Governing in the Shadow of Violence: School Governing Bodies and the Safety Governance Gap in High-Violence Secondary Schools in Gauteng

2.1 School Violence in South African Township Schools: The Governance Context

The prevalence and severity of school violence in South Africa is empirically well documented. National survey data consistently show that South African learners face rates of in-school victimisation substantially higher than comparable populations in other countries (Burton & Leoschut, 2012; Mncube & Harber, 2012). In Gauteng specifically, high schools in township communities record a concentration of violence, weapon carrying, gang activity, and drug-related assaults that reflect the chronic exposure of these communities to structural inequalities, unemployment, and the spatial disadvantages entrenched by apartheid planning (Chitsamatanga & Rembe, 2020; Ncontsa & Shumba, 2013).

Less examined in the empirical literature is the governance dimension of this crisis. Most South African research on school violence focuses on the psychological consequences for learners, the professional challenges for teachers, or the implementation failures of the SIAS policy (Makhalemele & Tlale, 2020; Nel et al., 2016). The SGB’s role in school governance has received comparatively limited systematic attention, despite its explicit statutory mandate. De Wet (2007) provides an early account of the SGB’s approaches to avert violence, including zero-tolerance policies, school-wide interventions, and targeted security measures, but notes that these approaches are frequently uncoordinated and under-resourced. More recent work on democratic school governance (Shaked & Schechter, 2020; Wolhuter & Van der Walt, 2020) emphasises the SGB’s accountability function, and studies of school leadership in South African contexts have shown how local learning environments shape what leaders are able to achieve (Zuze & Juan, 2020), but this body of work does not specifically examine SGB performance in a high-violence township context.

The gap in this literature is significant. If the SGB is the governance authority responsible for school safety, and the law is unambiguous on this point, then understanding how it functions (or fails to function) in the schools most severely affected by violence is a precondition for designing effective governance reform.

2.2 Legislative Architecture of School Safety

South Africa’s legislative framework governing school safety is extensive and, in formal terms, robust. The South African Constitution (RSA, 1996a) provides the foundational guarantee: every learner has the right to basic education, and, by extension, to the conditions that make education possible, including a safe learning environment. The South African Schools Act 84 of 1996 (RSA, 1996b), specifically Section 8, mandates the SGB of every public school to adopt a Code of Conduct for learners, enforceable through a graduated disciplinary process. The Regulations for Safety Measures at All Public Schools designate school premises as dangerous-weapon-free and drug-free zones, granting school principals specific search power and requiring protocols for confiscation and reporting.

Beyond SASA, multiple legislative instruments reinforce the school safety mandate. The Child Justice Act 75 of 2008 addresses the treatment of learners who commit violent crimes on school premises. The Children’s Act 38 of 2005 establishes child protection obligations that extend to school settings. The Policy Framework for the Management of Drug Abuse and its operational guidelines require schools to develop school-specific substance management policies, establish Learner Support Teams, and coordinate with health and social development departments. The African Charter on the Rights and Welfare of the Child (Organization of African Unity [OAU], 1990) and the Convention on the Rights of the Child (United Nations [UN], 1989) provide the International human rights architecture within which these local provisions operate.

The National Development Plan (NDP) identifies school safety as an integral part of South Africa’s educational development trajectory, and the DBE’s suite of safety-related policy documents, including the School Safety Framework and the values-focused governance manuals issued to SGBs and Representative Councils of Learners, provides operational guidance. The Integrated School Health Policy (ISHP) creates a coordinated health response structure that, in principle, includes mental health and trauma support for violence-exposed learners.

This legislative machinery is, on its face, comprehensive. The critical question examined in this study is whether it translates into operational governance practice in the schools that most need it.

2.3 The Whole-School Approach to Safety: Policy and Reality

South African policy documents, including the DBE’s School Safety Framework, consistently endorse a whole-school approach to violence prevention and safety management. A whole-school approach conceptualises safety governance, curriculum, physical environment, and community responsibility, not merely a disciplinary or care matter. Nyoni et al. (2023) identify the whole-school approach as an established benchmark of effective school-based violence intervention, precisely because it addresses the systemic conditions that produce violence rather than individual manifestations.

