CQ No. 25


SA Crime Quarterly no 27 • March 2009 23

violent crime. The June 2008 South African Police
Service national crime statistics3 cemented South
Africa’s reputation as one of the most violent
societies in the world, despite the fact that overall
levels of crime have consistently come down since
2001. In this and other developing countries,
crime exacts a high cost in terms of health and
security and has the potential to scupper the
attempts of governments to fulfil their
responsibilities in addressing poverty and
inequality. 

In this context there is an urgent need to explore,
understand and navigate the continuum between
criminal justice (and the role of the state), crime
prevention, and the struggle to adhere to
international human rights norms and standards.
In doing this we need to address a combination of

The recently published book: Action for a Safe
South Africa offers us a vision of a society in
which:

• Our children play safely in beautiful parks
• Our women walk easily to and from work
• Our public transport system offers safe 

passage to local and foreign tourists
• Our pavement cafes flourish and sound with 

relaxed laughter
• Our townships and suburbs reflect the 

amazing spirit of our people
• Our rural roads offer us the amazing vistas of 

our beautiful country2

A wonderful vision, indeed. In stark contrast,
however, our country still faces serious challenges
from unacceptably high levels of serious and

Finding the right
balance
Immediate safety versus
long-term social change

Louise Ehlers and Sean Tait

Open Society of SA
louise@ct.osf.org.za

South Africa has a reputation as one of the most violent societies in the world, despite the fact that overall
levels of crime have consistently come down since 2001. In this and other developing countries, crime
exacts a high cost in terms of health and security and has the potential to scupper the attempts of
governments to fulfil their responsibilities in addressing poverty and inequality.1 This article argues that
there is an urgent need to develop policies that balance the immediate need for safety and security with
the long-term objective of achieving systemic social change. It describes an intervention undertaken by the
Open Society Foundation for South Africa in collaboration with three provincial departments of Safety
and Security that aims to explore ways in which a dual approach of keeping citizens safe in the context of
current levels of crime, and simultaneously developing models for long term crime reduction, can be
achieved. 

 



24 Institute for Security Studies

theoretical and practical questions in relation to
criminal justice, such as: How do we negotiate the
tension between the project to promote safety and
the project to promote democracy? In other
words, how does one ensure the safety of the
citizenry without a concomitant negative impact
on its right to the free and unfettered enjoyment
of political and civil liberties? What are the main
social and political processes that initiate and
drive change in criminal justice systems? 

Current concerns in South Africa include the
introduction of increasingly repressive criminal
justice legislation; decreasing levels of access to
information held by criminal justice agencies
(notwithstanding enabling legislation offering the
contrary); and the weakening of institutions
established for the purposes of promoting human
rights (e.g. the Human Rights Commission).  

SOUTH AFRICA’S APPROACH TO
MANAGING THE CRIME PROBLEM

Crime prevention4 is recognised internationally as
a key component, along with law enforcement, of
an effective response to promoting safety.5 In
South Africa, however, despite having one of the
highest crime rates6 in the world, our support for
crime prevention has been patchy at best.7 In
contrast, strengthening the criminal justice
response to crime has been the focus of consistent
attention over the 14 years post apartheid. The
criminal justice budget has grown from 
R14 billion in 1995 to R71 billion in 2009.8 South
Africa’s per capita spending on criminal justice
(USD 130) is double the international average.9

While this investment may have contributed to
the rates of recorded crime in most categories
decreasing since 1995 (with murder down from
its high of 67,9 per 100 000 to 38,6 per 100 000 in
2008), this has not ensured a safe, secure and
peaceful environment for citizens. In addition to
being very expensive in monetary terms, the
primary focus on law enforcement as the way to
counter crime and ensure safety has resulted in an
increasingly militarised society. This is
demonstrated by the fact that police numbers are

set to increase (again) from the already high
current level of 183 000 to over 200 000 by 2012,
making South Africa a country with one of the
largest police agencies under single command in
the world. Yet, despite this, South Africans spend
ever greater amounts on private security. It is
estimated that the R14 billion private security
industry employs 300 000 ac tive registered
security guards.10

