The governance of security in
weak and failing states
IT DUPONT, PETER GRABOSKY AND
Universit´e de Montreal, Australian National
University and Australian National University
This article seeks to identify means of strengthening social control
and conﬂict resolution in weak and failing states. It begins by
discussing the governance of public security in stronger states, and
identiﬁes three basic forms of engagement between state and non-
state institutions that may contribute to the co-production of public
security: coercion, sale and gift. The article then seeks to identify
institutional arrangements that might be transplanted to those
settings where conventional state institutions of security may be in
decline, or non-existent. It also suggests how new institutions
might be invented in settings where states may be dysfunctional or
otherwise lacking in capacity. It develops a typology of security
provision, including auspices that are public; public under private
arrangements; collective or voluntary; private/national; private/
international; and criminal. By identifying new mechanisms for the
governance of security, it may be possible to arrest the
deterioration of states, or at least provide for a modicum of internal
security. The article concludes with a discussion of the
Zwelethemba model of peacemaking and peacebuilding that is
being developed in South Africa.
capacity building • civil society • policing • private security
• transnational institutions
© 2003 SAGE Publications
London, Thousand Oaks
and New Delhi.
Vol. 3(4): 331–349; 041818
Even in states that are, by most standards, ﬂourishing, the demand for
public security often exceeds the capacity of the state to provide it. So it is
that throughout most of the world, the state no longer has a monopoly on
policing. Bayley and Shearing (1996, 2001) speak of the multilateralization
of policing, by which they mean the variety of institutional forms, public,
private, non-proﬁt and hybrid, by which policing is delivered. Some of
these may be explicit instruments of policy, while others may operate more
or less spontaneously, driven by market forces. One might therefore speak
of the ‘governance of security’ to refer to the constellation of institutions,
whether formal or informal, governmental or private, commercial or
voluntary, that provide for social control and conﬂict resolution, and that
attempt to promote peace in the face of threats (either realized or anticip-
ated) that arise from collective life.
In states that are weak and failing, the situation is even more acute. By
weak state, we refer to circumstances in which the state is barely capable of
providing for the needs of its citizens. Health, education, national defence,
public welfare and other essential services, in addition to those related to
public security, are inadequate. One might look to states like Papua New
Guinea and Zimbabwe for examples. By failing state, we refer to a
situation in which the capability of the state to perform its essential
functions has largely disappeared. Basic state institutions, if they still exist,
are largely dysfunctional. Examples might include Somalia, Sierra Leone
and the Solomon Islands at the time of intervention there by Australia and
its South Paciﬁc allies in July 2003. Whether a state is ‘weakening’ or
whether formal institutions have ceased to exist and the state can be
deemed to have failed, social control and conﬂict resolution still happen. To
be sure, there will be less of them in a collapsed state, but what there is, is
Our goal is to identify means of strengthening social control and conﬂict
resolution in weak and failing states. We begin by discussing the govern-
ance of public security in stronger states, and by seeking institutional
arrangements that might be transplanted. We seek also to suggest how new
institutions might be invented in settings where conventional state institu-
tions of security may be in decline, or non-existent. By identifying new
mechanisms for the governance of security, it may be possible to arrest the
deterioration of states, or at least provide for a modicum of internal
The governance of security in advanced industrial societies
Since their invention by Sir Robert Peel in 1829, public police began to
cultivate an image of omnipresence and omnicompetence. While this may
Criminal Justice 3(4)
have been credible at some places and at some times, it is now illusory.
Nowhere is the ﬁnite capacity of public police more boldly evident than in
the growth of the private security industry in western industrial societies,
where private security ofﬁcers now outnumber public police by a factor of
at least three to one (for a recent review see Bayley and Shearing, 2001).
Governments of various political persuasions have consciously sought to
shift much of the burden of crime control back to the private individual.
The apposite if awkward term ‘responsibilization’ has been used by
Garland (1996) and by O’Malley and Palmer (1996) in this regard. An
important consequence of this has been a growing inequality of access to
services and inequalities in the distribution of human rights (Bayley and
The public/private dichotomy is itself illusory, with a number of
commentators noting the emergence of various forms of public–private
interface and hybrid institutional forms (Marx, 1987; Hoogenboom, 1991;
Johnston, 1992; Bayley and Shearing, 1996, 2001). Marx (1987) for
example, referred to ﬁve trends in criminal investigations:
1 joint public–private investigations;
2 public agents hiring or delegating authority to private police;
3 private interests hiring public police;
4new organizational forms in which the distinction between public and
private is blurred; and
5 the circulation of personnel between the public and private sectors.
Such organizational differentiation itself arises from the inability of public
police agencies to be everywhere and to do everything with their own
personnel and their own resources. This has given rise to some quite
creative partnerships with non-police agencies. In the city of Philadelphia,
the central business district is policed by an integrated team of public police
and private security professionals, called ‘community service representat-
ives’ (Greene et al., 1995). The latter, who co-ordinate their activities and
share ofﬁce facilities with Philadelphia Police, serve as ‘goodwill ambas-
sadors’ and engage in general surveillance. While deployed, the community
service representatives remain in radio contact with police, notifying them
when their intervention is required. These collaborative activities are
supported by city sanitation workers, who are available to address
problems of litter, grafﬁti and other physical indicia of disorder.
Today in the world’s more afﬂuent nations, a great deal of public life
occurs on private property. The most common example is that of the large
shopping malls of North America. Much of the policing of this space is
done by private organizations under contract to the property owner.