The international literature on whole-school safety approaches provides a consistent evidence base. Effective implementations are characterised by: clear, communicated safety policies with learner and parent input; physical environment audits that identify and address vulnerability hotspots; trained and coordinated safety teams with defined mandates; functioning relationships with external agencies, including police, social development, and health; and regular monitoring and evaluation of safety outcomes (Astor & Benbenishty, 2018; Bradshaw et al., 2021). What distinguishes effective from ineffective whole-school approaches is not the content of the policy framework but the quality of implementation governance, whether the SGB, principal, and management team are actively driving compliance and accountability.

In South African township schools, the evidence suggests that these governance conditions are frequently absent. Safety committees exist on paper, but do not function as intended; learner and parent participation in safety governance is nominal; and the critical connection between the school-level governance (SGB, SMT) and the school-level support structure (SBST) for violence-exposed learners is frequently missing (Cele & Mncube, 2022; Khanyile & Mpuangnan, 2024).

2.4 Family as a Constrained Partner in Safety Governance

The whole-school approach explicitly encompasses the family. South African policy, including Education White Paper 6 (DoE, 2001), the SIAS framework (DBE, 2014), and the NDP, positions parents as equal partners in learner support and school governance. The SGB, as a structure that includes elected parents’ representatives, is formally a vehicle for parental voice in school safety decisions. In practice, however, parents’ participation in school governance in low-income township settings is frequently compromised by structural barriers, irregular working hours, transport costs, limited literacy, and a legacy of institutional exclusion that prevents meaningful engagements (Singh et al., 2020; UNESCO, 2009).

More fundamentally, the family-as-a-safety-partner model assumes that families can provide home-based complementary support that reinforces school-level safety interventions. In high-violence township communities, this assumption is structurally compromised. Single parenthood, child-headed households, households’ exposure to violence, and the economic pressures of poverty collectively erode families’ capacity to function as active partners in their children’s safety (Danziger & Ratner, 2010). The resulting burden is displaced back onto schools and their governance structures, which are themselves often ill-equipped to bear it.

4.1 Research Design and Paradigm

A qualitative multiple case study design was employed, grounded in an interpretivist paradigm (Bryman, 2016; Creswell & Poth, 2018). A case study design was selected because the study’s interest is in the situated governance experience of principals and teachers across five bounded school settings, how they perceive, enact, and are constrained by the safety governance architecture surrounding them, and because the design accommodates the multiple data sources (interviews, focus groups, documents, and observation) used to build a rich account of each case. The interpretivist paradigm recognises that governance reality is constituted through situated meaning-making, and that policy-as-written and policy-as-lived are frequently contradictory.

4.2 Sites and Participants

Five public secondary schools in Gauteng Province were purposively selected. The schools are located in Diepkloof, Pimville, Westbury, Randburg, and Diepsloot. These areas vary in their socio-economic profile, but each school draws a substantial proportion of its learners from surrounding low-income and historically disadvantaged communities marked by high exposure to community violence and the spatial inequalities entrenched under apartheid. The school located in the Randburg area, although situated in a more economically mixed locality, serves a learner population drawn largely from adjacent informal settlements and low-income areas and was included on the same high-violence criterion as the other sites. Where the term ‘township schools’ is used in this article, it is applied in a broad sense, denoting schools whose learner populations are drawn predominantly from historically marginalised, low-income communities with high exposure to violence rather than as a strict geographic descriptor of each school’s immediate locality. Schools were selected on the basis of documented exposure to serious safety incidents, including weapon-carrying, drug-related assaults, and physical violence; recorded in their incident books over the preceding two years. Each school operates within the same formal legislative and policy architecture. Schools are identified as A through E to protect anonymity.

Thirty teachers (six per school) participated in focus group interviews, drawn from School-Based Support Teams, safety committees, Departmental Heads, and Life Orientation teachers. Five principals participated in individual semi-structured interviews. The total sample of 35 participants was selected through purposive sampling (Etikan et al., 2016) based on their direct knowledge of violence incidents and safety governance in their schools.

