Is the Stick Real? Trends from Concluded Prosecutions of Industrial Environmental Crimes in South Africa M Strydom

South Africa's suite of environmental laws contains many criminal sanctions and penalty provisions. Whether the criminal sanction is an effective tool that realises the constitutionally protected environmental right depends on how it is practically enforced and whether potential offenders become aware of such enforcement measures. This article reports on research aimed at collecting and analysing prosecutions for industrial-related transgressions (conducted mainly in Magistrates' Courts) and involving offences under the National Environmental Management Act (NEMA), the Waste Act , the Air Quality Act and the National Water Act . An analysis of 53 prosecutions shows that most cases resulted in convictions, half were concluded through plea and sentence agreements, half involved the conviction of individuals, no direct imprisonment penalties were imposed, and low fines were imposed in most cases. The findings include that there is some inconsistency in how different listed activities or water uses are treated as separate or consolidated criminal charges, and the exact number, outcome, or trends arising from such cases are difficult to determine as there is no central, readily accessible database of concluded prosecutions. Increased access to such information will improve knowledge, implementation and the effective use of the criminal sanction through prosecutions. In turn, this will contribute to the improved realisation of the constitutionally protected environmental right


Introduction
Since the mid-1990s, South Africa has enacted a comprehensive suite of environmental laws 1 to give effect to the constitutional obligation to protect the environment through reasonable legislative and other measures. 2hese laws 3 contain many criminal offences, and on conviction, a person may be liable to significant penalties, including imprisonment, fines, or both. 4iminal liability may be substantial depending on the nature of the offence.Maximum prescribed penalties up to R5 million or R10 million or imprisonment of up to five or ten years could be imposed. 5Section 34 of the National Environmental Management Act 107 of 1998 (NEMA) also provides for further possible consequences including that, on conviction, an accused can be ordered to pay damages or pay for losses suffered to rehabilitate or repair damages to the environment, reasonable prosecution costs, or damages equal to the advantage the accused may have gained from the transgressions. 6Liability may also extend to employers, including incorporated entities, being liable for the actions of their employees, and individuals (directors, managers, employees or agents) could be liable even where the offence benefitted an employer. 7ER / PELJ 2023 (26)  4 Effective enforcement is important when deterrence is the goal, and the public must be aware of penalties being utilized since "ultimately, one cannot fear what turns out to be a paper threat". 21us, the efficacy of prosecutions and the value of convictions require knowledge of what is happening in the criminal law system.
Holmes 22 famously asserted that the law consists of "prophecies of what the courts will do … and nothing more pretentious".Without a central, readily accessible database of concluded prosecutions, it is difficult to determine the number, outcome, or trends arising from such cases, i.e., what courts do.Although criminal measures have been evaluated in South African environmental law, 23 no one has undertaken an empirical assessment of concluded prosecutions for industrial environmental crimes in South Africa, i.e., prosecutions that reached the end of the criminal justice process, resulting in a conviction or acquittal.
This research set out to obtain information on concluded environmental prosecutions in the context of industrial operations, 24 focussing on offences in the NEMA, the National Environmental Management: Waste Act 59 of 2008 (the Waste Act); the National Environmental Management: Air Quality Act 39 of 2004 (the Air Quality Act); and the National Water Act 36 of 1998.
This paper aims to analyse the collected data on concluded environmental prosecutions, ascertain trends from such cases and report such findings.Ultimately, 53 concluded prosecutions under the mentioned laws were considered.A limitation of this research is that the outcome of such cases is not readily accessible and only published in limited instances; thus reliance was placed on obtaining the information from other sources, including research participants.
The remainder of this paper comprises Part 2, which sets out the methodology of the research.Part 3 outlines the relevant context in considering concluded prosecutions.Part 4 highlights findings and themes from the cases considered, and a discussion of the main findings appears in Part 5.

