Experimentation and innovation in police reform

Introduction

Although plural policing is now widely recognised as the norm in most developing (and developed) countries, this is rarely reflected in practical security and justice programming. ¹ In Timor-Leste, Solomon Islands and Bougainville, a wide range of actors provide ‘policing’ services. However, donor support remains almost wholly focused on the state police, despite the fact that they are seldom the first port of call for citizens seeking help. Overall, the results of years of international assistance have been underwhelming in terms of improving police performance, addressing the causes of insecurity or expanding the reach of police organisations to rural regions. These disappointing outcomes have been well documented in both policy and academic literatures in each of these places, as they have in many other parts of the world. ²

The literature has focused more on cataloguing failures than uncovering attempts to do things differently. Against a background of growing interest in innovation and hybridity in international policy discourse, ³ this article introduces three relatively rare instances of experimentation in plural policing practice in Australia and New Zealand’s near neighbourhood. These cases, all styled as ‘community policing’, involve setting up community–police councils in Timor-Leste, supporting a Community Officer (CO) pilot scheme in rural Solomon Islands and the Community Auxiliary Police (CAP) project in Bougainville that, in part, inspired it. The empirical material draws on evaluations and related research that both authors have been involved in. ⁴

We begin by reviewing some of the challenges of policing in so-called fragile and conflict-affected countries that have become a significant focus of international policy in recent years. We then look more specifically at the contexts of the three cases, identifying commonalities as well as the distinct social, historical and political endowments of each place. The bulk of the article is taken up with an analysis of each case study. Our conclusion sums up the findings and draws out the relevant policy implications.

Security governance in fragile and conflict-affected settings

There is now widespread recognition that policing is not, and never has been, the exclusive domain of uniformed police. Rather, it is provided by a range of actors, including state police, private security and an extensive constellation of community-based providers. This makes for a plural policing environment where security and order are, in effect, ‘co-produced’ by diverse and often fluid configurations of actors and everyday practices that vary considerably across space and time.

Acknowledgement of the plural realities in security provision in places like Timor-Leste, Solomon Islands and Bougainville has generated an assorted and sometimes confusing lexicon of phrases, which are often used interchangeably. These tend to be framed in oppositional terms such as ‘state and non-state’, ‘formal and informal’ and ‘modern and traditional’, and, among other things, this serves to obscure the critical interactions and relationships between them. Phrases in the academic literature that seek to overcome binary framings of the relations between different providers include ‘plural policing’, ‘hybrid security’, ‘multilayered security’ and ‘networked security’. ⁵ In the policing literature, the phrase ‘community policing’, despite its origins as a type of institutional policing and its innate elasticity as a concept, is being increasingly used in a broader sense to capture the co-production of security and order by the uniformed police in conjunction with other community-based actors. ⁶

Acquiring more nuanced and sophisticated understandings of security in these complex environments is a welcome development but does not make it any easier to formulate practical programming that can produce outcomes within the many imperatives and constraints of donor engagement. This is for a number of reasons. First, there remains a tenacious allegiance to an idealised Weberian conception of statehood premised on its monopoly of force. This is particularly prevalent within the community of practitioners, many of whom are either serving or former officers in metropolitan police organisations. Second, these individuals also tend to think and operate as technicians involved in a one-way process of institutional transfer. Living together in compounds and at considerable social distance from local populations, they often have only a superficial understanding of the environment around them. Related to this is an innate suspicion of irregular security providers, who rarely wear uniforms or belong to any recognisable accountability framework. Donors continue to ‘see like states’, ⁷ focusing on the realm of the formal with its familiar bureaucratic and organisational facade. In the administrative capitals of Dili, Honiara and Buka, the focus has remained on how the state and its machinery ought to work rather than how it actually does work. The progressive blurring of the security and development agendas over the past two decades has accentuated this predilection. ⁸ Such an approach is predisposed to identifying deficiencies in need of fixing rather than identifying existing strengths than might be leveraged. This has been recognised at a policy level by both the Australian and New Zealand governments but, for the reasons outlined earlier, has not consistently filtered down into programming. ⁹

