>From the viewpoint of traditional notions, this is much the same
as saying that we are different Selves in different situations or
activities (that we can be seen as such, just as we can be seen
in some other respects as the same from one to another).
But the first step really only considers one situation or
activity at a time, it is local and microsocial in nature. Yes,
we bring in the macrosocial and the historical and cultural
dimensions in terms of our accounts of the origin and nature of
the kinds of situations and activities that exist in a community
(and so the kinds of identities and behavior patterns, and
individuals, it produces). But we do not usually extend the unit
itself to higher levels of social organization than the local
situation, the immediate activity.
Much the same thing has been true in discourse analysis, and a
great change has come recently because of the growing emphasis on
'intertextuality' -- on the notion that the meaning here in this
text also depends on relationships with other texts.
Can we interpret an institutionally-mediated notion of person,
self, identity, behavior in similar terms? I think we can if we
also use something like Latour's actor-network model for our view
of what an institution is. Traditional sociological notions about
institutions often simply reify abstractions (the family is an
institution, but so is the English language), or they get lost in
'the middle range' between macrosocial categories and microsocial
interactions. Latour gives a 'bottom up' view of institutions in
terms of the networks of interdependent activities that
constitute them (e.g. scientific disciplines, natural history
museums, research laboratories, etc.). This view enables us to
say _which_ other activities belong to the same 'institutional
network', and _how_ they are connected. These are the same two
things one needs for a theory of intertextuality (which other
texts are relevant, and how).
Now we can say that an individual can be defined, analysed,
traced, described, theorized, etc. not just in relation to each
immediate situation or activity, and NOT in terms of some
ideology of invariant features across all possible activities,
but specifically in terms of self-construction and participation
in a _set_ of interlinked activities, where the linking is
precisely that defined by the constitution of a particular
institution. In this way we can get a useful notion (and
researchable method) of a 'family-individual', a 'school-
individual', a 'laboratory-individual' etc. These might all be
construable as 'the same' individual at another level of analysis
(what level? the societal-individual level?), but clearly they
represent something much closer to what many of us want to study
than does either the 'all-purpose individual' (which probably
exists only at so high a level of abstraction as to rather
useless, unless reconceived) or the 'activity-specific
individual' (even if defined in relation to a _type_ of activity,
rather than a particular activity-event; institutions consist of
heterogeneous networks of many types of activity). I think such a
notion represents a considerable advance over both traditional
role theory and our existing notions of individuals in units of
analysis.
JAY.
PS. Note, depending on your way of defining 'activity', that even
if we take an activity to be an extended set of actions unified
by a common purpose, or as a common social-cultural formation
(getting closer to Latour), an activity-network model will
generally include many such 'activities' as part of what defines
an institution and so what may define an institutional-individual
(and his/her institution-mediated participation in particular
specifically institutional activities).
---------------
JAY LEMKE.
City University of New York.
BITNET: JLLBC who-is-at CUNYVM
INTERNET: JLLBC who-is-at CUNYVM.CUNY.EDU