Methodology to collect and collate case information
The research involved gathering publicly available information on relevant concluded prosecutions, whether completed through a guilty plea, trial, or plea and sentence 25 (section 105A) agreement.The focus was on concluded prosecutions for "industrial" (brown sector) 26 environmental crimes, focusing on facilities established for producing or manufacturing products or commodities or treatment and disposal of waste, which activities are often undertaken in areas zoned for industrial use.
Industrial facilities are generally required to comply with various environmental laws, and their activities could have significant environmental impacts.To operate lawfully, they may be required to hold numerous environmental authorisations, permits, licences, and/or registrations (collectively referred to as "approvals"). 27The inquiry was limited to the NEMA, Waste Act, Air Quality Act and National Water Act, based on their overarching relevance to industrial facilities, and to keep the research within manageable bounds.Several other laws may apply to industrial facilities, 28 but these four are the principal laws considered.
There is no existing case law database containing the complete outcomes (with the underlying judgments, sentences or section 105A agreements) on concluded prosecutions for industrial environmental crime.We therefore had to turn to various sources to gather information on concluded cases, whether reported in the media, published by public benefit or other organisations or published in databases of court judgments.
The annual National Environmental Compliance and Enforcement Reports (NECERs) published by the Department of Forestry, Fisheries and the Environment (DFFE) 29  starting point for requesting case outcomes from relevant role-players.The number of industrial prosecutions instituted or adjudicated under the four selected laws is unknown as the NECERs provide overall statistics, not per sector.Information on concluded cases was received from participants in the underlying research 32 and is limited to what was obtained. 33There are likely more cases, including prosecutions predating or postdating the NECERs.The cases collected thus represent a starting point for the development of a more comprehensive database.
The period for the inquiry was limited to 2005 to 2020, as the first concluded prosecution under the selected laws appeared to be from October 2005. 34Although there are prior prosecutions, 35 their outcomes are not easily accessible.The 2005 start date also somewhat corresponds with the timing of the insertion of the approval-related offence provision in NEMA in 2005. 36The last concluded cases were from 2020, being the cutoff date for the research project. 37e prosecutions included in the data-set were mainly conducted in lower courts, 38 i.e., Magistrates' Courts, 39 as these courts also have jurisdiction over environmental crimes; NEMA grants magistrates' courts enhanced penal jurisdiction, but does not take away the high courts' jurisdiction to also hear such matters. 40It is not general practice in the Magistrates' Courts and not explicitly required by the rules of court for written judgments to be handed down by the presiding officer/magistrate for reporting purposes. 41The mechanical recording can be transcribed to In order to obtain case information for concluded prosecutions, other sources had to be consulted, including requests to relevant role-players, and some reliance has to be placed on the information received.Initially, when public information was considered and before more information was received from role-players, there appeared to be limited industrial-related prosecutions.The scope of the inquiry was thus extended to a wider range of cases under the four selected Acts, including prosecutions for mining or farming-related activities.This broader inquiry resulted in the collection of 53 cases, 49 of which 40 relate to industrial-type contraventions.
The thirteen additional cases comprised of the following: Although these additional cases fall beyond the remit of industrial environmental crimes, it was decided to include them in the dataset, as their inclusion advanced the objective of establishing a baseline archive of environmental criminal prosecutions.They also served as a separate subset to compare the findings emerging from an analysis of the industrial cases.For example, nearly half of the additional cases were also concluded through a section 105A agreement.
As stated, this collection of 53 cases is very likely incomplete because it relies on information from limited sources and is not a consolidated, official database.Other cases may be unknown to the authors, and more recent cases (involving municipalities) have since emerged of which the data is not readily available except for what is reported in the media. 63The available information obtained is often also limited; for example, although the NECERs contain case summaries or mention matters criminally investigated and prosecuted, they do not necessarily provide a comprehensive account of all concluded industrial-type prosecutions.
Notwithstanding the value of knowing how many cases have been successfully prosecuted, unfortunately, for 23 of the 53 cases, the information received was limited. 64 sentence annexures were received, i.e., after a guilty verdict the Magistrate will hand down a sentence usually comprising of a 1 to 2-page document.This sentence annexure cites the accused and the sentence but contains no (or very limited) further case information.From the sentence annexure, it may also be unclear which charges the accused faced.
The analysis and findings in this paper are limited by the challenges outlined above, particularly due to incomplete or difficult-to-access information.Nevertheless, the collection of cases in this research is the most complete current compendium of industrial environmental crime prosecutions assembled in South African environmental law scholarship to date. 65e bulk of this article highlights themes from the case analysis, with a view to determining the efficacy of the criminal sanction and its use as a reasonable legislative measure to protect the environment. 66 necessary pre-cursor to that analysis, Part 3 sets out the constitutional and legislative context for environmental crimes in South Africa.

Industrial environmental criminal sanctions as a legislative measure to realise the constitutionally protected environmental right
This part situates the environmental criminal sanction within the context of the reasonable legislative measures intended to advance section 24 of the Constitution; relevant role players and pathways to prosecution; and sentencing; and the effectiveness of criminal sanctions.

Role of the criminal sanction within a suite of legislative measures to realise section 24
Non-compliance with environmental law requirements can be met with enforcement action ranging from civil and administrative to criminal action (as "command-and-control" mechanisms). 67The criminal sanction is a tool the State can use to realise the constitutionally protected environmental right, as it provides for punitive consequences when there is noncompliance with legislative measures.Traditionally, the aim of criminal law is said to include: (i) prevention; (ii) deterrence, which may be on an individual or general basis; (iii) reformation or rehabilitation and (iv) retribution. 68ny consider the criminal mechanism a last resort69 that should be reserved for serious offences70 "when other means of dealing with offending and non-compliance fail". 71This reservation of the criminal law mechanism as a last resort can be construed as a last resort legislative measure, i.e. the last resort legislatures turn to in regulating behaviour.However, it could also be construed in a regulatory sense, as the last resort where legislation provides other mechanisms to address noncompliance.There is no requirement in South African environmental law to exhaust other mechanisms before proceeding to criminal prosecution; thus, administrative compliance and enforcement mechanisms and the criminal sanction can be used concurrently.Based on experience in practice, and the interviews conducted as part of the underlying research, the concurrent use of these mechanisms could be due to the seriousness of the contravention and to prompt corrective or remedial action.
Nurse Introduction to Green Criminology 84.

