I will try to have a look at the Cole&Cole chapter for more context, and
perhaps to find something useful to say about it.
Meanwhile the first thread that caught my eye was Mike's query about
meta-cognition, and its relationship to meta-discourse et meta-cetera, and
what might be more and less problematic thereabouts.
I quite agree with Mike's basic worry that if meta-cognition implies access
to some sort of a Cognition-Reality, that our meta-phors would be dragging
us back toward Cartesianism, if not Platonism. Both in their pejorative senses.
What that term mainly seems to name is whatever people do that they talk
about by using language resources for 'mental process' meanings. In
functional semantics, Mental Process does not refer to any reality as such,
only to the kinds of meanings we make when we say: I think, you feel, he
likes. These are transformed by very standard practices in Indo-European
languages into such variants as my thoughts, your feelings, his likes; and
from there in the play of discourse to locutions about my mind, your
sensitivity, his desires ... and on, via folk-theories of these matters, to
talk of will, intentions, and most of the 'mental faculties' of the
proto-psychology of a couple centuries ago. Some analysts of the history of
mentalist discourse would say that our folk-theories are often mainly
explicitizations of what is already implicit in the semantics of our
languages. The latter does change a bit historically, as we get more hooked
on the theories; there are of course other languages, in cultures
relatively insulated from these theories (e.g. aboriginal languages in
Australia) where one just does not find a mental process semantics that is
distinct from a Verbal Process semantics or a Behavioral Process one. There
are guesses that in the history of many language families, esp. I-E, that
there was just one original Process semantics, for something like what we
now call Material Process (actions and doings that have recipients,
including ourselves, other people, things), and that Behaviors (mostly
activity without a second participant, like walking or breathing) hived off
from the original semantic stem, and then a third differentiation came to
be prominent, with something like Verbal at its core, but including
functionally what we would call Mental today, and then the Mental split
off. In each case a 'split' means there are slightly different rules of
grammar and different classes of participants that can meaningfully engage
in the type of process.
What is at root here, I think, is that we cannot assume that language
reflects reality, or that it evolved certain meanings because there were
realities (at least material realities) needing to be named. Language
responds just as much to our theories about realities as to the realities
themselves; it responds in effect to our ways of coping, our ways of
getting by in life, which are perhaps constrained in some degree by
realities (or at least that sense of outermost limit of possible
meaning-making is all I can see 'reality' might mean/be), but very heavily
mediated by _Discourses_ that frame and interpret the Great Whatever. So
there must historically have been quite an interesting dance between the
deep semantics of a language (usually NOT explicit for speakers) and its
common Discourses (ways of talking about some topic).
In this view, meta-cognition, is a reification of locutions that use the
mentalprocess semantics available in the language, usually in mentalistic
Discourses (the upper case 'D' is Jim Gee's convenient reminder that
Discourses are social-cultural institutions, or what I call ecosocial
formations, while 'discourses' are just instances of people talking, and
usually tokens of some Discourse types). It commits a logical error: to
assume that because you can talk about something consistently, it must
exist. (There are also people of course who claim that you cannot talk
about mental processes consistently, that there are many contradictions and
incoherences if you follow the logic out ... but then that is the case with
most abstract discourses, I find.)
There are two levels of reification. One for cognition as a phenomenon (not
too bad; something is going on, I just prefer to call it meaning-making),
or (worse) for specific cognitive processes (distinctions between cognition
and affect, or between pattern-recognition and memory, etc. seem to me
badly misleading for the most part). The second for 'meta-cognitive'
processes. Here even the phenomenon is in question. Technically,
meta-cognition ought to mean cognition-about-cognition, and ought to be of
a different logical type (ala Russell and Whitehead, or Bateson, cf.
meta-learning) than first-order cognition. Neurologically, I can believe
there are brain-body processes which we interpret as mental processes that
involve different numbers of re-afferent loops (cf. Edelman), as for
example between conscious and unconscious processes, or perhaps between
sensorimotor processes and 'representational' or semiotic processes where
one can superimpose the imaginary on the sensory. I do NOT think there is
such a basis for a category of meta-cognitive processes, because so far as
I can tell both the phenomenon and the evidence for it all fall well within
the basic category of semiotic 'representational' processes -- though
whether of realia or irrealia (imagined processes, cultural fantasies about
how semiosis gets done) remains debatable, as above.
One can clarify this a bit by considering the usual wisdom that 'language
is its own metalanguage'. What this means is that we use words to talk
about the process of using words to talk about anything. The word 'word' is
a not quite trivial example (it implies a theory, probably a post-literacy
one, about language). If I say: "I used the wrong word in the last
sentence.' then I am talking about nothing but the use of language itself.