Another fact of life in many nations is the so-called ‘gated community’,
whether vertical or horizontal, where access and egress are controlled by
security personnel (Johnston and Shearing, 2003: ch. 8). Ironically, such
Dupont et al.—The governance of security in weak and failing states 333
‘modern’ developments are referred to as representing a ‘refeudalization’ of
While no one would suggest that public law enforcement agencies are an
endangered species, the resource constraints under which they operate are
real, and almost certainly likely to persist. The fundamental challenge then
is how best to resource public security in general, and public policing in
particular. One of us (Grabosky, 1996) has attempted to identify the
various legal and administrative strategies by which public law enforce-
ment agencies may harness private interests in furtherance of their mission.
These are summarized brieﬂy in Appendix A. While most of these are
applicable in situations where the state is in relative good health, the role of
private parties and market orderings will be proportionately greater in
Given the proliferation of forms of interaction between police and
private institutions, one could envisage a continuum of engagement ranging
from the most coercive to the least coercive. But it would help to simplify
this. Davis (2000) refers to three basic relational modes of exchange:
coercion, sale and gift. Let us look ﬁrst at coercion. Elsewhere, one of us
has used the term conscription to refer to the process by which the state
commands commercial organizations to engage in certain actions in
furtherance of law enforcement (Grabosky, 1995). Perhaps the most
prominent of these are cash transaction reporting requirements, where
banks and other deﬁned entities are required by law to report transactions
over a particular threshold, or those of a suspicious nature regardless of
their quantum, to law enforcement authorities. Similar requirements are
imposed on speciﬁed professionals in the case of suspected child abuse and
neglect, and on second-hand goods dealers and pawnbrokers. The future
may well see the conscription of Internet service providers in some juris-
dictions in certain circumstances.
Next, we can see the commercial acquisition of goods and services. This
is nothing new to public police organizations, which, since their establish-
ment in the 19th century, have purchased commodities from means of
transport to weaponry. What is new is the growing tendency to purchase
services that might otherwise have been provided from their own ranks.
The ﬁrst person whom one encounters upon entering the headquarters of
the Australian Federal Police in Canberra is a private security guard. Of
course, the new challenges facing public police organizations may require
specialized expertise which does not reside within police ranks: one thinks
of information technology, for example. And circumstances may arise
wherein special outside assistance may be required for a short period.
Security for the 2000 Olympics involved the collaboration of the New
South Wales Police Service with a number of public and private institutions,
including the Australian Defence Forces and private security ﬁrms
(Dolahenty, 2000). Police can be the providers as well as the purchasers of
services. Gans (2000) describes a variety of circumstances in which public
Criminal Justice 3(4)
police agencies act on a fee for service basis. These include the provision of
policing services at public events of a proﬁt-making nature such as concerts
and professional sporting events.
The third mode of exchange is the donation or gift. By this we mean
private sponsorship of public policing. This can entail the giving of cash
grants, or the provision of complimentary goods and services to the police
organization, usually in return for acknowledgement or recognition. This is
more familiar in the United States and South Africa, relatively uncommon
in Australia and unheard of, if not prohibited by law, in many other
•In Australia, the New South Wales Police Service Annual Report for
1999–2000 indicates that the Service received 28 separate donations of
$2000 or more from a diverse range of donors including Toyota Motor
Corporation, the Liverpool City Council, Lions and Rotary Clubs and
private individuals (NSW Police Service, 2000). The gifts in question ranged
for cash for speciﬁc crime prevention projects to ‘covert vehicles’, ﬁnger-
print cameras, petrol and pushbikes (2000: 61–2).
•The Westﬁeld Carousel Shopping Center in Perth donates computer, fax,
video interviewing facilities and ofﬁce space to the Western Australia Police
Service (Western Australia Police Service, 2000: 14).
•The Western Australia Police Service Macro Task Force, investigating a
series of murders of young women in Perth, was subsidized in part by
donations from members of the public.
•In the United States, local businesses in Crown Point, Indiana donate $1500
each to provide communications and other equipment for patrol cars.
Under the Adopt-a-car programme, the cars may bear the message ‘This
vehicle is equipped by “XYZ company”’ on the back of the car (Pilant,
•In the State of Virginia, insurance companies return a percentage of liability
insurance premiums to the criminal justice system earmarked for auto theft
reduction (Pilant, 1998: 44).
• In South Africa, an organization called Business Against Crime was estab-
lished to provide training and material support to police agencies. Projects
include the development of closed circuit television (CCTV) cameras,
improvements in courts administration and the enhancement of services to
victims of crime (http://www.bac.co.za_ consulted 17 July 2003).
Where they exist, public police will continue to seek means of engaging
private institutions in furtherance of their mission. It is therefore essential
to ask: whose interests are served by a particular mode of police/private
interface (Bayley and Shearing, 1996, 2001)? At its most basic, one can
identify three interests that might be served by a given exchange relation-
ship: the interest of the private party to exchange, the (public) police service
and that of the general public.
Dupont et al.—The governance of security in weak and failing states 335
New institutional conﬁgurations and opportunities for
public security in weak and failing states
In the developed world, new styles of governance have been characterized
by a diffusion of responsibilities previously monopolized by the state, such
as policy formulation and implementation. These two activities have been
disconnected and distributed among networks of public, private and hybrid
institutions. Thus, while the identiﬁcation of needs and formulation of
policies is done under a range of auspices that include government agencies,
civil society and economic interests, the requisite services can be provided
by private and public actors under a variety of arrangements including
those described in Appendix A (Grabosky, 1996; Bayley and Shearing,
2001). In this conceptual framework, any combination is conceivable and
is amenable to trial and experimentation in response to particular needs.