Cumulatively, these factors help explain why the default position continues to be that which has now been critiqued so extensively, namely, a focus on perceived deficiencies in recipient police institutions. There remains a misalignment between the solutions suggested by the analysis and the current repertoire of responses available to donors, which are predicated upon notions of linear institutional development and the universalism of institutional forms. Police-building work is typically conducted by specially trained cadres of serving police (the dominant model in Australia and New Zealand) or by contracted expertise managed by large consulting companies. The focus is on addressing perceived deficits in organisational capacity rather than one derived from a context-specific understanding of how governance and security actually work in the society concerned. This ‘deficit’ approach draws attention away from local strengths that might be mobilised in forging locally relevant solutions to problems of insecurity. ¹⁰ Few incentives exist for innovation and experimentation, while there are many for continuing with business as usual. It is against this background that our case studies are set.

Contexts of intervention

As Bruce Baker has noted frequently in African contexts, security provision ‘beyond the tarmac road’ looks very different, as, indeed, it often does within many capital cities. ¹¹ This observation applies equally to our three case studies, Timor-Leste, Solomon Islands and Bougainville. Available research amply demonstrates the plurality of security provision in each place. State police and justice agencies do not have a monopoly, but operate alongside and within wider assemblages ¹² of other individuals, groups and security practices.

While there are significant differences between each context, there are also important similarities that make them broadly comparable. Each carries legacies from earlier periods of colonial domination that continue to manifest themselves in the post-colonial present. For example, each place has experienced different policing traditions that reverberate in practice and in memory: Portuguese and Indonesian traditions continue to shape the practice of the Timorese police: nostalgia for a form of policing embedded in indirect rule under the British continues to infuse imaginings of how a police service should operate in Solomon Islands; while negative recollections of the Royal Papua New Guinea Constabulary (RPNGC) influence local discussions about the future of policing in post-conflict Bougainville.

In these small, poor and predominantly rural territories, the resources and reach of the state are severely constrained. While the beneficiaries of substantial windfall revenues from resource development, both Timor-Leste and Papua New Guinea (PNG) have circumscribed fiscal and administrative capacities. In each place, state authority coexists alongside and interacts with myriad forms of local social organisation. Hybridity characterises contemporary social and political order, and in all three cases, custom, Church and state overlap and intermingle as sources of authority and everyday regulation.

Two of these relatively peripheral contexts have experienced major international interventions, one in Timor-Leste under the auspices of the United Nations (UN), and the other in Solomon Islands through a regional mission. Policing was a major component of both interventions, in terms of initial peace and stabilisation operations and in the elaborate state-building efforts that followed. The Timor-Leste and Solomon Islands cases conform to the larger global trend of liberal peacebuilding. Both had a strong emphasis on (re)building state institutions, including security and justice systems.

For small police services, these interventions have generated a relatively abundant literature, particularly in Timor-Leste, where there has been one article published for every 100 Timorese police officers. ¹³ This literature has highlighted the deficiencies of prevailing international approaches, contributing to a by-now familiar critique. Harder to discern are stories of innovation, experimentation and positive change. We have already rehearsed some of the reasons why this might be so. Perhaps another is the dominant research perspective that privileges the organisational and political centre, and national-level outcomes. Such a focus is relatively blind to local complexities, including activities going on in the periphery. Rendering that complexity visible requires different analytical tools and a historically informed approach.

In each of the cases we write about here, local experts with this knowledge, familiarity and nuance have played a significant role in shaping their design. These small and modest projects, in many ways peripheral to the main game of the intervention, are also characterised by a degree of experimentation and adaptation that is noticeably absent from accounts of the larger-scale engagements. Although each one has its own particularities, they all share a wider conceptualisation of policing, going beyond a discrete organisational form, function and ethos. This involves deeper and broader interaction, with policing embedded and linked with other systems of governance, administration and service delivery. Each case involves positioning policing within a larger system of government and representation, with the police, in effect, serving as forward agents of the state in highly contested processes of state formation. Inevitably there are tensions, especially within donor and policing organisations, between this broader conceptualisation and more conventional views of policing. Progress made within this space is often the consequence of committed individual champions within donors and local police organisations rather than reflecting the priorities of the larger programmes of which they are part.

