institutionaldisequilibrium. 3 As such, rational choice and normative/sociological
institutionalism, which rely albeit for rather diVerent reasons on the assumption of
equilibrium, were theoretical non-starters. 4 Unremarkably, then, and by a process
of elimination, most routes to constructivist institutionalism can trace their origins
to historical institutionalism (see, for instance, Berman 1998 ; Blyth 2002 ; Campbell
2001 , 2004 ;Hay 2001 , 2002 ; McNamara 1998 ; Schmidt 2002 ).
Yet if historical institutionalism has typically served as an initial source of
inspiration for constructivist institutionalists, it has increasingly become a source
of frustration and a point of departure. For, whilst ostensibly concerned with
‘‘process tracing’’ and hence with questions of institutional change over time,
historical institutionalism has tended to be characterized by an emphasis upon
institutional genesis at the expense of an adequate account of post-formative
institutional change. 5 Moreover, in so far as post-formative institutional dynamics
have been considered (for instance Hall 1993 ; Hall and Soskice 2001 ; Pierson 1994 ),
they tend either to be seen as a consequence of path dependent lock-in eVects or,
where more ruptural in nature, as the product of exogenous shocks such as wars or
revolutions (Hay and Wincott 1998 ). Historical institutionalism, it seems, is incap-
able of oVering its own (i.e. endogenous) account of the determinants of the
‘‘punctuated equilibria’’ (Krasner 1984 ) to which it invariably points. This, at
least, is the charge of many constructivist institutionalists (see, for instance, Blyth
2002 , 19 – 23 ;Hay 2001 , 194 – 5 ).
If one follows Peter A. Hall and Rosemary C. R. Taylor ( 1996 ) in seeing historical
institutionalism as animated by actors displaying a combination of ‘‘calculus’’ and
3 Though hardly constructivist, the work of Robert H. Bates et al. ( 1998 ) is particularly interesting
in this regard. Operating from an avowedly rational choice institutionalist perspective, yet concerned
with questions of social and political change under conditions of disequilibrium which they freely
concede that rational choice institutionalism is poorly equipped to deal with ( 1998 , 223 ), they
eVectively import insights from constructivist research in developing a more dynamic but still
essentially rational choice theoretical model. Whilst the resulting synthesis can certainly be challenged
in terms of its internal consistency—ontologically and epistemologically—it does lend further
credence to the notion that constructivist insights have much to oVer an analysis of institutional
change under disequilibrium conditions (for a critical commentary see also Hay 2004 a, 57 – 9 ).
4 Strictly speaking, normative/sociological institutionalism does not so much assume as predict
equilibrium. For the ‘‘logics of appropriateness’’ that constitute its principal analytical focus and that
it discerns and associates with successful institutionalization are themselves seen as equilibrating.
The key point, however, is that, like rational choice institutionalism, it does not oVer (nor, indeed,
claim to oVer) much analytical purchase on the question of institutional dynamism in contexts of
disequilibrium.
5 Interestingly, this is something it seems to have inherited from the attempt to ‘‘bring the state
back into’’ (North American) political science in the 1980 s out of which it evolved (see, for instance,
Evans et al. 1985 ). For, in the former’s emphasis, in particular, upon the institutional and organiza-
tional capacity to wage war eVectively upon the process of state formation, it came to identify the
highly consequential and path-dependent nature of institutional genesis for post-formative institu-
tional evolution (see Mann 1988 ; Tilly 1975 ). In Charles Tilly’s characteristically incisive aphorism,
‘‘wars make states and states make war.’’
60 colin hay