Besides growing the police service, other
measures have been introduced to counter crime,
such as minimum sentences and restrictive bail
conditions. This has resulted in an extremely high
prison population: with 164 957 inmates our
prisons are hugely overcrowded, with all the
concomitant problems. South Africa’s
incarceration rates are among the highest in the
world. Approximately 350 in e very 100 000 S outh
Africans are in prison.11

In much the same way as the state has responded
to high levels of crime by increasing punitive
measures, citizens too have hardened their
attitudes to crime over the past six years. The
number of South Africans who believe that social
development should be prioritised to address
property crime fell from 62 per cent in the 2003
ISS National Victim Survey to 52 per cent in 2007.
Corroborating this, both the 2003 and 2007
surveys show that because South Africans
increasingly perceive crime to be motivated by
greed rather than real need, their response is to
suggest that there is a need to increase the
punitive response to crime rather than to spend
on social development.12

Yet, despite the massive amount spent on law
enforcement by the state, South Africans do not
feel safer. The 2007 victim survey also confirmed
a steady decline in perceptions of safety since
1998.  Almost 60 per cent of interviewees for the
2007 survey felt crime levels had increased over
the past four years with fewer people reporting
feeling safe walking around their
neighbourhood.13

Jody Kollapen, chairperson of the South African
Human Rights Commission, notes, 

 



SA Crime Quarterly no 27 • March 2009 25

If anything, this [punitive] stance has exacerbated

an over-reliance on criminal justice processes in

responding to the social ills of the nation. A more

balanced approach – one that recognises the need

to protect society but also the necessity to advance

social cohesion and development must be

adopted.14

He goes on to argue that while we need to retain
the capacity to protect society against those who
represent a threat, we should guard against the
tendency to overextend the criminal justice
system and the expectation that it should remedy
problems brought about by poverty and
inequality.15

Similarly, Marc Mauer, director of the Sentencing
Project in Washington DC, argues that one cannot
place crime prevention and law enforcement at
opposite ends of the spectrum in an either/or
situation. He makes the point that the discussion
about how to reduce crime has become polarised
between advocates of long-term solutions versus
those who seek immediate results, and points out
that this is not a useful dichotomy. 

On the one hand, we would be foolish to believe

that we can create a fair and just society without

addressing systemic issues such as poverty and

racism. However, if we only wait until those issues

are resolved, we will fail to meet the urgent needs

of improving safety on a d aily basis. In this regard,

we need to seek models of programmes and

policies that can provide short-term benefit while

building constituencies and approaches for long-

term systemic change.16

The classic example of this polarisation in South
Africa is the demise of the National Crime
Prevention Strategy (NCPS). Drafted in 1996, t he
NCPS explored an approach to addressing crime
and violence that sought to understand and
address the social, economic and historical drivers
behind the crime and violence plaguing the
country. In doing so it created a framework within
the spirit of the Reconstruction and Development
Programme of supporting interventions that
would address the risk and resilience factors
associated with many crime types, ultimately
seeking to prevent them.   

By 1999 the NCPS was for all intents and
purposes, shelved in a s eries of developments that
clearly demonstrated the ascendance of the law
enforcement agenda. These included the
downgrading of the National Secretariat for Safety
and Security, the lead agency in the development
of the strategy, its absorption into the police, and
the rising dominance of the National Crime
Combating Strategy of the SAPS. The non-
implementation of the subsequent 1999 White
Paper on Safety and Security, which gave specific
policy direction to many of the aspects of the
NCPS, and the fact that it has since lapsed,
effectively shut the strategy down in all but name.  

