What do courts do?

Most societies and certainly all modern societies recognise the fundamental importance of setting in place a system of impartial institutions, in which disputes can be heard by suitably qualified people who are both structurally autonomous and independent minded. Indeed, the case for such a system is so strong that it hardly bears repetition. Suffice it to say that such a system is the only basis upon which individuals can be persuaded to refrain from taking action themselves to enforce their rights, or to right wrongs (real or perceived) done to them. Simultaneously such a system reduces transaction costs by increasing the certainty that contractual commitments will be complied with or will be enforced, thereby promoting the development of more efficient economic production.⁴

The practical implications of how to give effect to the need for independent courts has been set out and interpreted in common law, statute and, more recently, in the South African constitution. The constitution states that “everyone has the right to have any dispute that can be resolved by the application of law decided in a fair public hearing before a court or, where appropriate, another independent and impartial tribunal or forum” (section 34). Thus, South Africa’s courts can hear any dispute which can be settled by the application of law in hearings that must be fair and, to that end, such courts (or other tribunals) must be independent and impartial. This independence is further guaranteed by section 165, which provides that the courts are “subject only to the Constitution and the law” and that they must apply these “impartially and without fear, favour or prejudice”. Furthermore, all decisions of the courts are binding.

Thus, philosophically, socially and legally, courts have been set up as impartial arbiters in legal disputes and, to the extent that goals for South Africa’s system of courts can be described, they should reflect this raison d’être. This definition of the role of courts emphasises that courts are a means to resolving disputes, but does not imply anything about the content of decisions made by the courts. However, it precludes a definition of the function of courts, even criminal courts, as having anything to do with the punishment or rehabilitation of criminals. Courts exist to hear criminal cases fairly and impartially, much as they do civil and administrative matters. They do not have a goal other than to process cases fairly and decide them impartially.

At the same time, however, the courts are also part of a system that does have a more outcome-oriented function. As anyone will tell you, the criminal justice system is supposed to do something about crime. Without question, this system—comprising police, prosecutors, magistrates, private defence counsel, public defenders, prison warders and support staff—is intended to result in some predetermined social goal. Unfortunately, it is not always clear that the goals of that system are easily defined or that they form a coherent unity.

The goals of criminal justice

For some, the objective of the criminal justice system is to punish those who violate the rights of others through their criminal deeds. To this, many would respond that the function of the system is to prevent crime by rehabilitating and reintegrating identified and convicted criminals, or by deterring those who may not yet have committed a crime, but who may consider doing so. Others point to the preventative effects of incapacitation, arguing that the imprisoned offender cannot rape your daughter. Equally, democratic societies set other goals for the system. It should, for instance, treat humanely everyone from victim to convict. It must uphold and defend the constitutional rights of suspects and accused persons. Moreover, in pursuit of these goals, the criminal justice system must retain public confidence, and perform its functions effectively and efficiently.

While there are circumstances in which one can conceive of the system achieving all of these goals, it is also clear that these diverse ambitions can be both confusing and contradictory. Is it true, for instance, that policies that increase the deterring and incapacitating effects of prison will not affect its capacity to rehabilitate and re-integrate? Will policies that enhance the system’s impact on crime in the short term, undermine its long-term legitimacy and, therefore, effectiveness? Can potential strategies achieve their aims while striking the appropriate balance between the costs of crime and the costs of fighting crime?

It has been established, therefore, that courts are institutions tasked primarily to hear cases impartially, but that, as components of the criminal justice system, they are also expected to contribute to the (somewhat uncertain) aims of that system with regard to the combating of crime. Nonetheless, even within that system, the function of the court is merely to hear cases brought against accused persons. More than this: precisely because they are impartial and are tasked with providing a fair hearing, they are expected to be a check on how the other agencies of the criminal justice system exercise their powers. In their function and their rules of procedure they are expected to ensure that any excesses committed in the pursuit of criminals go unrewarded so that, over time, such excesses are bred out of the methods used by the police and prosecution. Thus: courts hear cases, and are expected to do so fairly. But in doing so they are expected to act as a brake on other role-players in the system, to ensure that they fulfil their functions in ways that are consistent with the rules of due process defined in constitutional and common law.

The most important implication of this is that it is inappropriate to define or understand the role of courts as vehicles for the conviction of offenders, or to evaluate their performance on the basis of the number (or rate) of convictions achieved. This would be tantamount to evaluating the performance of civil courts on the basis of how many times they find in favour of the plaintiff. Rather, the performance of courts needs to be evaluated by reference to the number of cases completed and the extent to which the processes invoked in completing cases are fair and just.