The profiles of the participating schools and participants are summarised in Table 1 (presented at the end of the manuscript).

Schools are identified by a pseudonym (A–E) to protect anonymity. Participant codes follow the convention PA–PE for principals and TA–TE for teachers.

4.3 Data Generation

Four complementary data sources were used. Focus-group interviews with SBST teacher components at each school lasted 60–90 minutes and were audio recorded and transcribed verbatim. Individual interviews with principals focused on the governance perspective and the SGB’s safety management arrangements. The documentary review covered Codes of Conduct, incident and logbooks, and available safety committee records over a two-year period. Non-participant observation of the school environment, entry and exit routines, and informal governance interactions supplemented the interview and documentary data.

4.4 Data Analysis

Inductive thematic analysis followed the six-phase process outlined by Braun and Clarke (2006): data familiarisation, initial coding, theme collation, theme review, theme naming, and analytic reporting. Analysis was applied across data sources simultaneously, with documentary data used to corroborate or contextualise interview-based findings. Because the documentary review drew on existing school records, the absence of SGB safety records was interpreted with caution: such gaps may indicate non-engagement, but they may equally reflect informal or undocumented practice, uneven record-keeping, or records held elsewhere, and were therefore treated as one indicator among several rather than as conclusive proof of governance inactivity. Three themes were identified, each mapping onto a distinct dimension of the SGB’s safety governance function. This study and its data analysis plan were not preregistered at an independent registry.

4.5 Trustworthiness

The Lincoln and Guba (1985) four-criterion framework was applied throughout. Credibility was supported by purposive sampling, triangulation across interviews, focus groups, documents, and observation, and extensive use of verbatim quotations. Member checking was conducted by returning interview and focus-group transcripts to participants for verification, and by presenting a summary of the preliminary themes to a subset of principals and teachers at each school, who confirmed or refined the interpretations before the analysis was finalised. Transferability was supported by a thick description of the research context. Dependability and confirmability were ensured through a transparent audit trail, regular debriefing, and retention of verbatim data to anchor analytic claims.

Reflexively, the researcher approached this study through the lens of educational psychology, with a prior interest in school governance and learner safety. This orientation shaped the questions posed and the decision to foreground governance structures rather than, for instance, classroom practice. To prevent this lens from unduly steering the analysis, the researcher kept a reflexive journal, engaged in regular debriefing, and repeatedly tested emerging interpretations against participants’ verbatim accounts.

5.1 Theme 1: SGB Disengagement from Safety Governance

Across all five schools, participants described the SGB as, in their experience, functionally absent from the governance of serious safety incidents. No SGB members were interviewed for this study, so what follows is an account of how principals and teachers experienced the SGB’s role, read against what the schools’ own records did and did not contain. Where violence occurred, weapon carrying, drug-related assaults, stabbings, organised intimidation, the default institutional response was to direct the incident to the SBST or to individual teachers rather than to invoke the SGB’s authority or the SMT’s governance responsibility. This routing was experienced by teachers as both inappropriate and exhausting; the SBST’s mandate is support, not safety governance, and the displacement of serious safety incidents into support channels overloads the support structure while leaving the accountability gap in governance unaddressed.

Participants drew a clear distinction between who should manage serious safety incidents and who was actually being asked to manage them. Three accounts capture this pattern with particular clarity:

These learners who cause a lot of violence in schools and bully others should not be referred to the SBST. The principal must deal with them. What must the SBST do? There is nothing they can do. (Participant TD20).

There is so much that the SGB can do. There are corners where learners hide everywhere in the school, and the SGB must deal with that misbehaviour. They are governors, and they must deal with such. (Participant TA5).

The learners bring this bad behaviour they get from home and bring it to school. They bully each other, smoke dagga and nyaope, and bring knives to school. This is criminality and should be dealt with by the SGB. As teachers, we have a lot to deal with. (Participant TE25).

