Legitimising International Criminal Justice: The Importance of Process Control
Legitimising International Criminal Justice
The Importance of Process Control
International criminal tribunals have recently undergone two important evolutions: a procedural evolution and an evolution in the regulation of defence counsel. These two evolutions appear largely unrelated, and they appear to have stemmed from two very different motivations: the procedural evolution, from a desire to reduce the length and cost of international criminal proceedings, and the defence regulation evolution, from a desire to improve the quality of defence representation. This chapter summarises an argument I put forth in a recent law review article (Combs 2012), where I argued that although those motivations do explain the respective evolutions at a superficial level, a far more profound and far-reaching evolution stands at the heart of both of them. I show that the early international tribunals had little choice but to adopt party-driven adversarial procedures and to permit defendants maximum discretion in selecting the counsel to represent them. More importantly, by invoking decades of social psychology research involving process control, I show the way in which the tribunals’ early procedures and policies served to strengthen and legitimate the then-vulnerable institutions that employed them. Once international criminal justice had gained a measure of legitimacy, the tribunals were able to institute valuable procedural and regulatory reforms. Thus, the procedural and defence counsel evolutions reflect an underlying and far more fundamental evolution in international criminal justice as a whole: the evolution from a novel, distrusted criminal justice system to a respected, credible accountability mechanism.
The Procedural Evolution
The International Criminal Tribunal for the former Yugoslavia’s (ICTY) first set of procedural rules was substantially adversarial in character. The initial ICTY rules created a neutral, largely passive role for the judges and bestowed on the parties primary authority for developing their cases and presenting their evidence at trial. In particular, the rules adopted a “two-case” system in which the prosecution and defence each presented their best arguments and evidence to the judges and each challenged the other side’s evidence (see Langer and Doherty 2011; Safferling 2001: 223). The initial rules did permit tribunal judges to take some active fact-finding steps, by authorising them to summon their own witnesses (ICTY RPE 1994: r. 98), ask their own questions (r. 85(B)), and alter the order of evidence presentation “in the interests of justice” (r. 85(A)). However, the judges used these powers sparingly and were expected to do so.
Early tribunal proceedings under these rules were lengthier and costlier than anyone considered optimal, and many commentators placed a large share of the blame on the adversarial nature of the proceedings (Cassese 2003). More specifically, ICTY judges along with a UN Expert Group identified the judges’ failure to adequately control proceedings as causing substantial delay (UNGA 1999: 76–77). Consequently, the Expert Group advised the Chambers “to take a more active role in trials” (76).
The ICTY followed these recommendations. Specifically, between 1998 and 2003, the ICTY introduced into its proceedings pre-trial judges, status conferences, and pre-trial conferences, all of which were designed both to streamline the proceedings and to arm the Trial Chambers with information that would enable them to exercise greater control over their cases. The changes redistributed some process control from the ICTY’s parties to its judges. In particular, following the reforms, the parties had less control over which witnesses to call and how much time to spend questioning them; moreover, the parties had to contend with a better informed judiciary that had been strongly encouraged to take charge of the proceedings in a variety of both large and small ways. This trend toward greater judicial control was also accelerated with the creation of subsequent international tribunals. International Criminal Court (ICC) and Special Tribunal for Lebanon (STL) procedures contain far more non-adversarial elements than appear even in the reformed ICTY procedures, and Extraordinary Chambers in the Courts of Cambodia (ECCC) procedures are based almost entirely on the French non-adversarial system; consequently, ICC, STL, and ECCC judges are authorised to exercise far greater control over their proceedings.2 Thus, whereas international criminal procedures began at the ICTY with a “strong adversarial orientation,” (Langer and Doherty 2011: 245) the recent introduction of substantial non-adversarial elements into the procedures of all of the international tribunals has transformed international criminal procedure law into a “truly mixed” system (Ambos 2003: 34–37).