Commenting six years later, Frank refers to a
legacy characterised by lack of clarity in
government about crime prevention, the
devastating impact of a lack of skills and capacity
in expanding the quality and reach of social crime
prevention interventions, and the inability of the
country to make use of key leverage points, in
particular investing in children from an early
age.17

Despite the policy direction that government has
taken since 1994, the discourse on managing
crime has to some extent matured in recent years
and has started to acknowledge the role of both
law enforcement and crime prevention in making
South Africans safe. Networks of civil society
practitioners such as Action for a Safe South
Africa have continued to mobilise around the
prevention agenda, while government has also
developed and facilitated some efforts at
promoting a prevention agenda, ranging from
Urban Renewal to provincial Crime Prevention
Strategies. 

In his 2009 budget speech Trevor Manuel
announced a significant increase in social
spending. The largest adjustments to spending
plans go to poverty reduction: R25 billion has
been added to provincial budgets, mainly for
education and health care, and R13 billion for
social assistance grants and their administration.
R4 billion has been added to the school nutrition
programme and R2, 5 billion to municipalities for
basic services.18



26 Institute for Security Studies

Meanwhile, considerable knowledge has been
built up by civil society practitioners in a range of
prevention fields, including efforts to address
alcohol and drug abuse, promote diversion,
support early childhood development, improve
safety at schools, develop sport and recreational
facilities, and support young mothers and families
at risk. 

The challenge for many, however, remains how
this new knowledge can be brought to scale and
attract considered state investment over the time
periods required to demonstrate results. The
balancing act for the state right now is one of
keeping citizens safe against the backdrop of high
levels of serious and violent crime while
simultaneously providing committed resources
and support to long-term social change. 

Commentators such as Dixon19 warn against
conflating crime prevention and social policy – he
argues that issues like appropriate and adequate
recreational facilities should be prioritised and
addressed in their own right and should not only
receive attention because of a possible link to
their crime prevention utility. This said, efforts to
mitigate and respond to crime go beyond the
mandate of the police and must logically make
the link between crime prevention and social
spending. Social crime prevention must be a
priority that is integrated across a range of
government departments, most immediately the
departments of social development and
education.

THE OSF-SA PROJECT

Over the past ten years the Open Society
Foundation for South Africa (OSF-SA), has
invested extensively in building knowledge in
crime prevention in the fields of school safety,
violence against women, and local safety. The
Foundation is motivated by the fact that the
results of a singular law enforcement approach
have been mixed at best, and the increasingly
punitive stance adopted by criminal justice role
players could exacerbate underlying social and
economic issues that contribute to the high levels

of crime in South African society. It has therefore
employed the lessons emerging from its past work
in the development of a new strategy to support
crime prevention in three communities in South
Africa. 

The multi-year ‘Crime and Safety Project’
includes a substantial component on evaluating
safety. It is also an attempt to show that effective
partnerships between government communities
and civil society are essential for crime
prevention. The programme is informed by the
2002 United Nations Guidelines for Crime
Prevention20 and its eight core principles, namely:

• Government leadership
• Socio-economic development and inclusion
• Cooperation and partnership
• Sustainability and accountability
• Knowledge base
• Human rights and rule of law
• Interdependency
• Differentiation (recognising the different 

needs of communities and sectors of the
community) 

OSF-SA is working in partnership with provincial
governments in three provinces and has signed
Memoranda of Understanding (MOUs) with the
Department of Community Safety in Gauteng,
the Eastern Cape and the Western Cape. These
MOUs outline the roles and responsibilities of
each party in the implementation of the project in
three sites that were identified by the provinces as
being ‘crime hotspots’; namely Orange Farm in
Gauteng, Nompumelelo in the Eastern Cape, and
Elsie’s River in the Western Cape.

ORANGE FARM, NOMPUMELELO
AND ELSIE’S RIVER 

The three sites are diverse in terms of size,
demographics, structural design and social
context. However, they share many common
problems, including poverty, underdevelopment,
high rates of HIV/AIDS and social and economic
inequities. 