Running through these accounts is a single, consistent distinction: participants placed safety governance with the SGB and the SMT, and learner support with the SBST. The systematic displacement of governance responsibility onto support structures was experienced as a source of both role confusion and institutional overload. Participants were not simply frustrated with workload; they were articulating a structural mismatch between the mandate of the support structure and the nature of the incidents being directed to it. Mestry and Khumalo (2012) describe much the same dynamic, noting that SGBs in South African schools frequently lack the operational preparation to exercise their statutory safety governance function, which leaves accountability to default to school management and teaching staff.

Documentary evidence corroborated these accounts. Incident books and logbooks at all five schools contained records of serious safety incidents; weapon confiscations, drug-related altercations and physical assaults; but with minimal evidence of SGB-level governance responses. There were no records of extraordinary SGB meetings convened to address repeated safety crises, no safety-audit records, and no documented SGB resolutions directing the SMT to implement specific safety protocols in response to identified risks. The SGB’s presence in the documentary record was limited to Code of Conduct approvals and scheduled monthly meetings, with no evidence of the proactive safety governance role mandated by the South African Schools Act 84 of 1996 (RSA, 1996b) and the Regulations for Safety Measures at All Public Schools. This gap between documentary compliance and operational governance echoes the broader literature on policy implementation in resource-constrained South African schools (Engelbrecht et al., 2016; Nel et al., 2016).

Not all participants interpreted the SGB hand-off approach as passive indifference. Several noted that SGBs face their own capacity constraints, including limited training, irregular attendance by elected parent members, and a general unfamiliarity with the operational dimensions of their safety mandate. The principal of School E gave the following account:

We have problems of fights. Sometimes other learners are caught with cakes laced with drugs; others are caught with an assortment of dangerous weapons that they bring to school, allegedly to protect themselves. These drugs are a problem with all these violent acts. Really, it is scary. This is the responsibility of the governors. (Participant PE).

The attribution of responsibility to governors in this account carries analytical weight. The principal was not naming the SBST, individual teachers, or the district as the appropriate accountability structure for this level of safety crises; he was naming the SGB. That the governance structure was not responding to its mandate registered, in this principal’s account, as more than an operational gap; it was experienced as a form of institutional abandonment. Bronfenbrenner’s (1979) ecological framework would predict precisely this pattern: an absent or non-functional exosystem governance function pushes the burden downward, to the mesosystem and microsystem, where teachers and learners absorb consequences that governance should have addressed upstream.

Set against this absence is a countervailing institutional asset. The teachers and School-Based Support Teams to whom safety incidents were routinely redirected continued to absorb and respond to those incidents in the day-to-day life of the school, sustaining a measure of safety practice that the governance structure itself was not providing. Read through the Strength-Based Perspective (Hammond & Zimmerman, 2012; Saleebey, 2008), this frontline response is more than a symptom of an overstretched system; it is a form of existing capacity. The governance vacuum was, in practice, being partially held by committed practitioners whose informal safety work, though largely undocumented and unsupported, is precisely the kind of asset on which a governance-centred reform agenda could build.

5.2 Theme 2: The Gap between Legislative Framework and Operational Practice

All five schools possessed the statutory documents required by law: Codes of Conduct for learners, incident recording systems, and, in most cases, documents liaising with external agencies, including the South African Police Service (SAPS). The presence of these documents suggested formal legislative compliance. The gap between formal compliance and operational functionality, however, was extensive and consistent across all five school sites. Drug and substance abuse was the most consistently cited safety-management challenge, and the one that most clearly illustrated the policy gap. Participants described drug-related incidents; laced foodstuffs, nyaope use in toilet areas, and tuck shops selling controlled substances adjacent to school premises; that were both chronic and structurally embedded. South African law requires schools to be drug-free zones. Schools must have their own substance management policies and work with the Department of Health and Social Development, as set out in the Policy Framework for the management of Drug Abuse, the National Drug Master Plan, and the Integrated School Health Policy. Two participants from School C described the lived reality of this legislative gap:

Drugs are freely available. This is a sickness in our society. The informal tuck shops around the school sell drugs and alcohol to learners. The police were notified and closed them down. Soon, two more mushroomed elsewhere. Honestly, how do we support such learners? It would be better if this school were a drug rehabilitation centre because we clearly can’t cope with this many cases. (Participant TC18).