M STRYDOM & T FIELD
PER / PELJ 2023( 26) 12 Kidd aptly notes various challenges in achieving the aims of a criminal sanction, considering the widespread difficulties faced by a criminal justice system in crisis, 72 and its weaknesses, particularly in an environmental law context. 73Yet, criminal sanctions for environmental law transgressions are so widely prescribed 74 that the laws are described as being "littered with new criminal offences". 75r a mechanism to promote the section 24 constitutional right, it must be effective and used appropriately, within the confines of the other constitutionally protected rights.To understand the efficacy of the sanction, one has to consider its practicality and how it has been/can be used, i.e., by considering actual concluded cases and role-player concerns. 76

Role-players and pathways to prosecution
Environmental Management Inspectors (EMIs) are central role-players in investigating environmental crimes and referring cases to the National Prosecuting Authority (NPA) for prosecution.EMIs can be designated for different mandates and functions or to enforce different environmental laws. 77Their functions could include compliance monitoring or enforcement. 78In the case of the latter, EMIs usually investigate acts or omissions (within their mandate) where there is a reasonable suspicion that it might constitute an environmental law offence. 79They have broad general powers, including questioning people about suspected unlawfulness. 80Is must refer a criminal case to the NPA for a decision on whether or not to prosecute.In the context of industrial environmental crimes, if the NPA decides to proceed with a prosecution, the accused will ordinarily be summoned to appear in court.The underlying PhD research involved interviews with 32 key industry participants, the findings are not discussed in this article.crimes involve corporate entities, ordinarily cited as the first accused, as represented by corporate officers. 82iminal prosecutions can be concluded in different ways, if a matter proceeds to trial, it could culminate in a guilty plea, 83 or a section 105A agreement. 84If the accused pleads not guilty the matter will proceed to a hearing.When pleading guilty, an accused may explain their guilty plea through a plea explanation.The court may question the accused to ensure guilt and the absence of a valid defence. 85If an accused pleads not guilty, the presiding officer may ask if the accused wants to explain their defence. 86If it is unclear what is in dispute, the presiding officer may question the accused to establish which charges are contested. 87The matter then ordinarily proceeds to trial unless the prosecutor and a represented accused enter into a section 105A agreement (which is subject to formal requirements and judicial approval as explained below).
It is also possible for a prosecution to commence and not be concluded through any of these three avenuesa prosecutor could decide not to prosecute, or the matter could be provisionally withdrawn. 88This paper does not consider these cases as it is more challenging to establish the status of pending or provisionally withdrawn prosecutions.These cases are considered sub judice.Further, information is even less readily available because there is no definite outcome regarding the guilt or innocence of an accused.
As discussed below, nearly half of the cases considered were concluded through section 105A agreements, a section inserted into the Criminal Procedure Act 51 of 1977 (CPA) in 2001, formalising the "informal and longstanding practice of plea-bargaining". 89The section 105A provisions can be used in any criminal case and are not limited to environmental prosecutions. 90Certain formal requirements must be met before a court may accept a section 105A agreement, including that it must be in writing and at least state that the accused was informed of their rights to be presumed innocent until proven guilty beyond a reasonable doubt, to remain silent, not to testify during the proceedings, and not to be compelled to give self-incriminating evidence. 91The agreement must contain substantial facts, facts relevant to the agreed sentence, and any admissions made by the accused. 92The court may not participate in the negotiations, 93 and its role is limited to confirming that the accused entered into the agreement and that the prosecutor consulted the investigating officer and complainant. 94 satisfied that the accused admits the allegations in the charge and guilt, the court may ask relevant questions and hear evidence for sentencing purposes.If the court is satisfied that the sentence agreement is just and fair, the court sentences the accused accordingly. 95If not, the court informs the parties of the sentence it considers just.The agreement may be withdrawn by the parties, or they may abide by it subject to the right to lead evidence and present arguments relevant to sentencing. 96If the parties abide by the agreement and present evidence and argument relevant to sentencing, the court can convict the accused and impose the sentence it considers just. 97However, if they withdraw from the agreement, the trial will start anew before another presiding officer unless the accused waives this right. 98No record of the negotiations or agreement may be used unless the accused consents to recording any of their admissions. 99

Sentencing and level of penalties
High penalties, either R5 million-or five-years imprisonment, or R10 million or ten years imprisonment, are generally (currently) prescribed in NEMA, 100 the Waste Act 101 and Air Quality Act. 102This contrasts with the five years' imprisonment provided for in the National Water Act penalty, which does not specify an equivalent monetary penalty and converts to R200 000.Although high penalties are prescribed in statute, these are not necessarily the penalties imposed by courts or negotiated as part of a section 105A agreement.In considering what an appropriate sentence might be, courts have a wide discretion and consider "the triad" of the crime, the offender's personal circumstances, and the interests of society, as well as the proportionality of the sentence in relation to the crime, the offender's personal circumstances and the interests of society. 104e research considered how many of the 53 cases resulted in comparatively high or low sentences (considered further below).Echoing Holmes' assertion, 105 if courts do not impose such high penalties, it may indicate that the maximum prescribed statutory penalties (although an upper limit) are too simplistic or may give rise to disproportionality or disparity.

