Policing is essential to South Africa’s response to gender-based violence, but police action begins after violence has been reported or detected. It does not by itself address the conditions in which abuse develops, the reasons survivors may not report it, the availability of healthcare and shelter, economic dependence, social norms, education, or whether institutions can intervene before violence becomes lethal. South Africa’s National Strategic Plan on Gender-Based Violence and Femicide therefore treats justice and safety as one part of a six-pillar response that also includes prevention, survivor support, economic power, accountability and research.
The scale of the problem illustrates why a single institution cannot carry the entire burden. The Human Sciences Research Council’s first national GBV study found that 35.5% of women aged 18 and older had experienced physical and/or sexual violence during their lifetime, while 7.0% reported experiencing physical and/or sexual violence in the previous 12 months. Those figures describe experiences across the population, not simply crimes reported to police. A response focused mainly on arrests, prosecutions and convictions can therefore address an important part of the crisis without addressing the larger conditions that allow violence to occur and recur.
Policing is necessary, but it enters the problem at a particular point
Police have responsibilities that cannot be substituted by schools, hospitals, social workers or community organisations.
They investigate allegations, respond to emergencies, gather evidence, arrest suspects where legally justified, protect victims and witnesses within their mandate, and work with prosecutors and courts.
Effective policing can therefore make a direct difference to safety and accountability.
But policing has an inherent limitation as a prevention strategy.
An officer normally becomes involved after a crime has been reported, a threat has been detected or an emergency has occurred.
The causes of GBV can begin much earlier.
They can involve coercive behaviour, controlling relationships, harmful gender norms, economic dependence, exposure to violence during childhood, social isolation and failures across multiple institutions.
A police service cannot independently change all of those conditions.
South Africa’s government has itself acknowledged this distinction. In its 2026 police budget speech, the acting police minister stated that policing alone cannot solve crime and linked violence prevention to social and economic conditions, safer schools, youth development, substance-abuse prevention, urban planning and community partnerships.
For a broader explanation of why the problem extends beyond individual crimes, see Femicide vs Gender-Based Violence: What Is the Difference?. The distinction matters because femicide describes a specific form of lethal violence against women, while GBV encompasses a much wider range of violence and abuse.
GBV is broader than the cases that reach a police station
One of the most important distinctions in understanding the crisis is between violence that occurs and violence that enters the criminal justice system.
Not every survivor reports abuse.
A survivor may fear retaliation.
She may depend financially on the perpetrator.
She may have children with the perpetrator.
She may distrust the authorities.
She may believe nothing will change.
She may live far from services.
She may not recognise certain forms of coercive or psychological abuse as violence.
Or she may simply decide that reporting is too dangerous.
The national HSRC study is important partly because it attempts to measure experiences of violence at population level rather than relying solely on police statistics. It examined physical, sexual, emotional and economic violence as well as psychological abuse and controlling behaviour.
That broader measurement matters.
A police statistic can tell policymakers about cases entering the criminal justice system.
A prevalence survey can reveal violence that never becomes a police case.
Both types of information are necessary, but they answer different questions.
Justice matters because prevention also requires consequences
The argument for a broader response should not be misunderstood as an argument against stronger policing or prosecution.
Where crimes occur, the state has a responsibility to investigate them and pursue lawful accountability.
A weak criminal justice response can itself undermine prevention.
If perpetrators believe that reporting will produce little action, deterrence may be weakened.
If survivors cannot obtain protection, they may become less willing to seek help.
If cases are delayed for long periods, confidence in the system can suffer.
South Africa’s National Strategic Plan therefore places Justice, Safety and Protection alongside prevention, survivor support and other pillars rather than treating justice as irrelevant.
The practical question is not whether policing matters.
It is what policing can accomplish and what other institutions must accomplish alongside it.
For a closer look at what meaningful progress would need to look like, What Would Success Against Femicide Actually Look Like? The Measures That Matter Beyond Speeches examines the difference between public commitments and measurable outcomes.
Healthcare is part of the GBV response
Healthcare workers can encounter survivors before police do.
A woman may arrive at an emergency department with injuries.
A child may disclose sexual abuse to a healthcare professional.
A survivor may seek treatment without initially wanting to make a criminal complaint.
A health facility can therefore become an important point for identification, treatment, referral and support.
Healthcare also deals with consequences that policing cannot resolve.
These can include physical injuries, sexual and reproductive health needs, trauma and longer-term psychological effects.
South Africa’s national GBVF framework explicitly includes response, care, support and healing as one of its pillars.
This matters because surviving violence is not the same as receiving justice.
A person may obtain a successful criminal conviction and still need medical treatment, counselling, housing or economic assistance.
Shelters can create a practical route out of danger
Telling someone to leave an abusive relationship is not always a realistic safety plan.
