Cultural patterns

[From Bill Powers (2003.03.27.0845 MST)]

Bruce Nevin (2003.03.26 14:34 EST)--

>Variables of type 2 are not ordinarily controlled variables, although they
>are collective outcomes of individual control. True, it is possible for an
>individual to perceive and (attempt to) control e.g. a ring in a crowd, but
>this is unusual and not characteristic of them.

Check. These are variables an observer notices, which are actually
irrelevant to the operation of the collective system since they would not
be restored if a disturbance distorted them. It would be useful to know
which patterns that sociologists have observed belong in this class.

Variables of type 3 are not controlled variables, although they are
collective outcomes of cooperative or communicative control and in at least
some cases (e.g. the sound pattern of a language) they appear to be
necessary for successful cooperation and communication.

The first sentence just above doesn't seem to differ from the description
of type 2 variables, the only difference being in what is under control.
Type 3 appears to be a subset of type 2. I think that there could be
successful cooperation and communication in an infinity of different
universes in which the means of communication differed subtly or radically.
Look at the variety of methods human beings have invented for achieving
communication and cooperation, which take advantage of just about every
possible way of transmitting and receiving perceptions used as symnbols. If
one can conceive of cooperation and communication, then to achieve them one
uses whatever means is available. I'm sure the means that are available
have a strong influence on the structure of communication, and perhaps bear
on cooperation as well, but that's only to be expected. In any case, they
don't alter the concepts of communication and cooperating as things we
perceive and control, do they?

In short, I'm having doubts about variables of type 3 as an identifiable
class. When prisoners find they can communicate only by tapping Morse code
messages on the walls, the manner and speed of their communication is
certainly constrained by the physical world and by knowledge of the code,
so I can go as far as agreeing that we adapt our communications to use
whatever means are available. But you seem to be claiming something more
than that about type 3 variables. I guess I need to hear more about that.

Best,

Bill P.

[From Bruce Nevin (2003.03.276 23:21 EST)]

Bill Powers (2003.03.27.0845 MST)--

The first sentence [of the description of type 3 variables] doesn't seem to differ from the description of type 2 variables, the only difference being in what is under control.

Are you saying that the rings and arcs of type 2 are under control?

Bruce Nevin (2003.03.26 14:34 EST)--

... there could be
successful cooperation and communication in an infinity of different
universes in which the means of communication differed subtly or radically.

This is indeed one of the differences between them. A summary of differences:

1. Type 3 is an uncontrolled outcome of cooperative or communicative control.

But type 2 is an uncontrolled outcome of individual control by a number of individuals independently, no cooperation or communication required.

2. Cooperation or communication is enabled by the type 3 patterns. Indeed, type 3 must pre-exist in order for the cooperative or communicative control to be possible. If the two prisoners do not both know the code they cannot communicate. (Morse code, of course, depends upon the prior existence of a shared language.) A child is born into a community in which the type 3 patterns already exist.

But the autonomous control that gives rise to type 2 does not depend upon it. Control of proximity to a goal and avoidance of fellows by individual agents is blind to the existence or nonexistence of rings and arcs.

3. In this sense, type 3 patterns are controlled in order to use them as means of the cooperation or communication which gives rise to them as uncontrolled side effects in one's own actions. They are controlled in the same way that one controls the rising of the sun. Failure to meet expectations is a disturbance. (I emphasize here that I am talking about pattern, structure, not about e.g. controlling the pronunciation of a given word or syllable for which references are stored in memory. I am talking about how the set of words, or syllables, or phonemes, has a consistent structure.)

But type 2 patterns are not controlled in the course of or as means of creating them.

4. Even in this universe that we presumably share there can be an infinite variety of cultural patterns subserving any given human purpose, such as asking another person to pass the salt. (The "variety of methods human beings have invented for achieving communication and cooperation". But unlike Morse code, no individual has invented these. They are a collective invention.)

But when individuals pursue a goal while limiting their proximity to one another, the rings and arcs that they form are strikingly similar. The only variety is in their size, which depends upon their references for proximity to goal and to their fellows, and the "smoothness" of the rings and arcs, which depends mostly upon how nearly alike the individuals' references and gain are. (Assuming no other disturbances, including from their control of other variables.) It appears likely that this would be the same in any universe.

I don't think that you can say that either is a subtype of the other.