Effectiveness of the criminal sanction in SA environmental law
As previously stated, the deterrent value of environmental sanctions depends on "the likelihood of apprehension, prosecution, conviction and significant penalty". 106Deterrence relies on effective enforcement and requires public knowledge of penalties being imposed. 107According to Husak, whether criminal statutes produce any marginal gains in deterrence is doubtful, 108 and lengthy delays in the criminal justice system erode deterrence value. 109e critiques of deterrence theory 110 bring into question the value of environmental sanctions and prosecutions, particularly considering challenges in using criminal sanctions as an enforcement tool, the delays and difficulties within the criminal justice system, 111 and the lack of readily available information about concluded prosecutions. 112The effectiveness of criminal sanctions for environmental law transgressions is linked to the general efficacy of the criminal justice system, which has been described could equate to a fine of R200,000.GN 217 in GG 37477 of 27 March 2014 determines that the fine a court can impose is "R120 000 where the court is not the court of a regional division, and R600 000 where the court is the court of a regional division", i.e., and which could impose 3-or 15-years imprisonment respectively, equating to a R40 000 per year fine.as being in crisis. 113Any challenges and delays in that system undermine the aims of criminal law.In other words, prevention becomes less likely with lengthy time lapses whilst disputes are being adjudicated in court.
The impact of punishment and deterrence then also deteriorates.

Assessing the cases
With the above context in mind, the 53 concluded cases collected were analysed to identify emerging themes including (1) the pathways to criminal prosecution, either through a guilty plea, trial or section 105A agreement; (2) total convictions and whether a natural (individual) or juristic person was prosecuted; (3) the percentages of the sentences imposed compared to the maximum prescribed penalty for the charge(s) the accused faced and whether it was comparatively high or low; 114 (4) if additional sanctions were imposed in terms of section 34 of NEMA; (5) how charges relating to separate listed activities or water uses received different treatment in some cases; and (6) how similar penalties are prescribed regardless of the seriousness or actual environmental impact of a transgression.

Pathways to criminal prosecution
Twenty-six of the 53 cases were concluded through section 105A agreements and thirteen were concluded through guilty pleas or trials.For the remaining thirteen, it is unclear whether the prosecution was concluded by a guilty plea or trial due to limited or incomplete available information. 115Cases concluded through a section 105A agreement generally have the most information because the agreement contains relevant details and is more readily accessible.For some cases, written judgments were available; 116 for others, transcriptions or High Court appeal judgments provided some information. 117113 Cameron 2020 SALJ 32. 114 On the assumptions that the base data for the calculations are correct, although there may be variables that are not apparent from the information obtained. 115 For example, the information is often contained in brief summaries in the NECERs, SAP69 forms (the South African Police Service form that captures criminal convictions), sentencing annexures containing only the sentence imposed.Where limited documents are provided, it is not always possible to determine the charges the accused faced.Compared to the proportion of general crimes concluded with section 105A agreements, the use of such agreements for environmental crimes appear high.According to the NPA 2019/20 Report, 2166 section 105A agreements were entered into in that year out of 231 725 finalised criminal matters, 171 a ratio of just below one per cent.Based on the case information received, the ratio for section 105A agreements in environmental crimes is much higherat 49 per centbut this represents cases concluded over many years out of the total multi-year database of 53 cases (which is probably not an exhaustive database).
Several benefits (for the prosecution and accused) associated with the plea-and-sentence pathway may be driving the high ratio of section 105A agreements for environmental crimes.From the prosecution's perspective, environmental crimes are complex in terms of fact and law and require a lot of time, effort, and resources to pursue.Most prosecutors and magistrates (as experts in criminal or civil law and procedure) may have limited exposure to, or experience in, working with environmental laws. 172he evidence is often of a scientific and technical nature, comprising of extensive documents and reports.As a result, it may be difficult for the prosecution to achieve a conviction on all charges.A section 105A agreement can result in a quicker case resolution, which would otherwise involve a protracted trial, which also requires extensive resources.The benefits in terms of time and resources saved (within an already constrained criminal justice system) are often listed as mitigating circumstances in section 105A agreements.In other words, it states in mitigation that the accused pleaded guilty to the offences in the section 105A agreement thereby saving State resources and not wasting the court's time.It could (or at least should) also result in the transgressor's quicker attendance to the environmental impacts requiring remediation or authorisation.
The benefits of a section 105A agreement for an accused also involve some level of certainty over the conviction and sentence, as it is negotiated and agreed.In such negotiations, the accused may be able to select charges to which it is willing to plead guilty, thereby achieving a compromise.It is often also a compromise in the negotiation process that the juristic entity pleads guilty and is convicted, and then the charges against any natural persons who also appear as accused (often directors or employees) are withdrawn, preventing individuals from obtaining criminal records.It may be that natural persons are cited as co-accused as a tactic to encourage section 105A negotiations.In other words, where employees, directors or corporate decision-makers are cited as coaccused with the corporate entity, such employees would rather negotiate that the juristic entity pleads guilty, and a compromise is reached in the form of a section 105A agreement to avoid the prosecution of the individuals in their personal capacity.A criminal record likely has more severe consequences for an individual than it would have for a juristic entity.Individuals obtain a criminal record through a SAP69 which record is easily retrievable based on fingerprints and may affect various aspects of an individual's life (employment or travel prospects).Whereas juristic entities with such a record can be wound up, deregistered and reestablished.
When reading the facts and circumstances in the section 105A agreements, culpability/fault is not always apparent.Often the charges relate to negligent as opposed to intentional offences.Section 105A agreements contain lengthy explanations of the accused's reasonable measures to avoid environmental harm or offences.Of course, the accused takes a "best foot forward" approach in explaining the circumstances that led to the alleged contravention.However, this makes it difficult to assess the level of culpability thoroughly.Because not many of the cases considered have ended in a successful trial with available detailed records, it was also challenging to assess the level of culpability or how the court may decide on a matter. 173