Timor-Leste

Since 1999, when a UN transitional administration was established in the wake of the East Timorese vote for independence from Indonesia, the Timor-Leste case has been a mainstay in research and policy debates on police development in fragile and conflict-affected states. The template for reform endeavours has been liberal peacebuilding, with a focus on building state institutions. Over 10,000 UN police passed through Timor-Leste from 1999 until their eventual departure in 2012, trying to build up a national police service, the Polisia Nasional Timor-Leste (PNTL), which conformed to a Western model. A wide range of bilateral donors have also been involved. Australia has had a sizable programme since 2004 and New Zealand since 2009. Aside from these countries, a plenitude of other sources has funded training, study visits and symposia. It has been a truly global effort, ranging from courses at the Federal Bureau of Investigation (FBI) academy at Quantico to study visits of community policing in Bangladesh and almost everywhere in-between.

Scholarship on police development in Timor-Leste has focused on dissecting the efforts of the UN and bilateral donors, ¹⁴ with a consensus that there is very little to show for it all. One scholar characterised the approach adopted as one of ‘smoke and mirrors’, ‘smoke’ referring to a focus more on form than substance and the ‘mirrors’ connoting the observed phenomena of international police-builders wishing to develop a police force ‘in their own image’. ¹⁵

The focus on building liberal-democratic state institutions has been characterised as akin to ‘historical amnesia’ as it disregards the implications of both the country’s legacies – including the 24-year Indonesian occupation – and the realities of order maintenance in a still predominantly rural society. ¹⁶ Police development programmes paid less attention to alternate and localised sources of authority, which appear to have more substance and salience than offices of the metropolitan state in Dili. Nearly 10 years since the beginning of one of the most extensive state-building operations in the world, a 2008 survey found that only 12% of the public had come into contact with the Timorese police. ¹⁷ Security continues to be provided by an assortment of diverse security providers, not just the police. They include the leadership of Timor-Leste’s 442 village councils or sukus, and leaders who derive their authority from a combination of customary sources. As in Bougainville, another significant source of authority is derived from resistance to occupation. Groups known as seguransa voluntario and seguransa popular are often comprised of former resistance fighters, whereas other groups such as kablehan have origins in customary practices. Reflecting the diversity of Timor-Leste, groups operate in different ways from district to district and, oftentimes, suku to suku. According to a follow-up survey in 2013, people ranked citizens (51%) and community leaders (21%) higher than PNTL when asked who was primarily responsible for maintaining security within their locality. ¹⁸ The PNTL are a part of this security network, regularly using a range of different providers to extend their reach.

Few programmes have tried explicitly to engage with the plural realities of order maintenance in Timor-Leste. Commencing at the tail end of the large international presence in Timor-Leste, the Asia Foundation (TAF) programme is notable for trying to engage with this complexity. ¹⁹ Another innovation is that the programme is implemented largely by national staff rather than expatriate advisers.

Like the other case studies, the small TAF project is broadly framed using an elastic definition of community policing, a phrase in regular circulation in Timor-Leste. ²⁰

In implementation terms, the TAF project has entailed working ‘bottom-up’ with communities and, simultaneously, ‘top-down’ with the police institution. Activities include the establishment of community policing councils (Konsellu Polisiamentu Komunitária (KPKs)) and the provision of training to the lower tiers of the PNTL, a cadre that is frequently neglected by international police development programmes.

The most visible part of the programme is the KPKs, which cover about a quarter of the country’s sukus. In many respects, this is not a particularly new idea. Forums for the police and community to work together are staple features in many development interventions in societies with frayed or non-existent relationships between the police and community, and their value is the subject of continued lively debate here, as elsewhere. ²¹ What is innovative about this version is that the concept of a police community council has been adapted in a manner that aligns with the cadences and cultures of Timor-Leste, with its focus on consensus and group decision-making. Headed by the suku chief, the forums include representatives from the suku police, youth representatives and veterans. The KPKs ‘hear and resolve a variety of public safety concerns … such as fights, land disputes, theft and even domestic violence’. ²² The programme has collected data that point to the KPKs being used as vehicles for dispute resolution.