It is clear, therefore, that there is some conflict of goals in courts. They are part of a system that is intended to manage levels of crime in a society by dealing with people who have offended against social rules. At the same time they are also expected to protect the rights of the accused person and to dismiss cases against the guilty when due process rights have been violated. Thus, following the South African Law Commission, the goals of a criminal trial might be stated as:

These (diverse and conflicting) objectives are given philosophical and procedural flesh in South Africa’s adversarial system of justice, where the presiding officer’s task is to hear both sides of a case and to make a decision based on the facts and legal arguments presented to her. The process is driven by both the prosecution and the defence, and the presiding officer serves as an arbiter of both the admissibility of evidence and what weight to attach to it. In such systems it is the prosecution’s duty to furnish evidence—accumulated in a manner that is consistent with the rules of criminal procedure and the Bill of Rights—and to persuade the judge or magistrate of the guilt of the accused beyond a reasonable doubt. In the meantime the defence seeks to show that the evidence led is factually untrue, inadmissible or insufficient to prove guilt beyond a reasonable doubt. The presiding officer is, therefore, similar to an umpire or referee: she has no interest in proving or disproving the charges, merely in ensuring that all admissible evidence is presented and given due weight when reaching a verdict.⁶ For her, unlike the prosecution (and police) or the defence (and accused person) the trial is an end in itself, not a means to an end.

One of the implications of the ‘party-driven’ nature of the adversarial model of criminal courts is that the court itself needs to be conceptualised as an organisation in which conflict over the goals of the processes it manages is institutionalised. Given what we have said about the goals of criminal justice, the ‘goal conflict’ of the role-players at court is not, of course, a bad thing. It is, rather, part of the nature of a system that seeks to prosecute the guilty but protect the rights of those accused of committing crimes. Nevertheless, it does mean that there is a need to understand the management of courts and the finalisation of cases in a way which does not reduce the objectives of the process to the goals of any one of the parties. In particular, it is important to distinguish the aims of the parties (defence and prosecution) who regard a trial as a means to an end (the proving of their cases) and the magistrate, and those who administer the institution, who regard the completed trial as an end in itself.

The recognition of the importance of institutionalised goal conflict in courts is, of course, hardly fresh. Indeed, it is both trite and obvious. Nevertheless, it is important to build this into any attempt to evaluate the performance of courts. Equally, it should form the foundation of any conceptualisation of the court as an organisation.

Are courts organisational units?

The fact that the justice system serves and pursues multiple and contradictory objectives is reflected in the conflict over the goals of the actors in the drama of the court. It is also reflected in the character of a court as an organisation. This is of course, obvious in relation to the accused and his or her representative, neither of whom answer to any public authority. However, it is also the case for the presiding officer, prosecutor and investigating officer who may be members of different government departments, or who answer to different authorities within the same government department.

Thus, if one’s understanding of an organisation is that it forms a definite and rational hierarchy, then a court is not an organisation. This is so simply because most of the key actors of the court are independent of one another, and are not accountable to each other nor to a supervising authority.

Despite this obvious difference between courts and other organisations, it is submitted that it would still not be incorrect to characterise courts as organisations. Despite the absence of a fixed hierarchy in which tasks are delegated and through which actors account to a single boss, the court does have other characteristics of organisations. Following Feeley, we need to recognise that courts, like other organisations, institutionalise the interaction of a large number of actors whose roles are highly defined and who are required to follow a specific set of rules.⁷

More than this, and despite the institutionalised conflict over goals, we can identify a core goal which all parties share: the processing of the accused from being charged to being either convicted or acquitted. It is true that the actors at court have different interests in the final disposition of each case and, indeed, that the personalities playing the different roles change. Nevertheless, courts and their actors share an interest in the process of getting the arrestee through the court procedure.

However, even in this most simple of common denominators, matters are not all that simple, for there may well be circumstances in which there is a trade-off between the speed at which cases are heard and the fairness with which they are decided. The added haste can sometimes be achieved at the expense of the rights of the accused person. Nevertheless, as a general point of departure, one could say that perhaps the only goal shared by more or less all the role-players in a court is the processing of cases and accused persons.⁸ This is so despite the diverse and contradictory objectives of the system of justice as a whole, and despite the contrasts between the objectives of different role-players regarding the manner in which the case against any particular accused person is finalised.

Given these complexities, Clynch and Neubauer characterise courts as “informal workgroups” in which people work towards a common objective, interact with one another on a continuing basis, and perceive themselves to constitute a group. At the same time they have different levels of commitment to the group and its work, along with differing levels of group solidarity. This model, they argue, accounts for the character of the division of labour and the nature of working relationships between actors in trial courts.⁹ In any event, as Clynch and Neubauer point out, in spite of the fact that courts are intended to function as adversarial contests, studies repeatedly find relatively high levels of informal and, sometimes, illicit co-operation between the adversaries.¹⁰

The diverse and contradictory goals of courts and the actors who play out trial court dramas, mean that any evaluation of the performance of courts needs to be suitably nuanced, giving voice and weight to the enormously complex role that these organisations play. So, before looking at the role court specialisation may play in improving the performance of South Africa’s criminal justice system in general, and the courts in particular, we need to systematically set out the criteria against which one might evaluate the impact of reforms. In doing so, however, it is important to recognise that while players in a court include police, accused persons and defence counsel, it is the presiding officers, court administrators and prosecutors who are the most consistent members of the ‘informal workgroup’. It is the work of these people that forms the basis for the participation of the others.