To be more precise, if we take as a starting point Black’s general theory
of law (Black, 1976) we can expect that, all else equal, the strength of a
society’s formal legal institutions will vary inversely with the strength of its
informal institutions of social control. In dysfunctional and failed states, a
variety of non-governmental institutions may already exist whose resources
and energies might be harnessed in furtherance of public security. Not all of
these will be equally and universally appropriate. First, an institution must
possess the capacity to deliver the desired outcome. Second, even if it
commands the requisite capacity, the institution must be compatible with
the culture in which it operates. Elsewhere (Grabosky, 2000), one of us has
identiﬁed a variety of institutional forms that may serve as the vehicles for
the delivery of health services for the purpose of violence prevention in
developing countries. These include but are not restricted to:
• private enterprise, large and small;
• community organizations and voluntary associations;
• religious institutions;
• traditional health practitioners;
• medical professionals.
One will quickly notice that many of these are as likely to be part of the
problem as part of the solution. But in many settings, at least some of these
institutions can play a constructive role. There are those states in serious
decline that nevertheless are endowed with attractive resources and that
serve as attractive sites for domestic or foreign investors. Many such
investors, with signiﬁcant assets to protect, will rely upon their own
security arrangements. Whether these provide a ‘diffusion of beneﬁts
effect’, providing some degree of protection to neighbouring communities,
Criminal Justice 3(4)
or displace risk to interests less able to protect themselves, is a signiﬁcant
In a number of less advantaged nations whose authorities lack the
capacity or inclination to control copyright and trademark infringement,
copyright owners or their industry associations may play a leading role in
the investigation and prosecution of piracy cases. Large mining companies
that operate in remote areas may engage the services of local interests or of
multinational private security ﬁrms to protect their assets. Some large
multinationals will engage the services of local police and military institu-
tions. Others will contract in private security services. Johnston (2000)
provides a useful overview of such arrangements.
In weak states, the decay of public and private institutions might limit
the number of permutations possible, but it is still possible to outline a
typology of auspices and providers that can be used as a starting point to
design innovative security policies. Three main auspices that can be identi-
ﬁed are the state, civil society and transnational institutions. By civil
society, we mean the various residential and interest communities as well as
the individuals and non-governmental institutions that constitute the social
fabric. In the transnational institutions category, we include primary
commodities corporations, multilateral international institutions, aid
agencies and NGOs. Both civil society and transnational institutions take
over the role of auspice from the state in order to protect their constituents,
employees and the integrity of their operations. While each of these
auspices is far from being homogeneous, their constituents share a number
of features that make their approaches to security very similar in both
cases. These auspices activate providers that can be categorized as public,
public under private arrangements, collective—or voluntary, private/
national, private/international and criminal.
The state has traditionally been presented as the only institution willing
and capable of providing security to its citizens. However, if we turn our
gaze to the two remaining auspices outlined above, a legitimacy-resource
test clearly demonstrates that they also possess some of the required
features for the provision of security. Weak states are characterized by the
disintegration of their governmental institutions and the inability—or lack
of interest—of the latter to deliver basic services to the citizenry. This
incapacitation of the state can be attributed to a number of factors, such as
the colonial legacy of artiﬁcial state structures and boundaries that dis-
regard ethnic, religious and tribal divisions, the economic pressures of
international commodities markets, the monopolization of power by a
corrupt and incompetent kleptocracy or all of the above. As a result,
governmental institutions ﬁnd themselves with very limited supplies of
resources and legitimacy to authorize the provision of security. In countries
such as Burma, North Korea, Albania and Russia, where the state has co-
opted or has been co-opted by criminal organizations, this problem is
exacerbated and legitimacy is eclipsed by blatant illegality, while resources
are channelled towards activities fostering insecurity (Dupont, 2001: 174).
Dupont et al.—The governance of security in weak and failing states 337
Governments conducting systematic human rights abuses against minority
groups face the same legitimacy and resource deﬁcits. Civil society as an
auspice of security provision faces a different challenge. While it is often
very rich in legitimacy, it is generally poor in resources, which limit its
options for the authorization and funding of effective security arrange-
ments. Transnational institutions, on the contrary, tend to be legitimacy-
poor and resources-rich. Thus, the logic would suggest that the
establishment of close partnerships between civil society and transnational
institutions is likely to facilitate the birth of hybrid auspices, better adapted
to the constraints of weak states than the existing ones (see Table 1).
These hybrid auspices would generate a cluster of linkages between civil
society, the state and transnational institutions that would compensate for
the lack of strength with an increased adaptability, based on the power of
networks—what one of us describes as nodal governance (Shearing and
Wood, 2000). Furthermore, traditional culture-social groups that constitute
civil society in failing states are characterized by a narrow radius of trust,
which is the circle within which co-operative norms are operative
(Fukuyama, 1999). The hybridization of civil society and transnational
institutions as security auspices has the potential to widen this radius of
trust, and to increase the stock of social capital, strengthening governance
structures of security. The international community and aid agencies should
recognize the beneﬁts of this hybridization process and facilitate it, instead
of focusing too narrowly on state institutions as recipients of their capacity-
building programmes. Too often, state-centred strategies sponsored by
developed countries have exacerbated problems instead of resolving them,
particularly in the ﬁeld of security.
