19
First, the subjects examined here could have acted otherwise, they would not. It is
fallacious to equate human autonomy with unpredictability or the absence of learned
constraints upon action. The dichotomy of agent and structure is simply fallacious. It is
meaningful to speak of subjects as autonomous from specific other subjects, but not from
whole social structures. For example, its makes little sense to speak of persons as
autonomous from language. Instead, we develop here a model of the social constitution
of will. Structures contribute to the constitution of the subject, the entity which has
autonomy. The autobiographical scripts that define subjects arise from the assimilation of
actions produced by other autobiographical scripts. Communication and dramaturgical
presentations among subjects create the common knowledge that makes meaningful
interaction possible. The autobiographical scripts are the sources of subjects’ will, and
action according to that will is the condition of autonomy of the subject. Autonomy can
be conceived, not as the absence of causal relationships between structures and subjects,
but as a structural product itself.
Second, the problem of bracketing is overcome through the concept of
reproduction. There is a chronology of actions and statements, and there are cyclical
relationships within that chronology. The cyclical relationships, the traces of
reproduction, are the structure. One cannot perceive such structure without attending to
the chronology, to history. It is the analysis of reproduction as an empirical process that
overcomes the hermeneutic circle. That autobiographical interaction narratives generated
actions, and were generated by actions that were generated by other subjects’ scripts, in a
specific historical sequence, means that an ecology of scripts emerged that could
reproduce together. Like biological ecosystems, historical structures exist only over time,
only diachronically.
Finally, there is a great deal of intersection between subjects and structures in this
model. This is a continuation of the first point above. IR structurationists who seek
disjoint definitions of agents and structures are simply wrong to do so. There is no need
to define them separately to ensure the autonomy of the subject. The subject is formed
from structural semantic components. Its autonomy is an emergent property. Indeed, that
which reproduces itself, the structure, includes the whole of each interacting subject. The
process of structural reproduction is necessary for subject autonomy. Structural
Convention