such a clear-cut manner: ‘‘Each of the several disciplines that collectively constitute
the social sciences contained an older institutionalist tradition. In each case that
tradition has recently been resurrected with some new twist... .The new institu-
tionalism mean [sic] something rather diVerent in each of these alternative discip-
linary settings’’ (Goodin 1996 , 2 ). The characteristic meaning of institutions in each
discipline is then examined, and the author continues: ‘‘There is wide diversity
within and across disciplines in what they construe as ‘institutions’ and why. That
diversity derives, in large measure, from the inclination within each tradition to look
for deWnitions that are somehow ‘internal’ to the practices they describe’’ ( 1996 , 20 ).
A ‘‘central deWning feature’’ is then attempted: having adopted what he refers to as an
‘‘external’’ account of what institutions are and what they do, the author states that
‘‘a social institution is... nothing more than a ‘stable, valued, recurring pattern of
behaviour’ ’’ ( 1996 , 21 ), the formula being that of Huntington.
Yet it is probably more fruitful to look at diVerences among the social sciences in
this respect and in particular at what is suggested by economists and sociologists
alongside political scientists. There appears to be a dimension, with economics and
political science at the two extremes and sociology somewhere in the middle. As
was noted early in this chapter, W. R. Scott, a sociologist, stated that institutions
covered both organizations and activities: this is indeed the middle position
characterizing sociological analysis with respect to institutions. What was said
earlier suggests that in politics the traditional stress—including that which both
Lawson and Lane and Ersson indicated—was that institutions were Wrst and
foremost organizations. In economics, on the contrary, the emphasis is exclusively
on procedures. ‘‘Institutions are the rules of the game in a society or, more
formally, are the humanly devised constraints that shape human interaction:’’
this is how D. C. North begins his book onInstitutions, Institutional Change and
Economic Performance( 1990 , 3 ). No reference whatsoever is made to organizations
in this description.
Whether it is worth trying to reconcile the points of view of the disciplines about
institutions is debatable: it is unquestionably valuable to note that major
diVerences exist about the meaning of the concept. These exist because the three
disciplines are concerned with diVerent sets of problems. As Goodin points out,
economists are primarily concerned with solving the problem ofindividualchoice
( 1996 , 11 ) and are therefore concerned with rules. The individuals are the agents of
the economic ‘‘machine’’ (whether as physical individuals or in association with
each other inWrms): Individuals cannot be expected to achieve their goals unless
there are rules which determine how they are to relate to each other.
The situation is diVerent in the society at large: Individuals congregate to
form associations, unless theyWnd themselves in bodies which have existed for
generations, whether in traditional societies (tribes) or in modern societies (from
families to churches). These bodies constrain individuals. Institutions cannot just
be based on rules; they have to include the way collective arrangements, in groups,
about institutions, mainly, but not exclusively, political 721