Bureau of Justice Assistance: Pre-trial Services Project
EXECUTIVE SUMMARY

The Pre-trial Services (PTS) demonstration project is currently in operation in three criminal magistrates courts in South Africa. The project aims to improve the integrity of the pre-trial process, preventing dangerous, repeat offenders from being inappropriately released on bail while maximising the release on warning or affordable bail of accused persons who do not pose a threat. At the end of January 1999, an estimated 18 900 accused were interviewed at offices in Mitchells Plain (in operation for 18 months), Johannesburg (15 months) and Durban (9 months). The project is run, in this pilot phase by the Bureau of Justice Assistance (BJA) and involves collaboration between the Department of Justice, the South African Police Services and the Department of Correctional Services. The BJA is a partnership of the South African Ministry of Justice and the Vera Institute of Justice.

To assist these government departments in the policy decision about where to take PTS, the BJA has collected information about its strengths and weaknesses. A list of these are set out below, followed by an evaluation of the project as a whole and proposals about the way forward. For a more detailed description of the strengths and weaknesses, refer to page 45.

Strengths
Photo of accused (see page 13)
enables the Department of Correctional Services to reduce the number of remands by identifying prisoners that need to go back to court
prevents prisoners swapping warrants of detention (J7's) and escaping
assists investigating officers

CRIM linkage in PTS office (see page 24)
allows identification of wanted suspects
improves bail decisions at first appearance

Interview and verification (see page 14)
provides investigating officers with information that they often do not get from the accused
reduces remands for verification and reduces workload of investigating officers

First Appearance report with previous convictions (see page 31)
increases release on warning (with conditions) in appropriate cases, reducing awaiting trial prison overcrowding

Compliance reports with photos (see page 20)
improves efficiency and integrity of police monitoring of reporting conditions without increasing workload

Witness Waiting Room (see page 23)
permits (but does not assure) greater community involvement in courts

Electronic database of accused (Electronic Court Information System) (see page 21)
enables court to tell if accused has other matters pending
speeds up finding information about a case, such as date of next appearance

Weaknesses
Delays of first appearances until 11:00 (see page 17)
Resistance and distrust by some prosecutors (see page 29)
Low utilisation of supervision staff (see page 12)
Technology is not integrated into a national system (see page 23)

Conclusions
The strengths of PTS appear to outweigh its weaknesses. The strengths suggest that PTS has demonstrated that it is a credible and sustainable policy option for addressing a number of practical problems within the current bail administration system. The weaknesses that do exist, do not appear to be inherent in PTS and could be addressed separately or by the proposals set out below. PTS, in some form, should therefore be institutionalised, provided that it is given the necessary support and an appropriate policy and legal framework.

Recommendations
Options
There are number of options for the institutionalisation of PTS. For example, it is possible to roll out ceratin parts of PTS rather than all components. This requires a cost / benefit analysis of each option and it may be desirable to do this on a court by court basis. For example, the SAPS infrastructure may not allow the CRIM linkage to be established immediately for all courts, or the Witness Waiting Room may be successful and appropriate in some communities but not others.

Support and framework
Whichever option for institutionalisation, or combination of options, is selected, there is a need for national support and the creation of an appropriate legal and policy framework for PTS to operate. This is also important in order to address current weaknesses in PTS. To ensure the sustainable national role out of PTS, the following four things need to happen:

1) Legislation
The bail legislation should amended to formalise the status of PTS reports and to provide the legal framework in which PTS can operate most effectively. Specifically, two amendments are proposed. First, a provision directing magistrates, when deciding release, to consider imposing monetary conditions only if no combination of other conditions would be sufficient to protect the interests of justice. Second, a provision directing magistrates in courts with a PTS office, to take PTS reports into consideration in making bail decisions.

2) Directives
The National Director of Public Prosecutions should issue policy directives to prosecutors setting out the operation of PTS and, where appropriate, instructing them to have regard to PTS reports unless doing so will cause unreasonable delay. Unreasonable delay may be defined, for example, as not more than two hours after the arrival of the accused at court.

3) PTS task team
The PTS task team, set up within the Department of Justice, overseeing the role out of PTS, should continue and be supported in its work. There needs to be close liaison with or possible inclusion of the SAPS in this task team to ensure proper planning for and co-ordination of the CRIM component of PTS. The Department of Justice should support the Electronic Court Information System (ECIS) while evaluating the desirability of extending it nationally or at least until other viable options are implemented in the courts. Thereafter, ECIS may either be scrapped in its current form or parts of it incorporated into an enterprise level system along the lines proposed by the Integrated Justice System.

4) Training and support
There needs to be peer to peer training and support by a dedicated team of people. For example, one or two national prosecutors, as part of this team, will be available to work with other prosecutors to train them in the use of the system and deal with their concerns. Others in the team may include bail officers, supervision officers and technical support. An existing PTS office could be used as a national training centre to train staff from other courts.

INTRODUCTION
Overview
This report provides an assessment of the Pre-trial Services demonstration project. This assessment is divided into two parts: (1) an overview of what PTS does and how well it does it; and (2) what impact the project has had on the criminal justice system. The strengths and weaknesses emanating from this assessment are summarised and provide the basis for the conclusions and recommendations.

About Pre-trial Services
Pre-trial Services aims to improve the operation of the country’s bail system by providing the court with information that it needs to make more appropriate decisions and by offering alternatives to money-based bail. This involves submitting to court a report about each accused at their first appearance containing verified information about the accused’s community ties, employment, previous convictions, and other information needed for a bail decision. This information enables the court to make more appropriate bail decisions which means that high-risk, dangerous and repeat-offenders are more likely to be detained while awaiting trial. More appropriate bail decisions also mean that low-risk, petty, first time accused are more likely to be released from custody. In order to facilitate this release, the PTS project offers supervision of bail conditions as a viable alternative to money-based bail.