The Evolution in the Regulation of Lawyer Selection and Discharge
The procedural evolution just described is relatively well-known. By contrast, the focus of this section—the evolution in the regulation of the lawyer-client relationship at the international tribunals—has escaped scholarly attention. The ICTY, in its early days, offered defendants an almost limitless choice regarding who, if anyone, would represent them. It did so by imposing virtually no qualification requirements on counsel who wished to practice before the ICTY, by permitting defendants to fire their lawyers virtually at will, by allowing defendants to consent to even the most troubling conflicts of interest, and by accommodating defendants’ desire to self-represent despite considerable disruption to trial proceedings. In recent years, however, the ICTY has imposed restrictions in each of these areas. Subsequent tribunals have followed this trend, either by adopting more restrictive policies at their outset or by imposing similar or more robust restrictions than the ICTY.
The vast majority of international criminal defendants do not have the means to pay for retained counsel (Rohde 2003: 544; Tolbert 2003: 978; Wilson 2002: 170; UNGA 2006: 48–49),3 so virtually all of the lawyers appearing before the tribunals have been appointed at the tribunals’ expense.4 Although human-rights law does not require the tribunals to provide indigent defendants with a choice of counsel,5 the tribunals have endeavoured to do so in most cases (Kay and Swart 2002). Typically, the Registrar (or the Defence Office, where one exists) will provide defendants with a list of approved counsel from which the defendant may make a selection.6 The Trial Chambers have held that the Registrar is not required to appoint the defendant’s chosen counsel,7 but so long as that counsel is either on the list or eligible to be placed on the list, then he or she should ordinarily be appointed unless the Registrar has reasonable and valid grounds for denying the defendant’s request.8 These appointment practices have not changed significantly over the years. What have changed markedly, however, are the criteria for being placed on the Registrar’s list and the conflict-of-interest rules that now supply new grounds for a denial of the defendant’s selection.
In its early years, the ICTY required defence counsel assigned to represent indigent defendants to possess only the most minimal qualifications. In particular, a lawyer was eligible for assignment so long as the lawyer was either a University professor of law or admitted to the practice of law and could speak one of the tribunal’s working languages (ICTY RPE 1994: r. 45). Moreover, the ICTY soon reduced these negligible requirements still further by authorising the assignment of defence counsel who were unable to speak one of the Tribunal’s working languages so long as counsel spoke the language of the accused.9
In recent years, however, the ICTY has imposed additional, more burdensome entry requirements. The language-proficiency requirement now applies both to counsel appointed for indigent accused, as well as counsel retained by non-indigent accused, and it requires both “written and oral proficiency” in one of the Tribunal’s official languages.10 The ICTY’s Registrar has also introduced a more stringent vetting process to enable it to assess the language capabilities of counsel who claim to speak English or French.11 Finally, whereas in the past the Registrar could waive the language requirement for any defence counsel, now the Registrar is permitted to waive the language requirement only for defence counsel assigned as co-counsel.12
In recent years, the ICTY has also required counsel assigned to represent indigent defendants to have additional substantive experience. Although the ICTY initially permitted the appointment of any lawyer who was admitted to the practice of law or who was a University professor of law, in 1999, a UN Expert Group found these standards inadequate and recommended that the ICTY’s requirements “be elevated to require at least five years of criminal trial experience” (UNGA 1999: 210). It was not until 2004, however, that the ICTY began requiring appointed defence counsel to have both “established competence in criminal law and/or international criminal law/ international humanitarian law/international human rights law” and “at least seven years of relevant experience … in criminal proceedings” (ICTY RPE 2010: r. 45(B); Defence Counsel Directive 2006: art. 14).
