -- with people, it is probably what
makes cooperative control feasible at all (which I think is the point
of your sentence above).

The problem is that people are not inanimate objects -- and certain
individuals in certain circumstances may find that acting according
to a particular convention is impossible -- not because the person
is bad or contrary or immoral -- but because he or she is a hierarchical
control system that simply cannot act like the knob on a radio. So
my argument against "absolute" standards applies as much to standards as
social conventions (like grammars) as it does to standards as moral principles.
I am all for standards as conventions. I'm just saying that the notion
of absolute standards -- no matter how technologically and socially
helpful their existence might be -- are simply inconsistent with
human nature (if people are hierarchically organized perceptual
control systems). This does not mean that I believe everybody should
just go ff and do their own thing. I'm just saying that this fact about
human nature must be taken into consideration when we think about how
people can act cooperatively.

The people who want there to be absolute standards are not "bad" people
(from my point of view). The desire for absolutes is quite reasonable --
they are like people who say that "the lights should always go on
when I flip the switch 'up' and off when I switch the switch 'down'".
I can understand their desire -- especially with respect to people; people
should never kill each other or end a sentence with a preposition; people
want predictability. All I'm saying is that people are not switches;
they cannot abide by such absolute conventions even if they try.
This does not mean that social chaos is inevitable (homefully, a bit less
than what we now have); what I think it means is that we have to find ways to
cooperate that take into account the true nature of human control
systems. The fact that cooperation is possible in the context of this
reality (the inability of control systems to control relative to absolute
conventions) is evidenced (I think) by the general spirit of cooperation
found despite the diversity (in terms of many conventions) among
members of the Control System Group itself. It can be done.

Best regards

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)

2

[From Rick Marken (920612.1330)]

Martin Taylor (920612 17:00) says:

I wonder if anyone was equating "standard" with "reference value".

I thought you were proposing that "standards" be understood as a
convention for behavior. For example, there is a convention in
your country and mine that we drive on the right. So when I am on
a road I try to keep my car in a lane to the right of the center line.
This means (from a PCT point of view) that I set my reference for the
relationship between car and center line at "right of" rather than some
other value -- like "on" or "left of" of "perpendicular to". I
was agreeing that standards of this sort are quite useful for
successful social interaction.

intended was to suggest that a "standard" provides a convenient level at
which a reference value can be set, one that has often been found (perhaps
by other people over history) to result in a desirable percept.

Well, I agree, except that I think many of these standards are fairly
arbitrary (like which side you drive on) -- they work as long as there
is agreement among those that need to abide by them in order to aviod
interpersonal conflict.

But even with "absolute standards", there's no compulsion on anyone actually
to use them as reference values.

Well, there is some social coersion. People can have unpleasant run ins with
the LAPD out here if they pick the wrong side to drive on. Of course, one
is still under no compulsion to set their reference at the conventional level
since they are the one's setting it.

The existence of absolute standards depends on whether over evolutionary time
certain behaviours (in the PCT sense) have benefited the survival and
gene-propagation of the people (or others) using those behaviours. If they
have, then either by gene transmission or by social transmission, the
ordinarily effective behaviours will result in absolute standards.

If by "behaviors" you mean "references for certain inputs" then I agree;
there may be absolute (fixed, built into the individual, unvarying) references
for certain inputs. Such references are almost certainly at the cellular,
if not the genetic, level -- and they are what PCT fans would call "intrinsic
references". If, however, by "behaviors" you mean particular actions, then
I don't see how this can be correct; evolution could not possibly select
for actions that would have to produce their effects in a disturbance prone
environment. I think a lot of socio biology sounds like it imagines that
certain behaviors (in terms of actions) can evolve; for example, they
talk about evolution of "aggression". It sounds like they are talking about
the evolution of a certain visible patterns of outputs. I think the only
thing that might be able to evolve is a preference for a certain level of
sensory input resulting from these (and/or other) actions.

    But "absolute standard" as "that's what people have learned as a
usually effective way to behave" is simply a practical concept that improves
social interaction.

It sounds like you are saying that an "absolute standard" is only relatively
absolute (it is usually effective at improving social interaction, but not
always). If this is what is usually meant by "absolute standard" then it
turns out that I have been advocating a version of this approach to "absolute
standards" all along. I've just been saying that some standards are usually
effective for lots of people -- but not always (they don't work for some
of the people some of the time). I just wish some of the others
in the discussion of absolute standards would have clarified this point
for me. Does this mean that the 10 commandments are "absolute standards" in
your sense of absolute standards -- it is usually effective to not steal,
but not always? Is that what judeo christians think god meant? What
about that first one -- thou shalt have no other god before me -- usually?
Some people got stewed for not obeying that one. Are some standards more
absolute than others?

No wonder we need people like Dan Qualye to help us out with this stuff.
It gets really complicated.

Best regards

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)

3

[Martin Taylor 920612 17:00]
(Rick Marken 920612 11:00)

The problem is that people are not inanimate objects -- and certain
individuals in certain circumstances may find that acting according
to a particular convention is impossible -- not because the person
is bad or contrary or immoral -- but because he or she is a hierarchical
control system that simply cannot act like the knob on a radio. So
my argument against "absolute" standards applies as much to standards as
social conventions (like grammars) as it does to standards as moral principles.
I am all for standards as conventions. I'm just saying that the notion
of absolute standards -- no matter how technologically and socially
helpful their existence might be -- are simply inconsistent with
human nature (if people are hierarchically organized perceptual
control systems).

I wonder if anyone was equating "standard" with "reference value". I wasn't,
and if my posting gave that impression, I'll try to correct it. What I
intended was to suggest that a "standard" provides a convenient level at
which a reference value can be set, one that has often been found (perhaps
by other people over history) to result in a desirable percept. In function,
it's like the memory of a percept within an ECS.

But even with "absolute standards", there's no compulsion on anyone actually
to use them as reference values. As Rick says, such use may conflict with the
ability to achieve other reference values. Some day, you may have to try to
kill someone if you are to maintain other desired percepts, such as personal
survival or frredom.

The existence of absolute standards depends on whether over evolutionary time
certain behaviours (in the PCT sense) have benefited the survival and
gene-propagation of the people (or others) using those behaviours. If they
have, then either by gene transmission or by social transmission, the
ordinarily effective behaviours will result in absolute standards. (On social
transmission, see R. Boyd and P.J.Richardson, Culture and the evolutionary
process, U of Chicago Press, 1985).

I find no moral connotation to the idea of "standard", whether absolute or not.
The idea of "absolute standard" as "you have to do what I say is right" is,
I think, morally and practically repugnant, for many of the reasons adduced
by Rick. But "absolute standard" as "that's what people have learned as a
usually effective way to behave" is simply a practical concept that improves
social interaction.

Martin