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

Re: [xmca] Gratier, Greenfield, & Isaac

Thanks for your careful reading, Larry, but above all thanks for your reading list. We get long holidays over here, and mine starts next week; some of what you recommend is definitely going on my night-table.
We used to think of coding problems in terms of "high inference" and "low inference" categories, for example, coding whether an utterance is You know, I think we are actually talking about exactly the same thing when we speak of poor inter-rater reliability and when we speak of the ambiguity necessary for two people to describe the same experience as shared.
When you share an experience (and when you code a response to data), you have to abstract away the individuality of the experience, of course. But with it you also abstract away the reliability of the characterization, and as Mike points out this can only be reestablished with some hand-waving and renegotiation. Coding is really like Trobriand navigation; a general orientation, a certain amount of dead reckoning, and, most importantly, an endless process of minute course-corrections, each of which has to be negotiated with a more or less uncooperative crew and a very uncooperative environment.
There's a weird kind of dualism is Larry's use of "singularity" below. He appears to mean two things:
a) The DISCONNECTEDNESS of a particular experience, or a particular word or verbal expression in an utterance.
b) The UNREPEATABILITY of a particular experience, or a particular meaning in a context.
For example:
T: What time is it?
S: It's ten o'clock.
a) Some experiences (e.g. asking the time) are easily isolable from surrounding events. Similarly some words (e.g. "time" "o'clock" and "ten") are quite easily disconnected and listed in a dictionary with some kind of semantic meaning attached. Others are much less so. 
b) Some experiences are easily repeatable, and others are much less so. It is ten o'clock exactly twice a day, every single day.
In Chapter Two of Thinking and Speech, Minick translates as “dualism” the Russian word “двойственности”, a word translated as “ambiguity” by both Seve and Meccaci. The word DOES mean ambiguity in the sense of “double meaning”, but it contains the root “two”. “Duality” seems a good compromise. 
But of course there is a huge continuum here, between the repeatable and the unrepeatable, the isolable and the flowing, and we can only really speak of finding the right characterization of the data with respect to quite specific questions, and not in general. 
The problem with Gratier et al. is that one suspects the question is both not specific enough and a little TOO specific: "Which classroom is bettter?" 
David Kellogg 
Seoul National University of Education 

xmca mailing list