Although relatively early days, the KPKs appear popular and fulfil a community need. Focus group participants were extremely positive about having a forum for discussing issues of common concern. Many credited the KPKs as playing an important role in reducing community conflicts (by simply being there as a forum), although these claims appeared impressionistic and hard to verify. The KPKs also appear welcomed by the police, who view them as a way of sharing security provision. A police commander in the district of Baucau said: ‘It gives power and authority to the community to resolve the situation. And it makes them feel important. Many of the issues can be resolved according to tara bandu [customary law]’. ²³

The programme is working within the context of wider processes of administrative devolution in Timor-Leste. One particular unit of governance that is receiving a lot of attention is the suku, and forward deployment of state police is an important part of this overall move towards decentralisation. The strategy includes the establishment of 442 suku police officers (Ofisial Polisu Suku (OPS)) and is premised on the need for the PNTL to get closer to the people they serve. It is a popular initiative, supported by 97% of respondents in the 2013 TAF survey. ²⁴ Analogous to the CAP and CO to be discussed subsequently, these officers are conceived as the ‘front line’ of police interaction with the community. The TAF is among the organisations providing training support for OPS officers.

While the KPKs are the most visible features of the programme, there is engagement in a range of other activities, which perhaps are not as well heralded. The programme has supported the establishment of the Komisaun Diretiva Distrital (KDD), a forum to discuss pertinent issues at a district level. ²⁵ It is also supporting community outreach through television and radio and supporting training as well as other initiatives, with a focus on the lower ranks of the PNTL, a large subsection that police development initiatives have traditionally neglected.

Engaged, empowered and networked Timorese staff play an important part in ensuring that the programmes ‘fit’ with prevailing realities and are not considered an outside imposition. Unlike most police development engagements, which are dominated by expatriates, all but one of the staff on the Hametin Kooperasaun entre Kommunidade (HAKOHAK) initiative are Timorese. Through networks, contacts and a shared language, the staff are able to gain easy access to government officials and the police. This ‘Timor-first’ model contributes substantially to local ownership, engagement and, potentially, sustainability, as well as being a much cheaper delivery model.

An important part of the programme is the development and testing of hypotheses through evidence-based research. It has commissioned analytical pieces, managed a police–community survey and developed in-house contributions. The result is a substantial body of work that constitutes a valuable resource for researchers and a repository of practical information for donors engaging in this sector.

Solomon Islands

The small archipelagic nation of Solomon Islands was from mid-2003 the subject of a decade-long peacebuilding and state-building intervention with a strong accent on policing. The Australian-led Regional Assistance Mission to Solomon Islands (RAMSI) followed a low-level conflict, known locally as the ‘ethnic tension’. ²⁶ Government authority had broken down, with serious lawlessness in some areas and looming national bankruptcy. The Royal Solomon Islands Police Force (RSIPF) fractured along ethnic lines and effectively collapsed. Australian policymakers viewed these developments primarily through the global lens of state failure and the threat this posed to Australian and regional security. ²⁷

Conforming broadly to liberal peace interventions elsewhere, RAMSI was also unique in being a police-led mission. Its mandate was to restore security, repair the machinery of government, improve economic governance and build strong and peaceful communities. At its height, it comprised around 330 front-line police (known as the Participating Police Force (PPF)) from Australia, New Zealand and other Pacific Island states, supported by 1,800 military personnel. Civilian advisers were placed across a range of central government agencies. The PPF initially provided executive policing before assuming a predominantly advisory role with the RSIPF aimed at building ‘a high quality, professional and credible policing service’. ²⁸ This work entailed a familiar repertoire of training and mentoring programmes, physical infrastructure and equipment, and managerial and accountability systems.