The Bureau of Justice Assistance expects that, over time, this will change the profile of the awaiting trial prison population, reducing the proportion of people held in prison because they cannot afford to pay their bail. The move away from money-based bail towards release on warning with or without reporting conditions also reduces the economic injustice of incarcerating poor people, who should be released, but cannot afford to pay bail.

Study sites
The three sites where PTS has been implemented and where this evaluation was done are the criminal magistrates courts in Durban, Johannesburg and Mitchells Plain. For a description of each of these sites, see the baseline report.

METHODOLOGY
This report assesses both the process and the impact of PTS. The process review describes what PTS does and asks: How well does PTS function? The impact assessment uses a number of pre-defined indicators to answer the following question: What effect does PTS have?

Process assessment methodology
The process research for this part of the study used mainly qualitative methods including open ended interviews and on-site inspections. These were used to compile a table of strengths and weaknesses for each office structured under various sub-headings. These included (1) internal PTS operations; and (2) relations with the community and bodies outside the PTS office. More detail on each of these is provided below:

1) Internal PTS operations
The section starts with a description of the progress made in establishing offices and appointment of staff. Problems and successes with the core work processes in the offices are highlighted. These include interviews with accused persons, verification of the information that they provide, production of PTS First Appearance Reports for court, exit interviews, warrant for detention and tracking of cases. The supervision component describes what the office does to monitor and enforce compliance by accused with their bail conditions. This section also reviews progress made in implementing the computer application (Electronic Court Information System), the establishment and functioning of the witness waiting rooms and the linkage of the courts to the police criminal records database (CRIM).

2) Relations with the community and bodies outside the PTS office
This section discusses relations between the PTS office and people and organisations outside the PTS office. They include the community, other court-based staff and the police.

Impact assessment methodology
In order to measure the impact of the project a number of quantitative indicators were monitored. As a basis for comparison, baseline data were collected in each court before the PTS offices became operational. These were published in a report describing the baseline position. The methodology used for impact assessment is similar to that used for the baseline data, except where indicated. Refer to the baseline report for a description of the methodology.

Data samples
Court data
Impact data in this report are based upon the decisions of courts in first appearances as recorded in the charge sheets. Samples of charge sheets for at least 500 accused persons were collected on at least two occasions after the start of PTS for each court. The exact sample sizes and dates of evaluation are provided in the results section.

Prison data
Much the data regarding awaiting trial populations were obtained from the Department of Correctional Services in response to specific requests by the BJA. We also receive prison population data, by fax, directly from prisons. The DCS reports are broken down into sentenced and unsentenced prisoners while unsentenced prisoners are further divided between accused who have been denied bail and those who have been granted bail.

Indicators
The indicators used to measure PTS impact are:

1) Court bail decisions
The impact of PTS on bail decisions made by the courts is assessed by comparing baseline and impact statistics concerning bail (release) decisions, bail amounts, and use by the court of special bail conditions.

2) Compliance with bail conditions
The extent to which bail conditions are supervised in each PTS office is assessed and its effectiveness is measured in terms of compliance by accused with reporting conditions.

3) Awaiting-trial prison profile
The prison population profile is described by the proportion of the total waiting trial population who were granted bail compared with the proportion who were denied bail. The numbers of awaiting trial prisoners are also given.

RESULTS AND DISCUSSION

Results of PTS process assessment
PTS is a new concept in South Africa and the changes required in each court are often slow and painstaking. As expected, each of the offices initially experienced teething problems. Evaluating a project of this nature involves weighing up the strengths against the weaknesses. This evaluation therefore describes the successes but also points out where each office, or PTS in general, needs further improvement. These strengths and weaknesses of each PTS office are summarised and set out in an appendix to this report. These tables offer a concise summary of this section of the results.

One should bear in mind that the project, particularly some offices, are relatively new and are still undergoing processes of evolutionary change and working through their teething problems. This is a snapshot of the PTS offices at the time of this evaluation. Some of the challenges still facing some of the offices have already been solved at other PTS sites and the situation may have changed by the time you read this.

Internal PTS operations
Establishment of offices
PTS offices have been established in three South African Magistrates Courts: Mitchells Plain (since August 1997), Johannesburg central (since November 1997) and Durban (since May 1998). Each of these offices were planned and implemented by a local steering committee comprised of the Chief Magistrate, Senior Public Prosecutor, clerks of the court, SAPS, Department of Correctional Services officials and members of the community. They were each launched by the Minister of Justice.

PTS staff
The staff who operate the PTS offices on a day to day basis are seconded to the project by both the Department of Justice and the SAPS. At the end of October 1998, a total of 38 PTS officers were working full-time at the three offices. They are divided approximately equally between supervision officers, who are seconded from the police, and bail officers, who previously worked as either clerks or interpreters in the courts. See Table 1 for a more detailed description of PTS staff allocation.

Table 1 PTS staff allocation at the three PTS offices as at 1998-10-22.

One of the most heartening impacts of PTS has been the transformation within most of the staff seconded to the project. We have seen PTS staff, who are ordinary civil servants, take on their tasks with renewed enthusiasm. They talk of their work with a new sense of pride and are motivated by feelings of greater respect that their new positions have given them in their interactions with attorneys, prosecutors and magistrates. PTS has given them an opportunity for advancement which they were denied in the past.