Drugs and substance abuse at our school is a problem, but it involves less than ten percent of the learner population. We have to take action and prioritise it, as it could have a ripple effect in the school if not nipped in the bud. (Participant TC15).

Two interrelated dimensions of the legislative-practice gap surface in these accounts. The first concerns the inadequacy of enforcement-based responses: police closure of drug-selling tuck shops is followed within weeks by their replacement, demonstrating that a purely enforcement-oriented response to a structurally embedded supply chain is operationally insufficient. The second concerns the fundamental mismatch between the school’s legal designation as a drug-free zone and the operational reality of a school, which is embedded in a community where drug supply chains are persistent, territorial, mobile, and economically sustained. A similar conclusion is reached by Cele and Mncube (2022), who argue that school-related violence in South African townships cannot be effectively addressed through school-level legislative compliance alone. This is because the supply-and-demand conditions that sustain drug use lie in the wider community ecology rather than within the school’s jurisdiction.

Participants also described the legislative framework governing weapon-carrying in terms that foregrounded its enforcement limitations. The Regulations for Safety Measures at All Public Schools designate schools as weapon-free zones and empower principals to conduct searches of learners and their belongings. In practice, weapons were confiscated with high regularity, monthly in three of the five schools, according to document records, without any evidence of a corresponding change in the underlying pattern of weapon carrying. Confiscation, like tuck-shop closure, is a reactive response to a structurally embedded phenomenon: The legislation provides the authority for the response, but not the systemic conditions that would make it effective over time. One principal captured the cumulative experience of this cycle:

We sometimes get tired of these learners. Today it’s drugs, tomorrow it’s bullying, the next day it’s alcohol, and then after smoking all these drugs, they fight. But it’s not many learners, maybe ten percent, and we must get it fixed quickly, though this is tiring. (Participant PE).

This account is analytically significant not only for its content but for its tone. The expression of exhaustion by a school principal, a senior government official, points to a deeper problem than workload. It reflects the accumulated psychological burden of repeated reactive responses to the conditions that the available legislative tools are structurally unable to resolve. Khanyile and Mpuangnan (2024) identify this pattern as a distinguishing feature of high-violence school leadership, where principals operate in a permanent reactive mode rather than within a proactive governance framework that might reduce the frequency and severity of incidents over time.

Documentary review corroborated the reactive nature of the school’s legislative compliance across all five sites. Incident books recorded individual confiscations, suspensions, and referrals to external stakeholders, but contained no evidence of systemic safety audits, pattern analysis across incidents, or SGB-driven governance responses to identified trends. The legislative framework, in other words, was functioning as an incident-level reactive instrument rather than a governance-level proactive architecture. This distinction between reactive compliance and proactive governance is central to understanding why legislative robustness does not translate into operational safety improvement in these school contexts. As Wolhuter and Van der Walt (2020) argue, the effectiveness of school governance legislation in South Africa depends less on the content of the legal framework than on the institutional capacity and accountability mechanisms available to implement it at the school level.

5.3 Theme 3: Non-Operationalisation of the Whole-School Safety Approach

The policy aspiration for a whole-school approach to safety management was consistently recognisable in participants’ understanding of what an effective response to school violence should look like. When asked about effective safety management, participants described an integrated, multi-stakeholder, preventative model. When asked to describe the current model operating in their schools, they described a fragmented, reactive, and individual-teacher-dependent system. The gap between what participants understood a whole-school approach to require and what they experienced in practice was substantial and consistent across all five schools.

