[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [xmca] Gratier, Greenfield, & Isaac

I had originally asked about methodology in the CG&I article, and I'm happy to have had so much discussion over this.

I think that one of the problems with "linking micro and macro" is setting the issue up as just TWO levels. Even if we really mean each of these to stand for a range of scales of time, talk, action, and development/learning, the very contrast we use to define them as different and separate gets in the way of their proper "integration".

One alternative is to more explicitly construct several levels of analysis from the eye-blink (or wink, or "OK") to the sedimentation of a habitus over years. And of course in both directions. And to try to specify just HOW the connections are made across timescales, both adjacent ones and in those cases where we jump over larger gaps. I started a discussion of this several years ago in a piece in MCA, and I continue to find the basic approach helpful, if obviously not so simple to put into practice in analysis.

There is also the issue of contrasting an "individual" vs a "collective" cultural orientation. Or even an Anglo-MiddleClass culture vs. a Latino/a working class (rural ?) one. How can we imagine teachers bringing together these imaginary realms across the essentialized differences (not to mention the misleading stereotypes) we construct for them? The way to bridge across differences, which can be very real in their effects, is not to define them contrastively, but to construct the larger multi-dimensional space of possible variations, in which there may well be different average clusters between communities, but in which there are still wide ranges of OVERLAPPING features and paths across small gaps, like a path of stepping stones, that, in particular circumstances, make connections much more ready-to-foot. :-)

In her elaboration of Bernstein's coding orientations, Hasan's work, which I mentioned before, takes into account that while there are different dispositions toward how common and usual a way of expressing something (like a command/request) may be across social classes or cultural communities, it is still the case that both sides can and do use the mode of expression that is preferred by the other side (even unconsciously "preferred" as with habitus). Another parallel might be made with the ways in which different dialects, e.g. middle-class US English vs. African-American basolect ("street talk"), which can at their extremes be nearly mutually unintelligible (with some asymmetry due to power relations), in fact belong to a cline (the de-creolizing continuum) in which the AA acrolect overlaps much more substantially with middle-class "standard" (in speech, as opposed to writing, I rather doubt there is a "standard" dialect, just more and less prestige feature clusters). It is by deconstructing these reified forms into their component features that we find the paths for connection.

Such an approach requires valuing complexity over simplicity. That was not the path that led the natural sciences to their early successes. It is not the way to an easy tenure. It is what is supposed to underlie really insightful work in fields like history or literary criticism. If the humanities have methodological lessons for the study of human activity, I think they lie in ways of dealing with the complexity of things.


PS. On the specific point of inter-coder culture, we might imagine that it takes a village to analyze a village, and that in fact the collective ad-hoc "culture" of a group of researchers, long and deeply engaged with some other community, is the ideal tool to bring to its study. WITHIN that research culture/community, a consensus that took the form of inter-coder agreement could be useful insofar as it could be situated in relation to the rest of our mini-culture, including an understanding of our disagreements (and not just about coding some instances). When you can tell me something credible and interesting that arises from the disagreements, then I will take claims based on the agreements more seriously.

Jay Lemke
Professor (Adjunct, 2009-2010)
Educational Studies
University of Michigan
Ann Arbor, MI 48109

Visiting Scholar
Laboratory for Comparative Human Communication
University of California -- San Diego
La Jolla, CA
USA 92093

On Dec 5, 2009, at 11:25 AM, mike cole wrote:

I'll be very interested in what the authors have to say to your analysis, David (and no, Larry, emotions are all over xmca!- no problem unless they
serve ad hominim destructive aims - (my view).

I wanted to comment on another aspect of David's note that may be
overlooked, but which I think adds an important, additional cultural element
to the analysis. I am referring to the problem of efforts to
train coders and then test for reliability.

David describes the problem quite well. But I want to link it to the idea that what we do when we get (say) two people to code the same protocols from complex interactional scenes is to create a micro-culture among the coders
during the training.

