it shouldn't be a total loss

Mike Lieber (U28550@UICVM.BITNET)
Mon, 31 Oct 1994 08:13:19 CST

>From the ashes of this newest ugly debate on race and intelligence might come
something useful if we are determined to turn it into an opportunity. I do
see an opportunity, and Steve Mizrach has already gotten out in front on this
one. There is no consensus among psychologists and cognitive scientists on
just what intelligence denotes. Where might anthropological thinking take this
part of the debate?

One of the threads of discussion among psychologists and others involved in
intelligence research is the connection between intelligence and personality,
concentrating variously on intelligence as an expression of personality or on
personality as an organized set of constraints on intelligence and its
enactments. For example, take a kid who is used to solving problems on his/her
own, whether because of parental demand or parental negligence. Take that
kid's sibling, who is used to having things done and having problems solved for
him/her. When each one takes an I.Q. test and hits the problem solving part,
would you expect to see a difference in their scores? Yes indeed. Does this
mean that one is more intelligent than the other? Not necessarily. The
matter can be viewed as I.Q. scores being constrained by independence and
dependence for which each child has been "trained." Clearly personality
development _and the learning processes that shape that development_ can be
expressed in I.Q. scores, among other things. Rather than get into the
denotations of personality--which are also at issue--let's try a different tack
from this basic relation.

Gregory Bateson wrote a tantalizing article during his days at O.S.S. in the
1940s on the organization of learning, which he called "deuterolearning," or
learning to learn. He took the classical learning experiments of his day and
showed how such things as Pavlovian conditioning could be combined with, say,
instrumental punishment as an organized way in which children learn to learn
about the world around them. What you'd get in this way of learning are people
who see disaster as just around the corner and respond to that expectation by
carefully managed, rote enactments to prevent or postpone it. The Balinese,
according to Bateson, illustrate precisely this learning process. He applied
the organization of learning to what was then called "national character."
When he realized that deuterolearning was an instance of Russell and Whitehead,
theory of "logical type," he dropped his interest in national character (or
modal personality structure, as it came to be called) and concentrated on a
broader theory of learning, which you can find in his theory of learning and
evolution in _Steps to an Ecology of Mind_.

The trouble with dueterolearning was that it couldn't account for variability
in a population, but recast as part of logical types of learning, it could.
Russell and Whitehead tried to tackle the logic of classes using the relation
between a class and its members, a whole-to-part relation, which they called a
difference of "logical type." A class cannot be a member of itself. The
class of chairs is not a chair. This rule applies on up the chain of logical
types. A class can have other classes as its members--furniture is a class
that includes the class of chairs; a taxonomy is an ordering of classes of
classes of classes. Applied to learning, Bateson distinguished zero learning,
as when information that an organism uses is genetically programmed. The next
logical type is Learning I, including rote learning, Pavlovian conditioning,
instrumental reward and punishment, etc.. This is a change in zero learning.
Learning II is deuterolearning--learning to learn. Psychologists used to call
this "stimulus generalization." Give a monkey a puzzle, as Harry Harlow did,
and it takes about 20 minutes to solve the first one. But with successive,
different puzzles, it takes increasingly less time. The monkey has generalized
from separate puzzles to a class of puzzles which share similar strategies for
solution. Learning II is a change in Learning I just as acceleration is a
change in velocity. Then there is Learning III.

Learning III is learning about learning to learn. That is, it involves
learning about the context in which one learned to learn things. This
logical level of learning is what psychiatrists and psychotherapists try to
get their patients to achieve. Bateson's involvement with Learning III grew
out of his work with schizophrenics, where he discovered that the Russellian
paradox was what structured schizophrenic perception and communication. The
problem with Russell and Whitehead's logical types, you see, was that it
inevitably generated a paradox--the class of all classes that are members of
themselves. Kurt Godel showed that this must always be the case in any
_closed_ system of self-consistent axioms. One must always go outside that
system of axioms for their mathematical proofs. What schizophrenics learn is
that if they correctly perceive and respond to paradoxical messages of a parent
they will be punished. Example--mother comes to visit 112 year old Johnny, who
has been institutionalized. She says, "johnny, come kiss me," but as he bends
to kiss her, she pulls her face away. He responds to this by pulling back, and
momma says, "What's the matter Johnny, don't you love me?" Damned if he does,
and damned if he doesn't. Bateson saw this paradox as developing within and
supported by a larger system of family communication in which the schizophrenic
child occupies one role in that system. The family communication system is the
_context_ of learning to learn. This is where modern family therapy comes
from. Its founders were people like Don Jackson, Paul Watzlawick, and others
who worked on the original schizophrenia research with Bateson.

There has been a good deal of research on family communication systems since
the 1950s. Meanwhile anthropologists were busy with "kinship structure,"
churning out new terminology for genealogical patterning and group organization
and, finally debating whether kinship meant geneaolgy or conduct and whether
kinship was a subject or non-subject in the first place. That debate died away
as anthropologists like Carol Stack and Ruth Horowitz began to find out what
code for conduct actually meant in Black and Chicago Hispanic families
respectively. As studies like these began to be subsumed by the realization
that the concept of "the Person" was culturally variable, the issues of
ethnicity on the one hand (cf. Linnekin and Poyer, _Cultural Identity and
Ethnicity in Oceania_) and variability within an ethnic group or within a
family on the other went their separate ways as research issues. I want to
suggest to you that this is an opportune time to put them back together.

The issue to which these research threads most profoundly address themselves is
what is meant by "environment." One thing that I agree with Murray about is
that "environment" is a lot more than socio-economic position, class, and
locale. Why, for example, does there seem to be a significant correlation
between birth order and I.Q. scores in familes with three or more children?
There is plenty of data on family communication systems collected over nearly
40 years just waiting to be looked at. Is the environment of a first child
the same as the environment of the second, third, etc.? What are the
differences between them and what are the _patterned_ responses to these
differences? How are differences in ethnicity and class related to differences
in the microenvironments of members of a family and what sorts of relations
outside a family constrain relations within it? There are plenty of data to be
analyzed and plenty more to be gotten to answer these questions. As this sort
of research gets done, the issue of intelligence is reshaped into a new, hard
nosed kind of intelligibility; correlation is replaced by the details of
process; and social science becomes responsible to the larger public that so
badly needs it.

Don't bitch at Murray and Rushton. They are who they are. The best response
is better research with a John McCreery to make sure that the people who need
to know about it do.
Mike Lieber