RAMSI entered its transition phase in 2011, with the graduated drawdown of international personnel and resources. ²⁹ Development activities were absorbed into regular donor-managed aid programmes, and since 2013, RAMSI’s remaining presence has been a slimmed-down PPF advising the RSIPF. While there have been considerable achievements, the need for continuing PPF support indicates the slow and non-linear character of police-building. Solomon Islanders continue to lack confidence in the RSIPF and demonstrate anxiety about safety and security. ³⁰ There are also concerns about the sustainability of gains made under RAMSI as external assistance diminishes. With most of this support directed at rebuilding a centralised and urban-oriented model of policing, meeting the security needs of rural citizens remains another concern.

With the notable exception of the modest CO project, there have been few attempts to engage with the informal approaches to community security prevalent in rural Solomon Islands. Initiated by the RSIPF, with support from the PPF, on a pilot basis in late 2009, COs were appointed in 23 rural locations, covering every province. These locally based officials were lay persons appointed by the RSIPF in consultation with village chiefs. Their role was to act as liaison between the police and local leaders, reporting serious crimes upwards while working closely with community leadership to resolve minor disputes. The pilot was informed by recognition of the thin and uneven presence of state police in rural areas, the important role performed by local leaders in everyday order maintenance, and the need for better linkages between these different sources of authority and regulation. RSIPF and PPF officials behind the pilot were also aware of the old area constable, an official who performed a similar role in Solomon Islands’ recent past (see later), as well being familiar with the CAP in neighbouring Bougainville (see the following section on Bougainville). An evaluation of the short-lived pilot revealed its positive reception among rural communities and the RSIPF, and, more significantly, its potential as a mechanism for linking and coordinating plural policing and justice provision in rural areas. ³¹

An understanding of Solomon Islands’ contemporary context and recent history is critical to considering the significance of this relatively rare policing ‘experiment’. The island nation shares the socio-linguistic diversity and geographical fragmentation found in the wider Melanesian region, with its population of half a million people dispersed over six main islands and across 80 language groups. Around 85% of the population lives in rural communities that are based on complex interplays of kinship and exchange relations, friendships, Church membership, and myriad claims to customary land. Government resources and personnel remain concentrated in the national capital, Honiara, and a number of small provincial towns.

Government was already viewed by many Solomon Islanders as largely absent from rural areas well before the outbreak of the tension in 1998. ³² This perception of ‘state withdrawal’ has often been combined with nostalgia for the old British colonial system of indirect rule. The latter is viewed as having at least provided some tangible government presence in rural localities. As well as vertical links between the distant administrative capital and rural periphery, these reminiscences recall horizontal linkages between government officials and traditional leaders that enabled a degree of coherence between different forms of authority. Remnants of this old system were carried over after independence in 1978. These included elected Area Councils, which provided the basis of a rudimentary system of local-level government. Local courts dealt with a range of minor criminal, civil and ‘customary’ cases, while ‘area constables’ afforded a local police presence, helping to enforce council by-laws and local court decisions. These officials also served as critical intermediaries between traditional leaders, council areas and the wider state justice system.

Local-level government disappeared with the suspension of Area Councils in 1998, as did area constables in most places, while local courts became increasingly moribund in the absence of adequate government support, hearing only land cases rather than the range of local disputation managed previously. This is the context of local narratives of state withdrawal, including the perceived decoupling of state policing and justice from local community-based approaches. RAMSI’s top-down approach, with its focus on rebuilding central agencies in Honiara and relative neglect of rural areas, has done little to counter these perceptions.

As in Timor-Leste and Bougainville, most rural citizens continue to rely on community-based approaches for everyday dispute resolution and security. The prevalence and diversity of these local approaches are documented in recent research undertaken on behalf of the Solomon Islands’ Ministry of Justice and Legal Affairs. ³³ This research –Justice Delivered Locally (JDL) – identified three broad and overlapping systems of rules in rural areas, distinguished by the different sources of authority upon which each is primarily based – namely, state, Church and kastom. ³⁴ Rather than being uniform or discrete, these systems exist in diverse and relational configurations shaped by continuous interaction. Each is seen as having core spheres of operation. State justice tends to be viewed as the appropriate way of dealing with serious crimes and disputes. Kastom is preferred in the case of more socially embedded disputes, such as land matters, whereas Church-based approaches appear to be preferred in marital and family disputes, especially by women.