 



SA Crime Quarterly no 27 • March 2009 27

Orange Farm

Orange Farm is one of the largest townships in
South Africa. Located 45 kilometres south of
Johannesburg, with an estimated population of
one million, it is often referred to as one of the
most indigent urban settlements in South Africa.
Today the township is still considered to be fairly
underdeveloped with the majority of residents
facing significant socio-economic challenges,
despite considerable infrastructural investments
(e.g. the building of the new police station,
railway station and victim support centre) over
the past years. Recent statistics show that no less
than 40 per cent of the population is under the
age of 18 years and that many do not have access
to proper health care, education and recreational
facilities. Consequently a substantial proportion
of these young people are said to engage in crime
and violence as well as substance abuse.
Reportedly high numbers of teenage pregnancies
and HIV/AIDS infection are further obstacles to
the development of the Orange Farm community. 

Against this backdrop the dichotomy of crime
prevention and law enforcement is clearly
illustrated. Policing strategies to arrest
intoxicated teenagers for being drunk in public
has served to keep them safe over periods when
they are vulnerable to victimisation, but vital
access to treatment and support to address the
dangers of substance abuse is lacking, causing
residents and police to repeat the same patterns
of binge drinking and arrest weekend after
weekend.

Nompumelelo

Nompumelelo by contrast is a small settlement
located within the relatively wealthy suburb of
Beacon Bay in urban East London in the Eastern
Cape.  The township has come a long way from
its informal beginnings and has seen significant
development of low-income housing.  However,
its history as a temporary abode for those who
wanted to be nearer to their places of
employment continues to haunt it. The
population swells by thousands during the week.

These weekday migrants place a considerable
burden on an already overstretched
infrastructure; have a negative impact on
community cohesion (which was demonstrated in
residents’ response to the safety audit); and are an
eager market for the thriving shebeens.  

Elsie’s River

Elsie’s River is located approximately 12
kilometres from the Cape Town city centre and
has an estimated population of 150 000. A l arge
proportion of the area consists of low-income
housing such as council flats and semi-detached
homes, and the area is characterised by
overcrowding. It is also estimated that 34 per cent
of the population is unemployed and that the
majority of people who are employed are low-
income earners. Elsie’s River is renowned for its
gang activity and high levels of crime. 

Elsie’s River, like Orange Farm (and unlike
Nompumelelo), has seen significant support for
social crime prevention. It was the site of the
Cape Flats Renewal Strategy, which introduced a
comprehensive programme of environmental and
social programmes. Unfortunately its legacy is
largely a monument to our wavering and often
fickle support for crime prevention, as the plans
and interventions have not been evaluated or
followed up. Initial efforts have been superseded
by ever-new strategies aimed at social
transformation and gang reduction, making the
design of any subsequent intervention a
painstaking task of sifting through what has been
done, what has been achieved and where value
can be added.

HOW TO CREATE A VIABLE AND
REALISTIC SAFETY PLAN? 

In light of the significant challenges outlined
above, OSF-SA opted for a considered,
methodical approach to this work. In keeping
with the tested methodology of basing the
intervention on evidence and knowledge, safety
audits were undertaken at each of the sites.21



28 Institute for Security Studies

The researchers doing the safety audit were
required to be mindful of all elements of a crime;
namely the offenders, the victims, and the
locations in which crimes take place. Qualitative
and quantitative research methods were used to
collect data and statistical information;
complemented by well-placed interviews and
focus group discussions with community groups.

In particular the audit provided for: 

• The profile of the community (location, 
demographics etc)

• The nature and level of recorded crime in the 
identified community

• The specific safety concerns in the community
• The state resources available to address these 

concerns
• Current non-state resources to address these 

concerns
• Current interventions, the impact of these 

interventions and the challenges still faced 
• A database of stakeholders and role players in 

the community as well as the institutional
mechanisms in place to support broad multi-
stakeholder safety initiatives  

The information captured during the audits has
had a dual utility. It has elicited from the range of
partners, the buy-in and commitment necessary
for successful implementation, and it has also
provided baseline data for subsequent evaluations
in each of the sites. Unsurprisingly, the findings
from the audits support the need for a broad
approach to addressing crime that encompasses
not only the transformation of criminal justice
institutions, but also changing public attitudes
towards crime and addressing the underlying
socio-economic factors that feed both crime and
perceptions on the acceptability of crime. 