We wrestled with this problem (which may have been written up somewhere back in the LCHC newsletter days, or may have remained merely discussion within the lab) a good deal in, for example, work on categorizing forms of behavior in afterschool clubs. What we found was that we could get agreement up to a pretty high level for the behavioral categories we were using *so long as the coders stayed in contact with each other as they worked.* But at some point one of the people involved fell ill, and when s/he returned to work, the agreement level among coders took a tumble. I believe there was a good deal of work on this general issue back in the 1970's- 1980's -- we were not the only ones who stumbled across the fact that the coding reliability might be the result of what we would now call the creation of a microculture that
required ongoing maintenance.

Experimental procedures that script subjects' behaviors might be usefully viewed as "pre-coding" schemes that provide a patina of objectivity because they antecede the data collection phase, but run the danger of getting out of the observations what had covertly been built in. But that leads down a
long road we probably should not
pursue at present.

On Sat, Dec 5, 2009 at 10:28 AM, Larry Purss <lpurss@shaw.ca> wrote:


Your analysis of how we proceed from micro to macro VS from macro to micro is brilliant (is that too emotional a term for academic discourse where we
should remain reserved).
I have been trying to integrate the various cultural-critical discourses
with the micro-genetic "experience-near" acts of cognition and
You have suggested a framework which helps me begin to understand how the socio-cultural (historical-cultural) standpoint and the cultural- critical
standpoint take different positions on the same landscape.
I hope others have reflections on the way you have elaborated the contrasts
Thank you


----- Original Message -----
From: David Kellogg <vaughndogblack@yahoo.com>
Date: Friday, December 4, 2009 9:30 pm
Subject: Re: [xmca] Gratier, Greenfield, & Isaac
To: Culture ActivityeXtended Mind <xmca@weber.ucsd.edu>

Sometime in the late twentieth century, literary critics made
the discovery that all the qualitative methods developed for the
exegeses of literary texts could be readily applied to so-called
"non-literary" texts. This included biographical, formal, "new
criticism" based on the text itself, and a plethora of more or
less committed approaches like feminism, multiculturalism,
postcolonialism all of which share a cultural-critical stance of
one kind or another. Cultural critical discussion of the content
of texts, which could be readily confirmed by juicy quotes
(sorry, I mean, by the judicious application of cultural
analysis) led to a "postructuralist" critical discourse analysis.

There is a kind of classroom discourse analysis that I would
describe as qualitative literary criticism applied to teacher
talk, in which words like "enthusiasm", "excitement", and
"knowledge building" are used the way we used to talk about the
muscularity of Milton's metaphors or the crystalline structure
of Pope's verses.

But as Halliday points out, a discourse analysis of this type,
without some systematic and clear link to grammar on the one
hand and semantics on the other, is really no analysis at all;
it's just a running commentary on a text, with some selective
highlights tendentiously provided to bolster the critic's
argument. That's why Gratier, Greenfield and Isaac heroically
try to confirm their micro-analyses with a more macro-
logogenetic comparison.

How does an analyst manage to connect the kind of microgenetic
analysis that we really require to catch actual acts of
cognition and communication in the wild with something more
macrogenetic, even ontogenetic? I think there are two basic
The first tendancy is the tendancy we see in ethnomethodology
and conversation analysis influenced work, and it's one I would
describe as downward reductionist. The analyst takes SINGULARITY
as the key feature of communicative/cognitive events, and tries
to find the implicit communicative principle in a single
particular event (e.g. methods of turn-taking, the latching of
adjacency pairs, etc.). If these can be shown to inhere in the
actual meaning-making procedures of the participants themselves,
then presumably they can be generalized to the macrogenetic

The problem with this appears on p. 305 of Gratier, Greenfield
and Isaac, where the authors admit that their inter-rater
agreement rate was low: 68% for the collaborative completions,
75% for repetition, and only 62% (!!!) for nonverbal imitations.
For praise and criticism the inter-rater reliability was 78% and
71% respectively.

Raters are trained, and when that doesn't work (as it often
doesn't in my own work) they train each other until it does. But
if even trained and re-trained raters cannot agree on what
constitutes praise and what constitutes criticism, it's hard to
see how participants consistently do. One cannot help suspecting
that they simply don't, and that this may as responsible for
the discomfort reported in parent teacher conferences by
Greenfield, Quiroz and Raeff as anything else.