The JDL research also revealed how these local approaches are coming under increasing stress in light of larger processes of globalisation. New forms of contestation have emerged that local approaches struggle to manage. Four main types of contemporary disputation were identified: social order problems, usually associated with substance abuse; development and land-related disputes; disputes arising from non-governmental organisation (NGO), donor and government projects; and marital conflict and domestic violence. The standing of chiefs and other local leaders was seen as increasingly compromised in many areas owing to their involvement in controversial business and development schemes.

Despite the growing fragility of local approaches, the JDL work found evidence of extensive experimentation as communities seek to adjust and strengthen their coping strategies. These generally involve efforts to forge better linkages between different approaches and, in particular, between kastom and state justice, while retaining a measure of local control. They represent an attempt to reverse the decoupling noted earlier but, in contrast to the centralising tendencies of government and RAMSI, emphasise the need for alignment with existing local approaches, rather than their replacement. These local demands express popular aspirations for more effective articulation between local and state approaches that enhance the capabilities of both, and, more broadly, support for locally driven or ‘bottom-up’ approaches to security and development.

The positive reception afforded to the CO scheme in the rural communities where it was trialled needs to be seen within this larger context of social and political change. Community support provided another indication of local wishes to re-establish connections with the state system that were perceived as having largely disappeared over recent decades. One villager told the evaluation team that the CO marked ‘the return of government’ while another remarked that ‘the CO makes us feel that the government is now with us’. ³⁵ In large part owing to the failure of the RSIPF to provide adequate support, the role of the CO evolved organically over the lifetime of the scheme, adapting to the diverse configurations of local authority structures and priorities in different places. While retaining a quasi-policing orientation in most places, the CO operated as an enforcer of chiefly authority in some localities, while adopting a more direct dispute resolution role in others and a broader community governance role in yet others.

The evaluation found that the mere presence of the CO contributed to community perceptions of improved safety. With the RSIPF unable to deliver the anticipated support, it was the prospect that law enforcement might eventuate that appeared to vest the CO with a particular authority not shared by other local leaders. In this regard, the CO was able to evoke ‘the shadow of the law’, if not its substance. Another important finding was the catalytic potential of the CO as an agent of positive social change. Although broadly supportive of local power relations in most places, there was also evidence of individual COs working in subtle and creative ways to moderate the exercise of chiefly power and represent the interests of weaker groups, notably, women and youth. While it is important not to overstate the impact of this extremely modest and short-lived scheme, it nevertheless provides a tantalising glimpse of a local policing innovation with the potential to transform existing institutions and power relations, rather than simply reinforce them.

Bougainville

Part of the same island chain as Solomon Islands, Bougainville shares many sociocultural and geographic similarities with its southern neighbour. Its population of around 300,000 is overwhelmingly rural-based, and highly localised forms of social organisation and authority provide the basis for everyday governance and regulation. Deep historical grievances have contributed to a strong sense of Bougainvillean identity and a tradition of resistance to colonial and other forms of domination. ³⁶ There was strong local opposition to the construction of the large Panguna copper mine in the late 1960s, as well as to Bougainville’s subsequent incorporation into the newly independent state of PNG in 1975. While the establishment of a system of provincial government relieved tensions at the time, they resurfaced dramatically in the late 1980s and Bougainville experienced a bloody and protracted conflict from 1989 to 1997. A local disagreement over benefit-sharing arrangements at the Panguna mine escalated into a secessionist struggle with PNG; this was, in part, a result of heavy-handed tactics by the PNG police and military. In its later stages, the conflict fragmented into multiple small-scale disputes between various local factions.