It is from the analogous locutions for mental processes, derived I assume
historically from verbalprocess semantics, that we get the very concept and
discourse forms of mentalism and cognition (e.g. I imagined the wrong image
in my last meditation; I conceptualized the idea wrongly in my last
cognitive effort to solve the problem; etc.). While it is true that we use
the same language, more or less, to talk about language that we do to talk
about other matters (apart from a specialized vocabulary, which every topic
has), it is NOT true that language can meaningfully talk about itself with
no reference to anything other than language, i.e. without some other
semiotic system necessarily coming into play. Read on.
Slightly more interesting is meta-discourse. Now we are talking about how
_as activity_ and not merely as a system of forms, talk talks about talk,
or more properly people talk about their own talking, while they're talking.
Meta-discourse is self-reflexive _activity_ (discourse is an activity;
language is a formal system of semiotic resources -- note lower case 'd'
this time!), and it only makes sense _in context_, i.e. in a nonlinguistic
context, or at least a partially non-linguistic context. There is a
semiotics of action and activity that is quite distinct from (though it may
include) the semiotics of verbal meaning-making. Meta-discourse only works
through the contextualizations of setting and non-verbal activity, or of
activity _type_, even if that type is talk. It is only the whole semiotic
complex of 'communication' that can be self-reflexive (even the word
'communication' is not quite adequate here -- I mean meaning-making via all
the simultaneous semiotic modes normally deployed in human face-to-face
joint activity, including but never limited to verbal language). So if
there is a phenomenon here, it is meta-communication, or meta-activity, and
that in turn is the generalization, or perhaps the root origin, of the
representational-semiotic capacity to overlay the imaginary on the sensory.
The reflexive or meta- capacity enables us to 'frame' activity we are
participating in, to represent it (using this term loosely), so that we can
act on our interactions-activities semiotically as well as materially
(while they are still happening!). We can 'guide' ourselves and others in
action, and part of that guidance is the meta part and part of it is the
'imaginary' (or 'irrealis') part. I am sure someone could tease this out in
more detail, especially ontogenetically. (Lacan gives it a try, hence
Wilden's batesonian interest in Lacan). It should be very fruitful to study
the social-developmental trajectory of how kids acquire the use of various
semiotic resources (including language in general and various Discourses in
particular) in relation to how they acquire a sense and discourse of self
and identity. Some neo-Lacanians in Europe are starting to do just this.
Which brings me finally to note what I see as a wrong turn in some other
research programs on these issues. In the 1970s cognitive science
programme, gone badly wrong in the late 80s and 90s under the influence of
some schools of philosophy of mind (philosophy is a great critic of
science, but beware when it makes positive proposals!!), there is a
research tradition which aims to establish the reality of Mind by showing
that children younger than some age have no conception of their own or
others' minds, and that they acquire such an understanding of the reality
of Mind when their brains are developmentally ready to handle this (much as
Chomsky says we acquire language -- a story I also don't believe). I think
that some of the experiments Mike referred to about the ontogenesis of
meta-cognition belong to this research program. In effect there are two
kinds of evidence for kids having a theory of mind: indirect evidence that
they behave as if they did (as interpreted of course by researchers who
have this theory and moreover believe it's a truth about reality rather
than just a sometimes convenient story), and direct evidence -- and this
last is meta-discourse, talking about the meaning-making that the kids are
learning to talk about mentalistically. It is not surprising that the data
shows that they pick the discourse up from the experimenters. What these
experiments tell us, I think, is something of how well socialized into
mentalistic discourse the kids have become. But not what we'd really like
to know: what kinds of social interactions and input got them there. I'd be
willing to bet that there is a kind of 'ontogeny repeats linguistic
phylogeny' here: that first there is talk about material processes, then
about our own material behavior, then about our own verbal behavior, and
finally about 'mental process'. With all the usual complications of course.
The research program I am disapproving of (at least in its motivations and
interpretations) is very reminiscent of the worst excesses of Piagetians.
It assumes that the latest stage of development is also the highest, the
one closest to an understanding of how reality is; rather than just being
the one that is most culture-specific. (I don't say Piaget himself always
made this error, though I think sometimes he did . And we've discussed here
before whether LSV also made it in relation to 'scientific vs. everyday').
It is an interesting thesis of some complex-system theory that to make
progress beyond a dead-end, you have to back up to the last fork in the
developmental-historical road and try a path not taken. In this case to
back up from mentalism to language-ism (or semiosis-ism) and try again to
understand what's going on.
JAY.
---------------------------
JAY L. LEMKE
PROFESSOR OF EDUCATION
CITY UNIVERSITY OF NEW YORK
JLLBC who-is-at CUNYVM.CUNY.EDU
<http://academic.brooklyn.cuny.edu/education/jlemke/index.htm>
---------------------------