It seems to me that the socio-cultural (or cultural-historical)
approach to the methodological problem of linking micro- and
macro- levels of analysis is, as the name implies, a much more
UPWARDLY moving methodological maneuver. It involves trying to
find a point where microgenetic changes pass over into
macrogenetic ones, and often that involves working backwards
from macrogenetic differences to microgenetic ones.

That is why I am not so shocked as some participants by the
"essentializing" language of cooperation vs. competition,
"collective" vs. "individualistic" or even Latino vs. Anglo.
Actually, I think this kind of language is, as Larry says, a
heuristic; a form of "hypothesis and then research", and it's
excusable in a socio-cultural (cultural historical) attempt to
link the micro with the macro.

Where I think I really differ with the authors is in the way
that IRE is selectively employed to describe the latter and not
the former, and "criticism" is selectively employed to describe
the former and not the latter. It seems to me that BOTH of the
following interactions can be described in terms of IRE:

T:El mar es la colecion de much agua, pero agua fresca o agua salada?
S: Agua salada.
T: OK.

T: Tell us what an artist does. What does an artist do?
S: Draw.
T: Draws. OK? How about a photographer?
S: Takes pictures.
T: OK.

The usual way we think about IRE involves emphasizing the role
of the E, either as reward (in neo-behaviorist teaching) or as a
bridge to linking exchanges into sequences (in the work of Hugh
Mehan, and latterly in that of Nassaji and Wells, and Gordon
Wells generally). But here both teachers are moving from the
general to the particular, from definition to
exemplification, using "OK" as their E move.

It's pretty clear to me that the INITIATE is the key difference.
In BC the initiate is a STATEMENT, followed by a question with
only one degree of freedom. The result is a consensus, albeit
one based on very little choice (and one which confirms what one
child said earlier in the exchange, "ya me lo se todo").

In the Non-BC we have a much more open opening: a command which
actually REQUIRES participation, followed by a wh-question. The
child's response is the object of critical uptake (focussed on
verb-subject agreement). But the second question indicates an
implicit CRITICISM of this response, because a photographer can
also be said to be an artist, but a photographer does not draw.

So it appears to me that the teacher in the Non-BC classroom is
being a lot more critical, at least implicitly. The difference
between the two classrooms might actually be in the
demandingness ("prospectivity" is the proper technical term, I
guess) of the initiate.

The problem is that there is nothing inherent or universally
generalizeable in demandingness; it really does depend on the
needs and on the volition of the participants. In a classroom,
however, those needs and that volition are, either explicitly or
implicitly, explainable by that faith in a link between learning
and development which we all do confess.

David Kellogg
Seoul National University of Education

--- On Fri, 12/4/09, Larry Purss <lpurss@shaw.ca> wrote:

From: Larry Purss <lpurss@shaw.ca>
Subject: Re: [xmca] Gratier, Greenfield, & Isaac
To: "eXtended Mind, Culture, Activity" <xmca@weber.ucsd.edu>
Date: Friday, December 4, 2009, 5:21 AM

I wanted to amplify an aspect of the article that is implicit
and make it more explicit.
On page 312 various frames are offered to explain the reason for
the alternative communicative styles.
A) in the BC class the interactions are peer-group led rather
than teacher led
B)It might reflect a sense of partaking in a group voice and
sharing a coherent group identity.
C)It may denote a general OPENNESS TO NOVELTY and creative
responding and an orientation to verbal PLAY,NARRATIVE, AND HUMOR.

There is another discourse which focuses on creating "COMMON
GROUND" (p313) and "student attentional engagement and emotional
expression" (313) which "results in a rhythmic pattern that
creates ensemble" (313).
This is the dramaturlogical discourse of "enactments" and "PLAY"
and language "games"  This discourse emphasizes metaphors that
point to creating or opening spaces where common ground emerges
from novelty.
I share sympathy with the perspective of this article that
emphasizes collaborative emergence.  Yes the terms
collectivistic and individualistic are essentializing terms but
if we view them as heuristic language games to map the territory
of the relational patterns emerging and creating common ground
their are various discourses which can point us in that direction.