Total convictionsnatural persons convicted
Of the 53 concluded cases, only one resulted in an acquittal after going to trial. 174This is indicative of a 98 per cent conviction rate.A possible explanation for this could be that due to limited resources, only strong cases are selected for prosecution (or pursued).The authors are aware of some matters where the prosecution did not proceed due to the technical nature of some transgressions, the lapse of time, questionable culpability, the magnitude of relevant evidence and lack of State resources, including sourcing expert witnesses.
In about half (26 of 53) of the successful prosecutions, natural (individual) persons were convicted, 175 most involving a business entity where individuals were convicted either alone or with a business entity -as persons in authority, proprietors, or directors.Unfortunately, for most of these cases ( 17), limited information is available; for many, only a SAP69 form (the form that records the accused's conviction) and a short annexure indicating the sentence was received.Often, the information lacks the complete charges the accused faced.For these 17 cases, it was therefore impossible to assess each individual's culpability or factors relevant to the sentence imposed.
Nevertheless, the available information is useful in adding to the database of known concluded prosecutions.Albeit limited and constrained, the data was used to try and establish how the sentences imposed compare to the prescribed maximum penalties.Fourteen cases involving the conviction and sentencing of a natural person (excluding where a section 105A agreement was concluded or in the case of S v Frylinck) resulted in a comparatively extremely low sentence, equal to or below one per cent of the potential maximum penalty.This figure may indicate that when the court's discretion is exercised in sentencing, the punishment considered appropriate for individuals is at the lower end.

Percentage comparison of sentences imposed versus potential maximum statutory penalties
In the overwhelming majority of cases, comparatively low sentences were imposed.This finding is based on a comparison of the prescribed penalty for the known charges and sentences that were ultimately imposed. 176In 43 of the cases (81 per cent) sentences below 50 per cent of the potential maximum statutory penalty were imposed.Five cases resulted in comparatively high sentences/fines.For the remaining five cases, the comparative percentage is unknown because of limited data.
The five cases that resulted in comparatively high sentences, and the calculation/comparison, appear in Table 3 below. 177Notably, three of these cases were concluded through section 105A agreements (1, 4 and 5 in the table below).The other two (numbers 2 and 3 below) were concluded through a trial (Frylinck) or a guilty plea (Joubert). 178In respect of items 1, 4 and 5 in Table 3, the accused may have agreed to relatively high fines to keep evidence of its culpability out of the public domain and limit reputational risk; other factors unknown to the researchers may also have influenced these agreed penalties.Because a section 105A agreement was concluded, underlying evidence would not be presented in open court and may limit public attention.Without considering the underlying evidence, the basis for agreeing to high fines is speculative.The comparison appears in an annexure to the underlying PhD research, which is too lengthy to transpose here, but examples which illustrate the methodology applied in the calculations are presented in Table 3. 177 On the assumptions that the base data for the calculations are correct, although there may be variables that are not apparent from the information obtained. 178 Although this is uncertain as the documents received for this case are limited to a SAP69 form and a sentencing annexure containing limited particulars of the offence and sentence.Following the Adjustment of Fines Act 101 of 1999 in 2008, the equivalent monetary penalty for six months imprisonment was R10 000 at the time of the offence. 182 Count 1 could result in a maximum fine of two years imprisonment, or an amount prescribed under s 1(1)(a) of the Adjustment of Fines Act 101 of 1999 (i.e., R40 000) read with s 92(1)(b) of the Magistrates' Court Act 32 of 1944 which provides that a district court (i.e., where the court is not the court of a regional division) can impose a sentence of imprisonment for a period not exceeding three years, and a regional court can impose imprisonment not exceeding 15 years.GN 1411 in GG 19435 of 30 October 1998 determined that the fine a court can impose is "R60,000 where the court is not the court of a regional division, and R300 000 where the court is the court of a regional division".
183 Count 1 and 2 were both transgressions of s 20 of the Environment Conservation Act 73 of 1989, and offences in terms of s 29(4), at the time, which between 9 June 1989 and 18 September 2009, could result in a maximum penalty of R100 000 (under the said Act).This amount was increased by the National Environment Laws Amendment Act 14 of 2009, but its retrospectivity is questionable.There may have been other relevant facts/considerations, unknown to the researchers, which influenced the additional consequences/penalty agreed to in terms of s 34 of NEMA.Of the total cases, 41 resulted in fines of less than 20 per cent (most less than 5 per cent) compared to the potential maximum statutory penalty.
Eleven of these resulted in a zero per cent fine; the reasons for this could include that the accused did not have the financial means to pay a fine, and the court may have considered a conviction and criminal record a sufficient punishment.In other instances, remediation costs might have been paid or payable.One accused was sentenced to serve three years of correctional supervision and undertake community service. 184en considering all the comparatively low sentences imposed (or accepted by courts) it could be argued that the prescribed maximum statutory penalties are at the extreme high end.This is particularly so considering that when sentencing principles are applied such high sanctions were likely considered to be inappropriate (and lower sentences imposed).Although the prescribed sanctions are maximums, the fact that much lower sentences are imposed does bring into question the functionality of only prescribing maximum penalties.