In addition to meeting these requirements, lawyers who wish to represent ICTY accused now must be members of the Association of Defence Counsel, an organisation established in 2002 (see ICTY RPE 2002: r. 44(A);13 Defence Counsel Directive 2006: art. 14(A)(x)). The Association is authorised to expel members, which thereby prevents them from representing ICTY defendants (ADC 2004: art. 6(1)(c)(ii), (iii)), and to require them to attend training programs (ICTY 2002). The Association also has a Disciplinary Council that monitors its members’ representation and adjudicates complaints received against Association members for alleged misconduct (ICTY 2009: art. 16). The ICTY itself has also become involved in disciplining counsel. During the tribunal’s early years, it had no code of conduct for defence counsel. The ICTY adopted a code in 1997, but that code contained no disciplinary mechanisms. The current code, by contrast, creates a Disciplinary Panel to adjudicate ethical complaints made against defence counsel and requires counsel to report the professional misconduct of a colleague to the Panel (Code of Professional Conduct 2009: art. 36). The Panel has broad investigatory powers (art. 44) and also has the power to issue an order that counsel be suspended from practising before the Tribunal until the charge has been adjudicated (art. 45(A)). Moreover, lawyers who have been found to have violated their professional obligations may be disqualified from subsequently representing ICTY accused (ICTY RPE 2010: r. 44(A)(iv)).14
In addition, all of the other international tribunals have, from their outsets, required counsel to meet more rigorous entry standards than those initially in place at the ICTY. For instance, counsel who do not speak one of the respective tribunal’s working languages are not eligible to represent indigent defendants in any of the other international tribunals.15 In addition, like the International Criminal Tribunal for Rwanda (ICTR), the other international tribunals have required more extensive experience and qualifications of their defence counsel.16 Finally, not only have the other international tribunals shown a greater initial willingness than the ICTY to impose credible entry requirements on defence counsel, they also appear to take a more active role in monitoring the performance of their defence counsel. For instance, once assigned, an ICC defence counsel is required to “maintain a high level of competence in the law applicable before the Court” and to “participate in training initiatives required to maintain such competence” (ICC 2006a: art. 7(2)).
Conflicts of Interest
Imposing more burdensome entry requirements on defence counsel is not the only way in which international tribunals can restrict their defendants’ choice of counsel. Current ICTY defendants are less able than their predecessors to be represented by the counsel of their choice because changes to the ICTY’s conflict-of-interest rules make it more likely that counsel will be found to suffer from a disqualifying conflict of interest.17 Some revisions concern the lawyer’s general orientation toward her client and the tribunal,18 but the amendments that are most relevant to this study concern the defendant’s ability to consent to conflicts. For instance, whereas the ICTY’s original Code of Conduct for Defence Counsel permitted a client to consent to representation when his lawyer represented another client with materially adverse interests in a similar or identical matter (ICTY 1997: art. 9(3)(c)(iii)), the current code does not permit consent in such a situation (ICTY 2009: art. 14(D)(iii)).19
An even more significant change pertains to the defendant’s ability to consent to any conflict. The ICTY’s initial Code provided that, when a conflict arose, counsel either had to take all steps necessary to remove the conflict or had to “obtain the full and informed consent of all potentially affected Clients to continue the representation, so long as Counsel [wa]s able to fulfil all other obligations under th[e] Code” (ICTY 1997: art. 9(5)(b)). Pursuant to this rule, then, clients were permitted to consent to serious conflicts.
By contrast, the current ICTY Code limits a defendant’s ability to consent to a conflict because it permits consent only where it is not “likely to irreversibly prejudice the administration of justice” (ICTY 2009: art. 14(E) (ii)(2)). This vague restriction gives ICTY judges considerable discretion to hold that a conflict cannot be cured by a defendant’s consent. Indeed, pursuant to this provision, the ICTY has refused to permit consent in a series of cases presenting conflicts.20 Concededly, the conflicts at issue in these cases were serious, but under previous rules they would not have prevented defendants from retaining the counsel of their choice.21
Just as the ICTY initially afforded defendants wide discretion in selecting their first lawyer, it similarly afforded them free rein to fire the lawyer they had selected and to choose another one. In particular, ICTY defendants needed to make no showing of inadequate or wrongful behaviour on the part of their current lawyers in order to get those lawyers replaced. It was enough, that “the reasons for the accused’s dissatisfaction with the counsel … are genuine and that the request is not being made for frivolous reasons or in a desire to pervert the course of justice, e.g., by causing additional delay.”22 These lax standards led in some cases to “a bewildering series of changes of counsel representing an individual Defendant” (Greaves 2001: 185).23
Just as the ICTY has in recent years applied more stringent entry requirements and conflicts-of-interest rules on defence counsel, it likewise applied a more stringent standard to requests for counsel changes. For instance, by 2003, a defendant was allowed to change his defence counsel only upon a showing of “unusual and compelling circumstances” (UNSG 2003: para. 49) such as where there is “a complete breakdown in the relationship between counsel and client, or if counsel is ordered to withdraw from a case for ethical reasons.”24 Some cases, however, suggest that even a “complete breakdown in the relationship between counsel and client” may not be sufficient to constitute exceptional circumstances justifying a counsel change. (Combs 2012: 345). Thus, as things currently stand, a genuine breakdown in communication and a lack of trust are sufficient to justify a change of counsel only when they stem from counsel’s failure to fulfil his or her professional responsibilities.