Working with the audits, OSF-SA has set about
developing detailed site-specific safety plans. This
has been done in collaboration with a multi-
disciplinary project team of community members
and local and provincial government departments
at each site. In many ways the teams mirror the
multi-disciplinary Community Safety Forum
structures currently being mooted by government

as a potential vehicle for crime prevention
delivery. 

While many of the interventions identified in the
plans fall squarely within the mandate of local
stakeholders, OSF-SA will use its grant-making
capacity to support civil society interventions that
are aimed at meeting some of these objectives.
Even here, multi-utilities are being sought. Not
only will the investments seed the safety plans,
but the lessons for government in working with
civil society to implement aspects of the strategy
will be particularly useful in understanding the
elements of successful partnerships, and
integrating the experience and expertise of civil
society into general government service delivery.
Success in this instance is defined by targeted,
sustainable support that meets defined objectives
of a safety strategy with measurable impact.

THE PAUCITY OF ROBUST
LONGITUDINAL EVALUATION DATA 

A weakness in South Africa’s growing knowledge
and understanding of crime prevention is the
paucity of robust evaluation and particularly
longitudinal evaluations tracking progress over
time. This in turn has an impact on the
development of policy to promote crime
prevention as a viable strategy for making South
Africa safe. Recognising this, and as a part of its
contribution to developing knowledge on crime
prevention, law enforcement and safety, the OSF-
SA has introduced a parallel component to the
project: the development and implementation of a
longitudinal evaluation across all three sites. 

As the first step in setting up the evaluation
system, a comprehensive set of safety indicators is
being developed and tailored to the needs of each
of the provinces and individual sites, stakeholders
and service providers. This individualised
tailoring for role players will allow various service
providers, for example health services or the
police, to track and report on their progress
towards implementing the safety plan in a manner
that feeds into a system for tracking progress
across the safety plan generally. Using these

 



SA Crime Quarterly no 27 • March 2009 29

indicators, an assessment will be undertaken of
the crime and safety project at each of the sites on
an annual basis. This will provide the foundation
for a detailed longitudinal evaluation of the
intervention and impact over time. Importantly,
the indicators will be constructed off a generic
template that will allow for an easy transfer of the
methodology to other sites. 

The development and implementation of a sound
monitoring and evaluation system will have value
both for the evaluation of the impact of the safety
plans in Orange Farm, Elsie’s River and
Nompumelelo, but also, more importantly, for the
future management of safety strategies more
broadly. Through the implementation of a sound
monitoring system the OSF-SA hopes to be able
to clearly demonstrate the utility of a crime
prevention approach alongside that of law
enforcement.

CONCLUSION

International research tells us that efforts to
mitigate and respond to crime must be balanced
by longer-term systemic changes aimed at
addressing the causes and drivers of crime.
Bringing about these changes goes beyond the
mandate of the police, to encompass a range of
stakeholders and disciplines. Social crime
prevention must be a priority that is integrated
across a range of government departments, most
immediately the departments of social
development and education.

If we look at efforts to promote diversion, build
safety at schools and expand our social support
networks, it can be said that significant progress
has been made. We need to support these efforts
consistently over time and ensure that
interventions are evaluated so that we can
understand the type and level of impact being
made and adjust our strategies accordingly. We
should put behind us the days of simply replacing
one five-year crime prevention strategy with
another. Rather, we need to evaluate the successes
and challenges of our efforts and learn from the
experience. The wealth of knowledge in civil

society can make a significant impact if effectively
supported. It is in this area that the OSF-SA
project can provide new insights.  