Look at the variety of methods human beings have invented for achieving
communication and cooperation, which take advantage of just about every
possible way of transmitting and receiving perceptions used as symnbols. If
one can conceive of cooperation and communication, then to achieve them one
uses whatever means is available. I'm sure the means that are available
have a strong influence on the structure of communication, and perhaps bear
on cooperation as well, but that's only to be expected. In any case, they

the variety of methods human beings have invented for achieving
communication and cooperation [...]
don't alter the concepts of communication and cooperating as things we
perceive and control, do they?

The concepts of communicating and cooperating don't have much bearing on the doing, any more than the concepts of PCT are required for organisms to be able to control. I dare say the vast majority of humans rarely if ever perceive and control the concepts of communication and cooperating. They just do it.

In the course of communicating and cooperating, the means for doing so become conventionalized. We saw a primitive instance of this conventionalization in use in the description of ape infants. Among humans, the conventions are pre-established as children grow up learning with adults and then with one another how to use them, and over time in the course of use they change and are elaborated. I am thinking here especially of language of course, but I think this applies to all of culture.

you seem to be claiming something more
than that about type 3 variables. I guess I need to hear more about that.

There's a bit more.

         /Bruce Nevin

···

At 11:04 AM 3/27/2003, Bill Powers wrote:

[From Bill Powers (2003.03.28.1423 MST)]

Bruce Nevin (2003.03.276 23:21 EST)--

Bill Powers (2003.03.27.0845 MST)--

The first sentence [of the description of type 3 variables] doesn't seem
to differ from the description of type 2 variables, the only difference
being in what is under control.

Are you saying that the rings and arcs of type 2 are under control?

No, positions are under control, with the side-effect of producing apparent
spatial patterns which are not under control. In Type 3, linguistic
variables are under control, with the side-effect of producing apparent
patterns of communication which are not under control. So the only
difference I see is in the particular variables that are under control; in
both cases there are side-effects in the form of apparent patterns which
are not under control. That's why I said that type 3 seems to be a subset
of type 2, though I really should have said that the examples in type 2 and
type 3 are subsets of a single larger class of outcomes: controlling
perceptual variables in a way that produces apparent patterns of side
effects which, however, are not under control.

This doesn't seem to fit with your statement that the side-effects under
type 3 must pre-exist in order for control to occur. How can there be
side-effects of controlling that precede the controlling? Obviously I don't
yet understand your Type 3. There's something you're not saying.

Best,

Bill P.

···

At 11:04 AM 3/27/2003, Bill Powers wrote:

[From Bruce Nevin (2003.03.28 18:13 EST)]

Bill Powers (2003.03.28.1423 MST)--

[type 2]
positions are under control, with the side-effect of producing apparent
spatial patterns which are not under control. In Type 3, linguistic
variables are under control, with the side-effect of producing apparent
patterns of communication which are not under control. ... the only
difference I see is in the particular variables that are under control; in
both cases there are side-effects in the form of apparent patterns which
are not under control.

By this reckoning they are not different types at all. X variables are controlled, Y pattern is a byproduct that is not controlled.

This doesn't seem to fit with your statement that the side-effects under
type 3 must pre-exist in order for control to occur. How can there be
side-effects of controlling that precede the controlling? Obviously I don't
yet understand your Type 3. There's something you're not saying.

I've been saying but not yet successfully telling you. When an infant is born, the patterns of the language that it will learn already exist, immanent in the speech of those around it, and likewise other culture patterns. These patterns are part of the world the way it is, just like the sunrise and water running downhill.

Obviously they did not always pre-exist, but that is a question of evolution, not of learning or of control. These cultural patterns came into being in an evolutionary time scale that is intermediate between the time scale of biological evolution and that of individual learning - the time scale of culture change. Presumably the origination of language involved a bootstrapping sequence of the sort that Harris proposes, but that is immaterial for the situation of any individual, for whom the patterns are pre-existent.

This has been confusing and difficult to talk about because of these seemingly contradictory characteristics: controlled, not controlled, outcomes of control, pre-existing. They are not contradictory because they are in a way orthogonal to one another.

Outcome of control: Some of the patterning in speech is an outcome of controlling pronunciations of syllables, words, utterances in which that patterning is immanent. The references for producing pronunciations are constantly tuned for consistency with what is heard.

Not controlled: in the controlling of pronunciations in speaking, the patterning itself is not controlled. It is a self-organizing structure that arises in use, in the process of maintaining consistency between individuals in a communicating community.