The family dimension of the whole-school approach emerged as a specific and recurring point of failure. The legislative framework, including the South African Schools Act 84 of 1996 (RSA, 1996b), the SIAS policy, and the School Safety Framework, positions parents as equal partners in learner safety governance, including through their formal representation on the SGB. In the school communities studied, however, participants described family-level protective capacity as severely compromised by the structural conditions of township life. Working parents with minimal home supervision time, single-parent households stretched by economic stress, and child-headed households managing without adult oversight were identified as the dominant family patterns in these communities. Two participants described the results of this structural reality for the school’s support function:

We cannot take the role of teachers and parents at the same time. When we give support, we expect parents to continue with that kind of support at home. Unfortunately, that’s not happening; that’s why we get exhausted. (Participant TB8).

The parents do not have time to listen and support their children. They wake up in the morning to catch the earliest bus to work and knock off late. They barely see their children to listen to their stories. It is sad. (Participant TE26).

At the heart of the whole-school approach, as it is currently implemented in these schools, sits a structural incoherence. The whole-school model assumes a family partner who is available, present, and able to reinforce school-level safety and support interventions at home. In high-violence township communities characterised by economic marginalisation and limited parental availability, this assumption does not hold. When the family microsystem cannot fulfil its partnership function, additional pressures migrate to the already under-resourced and structurally overstretched school governance system. The same dynamic underlies what Bronfenbrenner (1979) calls risk accumulation, where several ecological systems fail at once and the developmental consequences for the child become qualitatively more severe than those of any single failure in isolation. Singh et al. (2020) similarly note that parent engagement models in South African township schools consistently fail to account for the structural barriers, economic, temporal, and spatial that prevent meaningful family participation in school governance.

The safety committee, envisioned by the DBE’s School Safety Framework as the operational vehicle for school-level safety governance, was described by participants as either nonfunctional or poorly integrated with other school governance structures. Where safety committees existed, their work was not systematically linked to the SGB’s governance decision-making, the SBST’s support work, or the school’s relationship with external stakeholders, including the SAPS and the Department of Social Development. Safety management, therefore, operated as a parallel and disconnected strand of school life rather than as an integrated governance function with clear accountability lines. Astor and Benbenishty (2018) put it sharply, arguing that whole-school safety approaches fail not for want of policy design but for want of a governance spine — the connecting institutional architecture that would link policy intent to operational practice at every level of the school system.

Documentary evidence from logbooks and codes of conduct corroborated the operational disconnect across all five schools. The codes of conduct reviewed articulated clear expectations for learner behaviour and graduated consequences for misconduct, and, in this respect, they satisfied the formal requirements of the South African Schools Act and the Regulations for Safety Measures. However, they contained no reference to the school’s whole-school safety framework, the designated roles and responsibilities of the safety committee or the SGB in safety governance, or the school’s external agency coordination protocols. The documents were legally compliant in that they contained what the law required, but they were operationally thin: they did not reflect a functioning, integrated safety governance architecture. This distinction between documentary compliance and governance functionality is critical. As Nyoni et al. (2023) observe, the presence of policy documents and whole-school structures is a necessary but wholly insufficient condition for effective school safety; what matters is the degree to which those documents are animated by active institutional leadership, clear role accountability, and monitored implementation at every level of the school system.

6.1 The SGB as a missing Governance Structure

The central finding of this study is that the SGB, formally the primary governance authority for school safety in South Africa, is, in practice, largely peripheral to the governance of school violence in the five schools studied. This finding has direct implications for how the school safety problem is framed and addressed. The dominant framing in South African school safety research and policy positions the SBST and DBST as the primary institutional responses to violence and its consequences. This framing is understandable: the SBST is the most visible support structure in the school, and the SBST-DBST partnership is the policy architecture most directly concerned with learner well-being.

This framing obscures an upstream governance problem. The SBST can only support survivors of violence when the SGB and SMT create the right conditions: working safety protocols, an enforceable code of conduct, physical safety measures, and coordinated links to external stakeholders. When the SGB neglects its safety mandate, the SBST is left managing the consequences without the conditions that would prevent them. As this study shows, the routing of serious safety incidents away from the SGB and onto the SBST is a symptom of this governance inversion.