Despite this overall positive impact, some staff in the Durban and Johannesburg PTS offices have not been as enthusiastic towards their work as their colleagues in the same offices and the work ethic appears to have flagged. For this reason, it may be advisable to introduce a system of individual performance evaluation to reward those staff who work the hardest in keeping the offices functioning.

A number of the weaknesses in all three offices reflect the difficulties of managing people and relationships in the rapidly evolving environment of a demonstration project. The Chief Bail Officers in each of the offices have had to deal with these challenges despite not having formal management training and often with relatively little support.

The automation of the PTS office through the use of ECIS in the Johannesburg court has speeded up the processing of PTS First Appearance Reports. The result is that the PTS office now appears to have more staff than it needed under the manual system. It may be possible to rotate some of the PTS bail officers, who are employed by the Department of Justice, into other offices in the court in the afternoons to help out with other work while still being available to assist with PTS interviews in the morning. This will reduce the overall salary cost to the Department of running PTS. A suitable range of numbers of PTS staff is set out in Table 2. The number of bail officers is dependant on the number of accused in the court that have to be interviewed, while the number of supervision officers is dependant upon the number of police stations which service the court and their geographical distribution.

Table 2. The current (demonstration) and the future (proposed) compliment of PTS staff

Photo of accused
Photographs are taken of each person interviewed by PTS. These are taken with a digital camera and the images are downloaded to a database of accused persons and printed out together with the forms using a laser printer. The photographs have a number of benefits, including:
Enabling the Department of Correctional Services (DCS) to reduce the number of remands by identifying prisoners that need to go back to court. A problem for most courts in South Africa are remands caused by prisoners not coming to court. This is happens when the prisoners do not respond when their name is called by the warders. The photographs of accused now make it far more difficult for a prisoner to "hide" from the warders in the relative anonyminity of large, overcrowded cells. The only complaint from DCS is that courts other than those serviced by PTS do not also supply photo’s.

Preventing prisoners swapping warrants of detention (J7's) and escaping. The J7 is a form which authorises DCS to detain a person until their next court appearance or, if granted bail, until they pay the bail amount set out on the form. Some prisoners who have been denied bail swap their J7 for someone who was granted bail, pretend to be them and escape by paying their bail. The photo on the back of the J7 by PTS makes this far more difficult.
Assisting investigating officers. Investigating officers find the photo very useful for their work, including finding the accused if they abscond.

Interviews, verification and production of PTS First Appearance Reports
Interview of accused persons
The interview process is going well at all three offices. Interviews are generally done in the language of the accused by the multi-lingual bail officers. Durban and Johannesburg have automated this process and interviews are done directly onto a computer instead of an interview form. This reduces the overall time taken to produce a PTS First Appearance Report and this system will be extended to Mitchells Plain shortly.

As the police have themselves pointed out, "More reliable information regarding the identity of the accused is obtained by Pre-Trial Services personnel due to reluctance to supply members of the SAPS with correct information. Members of the SAPS are seen as a threat".

The number of accused interviewed by PTS has shown a steady increase at all three offices since the start of each office. This may reflect a growing co-operation between PTS and other officials in the court (particularly the orderlies) rather than an increase in the numbers of people appearing in court. The PTS office only interviews people who are (1) in custody and (2) adult. An estimated 18 900 people were interviewed at the three office since the start of PTS until the end of January 1999 (Table 3).

Table 3 Start date, length of operation, current average number of interviews and the estimated total number of interviews until January 1999 for each PTS office

Note that these data do not represent the number of first appearances in court. As described above, the PTS system is not designed to interview all first appearances Even some accused who are in custody and are adult may bypass the PTS interview process if they are taken directly to the court by the police. Some accused who are interviewed by PTS do not appear in court where the control prosecutor nolle prosequi’s the matter before first appearance but after an interview has been done by PTS.

Although participation by accused is voluntary, and their right to decline the interview is explained to every accused, less than a hundred have ever refused an interview. Some accused ask for their attorney to be present. In such cases, where necessary the PTS office assists the accused to contact their attorney who invariably advise their client to make use of PTS.

Verification
Verification appears to be done well in all offices. In Johannesburg and Durban, verification is done directly into an ECIS computer which speeds the process up. The Durban PTS staff have taken the initiative in using innovative methods of verifying information supplied by the accused. They have established and use new sources (eg. Traffic Department, Telkom, municipal water and electricity departments, community colleges, etc.) to check the information supplied by the accused.

Three methods of verification are used: (1) telephone; (2) in court; and (3) in field. Heavy reliance is placed upon telephonic verification due to the time constraints of completing the verification of interview information in time for the accused’s appearance in court. The PTS office will not normally announce that they are calling from court. Open ended rather than leading questions are used, for example the PTS officer will ask "Where does she live?" rather than "Does she live at 41D Maxi Park?". This reduces the chances of conspiracy between the accused and their source. Although this method of verification is commonly used in other countries, some prosecutors do not trust the method.

In court verification is used when the friends or relatives of the accused are in court. The PTS staff in Mitchells Plain are assisted by the prosecutors who send family of accused to the PTS office so that verification can be done in court.

When the first two methods are not possible, then the PTS officers have to physically travel to the home or work address of the accused (in field). In field verification of information is used less than the former two methods. Besides the time constraints, the staff have either had restricted access to vehicles or funding for petrol. Although not used routinely, the PTS staff have often gone out and verified information on special request of the court. This has generated encouraging feedback. For example, see the attached letter from a firm of attorneys in Durban, Nhlanhla Mnculwane Incorporated. They comment that the PTS was able to avoid remands in a number of cases:
"... due to the neglect of the Investigating officer the addresses were not verified on the remand date. It was on these occasions when the assistance of Pre-Trial Services would be enlisted and details would be verified and the matter thereafter expeditiously dealt with."

