pia). The pursuit of limited numbers of perpetrators allows for more accurate planning and costing. Donor fatigue has set in as a result of the high budgets for ICTY and the International Criminal Tribunal for Rwanda (ICTR).20 A clearly defined prosecutorial strategy will enable more careful budgeting vis-à-vis competing priorities, such as rebuilding the domestic justice system or dealing adequately with ongoing crime. (b) The impunity gap The fact that only a few perpetrators can be prosecuted creates challenges of what to do with the remaining perpetrators (and their victims): the so-called impunity gap. The fact that many people will not be investigated, much less prosecuted, should not mean that they should escape any form of accountability. Wherever possible, some effort should be made to deal with large numbers of those responsible who are not likely to face criminal justice. For example, thousands of military and civil servants not prosecuted in Greece in 1974 were subject to a far-reaching vetting process. Another example is Timor-Leste, where low-level perpetrators have been subjected to community reconciliation processes and may be required to perform community service. In order to bridge the impunity gap, prosecutorial initiatives will need to build constructive relationships with other transitional justice mechanisms. It is generally accepted that massive human rights violations require a complex and integrated response comprising a variety of complementary mechanisms, including prosecutions, truth-seeking mechanisms, institutional reforms, reparations and programmes that reintegrate ex-combatants. In designing each mechanism an effort should be made to ensure that they complement rather than undermine one another.21 However, the reality is that, in situations where these mechanisms co-exist and, particularly, where their relationships develop in ways other than as part of a coherent policy approach or a common vision,22 ideological and practical tensions that have the potential to cause disputes if they are not properly managed may emerge. For example, truth commissions may seek to engage ex-combatants or facilitate their reintegration, but the possibility of subsequent prosecution may discourage perpetrators from coming forward and speaking about their crimes.23 20  lthough it is generally acknowledged that international tribunals are expensive, an overly simplistic approach to highlightA ing this problem should be resisted. As mentioned above, the goals of prosecuting system crimes are broader than effecting retribution on a few individuals, and much more effort and resources may go into the strategic indictment of a few than is reflected in such a figure. 21 See S/2004/616, sect. IX. 22 S ituations in which truth commissions and prosecutions have formed part of a single policy approach include Timor-Leste, Peru and South Africa (although very few prosecutions were pursued in that context). Sierra Leone is an example of a situation where a truth commission and prosecutions were pursued as part of distinct and unrelated policies. 23 S ee, for example, “Ex-Combatant Views of the Truth and Reconciliation Commission and the Special Court in Sierra Leone”, September 2002, available at www.ictj.org.