I'd like to add to Gordon's comments by first agreeing about how the unit of analysis is simply a methodological device for studying a particular phenomena. Expecting to find a central unit for the study of language use is dependent upon the purposes of the study.
Second, in regard to analyzing an activity structure, one could focus on the mediating artifacts, one or more of the agents involved, any of the points of Engestrom's triangles, and on it goes, but from my experience in the end the one focal point that pulls everything together is the object of the action. If you can figure out the object, everything else falls into place. Sometimes there is no object, however, and that's when things get really interesting for a theory of action.
>The discussion of the appropriate unit of analysis implies that there is one answer to the question. I don't think that there is such a universally appropriate best unit. Several suggestions have been proposed, ranging from 'word', through 'utterance' to 'sound envelope', and ranging in scope from the lexical item to the totality of modes of communicative action employed. All of these are attended to - at some level of conscious awareness - in the joint action and interaction of the participants, But which of these 'units' is chosen as the basic unit of analysis should depend on the question that the researcher is asking.
>I think this 'problem' and the varied solutions that have been adopted can be seen, for example, in A.N.Leont'ev's ways of analyzing 'activity'. Leont'ev proposes a tristratal account, in which activity (as 'motive') is realized in action (as a particular goal-oriented instance) and action, in turn, is achieved through operations, which are both relatively automatic and selected according to the particular exigencies of the situation. But, an action typically involves sub-actions, which involve further sub-actions... A vexing question, then, is which of these actions should be taken as the unit of analysis; another is, since operations equally consist of sub-operations, etc. which is the appropriate level of operation to focus on? And at what point does a particular form of 'behavior' cross from being an action to being an operation? His answer, as I understand it (or perhaps reconstrue it as) is that the way one maps the three strata onto a particular event depends on the !
purpose of the analysis. In spite of Engestrom's choice of 'action' as the level/stratum focused on in his triangle diagram of an activity system, I think he would give the same answer.
>Another example is Rogoff's argument that there are multiple 'zoom levels' that the researcher/camera can focus on in studying guided participation - individual, dyadic interactional, local community of practice, sociocultural - the choice on any occasion depends on the researcher's question. However, she also points out that, whichever zoom level is selected, the others are equally valid and are always part of the total picture.
>Ultimately, there is no best answer to the question. No one study can do everything and a choice has to be made - while recognizing that other choices afford alternative and equally valid perspectives,
>Dept of Education, http://education.ucsc.edu/faculty/gwells
>UC Santa Cruz.
This archive was generated by hypermail 2b29 : Fri Jul 01 2005 - 01:00:08 PDT