relatively large-n studies), it is important to remember that rules work very differently “in
action” in different institutional contexts. As I will argue below, just as institutional
measures intended to increase judicial independence may have effects that are limited in
numerous ways, judiciaries that are usually accepted as relatively independent sometimes
operate within the context of institutional rules that make them look relatively weak and
dependent on paper. The gap between formal rules and political consequences is particularly
relevant because the frequent unwillingness of legislatures to discipline the courts with
powerful formal tools. Section III provides a framework for considering ways in which the
judiciary can exercise power despite structural dependencies.
Limitations to Formal Studies of Judicial Power
Some of the problems that arise from using
formal rules to assess judicial independence can be observed in a study by Smithey and
Ishiyama (2000). In their study of courts in post-Communist European democracies,
Smithey and Ishiyama construct a variable of “judicial power” using six variables: the
presence of legislative or executive overrides of judicial decisions, the presence of a priori
review, the terms of judges compared to other political actors, the actors responsible for
creating court procedures, and the conditions for judicial removal. (2000: 170) Precisely
because these variables represent sound intuitions about judicial independence and power,
they illustrate some of the profound difficulties in empirically studying judicial independence.
There is, first of all, the problem of weighting the variables. It would seem clear that security
of tenure is a vastly more important component of judicial independence than the existence
of a priori review, and hence the decision of the authors to weight them equally is in this
respect unsatisfactory—but what is the appropriate weight? In addition, the variable of the
number of actors involved in judicial selection returns us to Shapiro’s paradox. Which