Enabling control: The patterning enables greater efficiency of memory, production, and recognition, supports error correction in transmission of information, and enables mapping of equivalent utterances across variable pronunciations, among other things. The systematicity of the phonemes of the language in a matrix of cross-combining features enables an "explosion" in vocabulary learning at about age 2. These efficiencies arise in cultural evolution much as structural and metabolic efficiencies of organisms arise in biological evolution: structures that work better survive better.

Controlled: This may be another difference between type 2 and type 3. In making use of these efficiencies (e.g. for recognition as noted) these patterns are presumably immanent in the set of reference values across a domain, rather than controlled in the sense of a person having a perception of the pattern and a reference value for it. However, it may be that type 3 are controlled in the same sense that the sunrise is controlled, or water running downhill instead of up, or your house being there when you drive around that last corner. Failure to be as expected is a disturbance. If type 2 rings or arcs do not form in a crowd gathering around a point of common interest, no one but a sociologist would notice. How could one test this, since disturbance to the pattern is simultaneously disturbance to the perceptual control of which the pattern is a byproduct? One way of resisting such a disturbance is to perceive the input as in the place in the pattern where it "should" be, correcting the input by imagination. There is a similarity here to illusions. Are there cases where the correction (i.e. the recognition of unclear or ambiguous input) is not explainable in terms of references for phonemes, features, syllables, but only in terms of fitting in a pattern?

Pre-existing: I discussed this briefly above.

         /Bruce Nevin

···

At 04:33 PM 3/28/2003, Bill Powers wrote:

[From Bill Powers (2003.03.29.0938 MST)]

Bruce Nevin (2003.03.28 18:13 EST) --

I've been saying but not yet successfully telling you. When an infant is
born, the patterns of the language that it will learn already exist,
immanent in the speech of those around it, and likewise other culture
patterns. These patterns are part of the world the way it is, just like the
sunrise and water running downhill.

OK, I agree with that. You seem to be saying that there are pre-existing
patterns of speech that are instrumental in making language possible, but
which people do not learn to control. In that case, how do they prevent the
patterns from changing?

Not controlled: in the controlling of pronunciations in speaking, the
patterning itself is not controlled. It is a self-organizing structure that
arises in use, in the process of maintaining consistency between
individuals in a communicating community.

This is beginning to look like an epiphenomenon. If consistency between
individuals arises in a communicating community, that must be an emergent
result of their interactions, just as the arc and rings are an emergent
result of interactions among system that avoid collisions and follow a leader.

You say:
Enabling control: The patterning enables greater efficiency of memory,
production, and recognition, supports error correction in transmission of
information, and enables mapping of equivalent utterances across variable
pronunciations, among other things.

Greater than what? Do you have examples of communities in which these
patterns did not appear, and which therefore were observed to have lower
efficiency of memory and production, etc.? I have no doubt that patterns
are observed -- what you call phonemic contrasts, for example -- but why
are they not simply emergent from the process of communicating and
correcting errors of communication? If one person does not understand
another, the other can correct the error by altering pronunciation until
understanding is more reliably produced. The particular kinds of
alterations that will accomplish this depends on how human beings recognize
and produce speech, and I expect on the particular kind of speech that is
in use (Chinese versus Spanish, for example). What more do we need? Or have
I simply succeeded, finally, in paraphrasing what you're saying?

Best,

Bill P.

[From Bruce Nevin (2003.03.31 13:18 EST)]

A lot going on, and that delays responding.

Bill Powers (2003.03.29.0938 MST)–

When an
infant is

born, the patterns of the language that it will learn already exist,

immanent in the speech of those around it, and likewise other
culture

patterns. These patterns are part of the world the way it is, just like
the

sunrise and water running downhill.

OK, I agree with that. You seem to be saying that there are
pre-existing

patterns of speech

and other behavior

that are instrumental in making language

and other aspects of culture

possible, but

which people do not learn to control. In that case, how do they prevent
the

patterns from changing?

Rather, how do they slow the rate of change and maintain consistency with
one another through change? Because they don’t prevent change. The
patterns do change, slowly, over time, but always in such a way that the
pattern is consistent for individuals in an intercommunicating society.
If the pattern itself is not controlled, then the control of CVs of which
the pattern is an epiphenomenon must have the effects, also as an
epiphenomenon, of maintaining mutual consistency and of slowing change to
the pattern. Contrast with type 2 rings and arcs which do not change or
vary in this systematic way.

There is also synchronic variation of detail within the pattern. If the
variation is too great you lose mutual intelligibility. But even the
variation is consistent, e.g. speakers shift from formal to informal
register in consistent ways.

Diachronic change originates in this variation. A variant of some aspect
of the pattern that is restricted to a certain group of speakers over
time becomes the preferred standard common to all, for example, because
that group is esteemed. (I’ve given references to Labov’s work on the
social motivation of language change.)

All this is consistent with maintaining consistency with one another, and
is not consistent with controlling with respect to reference values
retained from childhood for some established variables that everyone
happens to control.

If consistency with one another is a controlled variable, perhaps that is
why a settled culture (such as that described for Bali prior to WWII) has
a kind of thematic unity. If different aspects of the culture
(child-rearing, music, warfare, storytelling, conflict resolution, visual
arts, etc.) are consistent with one another in an overall pattern
(climaxless plateau), then it is easier to control for consistency with
one another.

This does not mean that the pattern is itself a controlled variable.
Suppose perception of the pattern in the behavior of others (perhaps in
terms of themes or features) sets references for the way many things are
done (music, warfare, etc.), and out of the doing of those many things
emerges an epiphenomenon which, being perceived, sets the references for
controlling the pattern. I’ll return to this loop at the end.

… a
self-organizing structure that

arises in use, in the process of maintaining consistency between

individuals in a communicating community

that is, maintaining mutual intelligibility, ability to communicate.

If consistency between

individuals arises in a communicating community, that must be an
emergent

result of their interactions, just as the arc and rings are an
emergent

result of interactions among system that avoid collisions and follow a
leader.

Yes indeed, I believe that is true, and that is what I said. There is a
difference, however, between type 1 and type 2. You can model navigation
of individuals among obstacles in pursuit of a goal, and when you have a
bunch of them with a common goal the epiphenomenon of rings and arcs
appears. The epiphenomenon need not pre-exist. The individuals do need
the example of rings and arcs being formed (nor the example even of
individuals in pursuit of a goal) in order to form rings and arcs (or in
order to pursue a goal). But, by contrast, you cannot model cultural
patterns such as the patterns in language without the pattern
pre-existing to be learned,and you must provide for the fact of variation
and change in language and culture, is that cultural patterns differ from
one community to another; rings and arcs do not.

You say:

Enabling control: The patterning enables greater efficiency of
memory,

production, and recognition, supports error correction in transmission
of

information, and enables mapping of equivalent utterances across
variable

pronunciations, among other things.

Greater than what? Do you have examples of communities in which
these

patterns did not appear, and which therefore were observed to have
lower

efficiency of memory and production, etc.?

No. No human communities are on record in which people lack a
common language, or in which more generally they lack mutually
predictable ways of behaving that are characteristic of that community
and that enable communication and cooperation.

But this answer (and your question) is sidestepping the issue. Without
phonological patterning, a large vocabulary is simply not possible. This
is because without discrete distinctions as fine-grained as syllables,
segments, and concurrent features a large vocabulary cannot exist. The
number of utterances that can reliably and repeatably be differentiated
from another without a phonemic system is rather small, such as is
observed in animal calls.

This says nothing about how such patterning arises, only that, having
arisen, it is part of the perceptual universe within which a new infant
in the community learns to control. The infant also learns to perceive
and control a stone, a door, fire, sunrise, shovel, mama, father’s
brother, etc. How you speak and act must exemplify the patterns of
language and culture in order for you to be a person who is a member of
that community, so they are means of controlling identity. They are far
and away the best means for controlling communication and cooperation
with other members of the community.

I have no doubt that patterns

are observed – what you call phonemic contrasts, for example – but
why

are they not simply emergent from the process of communicating and

correcting errors of communication? If one person does not
understand

another, the other can correct the error by altering pronunciation
until

understanding is more reliably produced. The particular kinds of

alterations that will accomplish this depends on how human beings
recognize

and produce speech, and I expect on the particular kind of speech that
is

in use (Chinese versus Spanish, for example). What more do we need? Or
have

I simply succeeded, finally, in paraphrasing what you’re
saying?