Leaving can mean losing housing.
It can mean losing income.
It can create childcare problems.
It can provoke retaliation.
It can leave a survivor isolated from family and community.
For some survivors, safe accommodation is therefore not an optional welfare service. It can be part of immediate violence prevention.
South Africa’s government has linked the country’s multisectoral GBV response to expanded access to shelters, safe spaces, psychosocial counselling and community-based prevention.
A police officer can respond to a complaint, but cannot by themselves create affordable long-term accommodation for every survivor who needs it.
That requires social services, housing systems, community organisations and funding.
Economic conditions can affect a survivor’s options
Economic dependence does not cause GBV, and poverty does not make violence inevitable.
But financial circumstances can affect whether someone can safely leave an abusive situation.
A person without independent income may have fewer practical options.
A person who depends on a partner for housing may face a different set of risks from someone with secure accommodation.
Unemployment can also increase household and community vulnerability without being a direct explanation for individual acts of violence.
South Africa’s own policy framework recognises the relationship between GBVF and wider economic conditions. The National Strategic Plan includes Economic Power as one of its six pillars, while the government has described GBVF as connected to poverty, unemployment and inequality.
The economic dimension is substantial enough to warrant separate attention. The Economic Cost of Gender-Based Violence in South Africa examines the wider financial consequences of violence for survivors, families, institutions and the economy.
That does not mean economic programmes should replace criminal justice.
It means economic security can form part of the prevention environment.
Education can influence behaviour before the police are involved
Schools cannot arrest perpetrators or prosecute crimes.
Their contribution occurs much earlier.
Schools are places where children and young people develop ideas about relationships, boundaries, conflict, consent, gender and acceptable behaviour.
Evidence-based prevention programmes can therefore address attitudes and behaviours before they become entrenched.
South Africa’s National Strategic Plan specifically calls for evidence-based prevention programmes and changes in social norms and behaviour.
But education has limits.
A lesson about respectful relationships cannot compensate for a dysfunctional justice system.
Nor can schools independently change unemployment, family instability, substance abuse or inadequate social services.
Education is therefore one component of prevention, not a standalone solution.
The difficulty of changing behaviour through education and messaging is examined more closely in Why Changing Attitudes About Women Is Harder Than Changing Laws in South Africa.
Families and communities influence what happens before formal intervention
Much GBV occurs in relationships and settings where other people may notice warning signs.
Family members may know about repeated abuse.
Friends may hear threats.
Neighbours may hear violence.
Colleagues may notice controlling behaviour.
Community leaders may receive requests for help.
Faith organisations may become involved.
That means prevention cannot depend entirely on formal institutions.
South Africa’s current GBVF framework explicitly includes social mobilisation and community participation as part of the response. The government’s 2025 Programme of Action called for prevention, social cohesion and engagement with men and boys to challenge harmful norms.
The challenge is to ensure that community involvement supports survivors rather than encouraging informal responses that place them at greater risk.
Gender norms matter, but changing them takes more than messaging
The causes of GBV are not reducible to one attitude or one cultural practice.
But ideas about gender, power, entitlement and acceptable behaviour can influence relationships.
The national GBV study examined gender norms alongside perpetration and victimisation, reflecting the importance researchers place on understanding how attitudes and behaviours interact.
Changing those norms is difficult because people do not form their beliefs in classrooms alone.
They encounter them through families, peers, workplaces, entertainment, religious communities, social media and intimate relationships.
That is why prevention programmes need to operate across multiple settings.
It also explains why changing a law can be much faster than changing behaviour.
Social services can address problems that policing cannot
A survivor may need a social worker rather than a police officer at a particular moment.
She may need help securing shelter, accessing benefits, arranging child protection, obtaining psychosocial support or navigating several government services.
Social workers can also help identify vulnerabilities that do not appear in police records.
The government has increasingly framed GBVF as requiring coordinated action across justice, health, social development and education.
This reflects a basic reality: survivors do not experience institutions separately.
A person may need several of them at the same time.
The institutions do different jobs
| Part of the response | What it can contribute | What it cannot do alone |
|---|---|---|
| Police | Emergency response, investigation, evidence gathering and protection functions | Prevent every act of violence or address underlying social conditions |
| Courts and prosecutors | Lawful adjudication, prosecution and legal accountability | Provide all health, housing or economic support a survivor may need |
| Healthcare | Treatment, identification, referral and recovery support | Investigate crimes or determine criminal guilt |
| Shelters and social services | Safety, accommodation, counselling, referrals and practical support | Replace criminal investigation or long-term economic policy |
| Schools | Prevention education, relationship skills and early identification | Resolve adult criminal behaviour or household poverty |
| Economic and employment systems | Increase financial security and reduce some barriers to leaving dangerous situations | Directly prevent every violent act |
| Communities and civil society | Awareness, support, referral and local prevention | Substitute for state responsibilities |
| Government coordination | Funding, policy, standards, monitoring and accountability | Guarantee implementation without capable institutions and oversight |
The important point is that these functions are complementary.