Let us now turn our attention to security providers, who in the absence
of strong auspices are left to themselves and can become central instru-
ments in the perpetuation of cycles of violence. Public providers of security
such as the police tend to lack the accountability mechanisms usually found
in strong states. They often resort to corrupt practices in order to supple-
ment their inadequate salaries. In this context, the coercive technologies
bestowed on police ofﬁcers by the state are used as private revenue raising
tools, contributing to the insecurity of those who experience the full impact
these public providers’ illegitimate and illegal authority. It should not come
as a surprise then that the population’s conﬁdence in public security
Table 1. Institutional auspices for the governance of security
State – –
Civil society (domestic) + –
Civil society (Int’l) (i.e. Red Cross) + +
Interstate Organizations (UN; NATO) + +
Multinational business ? +
Hybrid non-governmental auspices + +
Criminal Justice 3(4)338
providers is minimal in weak states: in Lagos for example, the biggest city
in sub-Saharan Africa, less than one-third of the population consider the
police effective (Adyemi et al., cited in P´erouse de Montclos, 1997). In
Russia, the proportion of the population expecting fair treatment by the
police is comparable (World Bank, 1997). Clearly, non-governmental
auspices ﬁnd in this lack of conﬁdence a strong incentive to explore
alternative options for security provision. Problems of legitimacy may
persist, or may even be aggravated, when public providers sell their services
to civil society and transnational institutions. In Indonesia for example, the
army (TNI or Tentara Nasional Indonesia) estimates that 70 per cent of its
soldiers’ revenues emanate from ‘non-ofﬁcial sources’ such as local com-
munities and foreign mining corporations (Bertrand, 2003). Entire
battalions are ‘rented’ to ensure the protection of extraction sites. Similarly,
in Papua New Guinea, the Government agreed to meet mining interests’
demands for more security on and around extraction sites and to create a
Rapid Deployment Unit only after the industry lobby group accepted to
ﬁnance most of the expenditures related to it (Dinnen, 2001).
The absence or inefﬁciency of state providers can also lead to the
institutionalization of self-help, under the guise of voluntary vigilante
associations. Cattle rustling, for example, is unlikely to be considered a
high priority problem for the police in a disintegrating society. However,
for rural communities, the effects can be disastrous and compromise their
future. In response to this threat, communities form vigilante arrangements
such as Rondas in Peru or Sungusungu in Tanzania (Hemed Bukurura,
1993; Yrigoyen Fajardo, 1993). In this case, the auspice and the provider
are identical, because of the lack of resources. One of the consequences of
this mode of security provision is its propensity to arbitrariness. Justice is
meted out on the spot, after an informal contradictory discussion between
suspect, victim and vigilantes, procedural fairness not being the main
preoccupation. The punishment is usually savage and brutal, in order to
make an example for others, and can include kneecapping (Northern
Ireland) or the burning alive of suspected shoplifters (Nigeria). If a sense of
security is restored among the communities that resort to self-help, human
rights are often a victim of these schemes, which tend to fuel rather than
weaken the circle of violence experienced by these communities.
Private security providers are not a new phenomenon, even if their
responsibilities have ﬂuctuated through history (Shearing, 1992). What is
new is the development of international private security providers, whose
size and expertise are equivalent, if not superior, to those of state security
providers (Johnston, 2000; Singer, 2001). Securitas, a Swedish security
company, employs for example 230,000 people around the world and
claims a global security market share of 10 per cent. MPRI, another major
player in the private provision of services to law enforcement agencies,
boasts among its associates a retired FBI assistant director, a former US
Assistant Attorney General and many former police chiefs. In September
2002, the US State Department announced that it would engage the
Dupont et al.—The governance of security in weak and failing states 339
services of a private company to protect President Hamid Karzai of
Afghanistan (Dao, 2002). Estimates place the number of these companies
at 100 worldwide. These multifunctional organizations offer a full range of
services to non-governmental and governmental clients. They have found a
booming market in weak and failing states, where economic actors are
willing to purchase products and services that hold the promise of reducing
those risks and uncertainties that hamper their economic activities. In
certain states, such as Angola, the Government makes it compulsory for
foreign investors to provide their own security, mainly through private
security companies (FCO, 2002). These private security providers do not
restrict their services to extracting companies and wealthy clients. They are
increasingly intervening to protect the activities of NGOs and relief agen-
cies. In other cases, they have even offered to act in place of state providers,
requiring in payment a portion of the mineral resources found in the
paciﬁed regions. If this option seems attractive and can provide some
temporary relief to the rampant insecurity found in weak states, it also
contains a few drawbacks. First, most of these providers mobilize para-
military modes of action and expertise, more suited to states of war and
counterinsurgency than to conﬂict resolution and peacemaking. Of course,
they are able to create ‘peace bubbles’ where they operate, but they do not
engage the community in the production of security and these bubbles
usually burst when they withdraw. In Sierra Leone for example, the South
African company Executive Outcomes was instrumental in stabilizing the
country, allowing the signature of a peace agreement and elections to be
held. However, its departure was followed by a coup and the exile of the
elected government, nullifying much of what had been achieved. The lack
of accountability and the absence of regulatory arrangements create some
moral dilemmas for those advocating a more widespread use of private
security providers. However, from a weak state’s auspice perspective, they
remain a comparatively cheap, quick and effective way of securing order.
Finally, the last and least desirable form of security provision lies with
criminal organizations, which can be seen as institutions offering security
and protection to their members. The Sicilian Maﬁa, Chinese Triads and
the Japanese Yakuza all originated from mutual protection schemes in
places where the state failed to perform this function (Gambetta, 1993;
Dupont, 2001). Today, in Brazil, drug gangs produce justice for commun-
ities, regulating order in exchange for the protection of their illegal
activities (Botelho Junqueira and Augusto de Souza Rodriques, 1993). In
the eyes of the marginalized communities that beneﬁt from the security
provided, the illegality of criminal groups does not in any way diminish
their legitimacy, in the absence of another alternative.