Sounds like a paraphrase as far as it goes, and paraphrase is the only
reliable test of understanding that I know. But there is a bit more. To
repeat what I said above, suppose there is feedback in social space.
Suppose control of cultural themes and phonological features etc. sets
references for the way things are done, and out of the doing of those
many things with one another emerges an epiphenomenon which, being
perceived, sets the references for controlling the themes and features of
language and culture. A slowly changing homeostasis in the systemic
interactions of autonomous control systems. No superordinate “social
control” system setting references is being proposed or suggested.
The fact that the epiphenomenon constitutes a structured system is a
consequence of the interactions of individuals that a mathematical
(information theoretic) analysis might account for - those ways of
interacting that work well are perceived and remembered and establish
reference values for control, and those that don’t work cannot be
perceived as how things are done for the very reason that they are not
means of successful control.

This is brought about in part through the reorganizing that each
individual does to reduce error in communication or cooperation
<http://www.ucomics.com/forbetterorforworse/2003/03/27/>

But that is not the whole story. We also set references by observing
others (hence my interest in a PCT account of imitation, repetition,
etc.) and to do so we observe what (apparently) “everybody
does” and what “this kind of person does” and how it
contrasts with what “that kind of person does”. Identifying as
“this kind of person” I of course avoid doing what “that
kind of person does”. Unless I find myself among them; and then my
control of “doing it the way everybody here does” comes in
conflict with “doing it the way my kind of people do”. This is
a simplistic beginning at issues of identity and social perceptions.

    /Bruce

Nevin

···

At 11:57 AM 3/29/2003, Bill Powers wrote:

If I may, humbly…French is my maternal language. I came here when I was 20. I keep my accent.

My son has no accent. He was born here 13 years after I came here.

A colleague psychiatrist noted how come the child does not pick up on the accent of the mother but that of the peers?

No answer, just a reflexion.

I know that a psychologist here has made such a thesis her baby…She got negative vibes. Sorry I do not remember her name

Paule A. Steichen. Asch, Ph.D.
IBIS Int’l
Individual Building of Integrated Success
2101 Grandin Road
Cincinnati OH 45208
voicemail: (513) 289-5998
fax: (513) 871-soul/7685
pasteichenasch@fuse.net

···

----- Original Message -----

From:
Bruce Nevin

To: CSGNET@listserv.uiuc.edu

Sent: Monday, March 31, 2003 1:19 PM

Subject: Re: Cultural patterns

[From Bruce Nevin (2003.03.31 13:18 EST)]

A lot going on, and that delays responding.

Bill Powers (2003.03.29.0938 MST)–
At 11:57 AM 3/29/2003, Bill Powers wrote:

When an infant is
born, the patterns of the language that it will learn already exist,
immanent in the speech of those around it, and likewise other culture
patterns. These patterns are part of the world the way it is, just like the
sunrise and water running downhill.

OK, I agree with that. You seem to be saying that there are pre-existing
patterns of speech

and other behavior

that are instrumental in making language

and other aspects of culture

possible, but
which people do not learn to control. In that case, how do they prevent the
patterns from changing?

Rather, how do they slow the rate of change and maintain consistency with one another through change? Because they don’t prevent change. The patterns do change, slowly, over time, but always in such a way that the pattern is consistent for individuals in an intercommunicating society. If the pattern itself is not controlled, then the control of CVs of which the pattern is an epiphenomenon must have the effects, also as an epiphenomenon, of maintaining mutual consistency and of slowing change to the pattern. Contrast with type 2 rings and arcs which do not change or vary in this systematic way.

There is also synchronic variation of detail within the pattern. If the variation is too great you lose mutual intelligibility. But even the variation is consistent, e.g. speakers shift from formal to informal register in consistent ways.

Diachronic change originates in this variation. A variant of some aspect of the pattern that is restricted to a certain group of speakers over time becomes the preferred standard common to all, for example, because that group is esteemed. (I’ve given references to Labov’s work on the social motivation of language change.)

All this is consistent with maintaining consistency with one another, and is not consistent with controlling with respect to reference values retained from childhood for some established variables that everyone happens to control.

If consistency with one another is a controlled variable, perhaps that is why a settled culture (such as that described for Bali prior to WWII) has a kind of thematic unity. If different aspects of the culture (child-rearing, music, warfare, storytelling, conflict resolution, visual arts, etc.) are consistent with one another in an overall pattern (climaxless plateau), then it is easier to control for consistency with one another.

