This concluding chapter focuses attention on the internal dynamics
of organizational life. We shall summarize the main points of our
foregoing analysis with particular emphasis on the processes of change
in formal organizations, processes that have sometimes been left
implicit in the earlier discussion. To do justice to problems of organi-
zational change, however, we shall not confine the analysis to material
already presented in previous chapters but shall draw upon additional
relevant sources whenever appropriate.

The concept of dilemma greatly contributes to an understanding
of internally generated processes of change in a social organization. If
we conceive of social systems as meeting basic needs and requisites, the
implicit assumption is that adjustment prevails and change does not
occur unless new external conditions require adjustments. But if we
conceive of social systems as being confronted by dilemmas, that is,
by choices between alternatives in which any choice must sacrifice
some valued objective in the interest of another, the implicit as-
sumption is that problems are endemic and, therefore, serve as a
continual internal source of change in the system. The concept of
dilemma by no means implies that improvements are not possible,
but it does imply that final solutions and perfect adjustments are
impossible. Dilemmas confront not only formal organizations, as we
shall see in this chapter, but also the study of formal organization, as
was mentioned in the first chapter.

A fundamental methodological dilemma in the study of social
organization is posed by the necessity to investigate the interdepend-
ence between units in a larger social system, on the one hand, and the
requirement to treat subunits of the totality under investigation as
independent cases in order to derive generalizations about them, on
the other. The investigator must choose between alternative approaches
and invariably sacrifice one of these requirements. Research that is
being conducted in one formal organization, for example, may either
treat the various sections in the organization as independent units for
analysis or focus upon their interdependence in the larger organization.
The first alternative furnishes independent cases that permit deriving
generalizations about work-group organization, but it ignores the

-222-