In looking at the ways in which one evaluates courts we will not focus unduly on the police, as evaluating them requires a far wider range of measures capturing activities in areas outside of the court. The defence, also, cannot be evaluated as an institution, because it is not. Our focus, therefore, will be on the functioning of the courts in the fair processing of cases. We will seek to register data on the rate at which convictions are obtained as a measure of the performance of the prosecution, as it is their task to identify and pursue those cases that have a reasonable likelihood of conviction.

The performance of courts and prosecutors

One of the clear objectives of courts and the justice system is the effective and efficient processing of cases. When we begin to look at court performance, we should therefore first consider the speed and cost-effectiveness of its processing of cases. But courts, as we have said, have other obligations to meet as well.¹¹

This is not to say that, when looking at a court’s performance, its impact on the wider crime-related goals of the criminal justice system should be ignored. It is legitimate to ask what impact the court (or a particular innovation) is making to society’s efforts to manage crime levels and to deal with offenders. It is submitted, however, that, in order to preserve the integrity and impartiality of courts, such measures should really be related to the performance of the prosecution and, therefore, the investigation, of cases. Indeed, even the core measure of the performance of the court—the processing of cases—is arguably affected most by the effectiveness and efficiency of the investigation and prosecution of cases. Thus, while the time required by the legal representative of the accused may delay trials, generally most delays are thought to be primarily due to delays in the investigation of matters or delays arising from the poor management of the case, resulting in the non-attendance of witnesses or defendants at court.¹², ¹³

That said, Olstrom and Hanson, comparing the timeliness of case management in a variety of courts in the United States, found that the single biggest factor influencing the pace at which cases were finalised was the local legal culture, and the extent to which it created expectations and demands on court actors to complete cases quickly. This, they insist, is a ‘cultural’ norm emerging from the interaction of the judges, prosecutors and defence attorneys, and could not be ascribed purely to any one of them. At the same time it did appear that if expectations for meeting deadlines imposed by the judiciary were credible, this would strongly influence the local culture. Other actors in the legal process would realise the need to meet deadlines. The police in particular (through the prosecutor) and local attorneys would learn the limits of the courts’ patience quite quickly, and moderate their own behaviour accordingly.¹⁴

How have South African courts performed?

Martin Schönteich, in looking at the performance of the prosecution service in South Africa over a number of decades, reviews a great deal of the data on South Africa’s courts.¹⁵ While this is not the place to review all those data, it is worth reporting some of the key findings in relation to the processing of cases. In this regard, he reports that:

The number of cases prosecuted peaked at the end of the 1960s, with nearly 620,000 cases prosecuted. This number fell to about 300,000 in 1995/96. Making the reasonable assumption that this decline was not accompanied by a proportional decline in the number of courts, this would suggest that the average efficiency of courts has declineddramatically.¹⁶



Against this, as Schönteich points out, changes in the portfolio of cases brought to court—drunkenness made up nearly 15% of charges brought to court in 1969/70—must be considered when evaluating changes in court efficiency. In the absence of such large numbers of petty cases, the average case is more difficult to prosecute, which means that each will, on average, take longer to prosecute.



The fall in the number of cases prosecuted to completion has, in recent years at least, been reflected by the rapid rise of both the number of prisoners in South Africa’s prisons who have yet to be convicted of the crime for which they are accused, as well as the increase in the amount of time that each prisoner awaiting trial spends behind bars. Thus the number of awaiting trial prisoners rose from 19,571 in June 1994 to 55,558 in December 2000. Similarly, their length of stay rose from an average of 76 days in June 1996 to over 130 days in December 2000.¹⁷



Schönteich also presents data on the processing of cases through the courts. Although these figures are available for a short period only, they suggest that the number of cases finalised in district courts and in regional courts every month, while somewhat variable, are ‘stuck’ around 23,000 and 3,000 respectively. These data suggest, therefore, that the average district court finalises 29 cases per month, while the average regional court finalises nine cases per month.¹⁸

Having reviewed the data on the performance of the prosecution service, Schönteich concludes that “the criminal justice is performing poorly”.¹⁹ Moreover, he believes that its performance “can be substantially enhanced only if the effectiveness of the prosecution service is also improved”.²⁰ In this regard, however, he identifies a number of reasons for the relatively poor—and deteriorating—quality and quantity of work delivered. These include:

inadequate numbers of prosecutors relative to the (growing) workload;


the loss of skilled personnel;


poor training methodologies; and


inadequate rewards, resulting in the service’s inability to retain qualified staff.²¹

Apart from these issues, however, Schönteich points to changes in the legal environment since the transition to democracy, which have complicated the prosecution of some kinds of cases, as so-called ‘reverse onus’ provisions have been overturned.²² In addition, new requirements governing the management of cases against juvenile accused persons also resulted in a slowing down of the processing of such cases. Moreover, he suggests that personnel and organisational problems in the police service have resulted in weaker investigations and, therefore, fewer prosecutions and convictions.