As we have shown in the previous paragraphs, the disintegration of
governmental capacity in weak and failing states creates a security vacuum
that can be ﬁlled by several conﬁgurations of auspices and providers. Some
of them might be morally objectionable, but the reality of their existence
should not be denied. We argue that the moral dilemmas that are intrinsic
Criminal Justice 3(4)
to some of these conﬁgurations should be considered in their local contexts,
which are, in every way conceivable, unlike the ordered world of demo-
cratic societies. Most of these conﬁgurations are unable to provide long-
term solutions for the governance of security, and to a certain extent
reinforce the processes that perpetuate the weakening of the state. They
rely on auspices that lack either resources or legitimacy, and on providers
that are fragmented, serve narrow interests and seem to concentrate their
activities on coercive measures. At the best of times, these arrangements
may provide a modicum of security for neighbouring residents, as well as
medical, welfare and infrastructure services, employment and other assist-
ance for community development. At the worst of times, they may contrib-
ute to repression, to the exacerbation of local conﬂict or to other
undesirable outcomes such as substance abuse and sexually transmitted
disease. Appropriation of land and other resources, environmental de-
gradation and the displacement of populations without satisfactory com-
pensation is not likely to win hearts and minds. However, there is no
fatality in this state of affairs, and all the elements are present—albeit in a
dispersed order—that would allow the design of new assemblages for the
democratic governance of security and the transfer of some of the policy
options already developed in strong states.
These assemblages should not be based on a unique ‘one-size-ﬁts-all
model’, but should instead value pluralism and accommodate the diverse
contexts, cultures and knowledges found in weak and failing states. Just
like community policing sought to bring state institutions closer to civil
society by way of consultative processes in order to co-produce security,
local knowledge can be harnessed in weak states and lead to the ‘invention’
of alternative governance mechanisms susceptible of complementing state
auspices and provision. This diversity should be encouraged, or at least
tolerated, by governmental and non-governmental (both local and inter-
national) actors, providing it does not undermine basic justice and account-
ability principles. The accountability of those new modes of security
governance is certainly the most intricate issue to be addressed. The blend
of public and private organizations involved in the production of security
in strong states generates inconsistencies between public and private instru-
ments of control (Chan, 1999; Stenning, 2000). In weak states however,
this thorny question is eclipsed by extraordinary and chaotic conditions
and the ensuing pre-eminence of an ‘ends over means’ mentality. As we
have demonstrated, efﬁciency tends to prevail over transparency, and rogue
paramilitary units, private security companies or organized crime
syndicates are not known for their receptivity to external regulation.
Nevertheless, some contribution towards accountability may be provided
by human rights NGOs such as Community Aid Abroad, or other ‘social
auditors’. It is also the responsibility of international development agencies
to ensure that the schemes they sponsor have high levels of accountability
embedded in them (Bayley, 2001) and that they shape the standards of the
private providers that protect their activities (Avant, 2003). These two
Dupont et al.—The governance of security in weak and failing states 341
complementary features of ﬂexibility and accountability might sound too
broad ranging to be useful at the operational level. However, in the next
section, we turn to a model developed in South Africa that seeks to realize
these requirements and that illustrates how these two features can be
creatively integrated in order to produce security and justice.
The Zwelethemba model for governing security
While South Africa is by no means a weak state there are many parts of the
country (urban, peri-urban and rural) where the governmental situation
mirrors many of the features of weak states that we have identiﬁed. Since
1998 an experiment—based on a partnership between a local NGO, the
Community Peace Programme (which has subsequently become part of the
School of Government at the University of the Western Cape), two
international governments (ﬁrst Sweden and then Finland) through their
aid programmes and various levels of government and governmental
agencies—has been taking place in poor urban collectivities (‘squatter
camps’) where state-based governmental service is weak to develop a model
for the governance of security that realizes the objectives noted earlier.
The work began in a community called Zwelethemba that is part of a
small country town, Worcester, a little over 100 kilometres north of Cape
Town. The model has been named after the community both to recognize
the crucial role it played in its development and because ‘zwelethemba’
means ‘country or place of hope’). The hope that motivated the develop-
ment of the Zwelethemba model was the development of institutions of
micro-governance (that would operate through the window of security but
not be limited to security) that would deliver effective and legitimate
governmental services. The programme to realize this hope began with a
two-year pilot project that produced a basic model which is now being
reﬁned and rolled out to new communities. At present it is operating in
some 20 communities across three South African provinces. The core of the
model is a process that gathers local people together to strive for a future-
oriented solution to disputes as well as solutions to generic problems within
The most recent iteration of the model is one that integrates the dispute-
resolution and problem-solving micro-institutions with the emergency
response capacity of the state police. This is done through the creation of
Community Peace Centres in which the police emergency response capacity
is integrated with the community dispute-resolution and problem-solving
capacity in ways that enable the police to spread their very limited
resources much more widely than would be the case without this integra-
tion. While the details of the model as it is evolving in South Africa are
likely to be culturally speciﬁc, the principles are transferable. This has
recently been demonstrated through an ongoing project in which the
principles have been transferred to two communities in Rosario, Argentina.