This does not mean that the pattern is itself a controlled variable. Suppose perception of the pattern in the behavior of others (perhaps in terms of themes or features) sets references for the way many things are done (music, warfare, etc.), and out of the doing of those many things emerges an epiphenomenon which, being perceived, sets the references for controlling the pattern. I’ll return to this loop at the end.

  ... a self-organizing structure that
  arises in use, in the process of maintaining consistency between

individuals in a communicating community

that is, maintaining mutual intelligibility, ability to communicate.

If consistency between
individuals arises in a communicating community, that must be an emergent
result of their interactions, just as the arc and rings are an emergent
result of interactions among system that avoid collisions and follow a leader.

Yes indeed, I believe that is true, and that is what I said. There is a difference, however, between type 1 and type 2. You can model navigation of individuals among obstacles in pursuit of a goal, and when you have a bunch of them with a common goal the epiphenomenon of rings and arcs appears. The epiphenomenon need not pre-exist. The individuals do need the example of rings and arcs being formed (nor the example even of individuals in pursuit of a goal) in order to form rings and arcs (or in order to pursue a goal). But, by contrast, you cannot model cultural patterns such as the patterns in language without the pattern pre-existing to be learned,and you must provide for the fact of variation and change in language and culture, is that cultural patterns differ from one community to another; rings and arcs do not.

You say:
Enabling control: The patterning enables greater efficiency of memory,
production, and recognition, supports error correction in transmission of
information, and enables mapping of equivalent utterances across variable
pronunciations, among other things.

Greater than what? Do you have examples of communities in which these
patterns did not appear, and which therefore were observed to have lower

efficiency of memory and production, etc.?

No. No human communities are on record in which people lack a common language, or in which more generally they lack mutually predictable ways of behaving that are characteristic of that community and that enable communication and cooperation.

But this answer (and your question) is sidestepping the issue. Without phonological patterning, a large vocabulary is simply not possible. This is because without discrete distinctions as fine-grained as syllables, segments, and concurrent features a large vocabulary cannot exist. The number of utterances that can reliably and repeatably be differentiated from another without a phonemic system is rather small, such as is observed in animal calls.

This says nothing about how such patterning arises, only that, having arisen, it is part of the perceptual universe within which a new infant in the community learns to control. The infant also learns to perceive and control a stone, a door, fire, sunrise, shovel, mama, father’s brother, etc. How you speak and act must exemplify the patterns of language and culture in order for you to be a person who is a member of that community, so they are means of controlling identity. They are far and away the best means for controlling communication and cooperation with other members of the community.

I have no doubt that patterns
are observed – what you call phonemic contrasts, for example – but why
are they not simply emergent from the process of communicating and
correcting errors of communication? If one person does not understand
another, the other can correct the error by altering pronunciation until
understanding is more reliably produced. The particular kinds of
alterations that will accomplish this depends on how human beings recognize
and produce speech, and I expect on the particular kind of speech that is
in use (Chinese versus Spanish, for example). What more do we need? Or have
I simply succeeded, finally, in paraphrasing what you’re saying?

Sounds like a paraphrase as far as it goes, and paraphrase is the only reliable test of understanding that I know. But there is a bit more. To repeat what I said above, suppose there is feedback in social space. Suppose control of cultural themes and phonological features etc. sets references for the way things are done, and out of the doing of those many things with one another emerges an epiphenomenon which, being perceived, sets the references for controlling the themes and features of language and culture. A slowly changing homeostasis in the systemic interactions of autonomous control systems. No superordinate “social control” system setting references is being proposed or suggested. The fact that the epiphenomenon constitutes a structured system is a consequence of the interactions of individuals that a mathematical (information theoretic) analysis might account for - those ways of interacting that work well are perceived and remembered and establish reference values for control, and those that don’t work cannot be perceived as how things are done for the very reason that they are not means of successful control.

This is brought about in part through the reorganizing that each individual does to reduce error in communication or cooperation <http://www.ucomics.com/forbetterorforworse/2003/03/27/>
But that is not the whole story. We also set references by observing others (hence my interest in a PCT account of imitation, repetition, etc.) and to do so we observe what (apparently) “everybody does” and what “this kind of person does” and how it contrasts with what “that kind of person does”. Identifying as “this kind of person” I of course avoid doing what “that kind of person does”. Unless I find myself among them; and then my control of “doing it the way everybody here does” comes in conflict with “doing it the way my kind of people do”. This is a simplistic beginning at issues of identity and social perceptions.

