Conclusion

The study of the state, as we have seen, is very different depending upon the kind of new institutionalism. Each has a different object of explanation - whether rational behaviour, historical structures, norms and culture, or ideas and discourse; a different logic of explanation - whether interests, path-dependency, appropriateness, or communication; a different emphasis on continuity or change - whether on continuity through fixed preferences, through path dependency, or through cultural norms, or on change through ideas and discursive interactions (see Table 5.1).

The result is that there are very different kinds of institutionalist studies of the state, many of which focus little on the state itself but, rather, on different kinds of action within the state.

To get a sense of how all of this fits together in a very general way, I conclude with a chart that situates the works cited above within each of the four institutionalisms while arraying the four institutionalisms along a horizontal continuum from positivism to constructivism - from interests to culture, with history in between - and along a vertical continuum from statics to dynamics, with interests, history, and culture at the static end, ideas and discourse at the dynamic end (see Figure 5.1). I put historical institutionalism between rational choice and sociological institutionalism, mainly because rational choice and sociological institutionalism are largely incompatible, whereas historical institutionalism can go either to the positivist or the constructivist side when it adds agency. I put discourse institutionalism underneath all three because, although it is distinctive, it

communities

Table 5.1		The four new institutionalisms
	Rational choice institutionalism	Historical institutionalism	Sociological institutionalism	Discursive institutionalism
Object of	Rational	Historical	Norms and	Ideas and
explanation	behaviour	structures	culture	discourse
Logic of explanation	Interest	Path-dependency	Appropriateness	Communication
Ability to	Static:	Static: emphasis	Static: emphasis	Dynamic:
explain	emphasis on	on continuity	on continuity	emphasis on
change	continuity through fixed preferences	through path dependency	through cultural norms	change and continuity through ideas and discursive interaction
Examples	Principle-agent theory; game theory	historical institutionalism process tracing varieties of capitalism	Constructivism; norms; cultural analysis	Ideas; discourse; constructivism; narratives; frames;

advocacy coalitions; epistemic

Figure 5.1 Scholars’ use of the four new institutionalisms: rational choice (RI), historical (HI), sociological (SI) and discursive (DI)

Positivism _____________________________________ Constructivism

interests * culture

history

ideas

discourse

RI	HI		SI
	Skocpol		Powell/Dimaggio
Shepsle	Krasner
	Katzenstein 1985		Dobbin
McCubbins/	Hall 1986		Soysal
Sullivan Hall 2001
Pierson
Martin Scharpf			March/Olson
Schmidt 2002 Pt II
	Thelen		Fligstein
Goldstein/Keohane
Weingast	Hall 1989		Katzenstein 1996
	Hall 1993 McNamara		Wendt
Berman		Blyth	Risse
Haas		Hay/Rosamond
Sabatier		Muller/Jobert
Schmidt 2000, 2002 Pt III
	DI

Static

norms

discourse

Dynamic

can rest upon the insights of any one of the three and because scholars often see themselves as continuing to fit in one or another of the traditions even as they fit best in discursive institutionalism. I have a darkened line under rational choice institutionalism to indicate its inability to handle ideas and discourse in a dynamic way

Among the questions that remain to be answered is one crucial one: can these four new institutionalisms fit together? Can empirical studies of any one issue mix approaches? Many of the most theory-driven of new institutionalists would answer in the negative, because their purpose is to demonstrate how their particular approach is the best way of explaining politics.

More problem-oriented scholars mix approaches all the time, using whichever approaches seem the most appropriate to explaining their object of study. More recently, some scholars have also been addressing the question of how to use the insights of all four approaches in their empirical work. In policy analysis, to take just one example, David Marsh and Martin Smith (2000) have proposed a dialectical approach to understanding policy networks which uses methods from all four new institutionalisms to explaining the ways in which policy actors in given institutional contexts under certain constraints with particular learning experiences and ideas choose different courses of action over time. But while this may help answer the practical question of how to meld approaches in empirical investigation, it does not resolve the question of how they may fit together theoretically. Indeed, a look at the responses to Marsh and Smith demonstrates this well, as all come back to defend their own approaches (e.g., Dowding 2001; Raab 2001). For a theoretical answer to the question of how the various approaches fit together, new institutionalists need first to stop seeing their relations with rival approaches as methodological wars where the battles are fought over conceptual territory. They would do better to declare peace, and begin exploring areas of mutual compatibility along their borders. This would surely move all four new institutionalisms forward theoretically, while providing the greatest benefits for empirical research.

Conclusion

More on the topic Conclusion: