Proxies for Progress


Richard Clements looks at Performance Indicators at the International Criminal Court

This month marks 15 years since the entry into force of the Rome Statute – the treaty establishing an International Criminal Court (ICC) which is designed to hold individuals criminally responsible for international crimes including genocide, war crimes, and crimes against humanity. Over the past 15 years, the ICC has received mixed reviews for its record, having investigated 10 situations in 9 countries, and hearing a total of 24 cases. In light of such statistics, calls have been made by States, victims and civil society groups for the Court to enhance its effectiveness and efficiency in dealing with mass atrocities. The Court has therefore begun to employ a range of methods to improve the monitoring of Court performance, securing cost-effectiveness, and generally bolstering its credentials as a ‘model of public administration’. One such method is that of performance indicators (PIs). This blog post analyses PIs, considering their benefits but also the difficulties inherent in using PIs in the context of international criminal justice.

In December 2014, the ICC Assembly of States Parties (ASP) requested that the Court ‘develop qualitative and quantitative indicators that would allow the Court to demonstrate better its achievements and needs’ and allow ‘stakeholders’ to more easily assess ‘Court-wide performance’. Critical issues were to be devised, data collected and targets set against which the performance of judges, lawyers and other institutional personnel could be measured. Two reports have been published in 2015 and 2016 providing information about measurement criteria, data collection and creation of additional indicators. Most recently, ICC judges and staff attended a retreat in Glion, Switzerland to discuss how best to measure Court performance and indicators featured prominently in their discussions. Given this broad turn towards indicators within the ICC, this method of evaluation deserves close scrutiny from interested actors, including States, victims, NGOs, and scholars.

Kevin Davis et al define indicators as ‘a named collection of rank-ordered data that purports to represent the past or projected performance of different units’.[i] To work effectively, indicators must be based on raw data which are used to measure the performance of actors over time against suitable benchmarks. Indicators emerged as part of the turn to management techniques within domestic public sectors in the 1970s and 1980s, when schools, hospitals, and courts began to be modelled on the private sector and its values of efficiency, cost-effectiveness, and performance oversight. PIs have emerged within international organisations only recently, having been implemented first in national jurisdictions including the United States, Australia, Canada, and many European countries which are monitored both internally and by the Council of Europe.

In her recent monograph, The Seductions of Quantification (2016), Sally Engle Merry conducts an ‘ethnographic analysis’ of the ‘production and use of global indicators’.[ii] Some of the advantages of indicators which Merry offers include the ability to hold States accountable via raw facts and figures which often reveal patchy implementation of international standards. Indicators are the product of expert research and are therefore ‘objective, scientific, and transparent’.[iii] This allows indicators to ‘stand above politics’ in an arena where politics appears omnipresent. The combined effect of such qualities is that indicators are able to describe the world we live in without the subjective or ideological baggage. Yet it is this claim to objectivity that also illustrates the power of indicators despite their drawbacks. As with any dataset, indicators can be incomplete and may not fully capture social complexities. The allure of scientificity and transparency allows them to go undetected as tools of power. Indicators are not free-floating data but man-made and therefore shaped by the personal and professional experiences, cultures, and beliefs of the experts who produce them. Rather than simply describing reality, they also have a role in constructing that reality thereby giving considerable power to those involved in the production of indicators. Ultimately, indicators ‘produce a truth about the world’ despite their flaws, a truth that more often than not entrenches existing distributions of global power and resources.[iv]

In addition to these abstract benefits and drawbacks of indicators, the unique features of the international criminal justice regime throw up some interesting questions about the purpose and utility of indicators, of which I illustrate three. Firstly, using PIs to assess judicial performance presumes a high degree of judicial control over the fate of the Court and the international justice project. However, the ASP and the Open Society Justice Initiative acknowledge that external factors such as money and cooperation play a significant role in the duration and expeditiousness of proceedings. Judicial control is therefore already incapable of being exercised over such factors, heightening the degree of scrutiny of factors over which judges do have control. Yet even here, assessment of individual performance seems to be misplaced. The idea that performance indicators can make judicial activity more transparent and, for example, less likely to be biased, presumes that through judicial bias or lack of commitment to the liberal values of international criminal justice, the entire project is capable of being upended. This assumption resembles the misconception, identified by David Kennedy, that institutions such as the ICC are capable of being ‘captured’ by individual and/or political agendas.[v] Judicial PIs are imposed out of a fear that judges might become ideologically partisan or that failure to monitor performance would close off certain variables from the public eye, reducing transparency and creating a ‘democracy deficit’. Implementing PIs presumes that judges possess a high degree of personal autonomy and thus responsibility for the fate of the international criminal justice system. Conversely, and to repeat a recent comment by Martti Koskenniemi, we should recognise that ‘it’s not the cases, it’s the system’.[vi] The underlying values, aspirations and institutional parameters – the structure – inside which judges operate provide little space for much judicial manoeuvring beyond finding that one of the core crimes is to be interpreted this way, or reparations are to be awarded in that fashion, never mind allaying fears of prosecutorial selectivity or an institutional bias against the African continent. A focus on PIs obscures the all-important structural dimension.

Secondly, indicators are reductive and do not contribute in any significant way to the improvement of Court performance. In its Report on the development of performance indicators for the ICC, the ASP stated that ‘the major features of proceedings before the ICC provide the obvious basis for a set of high-level strategic court-wide performance indicators’ (para. 7). Is this so self-evident? Is it possible to make a genuine link between the specific areas identified for improvement and the vague management-speak goals of ensuring ‘the Court’s proceedings are expeditious, fair and transparent’ or ‘the ICC’s leadership and management are effective’ or ‘victims have adequate access to the Court’ (paras 7a, b, and d)? It is unfortunate that any ‘external factors’ which might influence these goals are omitted from analysis. But this also begs the question whether indicators of, for example, the number of State referrals or the number of bilateral cooperation agreements are really sufficient to show heightened investment in and cooperation with the Court. At most, they can only hope to solidify the idea that information such as State referrals or bilateral treaties can act as proxies for the ambiguous and ultimately unattainable goals of international criminal justice. Indicators are thus unable to breach the divide between particular behaviours and general aspirations.

Finally, this alludes to what is perhaps the biggest problem with performance indicators and something the ASP acknowledged during its 13th session. Advocates of indicators claim that they can be used to hold States and other actors to account for their behaviours. Indicators therefore directly contribute to the effectiveness and efficiency of the international criminal justice system. Yet the ASP noted that indicators would allow the Court to ‘demonstrate better’ its achievements, suggesting that the real power of indicators lies in their legitimizing potential (Annex 1, para. 7b). Rather than improving Court performance as advertised, indicators are part of the public relations exercise that the ICC and other international courts now engage in. The elaborate process of suggesting, implementing, and improving performance monitoring techniques is not designed to improve Court performance, but to put a gloss on the organs’ activities and to justify the resources dedicated to them by States Parties. Given the difficulty in ensuring that justice is being done, it is sufficient that justice is perceived to be done. This is the ultimate purpose behind management techniques and language as embodied in PIs.

The problems identified above are designed to question the worth of using PIs at the ICC. Whilst these effects of PIs are not exhaustive, they query the motivations behind and utility of indicators, recharacterizing them as lazy and reductive proxies for social reality and progress, and tools for legitimating existing global power structures.

Richard Clements is a PhD Candidate, Girton College and Lauterpacht Centre for International Law, Cambridge University and incoming Editor-in-Chief, Cambridge International Law Journal 2017-2018.


The new Cambridge International Law Journal is available now, read the first issue free on Elgaronline

, , , , , , , ,


Subscribe to our RSS feed and social profiles to receive updates.

No comments yet.

Leave a Reply

%d bloggers like this: