I'm a skeptic about the notion of 'basic skills'. Not just
because of the conservatives' abuse of this notion, nor even
because -- not believing that 'transfer' is all it's cracked up
to be -- what is called 'basic' often really means only what some
analyst chooses to see as 'in common' among very different
situated activities, but mainly because I am not sure that it
makes sense to assign 'skill levels' as an explanation of
_performances_.
Of course there is some sense in which, totally lacking a skill,
we may not be able to perform. But then again, those poor cats
get skinned one way and another, and particular skills are often
only conventionally seen as required -- alternative practices
often suffice.
When we find that some group of people do better on a 'skills
test' than some other group, I tend to look for other sorts of
explanations. Mainly I look at the congruence/noncongruence of
the testing task and the dispositions of the poor testees. I
guess I learned to do this long ago from Mike Cole et al., the
Vai studies, and many other cross-cultural ones like them (both
intentionally and unintentionally so). Dispositions depend on
more than just 'cultural group' (on age, on gender, on class, and
on many more delicate positional-relational grids), and they
pertain to much more than just motivation-to-perform (perception
of task, sense of appropriate ways of performing, trial solution
prioritization, unconscious aspects of performance, etc., etc.).
Any analysis would of course ask how closely matched on grid
variables the 5D kids and the comparison group are. Assuming a
reasonable match, and that 5D kids are not self-selecting for
some obvious, congruence-relevant variable, there is then the
possible 'treatment' effect. But the treatment in this case need
not be at all directly related to the skills tested, it need only
have an effect on relevant dispositions. I assume that in the
coarsest analysis, skills tests and 5D activities have in common
that they depend on certain common, and often implicit and
undescribed, routines of middle-class practice. One could imagine
then that participation in 5D activities (of all sorts, including
just identification with the centers, ie. being the kind of kid
who does that kind of thing successfully) represents a certain
sort of reinforcing enculturation into these practices/routines.
Note that this way of looking at matters does not _contrast_
middle-class with working-class routines and dispositions; it
does not assume any necessary mutual exclusivity of practices
(though in some cases this may be true). I assume that all
occupants of non-dominant social positions have at least a
passing familiarity with ('passive' competence in) the practices
of dominant groups (though not vice versa; dominant groups only
_think_ they have the reciprocal familiarity, they more often
hold convenient myths). They often have little practice enacting
these practices, but do have at least a basis for doing so, and
in some cases (say school tasks) do have a fair bit of practice.
It may take only the removal of certain obstacles (e.g.
unwillingness to perform, a key misconception) or critical
additional experience to allow them to mobilize a largely
unenacted competence to the level of enactability.
So I am proposing, broadly, a more subtle sort of enculturation
model of good test performance, rather than a model that
attributes it solely to explicit teaching. I happen to consider
explicit teaching often necessary in the absence of more
automatic enculturation routes to performance, but I don't at all
suppose that knowing how something is done is nearly enough
reason for it to actually be done. On the other hand, having a
general sense of how to go about doing some class of tasks, and a
sense of yourself as the sort of person who can and does do that
sort of thing, probably goes a long way, with a little task-
specific competence, toward eventuating in functional
performance. JAY.
PS. My hypothesis would probably predict better-than-comparison-
group performance on a whole range of possible tests of middle-
class dispositions and routines. Reading and math tests are quite
notoriously caste-sensitive, and I'd not be surprised to find
even an IQ difference in the granddaddy of all measures of junior
membership in _our_ caste. It would of course also be interesting
to see longitudinal data.
JAY LEMKE.
City University of New York.
BITNET: JLLBC who-is-at CUNYVM
INTERNET: JLLBC who-is-at CUNYVM.CUNY.EDU