The statutes of all of the international tribunals provide an accused with the right to “defend himself in person or through legal assistance of his own choosing” (UNSC 1993: art. 21(4)(d)),25 and the tribunals have interpreted this language as providing defendants with a right of self-representation. The ICTY has considered the right of self-representation in several cases; however most of them featured both notorious defendants and starkly different factual circumstances, so it is difficult to trace a clear evolution in the ICTY’s treatment of this issue over time. Nevertheless, it is fair to say that whereas the tribunal initially showed great willingness to tolerate delay and to otherwise adapt its proceedings as a consequence of an accused’s decision to represent himself, that willingness substantially declined over time. Indeed, as early as 2006, Nina Jørgensen observed that the focus of the ICTY’s self-representation jurisprudence had “shifted from establishing the modalities for the exercise of the right … to establishing the circumstances under which the right may be qualified and the modalities of restricting it” (Jørgensen 2006: 65). In particular, whereas the ICTY initially bent over backwards to preserve defendants’ right of self-representation (even though doing so disrupted, delayed, and impaired the dignity of those trials), the ICTY has shown no similar solicitude for later defendants’ desire to self-represent (UNSC 1993: arts. 347–359).
Evolving Towards Legitimacy
We now know that the ICTY was the crucial first step on the road to an international criminal law revolution, but when it was created its significance—indeed its very survival—was in considerable doubt (Hazan 2004). Its eventual success was seen as even less likely. The ICTY was established in the midst of an armed conflict by Western powers that were reluctant to devote military resources to ending the conflict, so the tribunal was seen less as a principled effort to extend the reach of criminal accountability to mass murderers and more as a fig leaf designed to conceal Western unwillingness to take truly effective action (Scheffer 2004: 353; Hazan 2004: 21, 42, 49). Moreover, the tribunal’s legitimacy was questioned—even by those in the international community—because no tribunal had been established to prosecute the authors of the many atrocities committed during the fifty years following the World War II trials. Thus, the decision to prosecute Yugoslavian offenders was perceived by many—both inside and outside of the former Yugoslavia—as exemplifying biased and selective justice (Henley 1994; U.N. Council OKs Yugoslav War Crimes Court 1993: 8A; Time for a Global Criminal Court 1994; Jeu 2004: 424).
Within the former Yugoslavia, the ICTY’s legitimacy and impartiality was even more suspect. Serbs throughout the region considered the tribunal an illegitimate and biased court that had been established to persecute them. Bosnian Muslims were more favourably disposed to the tribunal, but they still viewed it with a healthy dose of scepticism because they perceived it to be yet another ineffective gesture from an indifferent international community (Coll 1994: J8). Finally, the very fact that the tribunal was established while the war was taking place meant that accurate information was scarce and that each party to the conflict was able to propound a self-interested narrative that situated its members as victims, not perpetrators (Clark 2009: 476; Saxon 2005: 562; Hazan 2004: 178).26 Thus, considerable factual uncertainty surrounded the tribunal’s early operations, and because no consensus existed regarding who was doing what to whom, the people of the former Yugoslavia could have little confidence that the ICTY would accurately assess the legal claims put before it.