Listed activity treatment -Splitting of charges, duplication of convictions and multiplicity of sentencing
Another aspect that arose from the case assessment is that there appears to be inconsistency in how cases involving approvals for listed activities or water uses are treated.For example, where an accused undertook an activity that triggered several NEMA listed activities or water uses under the National Water Act, sometimes the prosecution dealt with each listed activity or water use as a separate criminal transgression/charge.In other cases, these were consolidated into one charge.An illustration of these cases appears in Table 4 below.principle) the correct position is the separate listing of each activity or water use as a separate contravention and charge.There is no immediately apparent reason or trend explaining the different approaches.
A further aspect that may arise from this different treatment of listed activities or water uses is the possibility of duplication of charges and convictions, 188 but is excluded from the ambit of this paper.

Section 34 of NEMA
Section 34 of NEMA creates further possible consequences when a person is convicted of NEMA Schedule 3 offences, such as damages or cost orders, and additional liability provisions, including possible liability of directors, managers, agents and employees, even where an offence benefits an employer.As mentioned above, natural persons were convicted in 26 of the 53 cases.In at least eight of these, natural persons appear to have been convicted together with juristic entities. 189ction 34 was used in 17 of the 53 cases (32 per cent) to secure payment directly to environmental authorities.This is different to the normal procedure, where a fine paid at the Magistrates' Court goes to the National Treasury.In some instances, section 34(4) was used to recover the reasonable costs incurred by the public prosecutor and the organ of state concerned.In other instances, damages were ordered in terms of section 34.

. S v Vunene Mining
A portion of the R3 million fine imposed, i.e., R750 000, was payable to the Environmental Protection Agency, in terms of NEMA s 34B, which provides that a court may order that a fourth of the fine to be paid can be paid to an informant whose evidence led to the conviction or who assisted in bringing the offender to justice could.

S v Golfview Mining
The  The accused had to pay a fine of R4 million to DFFE for the execution of EMI functions.

S v Samancor Chrome
The accused agreed to pay R2 million (R700 000 earmarked for initiatives for the Steelpoort Primary School, R1 million to the EMIs to execute their duties), R100 000 for prosecution costs, R200 000 to the clerk for the offences, and R100 000 for prosecution costs 5. S v Rodeo Solve CC The accused had to pay R50 000 to the new tenant for costs incurred for remediation; and R30 000 to DFFE for investigation costs.

S v D Groenewald
The accused was ordered to pay R28 152.66 to the City of Tshwane for costs for removal and safe disposal of waste.

S v SAM Marie
Consulting CC The accused was ordered to pay additional compensation of R50 000 to the DFFE for the EMIs for the performance of their functions and duties.

S v EnviroServ Waste Management Ltd
The accused agreed to pay a R1,2 million fine to the DFFE and R1,2 to Upper Highway Air NPC.
Imposing such additional sentences or penalties may also raise questions regarding just desert and proportionality.However, the above cases were concluded through section 105A agreements where the accused agreed to these additional sentences.

Similar offences for less serious contraventions
South African environmental law prescribes high maximum penalties with no gradation or distinction between the actual impacts of such offences or their severity.A prosecutor should seek a just, proportionate and appropriate sentence after conviction, and a presiding officer will consider relevant evidence before imposing an appropriate sentence. 190Subjecting serious and less serious offences to similar high penalties could lead to disproportionality.In other words, for the legislature to catch all offences under possible prescribed maximum penalties of R5 million or R10 million, whether serious or not, provides no direction or guidance to those that enforce or impose penalties.Magistrates' courts, that may not regularly deal with cases where their monetary jurisdiction is extended to this extent, may be reluctant to impose high fines (that may be justified in the context of serious environmental offences In S v E Adams, 192 the accused pleaded guilty to 22 charges all related to waste management activities, the failure to comply with licence conditions and compliance notices; and norms and standards requirements (regarding access control, signage, failure to store waste in covered containers, or exceeding capacity requirements).The maximum potential statutory penalties under the Waste Act for fourteen of the charges 193 is R10 million, and for the other eight, 194 it is R5 million.Similarly, it appears that three of the charges related to contraventions with less or no direct environmental impacts, for example, a failure to comply with norms and standards requiring access control or signage.In comparison, the other offences have the potential to have severe environmental impacts, yet the same maximum penalty could be applied.
In S v Samancor Manganese, 195 the accused was charged with seventeen charges regarding the failure to comply with seventeen conditions across three waste management licences and failing to comply with a condition of its atmospheric emissions licence.The contravention of each different licence condition constitutes separate criminal acts, but the impact of each should be considered as some of these could be administrative noncompliances without direct environmental impacts.For example, not informing the authorities of changes in contact details is unlikely to have environmental impacts and a lower penalty would be more appropriate.

