Purging the Judiciary After a Transition: Between a Rock and a Hard Place

Judges play a key role in the implementation of transitional justice mechanisms. Yet, less attention has been paid so far to the question of how to address their collaboration with non-democratic regimes. In theory, judges can be subjected to virtually all transitional justice mechanisms ranging from criminal prosecution and lustration to truth-seeking, or even amnesties. However, we show in a case study of Czechia that these mechanisms are not well equipped to address the complicity of judges in past crimes for three reasons: (1) judges usually play different roles in past crimes from political elites, (2) the principles of the separation of powers and judicial independence preclude the easy replacement of judges, and (3) pragmatic exigencies, such as the shortage of lawyers who are not tainted by cooperation with the previous regime, further complicate the renewal of the bench. Nevertheless, we argue that the lack of recognition of the role judges have played in non-democratic regimes is dangerous, as it may negatively affect public confidence in the judiciary and taint its legitimacy. Examples from Hungary, Poland and Romania, moreover, show that populist leaders are tempted to abuse the transitional justice rhetoric use the failure to deal with the past of judges as a justification for their court-curbing practices. Post-transition purges are therefore stuck between a rock (interfering in judicial independence and practical exigencies) and a hard place (mental dependence of the judiciary on the previous regime, low public trust in courts). When the democratic opposition defeats the populist leader, such as in Poland in 2023, it unfortunately faces the same dilemma. Thus, the Czech way of dealing with the past within the judiciary in transition from communism to democracy (transition 1.0) provides important insights also for today's undoing of populist judicial reforms and transition from authoritarian populism to democracy (transition 2.0).