Even more destabilising perhaps than the factual uncertainty surrounding the tribunal’s early operations was the legal uncertainty that was just as pervasive. The ICTY’s Statute comprises a mere ten pages (UNSC 1993), and many of its provisions are summary and undetailed. For instance, although the Statute listed the crimes over which the tribunal had jurisdiction, it did not set forth the elements of those crimes; consequently, during the tribunal’s early years, it was frequently unclear what exactly the prosecution had to prove or the defendant defend against.27 Moreover, the ICTY Statute was completely silent as to defences. Only through case law, therefore, did the tribunal decide the applicability of duress28 and other defences.29
As a consequence of these circumstances, many early commentators— and even some tribunal judges—expected the ICTY to fail,30 and the international community initially seemed content to let it do so. The United Nations provided the ICTY with such scandalously inadequate resources during its early years,31 that the tribunal’s first prosecutor unofficially threatened to resign if funding were not increased (Tomuschat 1994: 21, n. 8). The ICTY received even less enforcement support, and this failure posed an even greater threat to the tribunal’s survival. During its first two years, the ICTY had no defendants in custody. By spring 1998, the tribunal had issued 205 arrest warrants, but the States of the former Yugoslavia had executed only 6 (Kirk McDonald 2004: 563).32 Although successive ICTY presidents presented numerous reports to the Security Council complaining about lack of state cooperation, the Security Council “failed to respond in a meaningful way” (Kirk McDonald 2004: 562).
The challenging circumstances and deep-seated distrust that surrounded the ICTY’s early work rendered it uniquely vulnerable and subject to perceptions of illegitimacy. It is this vulnerability and concomitant need to build legitimacy, I believe, that drove both the ICTY’s initial procedural and regulatory choices and its decision to retain many of those choices after other international tribunals had rejected them. Indeed, I will argue that adopting adversarial procedures and permitting them to be utilised by a lawyer of the defendant’s choice stood as the only viable option for an institution as weak and mistrusted as the ICTY.
To make this argument, I provide a brief description of adversarial and non-adversarial procedures. Criminal proceedings in an adversarial system are structured in the form of a contest between the defendant and the state. The adversarial model charges the parties with investigating the facts, researching the law, and presenting the case in the manner most favourable to their own positions (see LaFave et al. 2009: 43; Luban 1988: 57; Sward 1989: 302). That is, the parties decide which witnesses to call, in what order to call them, and what questions to ask them, among many other decisions. Trial judges and lay jurors are expected to passively receive the evidence presented by the parties. Although trial judges are authorised to ask questions, doing so is frowned upon, and judicial interventions are unlikely to be meaningful in any event because judges in adversarial systems are kept uninformed about the facts of their cases prior to trial (Damaška 1975: 1090–1091).
Proceedings in a non-adversarial system, by contrast, are driven by the judiciary. Some non-adversarial criminal justice systems place pre-trial investigations in the hands of an investigating judge, who is charged with collecting both inculpatory and exculpatory evidence (e.g., Van Den Wyngaert 1993: 9; Vogler 1996: 19).33 Others do not, but even in non-adversarial systems in which the parties play a more robust role in the investigations, at trial, it is the presiding judge who determines which witnesses to call and who takes the lead in questioning those witnesses (Damaška 1975: 1088; Herrmann 1992: 760; Daly 1999: 70; Tomlinson 1983: 143; Robbers 1998; Van Den Wyngaert 1993: 33). It is only after the presiding judge has concluded her questioning that the lawyers have the opportunity to suggest additional questions.34 The presiding judge is also authorised to raise any issues relevant to the charges and can even hear evidence not formally put forward by the parties (see Damaška 1973: 559). Not surprisingly, given this description, most commentators consider the key difference between adversarial and non-adversarial proceedings to be that the former are party-dominated, whereas the latter are judge-dominated.35 This key difference reflects the systems’ very different ideological underpinnings. By placing so much control in the hands of the parties, adversarial systems manifest respect for litigant autonomy and party participation,36 while expressing a concomitant distrust of the state and state officials.