----- Original Message -----
From: David Kellogg <vaughndogblack@yahoo.com>
Date: Friday, December 4, 2009 12:50 am
Subject: Re: [xmca] Gratier, Greenfield, & Isaac
To: lchcmike@gmail.com, Culture ActivityeXtended Mind
The part that Jay is puzzled by caught my eye as well. It's
right on the bottom of p. 297:

"Cultural conflicts between Latino family values and American
pedagogical values were also studied empirically by
Quiroz and Raeff (2000) through an analysis of patent-teacher
conferences. They found widely varying emphases on helping and
sharing as well as high levels of misunderstanding and
between Latino parents and US trained mainstream teachers.
Implicit cultural conflicts were shown to clearly (?) relate
underlying and nonverbalized cultural assumptions. In these
conferences, the teacher, having adopted the 'individualistic'
assumptions of US school culture, was verbally constructing an
'individualistic' child, whereas the parent was verbally
constructing a 'collectivistic' one. As an example, one
in the 'collectivistic worldview is a dispreference for
which makes one child stand out. In the 'individualistic' (p.
298) worldview, in contrast, praise is strongly preferred. In
one conference, teh teacher's
praise for the child made a father extremely
uncomfortable. Given that these parents were concerned with
socializing their children into tehir culture, we would
that high levels of praise in the classroom would cause
with the children's (?) more collectivistic worldview, based
their home socialization."

This paragraph is later transformed into a research hypothesis
on p. 303:

"H4: We predicted more use of praise in the non-BC classroom
more use of criticism in the BC classroom." (p. 303)

It is also the object of quantitative analysis on p. 304: "the
number of instances of praise and criticism directed at the
It seems clear, to answer Jay's query, that this means
praise/criticism of the child by the teacher. What is less
is how these can be "clearly" related to NONVERBALIZED and
UNDERLYING cultural assumptions,,

If they are underlying and assumed, why would they be
at all? If they are wholly or partly nonverbalized, how can
be quantified in the number of instances of praise and
directed at students?

In addition, it's not at all clear how or if research based on
parent-teacher conferences, which are performances of a rather
different nature in which the child does not take part, is
for classroom research.

I think, unlike Jay, I am rather sympathetic to the
assumptions that underly this kind of research. I do believe
that there is something called a restricted code and a more
elaborated one, and I even believe that up to a certain point
home-school mismatch can be debilitating for children.

But I also believe that after a certain point (say, fourth or
fifth grade) kids begin to talk like other kids and not like
their parents. So when we find restricted codes reproducing
themselves in learner language, it is not blameable on parent
cultures, but rather on the child's own emerging volition.

That is the bad news. The good news is that, like foreign
language codes (which are certainly elaborated), the
fossilization of restricted codes is highly susceptible
to teacher intervention.

David Kellogg
Seoul National University of Education

--- On Thu, 12/3/09, mike cole <lchcmike@gmail.com> wrote:

From: mike cole <lchcmike@gmail.com>
Subject: Re: [xmca] Gratier, Greenfield, & Isaac
To: "eXtended Mind, Culture, Activity" <xmca@weber.ucsd.edu>
Cc: "Patricia Greenfield" <greenfield@psych.ucla.edu>,
Date: Thursday, December 3, 2009, 4:51 PM

I am cc'ing authors in case they have not signed up for the
discussion. A
mixture of questions have been raised that perhaps
they can help to help us sort out.

On Thu, Dec 3, 2009 at 3:00 PM, yuan lai
<laiyuantaiwan@gmail.com> wrote:

Michael, I believe there are ways that mirror the “natural
way” to teach
cultural capital overtly. I’ve seen 3- and 4-year-olds from
families of
refugee status quickly appropriated the value placed on
showing> interest in print, wanting to write their names,
feeling proud of their own
attempts, not long after establishing a relationship with
preschool> teacher in various activities in a family literacy
program, which embeds
print in almost all its classroom activities. For example,
teacher read
to the children while they were eating, pointed out print
signs in the
environment for them as they went out for recess, and wrote
notes in front
of them to request materials needed for the classroom. The
transformation> of
the children’s attention, interest, and desire is amazing
given that the
children hardly understood English when they entered the
program and their
parents seldom read to them or pointed out print around due
low reading
and writing ability in English and in their first language.
I've since been
convinced of the importance of setting up a learning
environment that has
emphasis on relationship building.

Jay, until you revealed it, I didn't see it. I reread the
section leading
the hypotheses section and found that there is some
to praise,
not at all to criticism.

It appears that the same two classrooms (BC and non-BC) have
been studied
from different angles and the findings seem to be consistent
with Gratier
al.'s framework. This article certainly extends their work.
Terms such as
style and collectivism do connote essentialization; the
authors’ data
provide substantiation of the essentialzed norms and
communication styles
(although what one sets out to do confines what one looks
but I think
they could have gone a step further. The example of a
feeling> uncomfortable when the teacher praised his child does
not tell how he may
act or say to people in his in-group. There is also the
assumption that
socialization remains the same after immigration. Given the
contrastive> framework in Gratier et al., I see little reasons
not to include the
videotaping of the same groups of children (some of them,
likely)> interacting with their parents at home. Or is another
paper forthcoming?


On Thu, Dec 3, 2009 at 4:46 AM, Michael Glassman
<MGlassman@ehe.osu.edu> >wrote:

It seems to me a playing out - at least to some extent of
Bourdieu's> larger
theory.  That increasing the cultural capital of the
in relation
the class would increase the level of social capital,
would lead to
some of the findings they present.  A lack of cultural
capital (usually
assumed on the part of the students) would certainly lead
to more
difficulties in communication and the students feeling
uncomfortable> in

But this leads to a fairly radical assumption on the part
the authors
concerning habitus, even in terms of Bourdieu's theory.
That is that
cultural capital can be taught overtly, as cultural
capital -
to emphasize that we learn cultural capital more or less
unconsciously,> > through everyday experience in the right
situations (whether it is with
parents or in a school system where the type of cultural
capital that
to easy social capital is pervasive).  I'm not so sure
is possible.

I have another difficult which is that I read habitus as
defining class
distinctions rather than cultural distinctions, and that
not sure his
ideas translate between the two, or make that much sense
they do.  The
types of cultures like Latino/Latina cultures are going to
have class
distinctions defined by different habitas, defined most
easily by
levels of economic capital, and different recogntions of
symbolic capital
(and symbolic violence),  To say a population so large has
single type
habitus I think is problematic - especially when using a
terms such as
collectivist, which is both categorical and far too broad
think to be
really salient in describing classes, let alone entire
cultures (I think
level and type of social capital might be more appropriate
if you are
to use Bourdeiu's theory as a starting point).



From: xmca-bounces@weber.ucsd.edu on behalf of Jay Lemke
Sent: Thu 12/3/2009 12:16 AM
To: XMCA Forum
Subject: [xmca] Gratier, Greenfield, & Isaac

I don't know how many people have yet had a chance to look
at the MCA
article-of-the-month (Gratier, Greenfield, & Isaac on
communicative> > habitus and attunement in classrooms).

I must have missed something, so could someone explain to
how they
derive the hypothesis that the more culturally attuned
classroom will
have more criticism (by the teacher? or by everyone?) and
less praise,
than the mismatched classroom?

And what do you think generally about the methodology in
this work?


Jay Lemke
Professor (Adjunct, 2009-2010)
Educational Studies
University of Michigan
Ann Arbor, MI 48109
www.umich.edu/~jaylemke <http://www.umich.edu/%7Ejaylemke>
<http://www.umich.edu/%7Ejaylemke> <

Visiting Scholar
Laboratory for Comparative Human Communication
University of California -- San Diego
La Jolla, CA
USA 92093

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list

xmca mailing list