Production of PTS First Appearance Reports
The PTS First Appearance Report (FAR) gives the information supplied by the accused, indicates whether it was verified by the PTS office, includes an individual risk assessment of each accused and incorporates their digital photo. The FAR is produced by the staff online in Durban and Johannesburg, while still done manually in Mitchells Plain.

Delays of first appearances
In Durban and Johannesburg, some prosecutors complained that the PTS process caused a delay in the hearing of some first appearances. In response, the Chief Bail Officer reports that ECIS has speeded up interviews and the office is now able to produce first appearance reports for 50% of accused within two hours after the arrival of the accused at court. Their deadline for completion of all reports is 11:00, although this is contingent upon the timely arrival of accused at court.

The prosecutor’s current work to rule action and the late delivery of accused to the court by the police in the mornings are also sources of court delay. For example, a senior prosecutor in Johannesburg reported that courts generally only get started by the prosecutors after 11:00 and the District courts sit for as little as 20 minutes per day. These delays are unrelated to PTS.

A delay in hearing first appearances does not necessarily lead to a court delay if the court can get on with other business while the interviews are being conducted and PTS reports prepared. In Mitchells Plain, PTS does not delay the start of court because the court hears "old cases" first and then hears first appearances after 11:00 or 12:00. If for some reason, a first appearance needs to be heard urgently, then the prosecutors telephone the PTS office and request it to interview that accused first and prioritise the completion of their report.

Exit interviews and warrants for detention
After their appearance in court with a PTS office, each accused should have an exit interview where they are handed an exit notice. This exit notice is a document where the bail conditions and the next court date are recorded for the accused in a clear and simple format. During the exit interview, the contents of the exit notice should be explained by the bail officer in the language of the accused. Before the introduction of PTS to South African courts, accused were not systematically given this information in writing, making it more likely that accused persons would fail to appear simply because they forgot the correct court date. Despite this, the PTS baseline data indicates that not many accused fail to appear for court.

Only the Johannesburg PTS office conducts exit interviews for all accused on a regular basis. This reflects good co-operation between the orderlies, clerks and PTS staff in Johannesburg. Recently, the Durban PTS office began doing exit interviews for accused who have been granted bail and pay bail at court or when special conditions are ordered and the prosecutor specifically calls the PTS office. Exit interviews are still not done in Durban for accused who are remanded in custody (either because they have been denied bail or because they cannot afford to pay bail at court). The court orderlies do not bring prisons back to the PTS office before "loading" them onto the trucks destined for prison. No exit interviews are done in Mitchells Plain for the same reason.

The bail decision of the court at first appearance is recorded on the charge sheet and this information is needed to complete the exit notice. The charge sheets are held by the clerk of the court. The PTS officers in Durban and Mitchells Plain also claim that it is difficult to get access to the charge sheets before the accused are taken to prison, further hampering preparation of exit notices.

All three PTS offices have taken on the task of producing the warrants of detention (J7's). The J7 is the form that authorises the Department of Correctional Services to detain a particular prisoner. It also advises the prison authorities of the date for the next court appearance and states if bail has been denied or, if bail has been granted, gives the bail amount. Some prisoners, who were denied bail, have escaped from prison by swapping these forms and paying the bail of a person who was granted bail.

However, the chances of this occurring are minimised by the PTS office which generates the J7 from computer and prints the photograph of the accused on the form. The photograph also assists the prison warders to identify prisoners who are due to appear in court, thereby reducing the number of unnecessary remands.

Tracking of cases
Before PTS, no Magistrates Court kept proper track of every accused person throughout their case. For example, it was usually a time consuming and difficult task to find out the next appearance date of an accused. Tracking is essential for supervision of reporting conditions in the PTS office. It is also a pre-condition for any automated court management system. PTS offices initially attempted to track all its cases manually but this proved to be a burdensome process. Johannesburg and to a greater extent Mitchells Plain failed to do so adequately. However, the Durban PTS staff despite having over 30 court rooms to monitor, routinely tracked all PTS cases. This reflects the dedication and perseverance of the people involved.

ECIS now enables this tracking function to be carried out in the clerks office where court books and charge sheets are located. Provided the data entry is up to date, any court official with a computer on their desk is able to look up the status of any accused in the system. Electronic tracking of cases has at times proceeded well in both Durban and Johannesburg. However, the clerks complain that entering information on the computer duplicates their work because they also have to maintain the parallel paper-based system. Tracking is now behind in Johannesburg, particularly in the regional court. It appears that there maybe insufficient staff to cope with the added work load. The introduction of any system of automation will require appropriate staff adjustments and unless court management makes these urgently, it is likely that backlogs in Johannesburg will get worse.

Supervision
Supervision of conditions has proved difficult in all PTS offices. The main reason for this is that the criminal magistrates courts in South Africa do not track cases. For example, as a result there is no systematic way of knowing when bail conditions have been changed or withdrawn by a magistrate for a particular accused. A supervision officer also needs to know when an accused with a reporting order pays bail because only then do they need to start reporting, even though this may be months after the bail hearing. Regular access to all the charge sheets, where the bail conditions are recorded, is usually difficult.

PTS attempted to construct manual tracking and compliance monitoring systems. These generally failed and could not cope with the complexity of monitoring for every accused where they had to report, how often, when they had to start it and when to stop. It became apparent that the only feasible way of managing all this information was to use a computer application.