Viewed through Bronfenbrenner’s ecological lens, SGB disengagement is an exosystem failure. The learner does not experience it directly, yet it shapes the safety climate around them. When this governance function is missing, the mesosystem’s protective capacity weakens, teachers carry more, trust erodes, and the learner’s microsystem becomes less predictable. The framework makes clear that fixing SBST dysfunction without addressing SGB disengagement treats the symptom rather than the cause.

6.2 Legal Robustness, Operational Fragility

The second finding, the gap between the legislative framework and its operational implementation, raises a question that extends beyond South Africa. When is legislative robustness sufficient, and when does it create a false sense of policy effectiveness? South Africa’s school safety legislation is, by any comparative standard, comprehensive. Multiple statutory instruments create overlapping obligations, and the constitutional foundation is unambiguous. Yet, across the five schools studied, this legislative architecture produced reactive incident recording rather than proactive safety governance.

The explanation is structural rather than individual. The legislation assigns responsibilities (to the SGB, the principal, the SMT) but does not provide the resources, training, or monitoring mechanisms that would make those responsibilities operationally tractable. SGB members are elected, not trained; they receive minimal statutory preparation for their safety governance role. The Regulations for Safety Measures empower schools to act but do not fund them to create the physical security conditions that would make those powers meaningful. The coordination mandated by the drug abuse framework requires departmental resources that are routinely unavailable in the schools studied.

This pattern, a legal mandate without the resources to deliver it, is well documented in the inclusive education literature (Engelbrecht et al., 2016; Nel et al., 2016) and in the broader South African governance literature (Makhalemele & Tlale, 2020). How it plays out in school safety, though, has been examined less systematically. This study provides empirical grounding for what the literature implies: in South African township schools, the gap between policy aspiration and reality is not primarily a problem of intent but of implementation infrastructure.

6.3 The Whole-school Approach Requires Whole School Governance

The third finding is that the whole-school safety approach was never operationalised. This shows that such an approach is only as effective as the governance behind it. A whole-school safety approach is not mainly a curriculum or pastoral care model; it is a governance model. It requires an SGB that takes its safety mandate seriously, an SMT that turns governance decisions into working protocols, and a safety committee linked to both the SGB and the SBST. Without this governance spine, the approach stays a policy document rather than a living practice.

The family dimension deserves particular attention. This study confirms what the South African literature consistently shows. In high-violence township settings, families face structural conditions that limit their capacity to act as active partners in school safety. An approach that treats family participation as central but offers no scaffolding for it, outreach, accessible communication, translated materials, and flexible meeting times will systematically fail to reach the families who matter most. As the formal vehicle for parent voice, the SGB is the natural site for that scaffolding.

6.4 Towards a Governance-Centred Safety Architecture

The findings of this study point toward a governance-centred safety architecture in which the SGB is repositioned as an active, accountable, and resourced partner in learner protection. This architecture has four interlocking elements.

First, the SGB needs specific preparation for its safety role. The current model, where there are elections without training, produces bodies that are nominally responsible but practically unprepared. A mandatory SGB induction programme, focused on the safety mandate and giving members working knowledge of their statutory powers and responsibilities, is a minimum condition for accountability.

Second, the SGB must be formally connected to the SBST. Safety governance (preventing and managing violence) and learner support (caring for those affected by it) are distinct but interdependent. Without a formal SGB–SBST interface, as this study found absent, governance and support decisions are made in separate silos. An integrated model in which the SBST briefs the SGB on case-level trends and SGB safety decisions, and the SGB, in turn, resources the SBST, would close this gap.

Third, legislative compliance must shift from reactive to proactive. At present, safety regulations are used to respond to incidents rather than to plan for safety, a governance failure with structural roots. District offices must provide principals and SGB members with practical tools for safety audits, pattern analysis across incident records, and regular reviews of safety conditions. The DBE’s existing guidance, the School Safety Framework, the safety signposts manuals, and the drug abuse guidelines should be folded into an annual SGB safety cycle rather than left as stand-alone documents.

Fourth, the family partnership dimension needs active, not nominal, operationalisation. SGB parent representatives should be supported in reaching out to families, communicating safety protocols in plain language, and creating channels for parents to report concerns without fear of stigma or indifference.