A gap in one can undermine the work of another.
A survivor can encounter several systems in one case
Consider a hypothetical case.
A woman experiences repeated abuse from a partner.
A friend encourages her to seek help.
She visits a clinic.
A healthcare worker identifies injuries and refers her to support.
She contacts the police.
A criminal complaint is opened.
She needs somewhere safe to stay.
A shelter becomes involved.
A protection application proceeds through the legal system.
A social worker helps her access longer-term support.
If any link fails, the overall response can weaken.
The police cannot solve the accommodation problem.
The shelter cannot prosecute the alleged offender.
The clinic cannot issue a criminal conviction.
The court cannot provide every form of psychological support.
The system works only when the institutions connect.
Coordination is therefore as important as capacity
South Africa’s GBV strategy has long recognised coordination as a separate requirement.
The first pillar of the National Strategic Plan is Accountability, Coordination and Leadership, while the remaining pillars distribute responsibilities across justice, prevention, support, economic power and research.
That design reflects an important principle.
Having many programmes does not automatically create an effective system.
If departments operate independently, information can be lost.
If funding is fragmented, services can become uneven.
If responsibilities are unclear, institutions can pass problems between one another.
If there is no monitoring, implementation can diverge from policy.
More laws do not automatically mean more protection
South Africa has strengthened its legal framework on GBV.
But legislation requires institutions capable of implementing it.
That includes trained personnel, functioning procedures, accessible services, evidence collection, prosecutors, courts, monitoring and resources.
The government’s current policy documents recognise this implementation problem.
This distinction is crucial.
A law establishes a rule.
An institution makes the rule usable.
The same principle applies to femicide policy. A government can establish strategies, review mechanisms and commitments, but their significance ultimately depends on whether institutions can identify risks, respond to violence and learn from cases.
Prevention has to operate before a crisis
A system that only responds once a woman has been assaulted is fundamentally reactive.
Prevention asks different questions.
Can children learn healthier ways of resolving conflict?
Can communities identify escalating abuse?
Can survivors access support before violence becomes severe?
Can risk information be acted upon?
Can perpetrators be held accountable before violence escalates?
Can economic and social services reduce barriers to safety?
Can institutions learn from previous cases?
These questions do not eliminate the need for policing.
They extend the response further upstream.
Data can connect the different parts of the system
Better data is another reason policing cannot operate alone.
A police database may show reported offences.
Health records may show injuries.
Courts may record protection orders and criminal proceedings.
Social services may hold information about shelter or support.
Research can reveal patterns that none of these individual datasets captures.
South Africa’s National Strategic Plan therefore includes Research and Information Management as a dedicated pillar.
The country’s Femicide Review Pilot also reflects this approach by examining the circumstances surrounding killings and looking for systemic lessons.
The purpose is not simply to count victims.
It is to identify recurring opportunities for prevention.
For a deeper examination of this distinction, Why Is Femicide So High in South Africa? The Factors Behind a Persistent Crisis looks at the wider factors associated with the country’s persistent violence against women.
The scale of violence makes a whole-system approach unavoidable
The HSRC’s national study estimated that 7.31 million women aged 18 and older had experienced physical violence during their lifetime, while 2.15 million had experienced sexual violence. The figures are estimates from a national prevalence study, not counts of police cases.
That distinction matters.
A problem of this scale cannot plausibly be addressed only through the institutions that respond after a crime is reported.
It requires prevention, support and institutional change alongside criminal justice.
Economic empowerment is part of the prevention architecture
Economic empowerment is sometimes treated as separate from GBV policy.
But the relationship becomes clearer when prevention is viewed through the choices available to survivors.
A person with independent income, safe housing and access to support may have options that another survivor does not.
That does not guarantee safety.
Nor does poverty explain why an individual perpetrates violence.
The point is narrower: economic security can affect the practical feasibility of escaping dangerous circumstances.
South Africa’s government continues to identify women’s economic empowerment as part of its GBVF response, including training, financial support and preferential procurement.
The role of men and boys is also part of prevention
GBV prevention cannot focus only on supporting women after violence occurs.
Because much violence against women is perpetrated by men, prevention also needs to address men’s behaviour.
South Africa’s Programme of Action explicitly calls for greater engagement with men and boys to challenge harmful norms and toxic masculinity.
That does not mean treating men collectively as perpetrators.
It means developing interventions that address behaviour, relationships, conflict, consent, accountability and attitudes among those who may become perpetrators or witnesses to violence.
Prevention programmes need evidence, not slogans
The existence of a programme does not demonstrate that it works.