Criminal Justice 3(4)
While the core of the model (dispute-resolution and problem-solving
through gatherings) has remained intact, details differ. Similar explorations
of the model and its principles are being planned for Sao Paulo, Brazil and
Toronto, Canada. The essential features of the dispute-resolution and
problem-solving aspects of the model developed in South Africa are as
Peacemaking focuses on disputes. It is particularly concerned with smaller
everyday disputes that if left unresolved might escalate into serious prob-
lems. To facilitate peacemaking, a group of six to 10 people within a local
collectivity establish a Peace Committee. Disputants bring disputes to the
Committee. The Committee follows a set of tested steps and a Code of
Good Practice in responding to the dispute. The essential feature of this
response is a gathering together of local people (including the disputants)
thought to have the capacity and knowledge to envisage and then to
implement an enduring resolution to the dispute. A forward-looking
approach is encouraged throughout the Gathering. It is the people gathered
together, not the Committee members, who resolve problems and see to it
that solutions are acted upon. A central provision of the Code of Good
Practice is that force cannot be threatened or used.
In addition to seeking to contribute towards the resolution of interpersonal
disputes, the Committee also seeks to contribute to the resolution of more
generic problems that are often thought of as community development
issues—for example, issues of public health, education, security and the
environment. The facilitation process is similar to peacemaking—that is,
gathering people together to provide forward-looking solutions to prob-
lems. Needs-gatherings are held to identify generic problems. Solutions-
gatherings are held to explore how to respond to these problems through
the mobilization of local capacity and knowledge. Plans of action are
developed and implemented to create solutions—for example, building a
children’s park, developing dramas on health issues, support for primary
education centres, support for the aged, soup kitchens for children, youth
programmes and so on.
The ﬁrst level of partnership built is between the Committees and local
entrepreneurs of various sorts who can provide the capacity required to
implement peacebuilding plans. This has been the focus to date. The
Community Peace Centres mentioned above are seeking to build on this
level of partnership by including the police as a partner within an institu-
tional and regulatory framework that was developed independently of the
Dupont et al.—The governance of security in weak and failing states 343
police. A partnership has also been built with local governments (see
below) and with the South African Law Commission that is developing
regulatory guidelines that draw upon the guidelines developed with and for
The work of Peace Committees is sustained both through the involvement
of international agencies (consistent with the strategy outlined earlier in the
article) in the form of support through the Finnish Embassy and through
local governments. The Finns provide support for the administrative
services provided by the Community Peace Programme as well as the
research and reﬂection that has and continues to guide the model building
while local governments along with the Embassy provide direct support for
Peace Committees and the Organizers and Co-ordinators who facilitate
their work. The support for Peace Committees is provided according to the
following formula that directs a portion of tax resources available to local
municipalities directly into poor communities in ways that both promotes
self-direction and contributes to community development. Committees are
paid Rand 400 (approximately US$40) for every gathering held in accord-
ance with agreed upon principles. Half of this goes ‘into the pockets’ of
Committee members who organize the gathering, in recognition of the
value of their knowledge, capacity, energy and time. The other half goes
into a community-building fund to support peacebuilding projects. If a
Committee arranges 10 gatherings a month (that is the ﬂoor), aimed at this
brings what is in local terms a signiﬁcant amount into the community in
which the Committee works. For Committees that exceed this norm, for
example, by arranging 30–40 gatherings a month, the money that ‘puts
food on the table’ of members’ households and that is available for
peacebuilding can be quite sizeable in local terms.
The programme is managed at the community level through people drawn
from Peace Committees, who organize, review and gather data. This is
essential to the regulatory process (see below) and provides data required
for both the outcome-based payment scheme and the research and re-
ﬂection noted above. These local Organizers and Co-ordinators are
supported by a small professional staff (one full time and three part
Regulation is embedded in the model rather than added on. Reviews are an
essential part of both the peacemaking and peacebuilding processes and are
required as part of the remuneration formula. In addition, persons attend-
Criminal Justice 3(4)
ing gatherings are asked to participate in ‘exit interviews’ and regular base-
line surveys are conducted.
Principles of the model
At the heart of the model is the premise that good governance requires the
mobilization of local knowledge and capacity and its integration with
professional knowledge and capacity. This mobilization is used both to
establish auspices that direct governance and establish sources of provision.
To use a nautical metaphor both steering and rowing take place at the
micro level. This serves to deepen democracy by involving local people
directly in their own governance. In doing so it enhances self-direction as
well as the quality of service provision. Mobilizing local knowledge and
capacity requires sustainable institutional arrangements that bring people
with local knowledge and capacity to respond to issues. Integrating local
knowledge and capacity with other knowledges and capacities—
particularly those mobilized by state agencies—in ways that will retain self-
direction and local capacity, requires institutional structures that will
ensure that state agencies do not dominate. Sustaining micro-governance
requires new ways of channelling tax-derived resources to local commu-
nities, as well as ensuring ongoing support from international bodies. This
will assist in weathering local political ups and downs, and will provide for
ongoing research, reﬂection and international legitimacy. Regulation that
will ensure compliance with internationally recognized standards of good
governance must be built into the governance processes and supported
through the remuneration template (see Roche, 2002 for a recent assess-
ment of the Zwelethemba model).
Reconﬁguring the governance of security in weak and failing states requires
a dose of institutional creativity and audacity. Traditional models of
governance, which are undergoing considerable transformations in strong
states, and whose transplant has often been instrumental in the dis-
integration of ﬂedgling states, should no longer remain the only available
option for the provision of security.
