from—and indeed much less important than—the one which it has in political
science. Referring to a work by Selznick, Scott says that ‘‘[o]rganisations with
more precisely deWned or with better developed technologies are less subject to
institutionalization than those with diVuse goals and weak technologies’’ ( 1995 ,
19 ). He then notes the diVerence between ‘‘oYcial’’ and ‘‘real’’ goals of organiza-
tions and the part played by power. Quoting Stinchcombe, who stated that
institutions are ‘‘structures in which powerful people are committed to some
value or interest’’ ( 1968 , 107 ), he emphasizes that ‘‘values are preserved and
interests are protected if those holding them retain power’’ (Scott 1995 , 19 ).
Referring to experiments conducted by Zucker ( 1977 ), Scott states that ‘‘to
create higher levels of institutionalization, the subject [of the experiment] was
told that she and her co-worker were both participants in an organization and
the co-worker (the confederate) was given the title of ‘light operator’ ’’ (Scott
1995 , 83 ). Institutionalization is ‘‘manipulable,’’ so to speak. It is not a state
acquired over time, but a state which an institution has when certain conditions
are fulWlled, conditions which may or may not, more or less at will, be
introduced.
Why is it the case with the kind of social organizations which Scott examines,
and apparently not the case in politics? The point is that Scott is concerned with
what occurs among the members of an organization and not with the eVect
which the organization may have on persons outside the organization. Such
a point of view is justiWed from the point of a sociologist, particularly of
a sociologist who focuses on relatively small organizations or onWrms which do
not have to ensure that their decisions are applied outside the organization.
For those in the organization, the organization is indeed institutionalized:
Employees have to abide by the rules, not merely because if they do not,
they are likely to be dismissed, but because they cannot relate to other
employees unless there is some agreement, that is to say unless some rules are
institutionalized.
The case is diVerent in politics, as was pointed out earlier, as it is diVerent in
‘‘social’’ organizations which attempt to impose their views on non-employees,
for instance trade unions, employers’ organizations, or many NGOs. When what
can be described as an ‘‘external constituency’’ plays a part in the life of an
organization, it cannot be taken for granted that people will abide by the rules
and therefore that institutionalization will be ‘‘automatic’’ and ‘‘instantaneous:’’
It has to be built. The fact that institutionalization has to be built (and,
conversely, can be ‘‘unbuilt’’) explains why institutionalization is such a central
concept in the analysis of political scientists. Yet, although it is central for
political scientists—and somehow perceived as central—it is surprising that the
basis for the development of institutionalization has not been systematically
explored.
about institutions, mainly, but not exclusively, political 725