NOTES

1 J Irish-Qhobosheane, See no evil, hear no evil, the 
social economy of organised crime in Mozambique,
SAIIA, 2008.

2 What every South African can do to build a safe 
community, Action for a Safe South Africa, SOUTH
AFRICA – the Good News (Pty) Ltd, December 2008, 25.

3 See http://www.saps.gov.za/statistics/reports/crimestats/ 
2008/crime_stats.htm (accessed February 2009).

4 L Sherman, Thinking about crime prevention, in 
D Gottfredsen, L Sherman, D MacKenzie et al, Crime,
what works, what doesn’t, what’s promising, University of
Maryland, 1996. Sherman makes the point that the debate
over crime often treats ‘prevention’ and ‘punishment’ as
mutually exclusive concepts, polar opposites on a
continuum of ‘soft’ versus ‘tough’ responses to crime. He
says: ‘The science of criminology, however, contains no
such dichotomy. It is as if a public debate over physics had
drawn a dichotomy between flame and matches. Flame is
a result. Matches are only one tool for achieving that
result. Other tools besides matches are well known to
cause fuel to ignite into flame, from magnifying glasses to
tinder boxes. Similarly, crime prevention is a result, while
punishment is only one possible tool for achieving that
result.’ This paper argues that there needs to be a balance
between these various tools. 

5 Guidance on local safety audits, European Forum for 
Urban Safety, 2007.

6 The murder rate at 38,6 per 100 000 p opulation is far in 
excess of an international average of five per 100 000. S ee 
J Burger and H Boshoff, The state’s response to crime and
public security in South Africa, ISS Research Report,
Pretoria: ISS, 2008,  h ttp://www.issafrica.org/dynamic/
administration/file_manager/file_links/STATECRIME.PD
F?link_id=24&slink (accessed February 2009).

7 See C Frank, The role of education, health and social 
development in preventing crime, ISS Monograph 126
October 2006. 

8 National Treasury. Estimates of national expenditure 2009, 
Pretoria: National Treasury, 2009.

9 In 2005 South Africa spent US$130.00 per person on 
criminal justice, compared to an average by the rest of the
world of US$ 66.00. S ee A Altbeker, Paying for crime:
South African spending on criminal justice, ISS
Occasional Paper 115, Pretoria: ISS, July 2005.

10 R Taljaard, Private and public security in South Africa, ISS 
Monograph 146, Pretoria: ISS, July 2008. 

11 International research indicates that rates of incarnation 
have no significant impact on levels of crime – see M
Tonry, Crime does not cause punishment, Keynote
address at the OSF-SA Conference on Sentencing in South
Africa, Sentencing in South Africa conference report,
Open Society Foundation for South Africa, 2006, 4-6.

12 R Pharoah, National Victims of Crime Survey, ISS 
Occasional Paper 175, Pretoria: ISS, December 2008. 

13 Pharoah, National Victims of Crime Survey.

 



30 Institute for Security Studies

14 J Kollapen, Approach to crime fails democracy, Mail and 
Guardian 22–28 June 2007. 

15 Kollapen, Approach to crime fails democracy. 
16 See the OSF-SA conference report: Sentencing in South 

Africa, October 2006. 
17 Frank, The role of education, health and social 

development in preventing crime.
18 Budget Speech 2009 by the Minister of Finance, Trevor A 

Manuel, available at http://www.info.gov.za/speeches/
2009/09021114561001.htm (accessed February 2009).

19 B Dixon, Development, crime prevention and social 
policy in post-apartheid South Africa, Critical Sociology,
26(1), 2006, 169-191.

20 UN Economic and Social Council Action to promote 
effective crime prevention, Guidelines for crime
prevention UN ECOSOC, July 2002. 

21 The safety audit is an important tool in reaching a 
common understanding on crime and its causes in a
particular area. See Making South Africa safe, a manual
for community based crime prevention and guidance on
local safety audits, European Forum for Urban Safety,
2007.