      /Bruce Nevin

[From Rick Marken (2003.03.31.1510)]

Bruce Nevin (2003.03.31 13:18 EST)--

Suppose perception of the pattern in the behavior of others (perhaps in terms of
themes or features) sets references for the way many things are done (music,
warfare, etc.), and out of the doing of those many things emerges an
epiphenomenon which, being perceived, sets the references for controlling the
pattern. I'll return to this loop at the end.

Sorry, sir, but as the PCT policeman it is my duty to cite you for a minor PCT
violation. Perceptions don't set references in PCT. If they did, external events
would exert far more control over behavior than they actually do.

You'll be appearing in court before judge William T. Powers, probably in a couple
hours.

Best regards

_Petty_ Officer Marken

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[From Rick Marken (2003.03.31.1630)]

Bill Powers (2003.03.31.0449 MST)--

Rick Marken (2003.03.31.1510)]

> > Bruce Nevin (2003.03.31 13:18 EST)--

> > Suppose perception of the pattern in the behavior of others (perhaps in
> >terms of themes or features) sets references for the way many things are done

> Sorry, sir, but as the PCT policeman it is my duty to cite you for a
> minor PCTviolation.

I gave him a pass on this one because he contributed to my political
campaign for Bruce Abbott's robot. If interpreted generously, this could be
taken to mean that there was a control system trying to make some perceived
variables that the person controls behave the same way as when other people
control them.

Another possibility is that the person varies the way some controlled
variables behave until another person indicates that the result is
satisfactory -- two-level control.

I like this. I nab 'em and you forgive 'em.

Special on indulgences this week.

Hey. Wait a minute. I thought you were a Protestant;-)

Best regards

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[From Bruce Gregory (2003.0331.1734)]

Bruce Nevin (2003.03.31 13:18 EST)

But that is not the whole story. We also set references by observing
others (hence my interest in a PCT account of imitation, repetition,
etc.)

I am not sure that at PCT account of imitation differs at all from a PCT
account of any other action. Higher level systems set the references of lower
level systems based on observation or memory.

[From Bill Powers (2003.03.31.1604 MST_]

Bruce Nevin (2003.03.31 13:18 EST)--

Or have

I simply succeeded, finally, in paraphrasing what you're saying?

Sounds like a paraphrase as far as it goes, and paraphrase is the only
reliable test of understanding that I know. But there is a bit more. To
repeat what I said above, suppose there is feedback in social space.
Suppose control of cultural themes and phonological features etc. sets
references for the way things are done, and out of the doing of those many
things with one another emerges an epiphenomenon which, being perceived,
sets the references for controlling the themes and features of language
and culture. A slowly changing homeostasis in the systemic interactions of
autonomous control systems. No superordinate "social control" system
setting references is being proposed or suggested. The fact that the
epiphenomenon constitutes a structured system is a consequence of the
interactions of individuals that a mathematical (information theoretic)
analysis might account for - those ways of interacting that work well are
perceived and remembered and establish reference values for control, and
those that don't work cannot be perceived as how things are done for the
very reason that they are not means of successful control.

Getting there, I think. It's still a bit convoluted but the basic idea is
taking shape each time you describe it a little differently. I think you're
saying that the slow and continuing reorganizing process is finding local
minima in the systemic errors, so over a period of time the system seeks a
certain configuration, but that configuration is not itself under control.
It's just whatever configuration results from repeatedly altering the
system parameters and reducing total error. This brings the environment
into the act, too -- the physical environment, and other control systems
that push back when they are disturbed ("feedback in social space").

I saw the cartoon in our paper and almost posted it.

Best,

Bill P.

[From Bill Powers (2003.03.31.0449 MST)]

Rick Marken (2003.03.31.1510)]

> Bruce Nevin (2003.03.31 13:18 EST)--
>
> Suppose perception of the pattern in the behavior of others (perhaps in
terms of
> themes or features) sets references for the way many things are done

><Sorry, sir, but as the PCT policeman it is my duty to cite you for a
minor PCT
>violation.

I gave him a pass on this one because he contributed to my political
campaign for Bruce Abbott's robot. If interpreted generously, this could be
taken to mean that there was a control system trying to make some perceived
variables that the person controls behave the same way as when other people
control them.

Another possibility is that the person varies the way some controlled
variables behave until another person indicates that the result is
satisfactory -- two-level control.

Special on indulgences this week.

Bill P.