A school curriculum, community campaign or men’s dialogue programme should be evaluated against measurable outcomes.
Those might include changes in knowledge, attitudes, behaviour, reporting, repeat violence or other appropriate indicators depending on the intervention.
South Africa’s National Strategic Plan specifically calls for evidence-based prevention programmes.
This is important because attractive campaigns can generate visibility without producing measurable reductions in violence.
The challenge is implementation across unequal communities
South Africa is not a single service environment.
Urban and rural communities can have different access to police stations, courts, healthcare, shelters, public transport and social services.
A national policy may therefore exist on paper while its practical availability varies significantly.
This makes implementation a central part of the problem.
The government has acknowledged continuing structural and implementation challenges alongside the country’s policy and institutional progress.
The police are one part of a chain
It is tempting to ask whether the police are doing enough.
That is a legitimate question.
But it is incomplete if the wider objective is to reduce GBV.
The more useful question is whether the entire prevention and response chain works.
A police officer may respond correctly to a complaint, yet a survivor may still lack somewhere safe to go.
A court may issue an order, yet enforcement may fail.
A hospital may treat an injury, yet the underlying abuse may continue.
A school may teach prevention, yet a young person may return to a violent environment.
A social worker may identify danger, yet resources may be insufficient.
A government may adopt a strategy, yet implementation may remain uneven.
Each is a different point in the system.
What a genuinely comprehensive response would measure
A whole-system response should not be judged only by arrests.
Relevant measures can include:
- whether survivors can report safely;
- how quickly serious threats receive attention;
- whether protection mechanisms are accessible and enforced;
- whether survivors can obtain medical care;
- whether shelters are available where needed;
- whether cases progress through the justice system;
- whether repeat offending is identified;
- whether prevention programmes change behaviour;
- whether survivors can access economic support;
- whether institutions share relevant information appropriately;
- whether rural and underserved communities can access services;
- and whether recommendations from reviews and inquiries are implemented.
No single measure captures the whole response.
The question is whether the system learns from killings
Femicide makes the weaknesses of fragmented systems particularly visible.
When a woman is killed, the most immediate question is what happened to her.
But prevention requires additional questions.
Were there previous reports?
Were threats documented?
Had she sought help?
Was there a protection order?
Did institutions have information that could have changed the response?
Were warning signs recognised?
Were previous cases handled in ways that provide lessons for future cases?
These questions are central to the value of femicide reviews.
A case should not become only a statistic or an isolated criminal investigation. Where appropriate and legally possible, it can also become a source of institutional learning.
For a closer look at how this broader measure of progress can be understood, What Would Success Against Femicide Actually Look Like? The Measures That Matter Beyond Speeches examines what should be measured beyond official statements and public commitments.
Why the distinction between femicide and GBV matters
The broader GBV system cannot be designed around killings alone.
Femicide is the most lethal end of a wider spectrum.
Women may experience harassment, coercive control, sexual violence, physical assault, economic abuse, stalking and other forms of harm without being killed.
Those experiences can still have serious consequences and can sometimes form part of escalating patterns.
Understanding that continuum is important for prevention.
Waiting until violence becomes life-threatening means intervention has already come late.
The distinction is explored further in Femicide vs Gender-Based Violence: What Is the Difference?.
The challenge is not choosing between policing and prevention
South Africa does not have to choose between stronger policing and broader prevention.
It needs both.
Police investigations are essential when crimes occur.
Courts are essential for lawful accountability.
Healthcare is essential for treatment and recovery.
Shelters can be essential for immediate safety.
Social services can help survivors navigate practical barriers.
Schools can contribute to prevention.
Economic policy can affect people’s options.
Communities can support early intervention.
Government has to coordinate, fund, monitor and correct the system.
The central problem is therefore not that South Africa has too much policing.
It is that policing cannot perform the functions of every other institution.
The harder test is what happens before the next emergency
South Africa has developed policies, institutions and strategies specifically because GBV cannot be treated as an ordinary crime problem.
The National Strategic Plan’s six-pillar architecture reflects that understanding.
The continuing challenge is implementation.
For survivors, the system is experienced not as six pillars or separate government departments, but as a sequence of practical encounters.
Can someone get help?
Will the report be taken seriously?
Can she reach healthcare?
Can she find somewhere safe?
Can she obtain legal protection?
Will the case be investigated?
Will institutions communicate?
Can she rebuild an independent life?
And, crucially, can warning signs be recognised before violence becomes fatal?
That is why the most meaningful measure of a GBV response is not simply how many arrests police make. It is whether the institutions surrounding the police can prevent violence earlier, protect survivors when it occurs, hold perpetrators accountable, and learn from cases well enough to reduce the likelihood that the next woman will need an emergency response at all.