It is because support cannot fail but to play a part in the context of politicalinstitutions that institutionalization has a peculiar and indeed a peculiarly import-
ant character in the political context. Yet, while the introduction of support in the
equation renders the analysis of institutionalization in politics more realistic, it
seems to complicate further the question of a deWnition of institutions in
the political context: This is because the question arises as to whether political
institutions can be examined independently from the support which they might
enjoy. Is a political organization or procedure still an institution even if it does not
have support or has only very little support? Are political institutions conditional
on them enjoying support?
It seems prima facie unrealistic to tie the very existence of political institutions to
the support which they might have: Institutions are organizations or procedures, as
we saw, characterized by ‘‘stable, valued, recurring patterns of behaviour’’ (Goodin
1996 , 21 ). Support seems extraneous to these characteristics: The way an institu-
tional arrangement is shaped does not seem to depend on the support for that
arrangement. Moreover, if support is brought into the picture, since support is
never ‘‘total,’’ the question arises as to what is the threshold below which the extent
of support would be too small for the arrangement to be an institution. The
diYculties are such that one is tempted to conclude that what makes an arrange-
ment an institution, in politics as elsewhere, is merely whether that arrangement is
a ‘‘stable, valued and recurring pattern of behavior.’’
It does seem prima facie reasonable to claim that governments, legislatures,
parties, indeed constitutions, exist as institutions even with very few followers and
need coercion to remain in being; But it is also doubtful as to whether, in the
extreme case of the near-complete collapse of such bodies, one can still refer to
them as ‘‘institutions,’’ unless one comes to the conclusion that one must distin-
guish between institutions and institutionalization. The government of a regime on
the verge of collapse is clearly ‘‘de-institutionalized:’’ Such a government seems
therefore to be no more than a ‘‘pseudo-institution.’’ The point needs to be
registered somewhere in the description of the organization or procedure under
consideration. Perhaps this is the reason why, deep down, political scientists have
found it diYcult to deWne what an institution is and why they have in some
ways felt more at home with the concept of institutionalization than with the
concept of institution.
Much work manifestly needs to be done before a coherent conception can be
expected to be found of what institutions consist of. Given the major diVerences
among the social sciences as to what institutions appear to be, however, it is
probably more realistic to undertake disciplinary eVorts before an overall social
science view of institutions is elaborated. There is no doubt that the notion refers to
highly distinct realities in economics, sociology, and politics. True to their tradi-
tions, economists are able to simplify the concept and reduce it to what seems a
homogeneous viewpoint. True to their traditions, too, political scientists are
728 jean blondel