ECIS has been designed to collate the reporting information into a weekly list for each police station with names and photos of who is required to report there. Currently, this is only used in Johannesburg where it appears to be a good solution to the problems of the manual systems. Captain Koekemoer who works in the charge office of the Johannesburg Central police station pointed one advantage of the new forms: "Prior to the submission of these reports, people swapped around. There is only a problem with people who do not go through PTS ... since there is no photo". Captain Makhubela of the Maroka SAPS concurs by saying that, "the photo’s are valuable and it doesn’t increase workload".

In Durban, PTS supervised compliance in cases where the clerk of the court informed them that an accused had been ordered to report. However, supervision has flagged in the past two months. The inability to travel to police stations was attributed to the police running out of money for petrol. Special reporting conditions are seldom ordered by the court in any event.

Supervision of reporting conditions is in a dismal state in Mitchells Plain with no effective supervision is currently taking place. This means that even if accused do report to police stations then it is not directly because of ongoing supervision by the PTS office: They probably would have reported anyway, without the intervention of PTS. Therefore, at this point, it is unlikely that the supervision component of PTS in Mitchells Plain is having a significant impact on compliance. It is possible that some accused, who would not otherwise have reported, may report because they know that their photo was taken and thus (mistakenly) think that a failure by them to report will be followed up by the PTS office. It is probable that this will improve once ECIS is implemented.

Electronic Court Information System
The Electronic Court Information System (ECIS) is a computer application designed to perform a number of tasks in a South African criminal magistrates court. It aims to automate and integrate these tasks to make the courts more efficient. It comprises a number of modules, some of which are still under development. Although it contains elements of a court management system, ECIS is not, nor does it pretend to be, a full court management system.

What is the difference between PTS and ECIS?
ECIS is more than PTS and PTS is more than ECIS. While ECIS was first designed only to be used for PTS and thus it was originally called the "PTS application", soon other modules were added to the design and it became more than just a system for PTS. This led to it being renamed ECIS. Currently its functions extend beyond the PTS offices where it has been implemented and future enchantments will mainly be for non-PTS functions. Although the PTS offices in Johannesburg and Durban use it to automate many of their processes, the Mitchells Plain PTS office currently operates without ECIS.

ECIS modules
Development of ECIS has taken place in two stages: version 1 and version 2. The modules comprising these two modules are set out below:

version 1
bail administration module.
This automates a PTS office, enabling it to both conduct accused interviews and verify information online and monitor compliance with bail conditions. Warrants of detention (J7's) are printed together with a photo of the accused to minimise fraudulent swapping of forms and improve identification of accused in prison.

case tracking.
This comprises a database of each accused person which is updated by the clerk of the court every time an accused appears in court or pays bail. This enables court officials to rapidly look up the status of a case, including remand date, bail decision, custody status, the history of the case and a digital photo of the accused.

Charge sheet production.
The prosecutor can produce the charge sheet online, linking it to information obtained by the PTS office, and the database maintained by the clerks office.

Witness notification
Witnesses can be automatically notified of the outcome of a case or if the accused is released on bail. The amount of information, if any, sent to each witness can be controlled by the prosecutor or other court officials.

Statistics
Valuable information from the database and tracking system can be used for court management, research, and to automatically generate the statistical reports required by the Department.

Version 2
Docket management
Dockets will be registered by the prosecutor and their decision on each case will be recorded. This will be linked to information on the case captured in the other modules.

case scheduling
The clerk of the court will be able to schedule trials electronically according to the availability of appropriate courts.

Progress with development
Except for the statistics module, version 1 is complete and has been implemented in Johannesburg and Durban. Version 2 is under development and will be completed shortly. Once version 2 is complete, both versions will be extended to Mitchells Plain court.

Integration of ECIS into a national structure
ECIS does not have a national support and maintenance infrastructure and its management is not yet integrated into Departmental decision making processes. A task team set up by the Department is however addressing this.

Witness waiting room
Witness waiting rooms (WWR) have been established in all three courts. The Mitchells Plain WWR is staffed by a committed core of community volunteers, some of whom have been working every court day since PTS started, 18 months ago. A NICRO social worker refers witnesses for counselling, where necessary. The WWR is widely used by state witnesses as a secure and a safe environment in which to wait for their court appearances.

The Durban and Johannesburg WWR’s are not fully utilised by witnesses. It is possible that many people do not know about it in Johannesburg. In Durban this is because of its remote location, although there are plans to move it to a more central location.

CRIM linkage
This is a joint project between the police and the Department of Justice which significantly assists the courts to make better bail decisions. This new system will help the court to identify people who are dangerous or wanted by the police as suspects in other crimes and thereby reduce the mistaken release on bail of such people. A computer link provides the courts with rapid access to the police criminal record information for the first time.

The link up came about as a result of the initiative of the of the SAPS Criminal Records Centre (CRC), and through co-operation with PTS. Mitchells Plain is the first court to have this kind of access to the 26 million records criminal records held by the police central database. The link has been running on a pilot basis since May 1998.

Normally, the courts have to wait six to eight weeks for this kind of information to come from Pretoria. This has meant that bail decisions in South Africa are generally made without the benefit of considering the criminal record of the accused or if that person is wanted as a suspect in any other matter.

As a result of this linkage, this information is now available to Mitchells Plain court within three hours. When the person is arrested, their fingerprints are taken and given to the Local CRC (LCRC). A fingerprint expert classifies the prints and calls up the criminal record on computer. The reports on previous convictions can be sent via a high-speed digital link to a computer in the court’s PTS office which delivers it to the prosecutor in the case.

This report refers to "possible"previous convictions. This is because the method used to match the fingerprint with the computer record does not result in an absolutely unique identification of the accused. Rather, a small number of people (up to 30) may share the reference number produced by the manual fingerprint classification and computer search. Taking into account demographic information such as age, race, sex and name reduces this uncertainty.