In addition to this list, it is also submitted that an under-researched factor explaining the decline in the number of cases completed every year is the increase in the proportion of accused persons who are now represented in court. This change, a result of the constitutional right to state-funded legal representation in serious cases, must, at a minimum, have lengthened the time taken to complete cases, while, in all probability, increasing the number of accused persons eventually acquitted of charges.

There can be little doubt that democracy and the transformation of both South African society and the institutions of the South African state have created enormous challenges to the functioning and performance of the criminal justice system as a whole. Courts are suffering deteriorating performance, as the workload of the system, measured by the number of crimes recorded by the police, has risen; personnel numbers have declined or, at best, average levels of experience have fallen; and the rules and procedures of criminal justice have changed. Add to this the high levels of insecurity generated by the restructuring and transformation of the departments concerned, as well as the rapidly rising level of public pressure on the system, and the only surprise is that the system has, for the most part, not crumbled under the weight of these combined factors. These points are, in general, well made by commentators like Schönteich.

The point of our preceding analysis, however, is that this one-sided evaluation of the performance of the criminal justice system is inadequate, for, as we have seen, its nature and structure embodies the pursuit of objectives other than the successful prosecution of offenders.

To be sure, the transition to democracy has also created some opportunities for improving the system. There is little doubt, for instance, that, measured against the criteria of fairness and justice, the rules and procedures of criminal justice are today much improved compared to those that existed under apartheid. The overturning of ‘reverse onus’ provisions, the constitutional protection of due process rights, and the greater independence of the magistracy all serve to ensure that the process of achieving convictions is substantially more conducive to judicial certainty than may have been the case in the past.

In addition, legal assistance is, for the most part, available to indigent people accused of serious crime. Juvenile accused persons are entitled to better treatment, while victims of crime, particularly those from the so-called vulnerable groups, are beginning to be better served by the system than they were in the past. While the legitimacy of the system in the past was undermined by the fact that the people who worked for it were not representative of the greater population, the present staff complement is likely to have begun to turn those issues around.

One must, therefore, reach a mixed judgement in relation to the performance of the criminal justice system, its individual agencies and the institution of the ‘informal workgroup’ that is the court at the heart of the process of administering justice in South Africa. On the one hand it is clear that there are severe problems in relation to the achievement of the system’s crime control objectives. On the other, the system is substantially fairer and its decisions ought to enjoy far more credibility than was the case under apartheid.

How to evaluate the workings of South Africa’s courts

In showing that the evaluation of the functioning of South Africa’s courts is no easy matter, much ground has been covered. This paper has tried to show, for instance, that although one might say, from a social policy perspective, that the function of the criminal courts is to ‘do something about crime’, it is, in fact quite difficult to say what that ‘something’ is. Equally, whether an individual court achieves anything with respect to overall crime levels, or not, is, to some extent, immaterial. After all, it is not the court’s job to think about crime or crime levels, but to process cases brought before it. Its micro-level objectives—the rapid but fair hearing of cases—may have something to do with the macro-goals of the criminal justice system, namely the control of crime levels, but its function is to simply process cases as rapidly and fairly as possible, all the while recognising the potential conflicts between these goals.

Courts are therefore peculiar organisational entities, for they embody conflicting and incompatible goals. Indeed, while courts facilitate the work of investigators and prosecutors, they also put brakes on the excesses to which these functions may be prone. For this reason, courts do not consist of stable, organisationally rigid hierarchies, but are better understood as more-or-less informal workgroups, coming together for the processing of cases in which the identities of the role-players differ and each belongs to a different organisation or is entirely independent.

Given these complexities, it is clear that defining a set of criteria against which to measure the performance of courts or, indeed, of the impact of innovations in the delivery of justice, is no easy matter. Furthermore, even if a set of criteria were developed to assess the performance of courts or the impact of innovations, it is possible that the impact may be positive in relation to one set of criteria and negative in relation to another, incommensurate set of indicators. In these circumstances it would be hard to determine whether a court or innovation were doing well or poorly overall, and reasonable people might differ on how to interpret the results.

Thus, when reviewing the nature and effect of court specialisation, one must ensure that account is taken of the effects not just on conviction rates and case processing speed, but on the fairness, accessibility and independence of the courts concerned.