In their place, models that rely on local knowledge and the mobilization
of a broad range of resources have the potential to facilitate a de-escalation
of violence and disorder, while at the same time allowing communities to
stabilize and recover some of their lost strengths. The state, which is very
often as much part of the problem as of the solution, cannot remain the
sole conduit by which security is authorized and provided to vulnerable
communities. Increased involvement of non-state actors need not eclipse
the state, but rather, can ‘buy time’ and relieve pressure in a manner that
can allow legitimate state institutions to emerge or regenerate. They can
also help arrest the deteriorating quality of life at grassroots levels.
Dupont et al.—The governance of security in weak and failing states 345
Non-state actors can make a valuable contribution. International aid
agencies, religious organizations (Alger, 2002) and multinational extracting
companies can bring to weak states the resources and the organizational
infrastructure of the global. Residential and interest communities, through
their adaptive capacities and their intimate knowledge of the local, ensure
the relevance of the model and the equitable diffusion of its beneﬁts to their
members. States, to the extent to which they are genuinely democratic,
bring to this mix of resources and know-how the possibility of democratic
regulation and access to state taxes. There is no standard formula for this.
Settings will vary in terms of what remains of the state, what domestic
institutions of civil society are viable and in cultural receptivity to proposed
solutions originating from overseas.
The rethinking and reordering we have proposed here challenges mental
schemata with which we are familiar and comfortable. This, however, is
precisely what is required.
Adapted from Grabosky (1996).
1Conscription: Governments may simply command third parties to assist with
one or more processes of law enforcement, as in cash transaction reporting
requirements or mandatory reporting of suspected child abuse.
2Required private interface: In contrast to conscription, where commands are
directed at third parties, this approach entails governments requiring that
targets of crime control engage the machinery of private institutions. Finan-
cial auditing requirements represent one example.
3Required record keeping and disclosure: In order to encourage introspection,
or with a view to informing markets or other private institutions in a
position to foster compliance, governments may require disclosure of certain
aspects of a regulated entity’s activities.
4Co-optation of private interests: In some settings, public agencies may
actively seek the co-operation of private interests in furtherance of surveil-
lance and detection. Neighbourhood watch programmes are one example.
5Conferring entitlements: There are two basic avenues by which governments
can empower private interests to enforce the law. The ﬁrst entails the
creation of certain speciﬁed rights, conferring them upon private parties, and
leaving it up to those private parties to enforce. Many systems of patent,
trademark and copyright rely on such private enforcement
A second avenue of private enforcement entails empowering third parties
to undertake enforcement actions on behalf of the state. In many juris-
dictions, cases of cruelty to animals are investigated and prosecuted by
societies for the prevention of cruelty to animals.
6Incentives: Governments may offer incentives directly to targets of regulation
to induce compliance, or to engage in a desired course of conduct. They may
also offer rewards and bounties for assistance in criminal investigations.
Criminal Justice 3(4)346
7Contracting out: Governments might seek to engage private consultants
rather than rely upon in-house resources.
8Delegation or deference to private parties: In some instances, governments
may become aware that certain functions relating to compliance are already
being performed or could be performed by private parties. Representatives of
the relevant ﬁrm or industry association investigate many cases of insurance
fraud and of copyright infringement.
9Abdication: The state may simply abdicate some regulatory functions, and
leave allocative and ordering decisions to the market. The growth of the
private security sector is one example.
The authors would like to thank the participants in the Failed States Workshop
V, University of California, Santa Barbara, 9–11 September 2002, sponsored by
the US Army War College, and anonymous reviewers for their comments on
earlier drafts. Opinions expressed in this article are those of the authors.
Alger, C. (2002) ‘Identifying Opportunities for Coping with “Failed States” in
Emerging Global Governance’, paper presented to Failed States Workshop V,
University of California, Santa Barbara, 11 September.
Avant, D. (2003) ‘Beyond Regulation: Institutional Lessons and the Trade-Offs
of Private Security’, paper presented to In Search of Security International
Conference, Montreal, 19–23 February.
Bayley, D. (2001) Democratizing the Police Abroad: What to Do and How to
Do It. Washington, DC: National Institute of Justice.
Bayley, D. and C. Shearing (1996) ‘The Future of Policing’, Law and Society
Review 30(3): 585–606.
Bayley, D. and C. Shearing (2001) The New Structure of Policing: Description,
Conceptualization and Research Agenda. Washington, DC: National
Institute of Justice.
Bertrand, R. (2003) ‘Le Militaire et le Milicien: L’Indonesie en Route vers la
Violence?’, in T. de Montbrial (ed.) Ramses 2003, pp. 197–214. Paris:
Black, D. (1976) The Behavior of Law. New York: Academic Press.
Botelho Junqueira, E. and J. Augusto de Souza Rodriques (1993) ‘Pas´arga
Revisited: Policing Favelas in Brazil’, in M. Findlay and U. Zvekic (eds)
Alternative Policing Styles: Cross-Cultural Perspectives, pp. 123–30.
Chan, J. (1999) ‘Governing Police Practice: The Limits of the New Account-
ability’, British Journal of Sociology 50(2): 251–70.
Dao, J. (2002) ‘U.S. Company to Take Over Karzai Safety’, New York Times
19 September, URL (consulted 23 September 2002): http://www.nytimes.
Dupont et al.—The governance of security in weak and failing states 347
com/2002/09/19/international/asia/19SECU.html?ex = 1033539498&ei = 1
&en = 4843e13930012284
Davis, N. (2000) The Gift in Sixteenth Century France. Madison, WI:
University of Wisconsin Press.
Dinnen, S. (2001) Law & Order in a Weak State. Adelaide: Crawford
Dolahenty, Adrian (2000) ‘Protecting the Village’, Security Australia 14–16
Dupont, A. (2001) East Asia Imperilled: Transnational Challenges to Security.