How well does the process work?
The feedback from magistrates and prosecutors has been very enthusiastic and they have said that they place great value in the reports that they receive. The only potential problem areas that need further consideration are whether the reports are received in time to be used by court and the question of cancellation of warrants.

Criminal reports received timeously for court process
The LCRC relies on the investigating officer or arresting officer to take fingerprints immediately after arrest and deliver them to the LCRC before 07:00 on the morning that the accused is due to appear in court. The LCRC and PTS identified the problem of late delivery and non-delivery early on in the project and have tried a number of measures to address it. To measure how often reports were received late or not at all, data were collected for a period of two weeks, 19 to 30 August 1998. These are presented below.

Table 4 Receipt of fingerprint forms divided into those received in time for court, those received too late for court and those not received at all.

Fingerprint forms are sent to the LCRC 72% percent of the time. In 51% of cases the fingerprint form is received on the morning of an accused person’s first appearance and therefore a bail report is available on time at the court. In 21% of cases the fingerprint form is sent late to the LCRC, and thus no report is available at court. This information was ascertained by cross-checking the names on the fingerprint forms with the names on the body receipt forms from the cells at court.

According to the PTS data, criminal reports are received 41% (n=168) of the time before an accused’ first appearance. This information was ascertained after the accused’ first appearance, when the first appearance report and the bail report was returned to the PTS office.

Most police stations do not send fingerprint requests on time to the LCRC for processing. Of the 6 police stations in a 10km radius of the Mitchells Plain court, only Mitchells Plain police station sends more than half of the reports on time. This is still a large improvement in light of the fact that none were previously available. However, there is room for further improvement.

Cancellation of warrants
of warrantsThe second problem experienced by the project related to LCRC bail reports is the non-cancellation of warrants. Sometimes an accused is correctly identified but the computer incorrectly lists them as wanted. This may happen when they were previously wanted for a matter which has since been resolved but where the police did not cancel the previous warrant. The accused then has to explain the mistake to court in order to prevent it unfairly prejudicing their bail hearing.

Non-cancellation of current warrants is also problem. This is where an accused person is correctly identified and indeed wanted as a suspect in another case but where the warrant is not canceled. When the investigating officer fails to cancel the warrant, the police system still lists the person as wanted and the success for the LCRC in having found the wanted person does not translate into clearing the backlog of wanted persons on their system.

Impact of CRIM on bail decisions
See the section on bail decisions later in this report for a description of the impact of the CRIM linkage.

Relations with the community and bodies outside the PTS office
Much of the success of a PTS office is dependant on the co-operation of people outside the PTS office. These include court officials, community volunteers, other NGO’s the prosecutors, magistrates, clerks of the court, court orderlies, court management, and other court officials. Police and Correctional Services personnel involvement is also key to success. Some PTS offices have been more successful than others in winning the respect and co-operation of people these officials. This is particularly so where the Local Steering Committee, referred to earlier, has remained active. Where a PTS office has not yet won or has not retained external support and co-operation, its ability to do its work has been adversely effectively.

Community relations
Victims of crime also come to the PTS offices and obtain information about their case and complain if they have been intimidated. They have a safe and sympathetic place to wait before they have to go to court. Attorneys have written letters thanking PTS staff for their courtesy and efficiency (see the attached letter).

Court integration and relations with court other court-based staff
The approach used by the BJA in piloting PTS is to build policy from the bottom up. This means that a formal policy making process or legislation did not precede PTS. PTS was deliberately designed to operate with the flexibility to evolve into a model that best suited the courts, unconstrained by the rigidity of regulations. The rationale was that this approach gave more opportunity for people working in the courts to influence the outcome and that the resultant model would therefore probably be more appropriate to those courts. Once a model had emerged, this could then be formalized into national policy, including legislation, if necessary.

The trade off with this approach is that PTS offices do not operate under the authority of national legislation backing up their work. In the very formal and legalistic environment of the magistrates courts, some officials will not take cognisance of something unless they are required by law to do so. This makes the work of the PTS staff more challenging. To address this, legislation should be amended to formalise the status of PTS reports and to provide the legal framework in which PTS can operate most effectively.

Specifically, two amendments are proposed. First, a provision directing magistrates, when deciding release, to consider imposing monetary conditions only if no combination of other conditions would be sufficient to protect the interests of justice. Second, a provision directing magistrates in courts with a PTS office, to take PTS reports into consideration in making bail decisions.

Prosecutors
Co-operation by prosecutors with PTS differs significantly among the three courts. In Mitchells Plain, there is a close working relationship between the PTS office and both senior and junior prosecutors. The prosecutors have come to rely upon the FAR’s and the police bail reports. The Mitchells Plain Senior Public Prosecutor is actively involved in managing the PTS office and the court has taken ownership of the PTS process.

However, in Durban and Johannesburg, the attitude from many prosecutors towards PTS is negative. Based upon interviews conducted with prosecutors in Johannesburg and Durban, reasons given by them for not co-operating with or opposing PTS fall under the following categories:
Perception that PTS causes court delays.
Prosecutor does not understand or know what PTS is or does.
Apathy and not caring about their work.
The feeling that using the FAR involves extra work for them and their unwillingness to do extra work without compensation.
The perception that PTS is a project of the Minister of Justice and because he is allegedly not helping prosecutors to get more money they would not co-operate with "his" project.
The view that the FAR does not add value to the work of prosecutors because PTS is either unable to verify information or verifies information by telephone and telephonic verification is not as reliable as information verified in person.
The view that PTS does not assist with the administration of justice, is "useless" or even harms the administration of justice.