Discussion and conclusion
This paper set out to establish trends from the concluded environmental prosecutions in South Africa, and found that most of the cases resulted in convictions, half were concluded through section 105A agreements, half involved the conviction of individuals, and no direct imprisonment penalties were imposed.Comparatively low fines (compared to the possible maximum prescribed statutory penalties) were imposed in most cases.In about a third, the mechanisms or consequences contemplated in section 34 of the NEMA were invoked.There is some inconsistency in how different listed activities or water uses are treated as separate or consolidated criminal charges.The exact number, outcome, or trends arising from such cases are difficult to determine as there is no central, readily accessible database of concluded prosecutions.
The foregoing analysis informs the question of whether the criminal sanction is a 'reasonable measure' for purposes of realising section 24 of the Constitution.In revisiting the questions a prospective offender may ask themselves in considering the risk of facing an environmental prosecution (as posed in the introduction), it is clear that such prosecutions occur, more than we may be aware of (considering that 53 cases were gathered as part of this research) and 98 per cent of these cases resulted in convictions.The risk of being caught and successfully prosecuted may not be as low or unlikely as may have been perceived before knowing of these 53 cases.Although the sentences ultimately imposed may not result in direct imprisonment (at least not for first offenders), many individuals have obtained criminal records in these cases (in 26 of the 53 cases).Nevertheless, awareness of these prosecutions is a key to attaining improved protection of the section 24 environmental right.
The trend to conclude environmental prosecutions through section 105A agreements 196 appears to have benefits, including a quicker resolution of a case and some certainty over the outcome.A further benefit is that authorities can use it to negotiate funds earmarked for specific purposes, as contemplated in section 34 of NEMA, i.e., to be used for remediation costs, and investigation costs, to support the performance of the functions of EMIs.
For large organisations, the penalties imposed may not break the bank, but for smaller enterprises, it may be more of a punishmentconsidering For the section 24 constitutionally protected environmental right to be realised using criminal sanctions, environmental offences must be met with retributive action, punishing offenders proportionally to the harm caused, deterring offenders and others from committing such offences.Proportionality is influenced by the nature of the accused (corporate entity or an individual) and the nature of the transgression (serious or administrative).To be reasonable, criminal sanctions should distinguish between serious and less serious offences depending on the actual environmental impacts.Prescribing the same potential maximum penalty for all offences (or offenders) may lead to disproportional sentences or unequal enforcement. 197There must be consistency in prosecutorial approachessuch as with listed activity treatment and depending on the underlying facts.Of greater importance is that the harm to the environment is repaired and rehabilitated.These are considerations that should guide criminal enforcement and the creation of criminal sanctions to achieve the ultimate aim of realising the constitutionally protected environmental right.
The research has also highlighted the need for a publicly accessible database with detailed case information.The lack of such a database undermines the deterrent objective of criminal law and arguably impairs the efficacy of law enforcement efforts.
Knowledge of practical challenges in the application of the law (through case outcomes) will positively influence legislative drafting, thereby promoting more practical and accurate laws.Key role-players (regulators, enforcement authorities, prosecutors, magistrates, defence counsel, the regulated community, and the general public) may learn from how others have applied the law to facts, and become more familiar with environmental laws, which will develop and enhance its application.
The accessibility of such cases will broaden general awareness (enhancing its deterrent value).Justice would be more visible to the general public and would be "seen to be done".successful prosecutions will increase deterrence and contribute to the improved implementation and the effective use of the criminal sanction through prosecutions.
The benefits of such a database will ultimately extend to a better-protected environment.