Merely to describe non-adversarial procedures as conceptualising the State as a benevolent protector of public interest, capable and willing to police itself is to show how extraordinarily unsuitable those procedures would have been for early ICTY proceedings. If the use of the adversary system in the United States is understood to reflect Americans’ notorious distrust for their governmental officials (see Huntington 1981; Kagan 2001), consider the procedural implications of the far more virulent scepticism and mistrust that pervaded the ICTY at its inception. Uncertainty and scepticism about every aspect of the tribunal abounded during its early years, and that rendered it impossible to adopt a set of procedures that presupposed trust between citizen and state and between litigant and the judicial system. Under these circumstances, the only viable procedural system for the early ICTY was the procedural system that bestows upon the parties maximum control over their cases. That is, the only viable procedural system for the early ICTY was the adversary system.
Further, it is not merely that the ideological underpinnings of the adversary system are in far better alignment with the perceptions and reality of the early ICTY. In addition, social psychology research of the last few decades shows that adversarial procedures had considerable potential to provide the ICTY with other key benefits. In the past, social psychologists assumed that people assessed the desirability of procedural systems primarily on the basis of the outcomes they received under those systems (Lind and Tyler 1988), but the seminal work of John Thibaut and Laurens Walker (1975) showed that individuals’ preferences for one set of procedures over another are in fact substantially predicated on their perceptions of the fairness of those procedures. Following upon Thibaut and Walker’s studies, subsequent researchers have shown that individuals who believe that the procedures used in their cases were fair view the legal officials, judicial institutions, and the specific outcomes of their cases more favour-ably. Indeed, as a general matter, assessments of procedural fairness lead to “greater overall satisfaction with the legal experience and more positive affect with respect to an encounter with the justice system” (Thibaut and Walker 1975: 70). Two studies in this vein, for instance, examined the views of convicted felons and showed that a defendant’s evaluation of his overall experience in the criminal justice system was heavily influenced by his assessment of whether his case was handled fairly (Thibaut and Walker 1975:73; Landis and Goodstein 1986: 675, 706–707; Casper et al. 1988: 483). Researchers conclude, therefore, that when procedures perceived to be fair are used, individual views about authorities remain positive whereas when procedures perceived to be unfair are used, “negative outcomes lead to negative affect toward the authorities involved” (Thibaut and Walker 1975: 72). Finally and most relevantly here, research shows that people are more likely to accept and obey negative decisions when they believe those decisions were made pursuant to fair procedures (Tyler 2000: 119; Lind and Tyler 1988: 79).37
The research just canvassed took place within the context of stable groups that have existing authority structures. These authorities are widely considered to be legitimate and therefore entitled to obedience. However, studies show that the fairness effect just described is less influential in the context of authorities with questionable legitimacy. In particular, when individuals doubt the legitimacy of an authority, they are less willing to defer to the authority’s decisions on the basis that the decisions were made fairly. Rather, they will focus more on the favourability of the decision (Tyler 2000: 120). Thus, a central goal for new authorities must be to gain legitimacy, and in this quest, fairness assessments also play a key role: in particular, if “people view or personally experience the authorities as making decisions fairly, they increasingly view them as legitimate. Over time, this legitimacy shapes deference, which becomes increasingly independent of the favourability of policies and decisions” (Tyler 2000: 120). In other words, the key ingredient that shapes an individual’s assessment of an institution’s legitimacy—which legitimacy leads individuals to defer to the decisions of that institution—is the fairness of the procedures through which institutions exercise their authority.
Because the perception of fair procedures is so influential to so many realms, it becomes crucial to determine what procedures are perceived to be most fair. Through a series of studies, Thibaut and Walker (1975: 78–80) concluded that individuals across legal cultures consider adversarial procedures to be fairer than non-adversarial procedures.38