As an illustration of the lack of co-ordination in Johannesburg, the control prosecutor makes their bail decision from 08:00 when dockets arrive at court and will not wait for the PTS first appearance report to be prepared. Often the dockets arrive before the prisoners and it is impossible for PTS to do the interview and prepare the report before the control prosecutor makes their decision. By contrast, prosecutors in Mitchells Plain and Durban have adjusted their processes to accommodate the production of PTS reports by not processing first appearance dockets before 12:00.

In a number of specific cases in Johannesburg, the PTS office discovered through their database search that an accused had been arrested while out on bail on the same charge. This was put into a report and given to the prosecutor but the accused was released on bail again for the same amount. The prosecutor had failed to use the information supplied by PTS and did not inform the court of the pending matter.
In February 1999, the author found about 250 FAR’s dating back to October 1998 dumped and unused in Johannesburg’s court 13. It appeared that the prosecutors had stopped using the reports.

Following this, the BJA informed the Director of Public Prosecutions for the Witwatersrand, and the Senior Public Prosecutor and Control Prosecutors for the Johannesburg Magistrates Court of this discovery. There has been a reported increase in the use of the PTS reports by the prosecutors in Johannesburg since this intervention.

Relations with the police
At a national level, there are excellent relations with the Criminal Records Centre. This culminated in the CRIM linkage in Mitchells Plain and the planned extension to other courts.

In Durban, the police are very supportive and co-operative. "Investigating officers are making more use of PTS to establish the identity of accused through the readily availability of photographs ... [The PTS procedure] is conducive to a good working relationship between the investigating officer and the Court. The planned installation of a direct link to Pretoria CRC ... will assist Investigating officers tremendously in ensuring that wanted suspects can be arrested on their first appearance ... " (see the attached letter from the SAPS).

The Mitchells Plain Local Criminal Records Centre (LCRC), which supplies the bail reports containing previous conviction and wanted information, is extremely co-operative. However, this level of support is not matched by other elements of the SAPS in Mitchells Plain. Fingerprint forms are delivered timeously for processing by the LCRC in 51% of cases and there is poor supervision of conditions in the charge offices.

In Johannesburg, the police have been co-operative with the monitoring of compliance with bail reporting conditions.

Relations with accused
Accused persons speak of being treated with basic dignity by the PTS staff, a result of the service culture that has been created in these offices. The mostly unrepresented accused perceive PTS to give them a voice in court.

Court bail decisions
bail decision
The impact of PTS on bail decisions was assessed by comparing the decisions made by the courts before the start of PTS (baseline data) and at intervals after the start of PTS (impact data). These are presented below.

Durban
Base line and impact data regarding bail decisions for Durban are provided in Table 5. Note that the baseline data are only for those accused in custody which allows direct comparison with the impact data sourced from the PTS office.

Table 5 Bail decisions for Durban magistrates court at first appearance for accused in custody.

There has not been a significant change in the use of warning after the introduction of PTS. This remains in the region of 2-3% of all first appearance decisions. There was a reduction in the number of people granted bail (from 53 to 33 %) with a concomitant increase in the number of people being denied bail. The high number of "other"decisions in Durban reflects the practice of transferring cases to other courts.

Johannesburg
Base line and impact data for Johannesburg are provided in Table 6. Note that the baseline data are only for those accused in custody which allows direct comparison with the impact data sourced from the PTS office.

Table 6 Bail decisions for Johannesburg magistrates court at first appearance for accused in custody.

Although there was an initial increase in the use of waring, this has fallen back to the base line position. Seven to nine months after the start of PTS, the use of conditions had increased from 5% to 9%. However, 14 - 16 months after the start of PTS, this level has decreased to 3%, slightly below the baseline position. There has been a significant increase in the denial of bail in Johannesburg increasing from 24% (baseline), through 50% (7-9 months) to 55% (14 - 16 months).

This change in bail decisions in the Johannesburg Magistrates Court may reflect the impact of the 1997 amendments to the bail law. The new law obliges the court to deny bail for certain serious charges unless the accused can prove that there are "exceptional circumstances".

Mitchells Plain
Base line and impact data for Mitchells Plain are provided in Table 7. Note that this sample includes all first appearances. It does not separate accused who were in custody and those who were not in custody.

Table 7 Bail decisions for Mitchells Plain magistrates court at first appearance for accused in custody.

There has been increased use of warning after the start of PTS. The percentage of accused released on warning has increased from 40% before the start of PTS to 50% after 18 - 19 months of PTS operation. It appears that this increase in the use of warning may be related to a decreased use of money bail. There was a 13% decrease in the use of money bail over the same period. The number of accused denied bail did not change significantly for 14 months, but the most recent data suggest a decrease from nine percent to four percent.

Bail amounts
The bail amounts ordered by the courts in the three courts provide a further indicator of impact of PTS on the court decisions.

Durban
The bail amounts in the Durban Magistrates court before and after the start of PTS are provided in Table 8.

Table 8 Bail amounts for Durban Magistrates Court. The Table gives the range within which 75% of bail amounts fall (indicated by the 75% low and the 75% high) and the median bail amounts before the start of PTS (baseline) and a successive periods after implementation (impact). Amounts are in Rand.

Bail amounts have not changed significantly since the introduction of PTS in Durban. The median amount has constantly remained R500.

Johannesburg
The bail amounts in the Johannesburg Magistrates court before and after the start of PTS are provided in Table 9.

Table 9 Bail amounts for Johannesburg Magistrates Court. The Table gives the range within which 75% of bail amounts fall (indicated by the 75% low and the 75% high) and the median bail amounts before the start of PTS (baseline) and a successive periods after implementation (impact). Amounts are in Rand.