Bibliography
and (2) of the Waste Act.102 Section 52(1) of the Air Quality Act.103 In terms of the Adjustment of Fines Act 101 of 1999 s 1(1)(a) read with the Magistrates' Court Act 32 of 1944 s 92(1)(b), a five-year imprisonment penalty M STRYDOM & T FIELD PER / PELJ 2023(26) 15 from 2008 30 to 2020 31 were considered as a National Heritage Resources Act 25 of 1999, the National Environmental Management: Biodiversity Act 10 of 2004, the National Environmental Management: Protected Areas Act 57 of 2003, the National Forests Act 84 of 1998, the Water Services Act 108 of 1997, the Hazardous Substances Act 15 of 1973, provincial laws and municipal by-laws.
25 In terms of s 105A of the Criminal Procedure Act 51 of 1977 (the CPA).26 Defined in the NECER as the "pollution, waste and EIA (brown) sub-sector".27 This may include a NEMA environmental authorisation, a Waste Act waste management licence, an Air Quality Act atmospheric emissions licence, a National Water Act water use licence or registrations.28 Including the 29 As it is now known, and previously known as the Department of Environment, Forestry and Fisheries and before that the Department of Environmental Affairs.30 The first publicly available report.31 This is the last report considered for this research.
record of the proceedings and details of the judgment.A limited number of outcomes in prosecutions relevant to this research are accessible online.For example, section 105A agreements, judgments, or transcripts have been published by the Centre for Environmental Rights for eleven cases concluded in Magistrates' Courts.42Theoutcomes of prosecutions in Magistrates' Courts are not reported in a central academic or public database (e.g., in case law reports) and are not accessible on research platforms.Concluded case information is public unless otherwise directed by a Court, 43 but it was impossible to visit each Magistrates' Court in South Africa to source cases due to time and resource constraints.High Court judgments relating to environmental matters are generally accessible, particularly when the judgment is marked reportable, 44 but prosecutions rarely take place in these courts.Generally, the High Court becomes involved where a stay of prosecution in the lower court is sought, 45 or where the outcome of a prosecution or warrant, or the seizure of items or confiscation orders are challenged.
32Strydom Use and Impact of Criminal Sanctions, subject to ethics clearance(protocol number H18/11/31, 16 November 2018).33Thepurpose of the research was not to contact each official that may possibly have access to case information for concluded prosecutions.Considering the magnitude of officials tasked with enforcement, this was impossible from a time and cost perspective, and also not within the bounds of the underlying research.This case collection is thus limited to what was received.34SvAcker (Hermanus Regional Court) (unreported) case number ECH 100/05 of 1 October 2005 (hereafter S v Acker).35Forexample, the Sappi Ngodwana case (Mpumalanga) (unreported) SH 158/190 (date unknown) referred to by Kidd 2004 SAJELP 53.36 Proc R1 in GG 27161 of 6 January 2005.37Aspart of the underlying PhD research.40NEMAs34H(1).41RulesRegulating the Conduct of the Proceedings of the Magistrates' Courts of South Africa, published in GN R740 in GG 33487 of 23 August 2010.46Theonlyknown environmental prosecution in the High Court, relevant to this research, 47 was the private prosecution in Uzani Environmental Advocacy v BPSA.48

Table 2
below provides a summary of how the 53 cases were concluded.

Table 3 : Comparative high percentage between prescribed penalties and sentences No Case name
180 Atmospheric Pollution Prevention Act 45 of 1965.181

Table 4 : Illustration of different treatment of activities requiring approval Separate criminal charges pursued Consolidated charges pursued
184S v T Mosekidi, presumably because the accused could not afford to pay a fine and the court may have deemed a wholly suspended sentence insufficient punishment.

Table 4 : Illustration of different treatment of activities requiring approval Separate criminal charges pursued Consolidated charges pursued
186Sections 20 and 67(1)(a) of the Waste Act.187 Sections 22 and 151(1)(a) of the National Water Act.

Table 5 : Examples where NEMA section 34 was used to impose additional penalties
if all offences are tarred with the same brush.If there is no gradation in the law as to what offences are considered more serious and no gradation or guidance it may result in absolute inconsistency in fines or sentences imposed, specifically as there is no central database where these case outcomes are recorded.
the cases (81 per cent) resulted in comparatively low sentences/fines.Only five of the cases considered resulted in comparatively high sentences.As courts consider appropriate sentences and factors relevant thereto, natural persons received comparatively extremely low sentences (in fourteen cases), equal to or below one per cent, indicating that courts' discretion may temper disproportionality as they do not impose fines commensurate with the maximum prescribed penalties.The more common imposition of comparatively low sentences may weaken the retributive and deterrence value of the criminal sanction.
196NPA 2020 https://www.npa.gov.za/sites/default/files/media-releases/ Burns and Kidd "Administrative Law" Burns Y and Kidd M "Administrative Law and Implementation of Environmental Law" in King ND and Strydom HA (eds) Fuggle and Rabie's Environmental Management in South Africa 2 nd ed (Juta Cape Town 2009) 222-268 Cameron 2020 SALJ Cameron E "The Crisis of the Criminal Justice System" 2020 SALJ 32-71 Craigie, Snijman and Fourie "Dissecting Environmental Compliance and Enforcement" Craigie F, Snijman P and Fourie M "Dissecting Environmental Compliance and Enforcement" in Paterson A and Kotzé L (eds) Environmental Compliance and Enforcement in South Africa: Legal Perspectives (Juta Cape Town 2009) 41-61 Craigie, Snijman and Fourie "Environmental Compliance and Enforcement Institutions" Craigie F, Snijman P and Fourie M "Environmental Compliance and Enforcement Institutions" in Paterson A and Kotzé L (eds) Environmental Compliance and Enforcement in South Africa: Legal Perspectives (Juta Cape Town 2009) 65-102 Duff et al Boundaries of the Criminal Law Duff RA et al The Boundaries of the Criminal Law (Oxford University Press Oxford 2010) Fourie 2009 SAJELP Fourie M "How Civil and Administrative Penalties Can Change the Face of Environmental Compliance in South Africa" 2009 SAJELP 93-118 S v Anker Coal (Ermelo Regional Court) (unreported) case number ESH 8/11 of 3 April 2012 S v Arbac Services CC (Germiston Magistrates' Court) (unreported) (unknown case number) (date unknown) S v SS Ngcobo, ML Serite and Iseewaste (Pty) Ltd (Kempton Park Regional Court) (unreported) unknown case number (date unknown)