Cambridge: Cambridge University Press.
FCO (Foreign & Commonwealth Ofﬁce) (2002) Private Military Companies:
Options for Regulation. London: The Stationery Ofﬁce.
Fukuyama, F. (1999) Social Capital and Civil Society. Washington, DC:
International Monetary Fund. Available at http://www.imf.org/external/
Gambetta, D. (1993) The Sicilian Maﬁa: The Business of Private Protection.
Cambridge: Harvard University Press.
Gans, Jeremy (2000) ‘Privately Paid Public Policing: Law and Practice’,
Policing and Society 10(2): 183–206.
Garland, David (1996) ‘The Limits of the Sovereign State: Strategies of Crime
Control in Contemporary Society’, British Journal of Criminology 36(4):
Grabosky, P. (1995) ‘Using Non-Governmental Resources to Foster Regulatory
Compliance’, Governance 8(4): 527–50.
Grabosky, P. (1996) ‘The Future of Crime Control’, Trends and Issues in Crime
and Criminal Justice, No. 63. Canberra: Australian Institute of Criminology.
Grabosky, P. (2000) ‘Managing Violence and Health: Strategies Solutions,
Research and Methodological Issues’, in World Health Organization Centre
for Health Development (ed.) Violence and Health: Proceedings of a WHO
Global Symposium, 12–15 October 1999, pp. 415–28. Kobe, Japan: WHO
Greene, J.R., T.M. Seamon and P.R. Levy (1995) ‘Merging Public and Private
Security for Collective Beneﬁt: Philadelphia’s Center City District’, American
Journal of Police 14(2): 3–20.
Hemed Bukurura, S. (1993) ‘Vigilantism in Tanzania’, in M. Findlay and U.
Zvekic (eds) Alternative Policing Styles: Cross-Cultural Perspectives,
pp. 131–8. Deventer: Kluwer.
Hoogenboom, Bob (1991) ‘Grey Policing: A Theoretical Framework’, Policing
and Society 2(1): 17–30.
Johnston, L. (1992) The Rebirth of Private Policing. London: Routledge.
Johnston, L. (2000) ‘Transnational Private Policing: The Impact of Global
Commercial Security’, in J. Sheptycki (ed.) Issues in Transnational Policing,
pp. 21–42. London: Routledge.
Johnston, L. and C. Shearing (2003) Governing Security: Explorations in
Policing and Justice. London: Routledge.
Marx, Gary (1987) ‘The Interweaving of Private and Public Police in Under-
cover Work’, in C. Shearing and P. Stenning (eds) Private Policing, pp. 172–
93. Newbury Park, CA: Sage Publications.
Criminal Justice 3(4)348
New South Wales Police Service (2000) Annual Report for the Year
1999–2000. Sydney: New South Wales Police Service. URL (consulted 17
July 2003) http://www.police.nsw.gov.au/annual_report/ar.cfm
O’Malley, P. and D. Palmer (1996) ‘Post-Keynesian Policing’, Economy and
Society 25(2): 137–55.
P´erouse de Montclos, M.-A. (1997) ‘Does Africa Need the Police?’, Le Monde
Diplomatique, URL (consulted 8 August 2002): http://mondediplo.com/
Pierre, J. and B.G. Peters (2000) Governance, Politics and the State. New York:
St Martin’s Press.
Pilant, L. (1998) ‘Creative Funding’, Police Chief LXV(3): 42–6.
Roche, D. (2002) ‘Restorative Justice and the Regulatory State in South African
Townships’, British Journal of Criminology 42(3): 514–33.
Shearing, C. (1992) ‘The Relationship between Public and Private Policing’, in
M. Tonry and N. Morris (eds) Modern Policing, pp. 399–434. Chicago, IL:
University of Chicago Press.
Shearing, C. and J. Wood (2000) ‘Reﬂections on the Governance of Security, a
Normative Enquiry’, Police Practice 1(4): 457–76.
Singer, Peter W. (2001) ‘Corporate Warriors: The Rise and Ramiﬁcations of the
Privatized Military Industry’, International Security Winter 2001–2, URL
(consulted 29 July 2003): http://www.brookingsinstitution.org/dybdocroot/
Stenning, P. (2000) ‘Powers and Accountability of the Private Police’, European
Journal on Criminal Policy and Research 8(3): 325–52.
Western Australia Police Service (2000) ‘Cannington Police Ofﬁce’, Newsbeat
World Bank (1997) Expanding the Measure of Wealth: Indicators of Environ-
mentally Sustainable Development. Washington, DC: World Bank.
Yrigoyen Fajardo, R. (1993) ‘Rural Rondas in Peru’, in M. Findlay and U.
Zvekic (eds) Alternative Policing Styles: Cross-Cultural Perspectives, pp. 109–
21. Deventer: Kluwer.
IT DUPONT is assistant professor of criminology at the University of
Montreal and researcher at the International Center for Comparative
Criminology. He is the author of Construction et R´eforme d’une Police: Le Cas
Australien (L’Harmattan, Paris, 2003).
PETER GRABOSKY is professor and Co-Director of Security 21, the
International Centre for Security and Justice, at the Australian National
University. His central interest is the involvement of non-governmental
institutions in furtherance of public policy.
CLIFFORD SHEARING is professor and Co-Director of Security 21, the
International Centre for Security and Justice, at the Australian National
University. His research and writing focus on developments in the
governance of security.
Dupont et al.—The governance of security in weak and failing states 349