Johannesburg Magistrates court bail amounts remain unaffected by PTS. The range of 75% of the bail amounts has remained constantly between R500 and R5000. The median amount before the start of PTS of R2000 increased to R3000 seven months after the start of PTS but has maintained this level after 16 months.

Johannesburg bail amounts are still significantly higher than the other two courts, despite a similar or lower proportion of serious, violent crime compared with the other courts.

Mitchells Plain
The bail amounts in the Mitchells Plain Magistrates court before and after the start of PTS are provided in Table 10.

Table 10 Bail amounts for Mitchells Plain Magistrates Court. The Table gives the range within which 75% of bail amounts fall (indicated by the 75% low and the 75% high) and the median bail amounts before the start of PTS (baseline) and a successive periods after implementation (impact). Amounts are in Rand.

In Mitchells Plain, the median bail amount has fallen from R500 to R300 since the implementation of PTS. The 75% range has also narrowed.

Changes in the ability of accused to pay bail over this period provides valuable insight into the extent to which monetary bail conditions affects the release from custody of accused persons. The September 1998 sample showed that fewer accused persons were able to pay bail compared with the sample collected in June 1997. Only 37% of accused were able to pay bail at court in September 1998 (median of R300) compared with 47% in June 1997 (median of R500). Fewer people were able to pay bail despite a lower median bail amount. Perhaps accused became poorer over 14 months between the samples and this was why they could not pay.

Whatever the reason, these figures indicate that lower bail amounts do not necessarily mean a lower awaiting trial population. In September 1998, 63% of accused persons could not afford to pay a median bail amount of R300. These people were sent to Pollsmoor prison despite having been granted bail by the court. The low bail amounts tend to indicate that the court considers it in the interest of justice for these accused to be released but the imposition of monetary bail, even low bail amounts, means that the accused remain in custody.

Use by the court of special bail conditions
Upon release of an accused, certain "automatic" conditions apply (such as the condition that the accused must return to court). In addition, a magistrate may impose "special" conditions of release, which are divided into two classes: the first is to ensure that the accused stands trial and therefore to make it difficult for them to abscond or flee (for example, the accused must hand in their passport or must report regularly to a police station). The second class of conditions are designed to prevent recurrence of or persistence in unlawfulness while awaiting trial (for example, a condition that the accused may not communicate with state witnesses). The use of special conditions by the three courts are set out below.

Durban
The use of special conditions by the Durban court is set out in Table 11.

Table 11 The use by Durban court of special conditions as a percentage of accused who were released on warning or granted bail.

Since the introduction of PTS, there has been a slight increase in the use of conditions in Durban. However, the overall use of conditions remains low.

Johannesburg
The use of special conditions by the Johannesburg court is set out in Table 12.

Table 12 The use by Johannesburg court of special conditions as a percentage of accused who were released on warning or granted bail.

Before PTS started, special conditions were seldom used in Johannesburg. After nine months, the use had picked up. Special conditions were used in 12% of cases where the court released the accused on warning or granted bail. Even more significantly, almost half (49%) of all accused who were released on warning were ordered to report. However, the most recent data (after 16 months) indicates that the use has declined to an overall rate of 4%

Mitchells Plain
The use of special conditions by the Mitchells Plain court is set out in Table 13.

Table 13 The use by Mitchells Plain court of special conditions as a percentage of accused who were released on warning or granted bail.

Mitchell Plain court shows a similar pattern to Johannesburg: An initial increase in use but a tapering off lately. The use of special conditions increased significantly from 0.3% before PTS to 10% 14 months after the start of PTS. The February 1999 sample shows that use is down to 1% of all accused released on warning or granted bail.

Impact of previous record information on bail decisions
The police bail reports provided to court show that a significant proportion of accused have previous records or are wanted by the police. Up to 30% (n=525) of accused "possibly" had a previous record and as many as 6% (n=1747) of accused were "possibly" wanted by the police.

The impact of the bail reports on court decisions is set out in Table 14.

Table 14 Bail decisions divided between accused who had "possible" previous convictions, those who did not and those where this was unknown because no bail report was before the court.

Being aware of an accused’s possible previous convictions seems to significantly influence the court’s decision. Of the accused that had a no previous conviction, 74% were released on warning or granted bail, compared with just over 50% who had a possible previous conviction. Accused who have a possible previous conviction are twice as likely to be denied bail, as those with no prior convictions. Accused who have no previous conviction are twice as likely to be released on warning as those who have no previous conviction. When an accused’ possible previous convictions are unknown, they stand a similar chance of being released on warning or granted bail (when these are combined) or having their bail denied, as when they have no previous convictions. See Table 14.

Bail decision for accused that are possibly wanted
Bail was denied to all three accused who were wanted by the police in the study period.

Post-release supervision
The extent to which accused comply with their reporting conditions gives some indication as to the usefulness of this condition and provides one basis for assessing the impact of PTS supervision. However, it has proved difficult to monitor this indicator and data are not available for all courts. The available results are reported below.

Durban
The compliance data for Durban are set out in Table 15.

Table 15. Compliance with reporting conditions in Durban

Of the people required by the Durban courts to report to police stations, very few fail to comply with this condition. In the two samples, 2% and 0% respectively failed to report. Accused are in partial compliance if they failed to report on at least one occasion. Accused in partial compliance still reported most of the time and resumed reporting after a lapse in compliance. The vast majority of accused were in full compliance with their reporting conditions by reporting on every occasion that they were required to report.

Johannesburg
Data regarding reporting in Johannesburg are set out in Table 16.

Table 16. Compliance with reporting conditions in Johannesburg.