‘‘cultural’’ logics, then it is perhaps not diYcult to see why. For, as already noted,
instrumental logics of calculation (calculus logics)presumeequilibrium (at least as
an initial condition) 6 and norm-driven logics of appropriateness (cultural logics)
are themselves equilibrating. Accounts which see actors as driven either by
utility maximization in an institutionalized game scenario (rational choice insti-
tutionalism) or by institutionalized norms and cultural conventions (normative/
sociological institutionalism) or, indeed, both (historical institutionalism), are
unlikely to oVer much analytical purchase on questions of complex post-formative
institutional change. They are far better placed to account for thepath-dependent
institutional change they tend to assume than they are to explain the periodic, if
infrequent, bouts of path-shaping institutional change they concede. 7 In this
respect, historical institutionalism is no diVerent than its rational choice and
normative/sociological counterparts. Indeed, despite its ostensible analytical con-
cerns, historical institutionalism merely compounds and reinforces the incapacity
of rational choice and normative/sociological institutionalism to deal with dis-
equilibrium dynamics. Given that one of its core contributions is seen to be its
identiWcation of such dynamics, this is a signiWcant failing.
This is all very well, and provides a powerful justiWcation for a more construct-
ivist path from historical institutionalism. It does, however, rest on the assumed
accuracy of Hall and Taylor’s depiction of historical institutionalism—essentially
as an amalgamation of rational choice and normative/sociological institutionalist
conceptions of the subject. This is by no means uncontested. It has, for instance,
been suggested that historical institutionalism is in fact rather more distinctive
ontologically than this implies (compare Hay and Wincott 1998 with Hall and
Taylor 1998 ). For if one returns to the introduction to the volume which launched
the term itself, and to other seminal and self-consciously deWning statements
of historical institutionalism, oneWnds not a vacillation between rationalized and
socialized treatments of the human subject, but something altogether diVerent.
Thelen and Steinmo, for instance, are quite explicit in distancing historical
institutionalism from the view of the rational actor on which the calculus approach
6 This is, of course, not to deny that standard rational choice/neoclassical economic models can
describe/predict disequilibrium outcomes (think, for instance, of a multiplayer prisoner’s dilemma
game). Yet they do, assuming initial equilibrium conditions.
7 The distinction between path-dependent and path-shaping logics and dynamics is a crucial one.
New institutionalists in general have tended to place far greater emphasis on the former than the latter.
This perhaps reXects the latent structuralism of the attempt to bring institutions back into contem-
porary political analysis (see Hay 2002 , 105 – 7 ). For institutions, as structures, are invariably seen
to limit, indeed delimit, the parameters of political choice. As such, they are constraints on
political dynamism. This is certainly an important insight, yet there is a certain danger in tilting
the stick too strongly in the direction of structure. For, under certain conditions, institutions, and
the path-dependent logics they otherwise impose, are recast and redesigned through the
intended and unintended consequences of political agency. Given the importance of such moments,
the new institutionalism has had remarkably little to say on these bouts of path-shaping institutional
change.
constructivist institutionalism 61