Much has been made of National Prosecuting Authority (NPA) head Shaun Abrahams exercising his prosecutorial discretion inappropriately. But what if the abuse of discretion is mirrored lower down the ranks?
The evidence suggests an increasing tendency to decline to prosecute in favour of "informally mediating" ordinary cases. Decisions not to prosecute are not routinely reviewed and no policy guides informal mediation. The result is fewer convictions in cases that matter despite increased resources.
NPA policy says prosecution should occur when there is "a reasonable prospect of success" and no public interest reason for not prosecuting. Figures reported in NPA annual reports show that "a reasonable prospect" seems to mean a near-certainty: in 2015-16 the NPA achieved a 93% conviction rate. The rate has gone up steadily from 82% in 2002-03. Unfortunately, the increase has been associated with a drop in the number of convictions — 13% fewer since 2002-03. Why have convictions dropped if the conviction rate is so high?
The NPA has stopped reporting on new enrolments. So it is not immediately clear whether the drop in convictions is due to fewer referrals of cases from the police. The annual reports of the South African Police Service (SAPS), however, show that the number of priority crime arrests has almost doubled since 2002-03. And since 2010-11, the SAPS has reported on the number of "detections" and the number of "trial-ready dockets". These have gone up rather than down — the latter have more than doubled. This suggests lack of SAPS throughput cannot be the reason for the reduction in convictions.
Could the drop in convictions be the result of the NPA prioritising more serious cases for prosecution? The SAPS (not the NPA) records convictions per crime type. The SAPS-recorded convictions for the 20 most serious crimes have dropped sharply — there were a third fewer in 2015-16 than there were in 2008-09. The SAPS-recorded convictions usually exceed NPA case convictions, because each NPA case conviction may involve more than one crime conviction. But in 2015-16, there were fewer SAPS-recorded serious crime convictions than NPA case convictions. This must mean more NPA case convictions relate to crimes that are not among the SAPS 20 most serious crimes.
Furthermore, convictions for "crimes dependent on police action" are increasing. These crimes — 90% of which are drug crimes, 6% "driving under the influence" and 4% firearms offences — are easier to prosecute because the accused is often caught in the act, and there is usually no "victim". In the past two years, more than half of serious crime convictions recorded by SAPS related to this category. Serious crime convictions other than these three crimes have halved since 2009-10. This suggests the reduction in convictions is not due to a focus on other serious crime.
Is the reduction in convictions due to lack of court time as a result of more time-consuming cases like that of Oscar Pistorius? The NPA reports that average court hours per day dropped from four hours and nine minutes in 2002-03 to only three hours and 16 minutes in 2015-16. Courts are empty most of the day.
Is the reduction in convictions due to dwindling human resources and funding? The inflation-adjusted budget for general prosecutions has almost doubled since 2002-03, while the NPA as a whole enjoys an inflation-adjusted budget that is 185% of what it was in 2002-03. The NPA employed 38% more prosecutors in 2011-12 than in 2005-06.
The NPA has argued that the reduction in convictions is accounted for by the increase in resolutions via "alternative dispute-resolution mechanism" (ADRM) cases, which have indeed increased 900%. Adding ADRM to cases finalised in court shows some increase after 2007-08, but not as large as the increase in funding or prosecutors.
There are many reasons to prefer a fair ADRM to prosecution. But most ADRM (as much as 75%) is "informal mediation". " This appears to be the ultimate in the exercise of prosecutorial discretion, where the prosecutor mediates between the complainant and accused — eventually leading to withdrawal of the case.
Anecdotal evidence suggests informal mediation sometimes involves payment of compensation to the complainant. This scenario holds potential for abuse. ADRM should occur only when there is an answerable case.
If there is no case to answer, the prosecution should simply withdraw, and not subject an accused to ADRM. Yet the SAPS includes among its "trial-ready dockets" those cases "where the public prosecutor decided to finalise the case by means of alternative dispute-resolution programme while investigations are still outstanding".
A further concern is that if a case is resolved by ADRM, the accused has no criminal record. The sheer volume of ADRM (180,000 in one year) means that the absence of a previous conviction is no longer a reliable indicator.
There are no policies guiding informal mediation and it is unclear whether prosecutors are trained in mediation. In his 2016 budget vote speech, Justice Minister Michael Masutha said in relation to ADRM: "We are currently working with the NPA to refine the system to ensure that it is fair and just, not only to the accused but equally to the victim as well." This suggests that in its current "unrefined" form, it may not be fair and just.
The reason for the reduction in convictions is clearly an increasing tendency not to prosecute. Is this resulting in some people getting away with crimes, some of which may be serious? Is it resulting in some who have no case to answer being pressured into settling informally, or victims being pressured to withdraw? Are serial offenders getting off with no criminal record?
Combining this general tendency not to prosecute with what has been occurring at the highest levels of the NPA raises serious concerns. Abrahams may be the least of our worries. It is not enough for the NPA to say "trust us". Independent oversight mechanisms are needed. The conversation about what future oversight might involve is just starting.
• Redpath is a Civil Society Prison Reform Initiative researcher at the Dullah Omar Institute at the University of the Western Cape. She is a panellist for a discussion on future oversight of the NPA to be held on November 22.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.