The peace process that began in 1997 was largely locally initiated and led. International assistance comprised a small UN political office and unarmed truce, and, later, peace monitors from New Zealand, Australia and neighbouring Pacific Island states. In contrast to the bigger engagements in Timor-Leste and Solomon Islands, that in Bougainville has been described as a ‘light intervention’. ³⁷ A comprehensive peace agreement was signed in 2001 under which Bougainville was granted a unique status as an autonomous region within PNG, as well as the prospect of a referendum on its future political status within a 10- to 15-year period. The newly autonomous region adopted its own constitution and established an autonomous government in 2005. Various powers were to be transferred from the national level to the autonomous government, including those in relation to police, courts and criminal laws. ³⁸

Emerging from a devastating decade-long conflict, Bougainville’s leaders were intent not only on forging a different political relationship with PNG, but also on reconfiguring state–society relations within Bougainville itself. Bougainville’s foundational documents ³⁹ envisaged a ‘hybrid political order’ ⁴⁰ that could accommodate and build on local strengths, including ‘traditional’ forms of leadership and authority. This vision was particularly evident in deliberations around the future of policing and justice. ⁴¹

The preference for hybrid governance manifested early in the peace process with the establishment of Councils of Elders (CoEs), initially as an instrument of reconciliation and now the second tier of government in the autonomous region. CoEs are authorised to maintain security, settle disputes and facilitate reconciliation in designated council areas. Forging closer links between (state and non-state) actors at different levels of Bougainville society was seen as critical to the development of more socially embedded approaches to policing and justice.

Most existing police infrastructure was destroyed during the conflict. In early 1998, Bougainville authorities requested international assistance to train community-based police. This resulted in the Bougainville Community Policing Project (BCPP), jointly funded by Australia and New Zealand before being taken over by the latter. An early focus was on the establishment of community-based officers, later known as the ‘Community Auxiliary Police’, who were to be supervised by the newly named ‘Bougainville Police Service’ (BPS), made up of remaining Bougainvillean members of the RPNGC. ⁴² The new name was suggestive of the strong community orientation envisaged for the BPS and a decisive break from the approach attributed to their RPNGC predecessors. As the Bougainville Constitutional Commission stated:

it is also clear that people do not want to go back to the kind of situation that existed before the conflict, when there was a concentration of large groups of police in the urban centres, and little effort to integrate police into communities. ⁴³

Donor efforts to reorient and develop the BPS have relied on international advisers on six-month deployments and have engaged in various ‘capacity-building’ activities such as mentoring, preparing and delivering training modules, and writing procedural manuals. The BPS has been described as ‘a weak institution, suffering poor morale, bad discipline and low effectiveness’, ⁴⁵ with these characteristics attributed to inadequate numbers and resources, as well as deficiencies in leadership, management, accountability and professionalism. Public confidence in the BPS remains low, reflecting, in part, antipathy to what many view as the continuation of the old RPNGC policing approach. This perception is particularly marked in the southern region, where the BPS is often referred to disparagingly as the ‘PNG police’. ⁴⁶

By contrast, the BCPP’s engagement with the CAP has demonstrated more promising results. International advisers have adopted an enabling rather than a prescriptive approach and have attempted to work with the grain of local beliefs and practices. Significant time spent in the country has also helped attune key programme staff to on-the-ground realities. The BCPP has progressively extended the presence of CAP officers in different parts of Bougainville, providing greater access to basic policing services than ever before. It has also placed a priority on female participation in recruitment courses, leading to a reported increase in women recruits from 5% in 2008 to 21% in 2012. ⁴⁷ This provides aggrieved women with the opportunity to interact with female CAP officers regarding sensitive issues, such as family and sexual violence. This is also a practical example of how one of the obvious risks of engaging with ‘traditional’ justice actors can be mitigated. Instead of simply reinforcing existing leadership structures and power relations, the CAP, like the COs in Solomon Islands, can potentially be mobilised as a catalyst for social change.

Whereas the institutional form of the ‘new’ BPS makes few obvious concessions to Bougainville’s social and geographical endowments and recent history, the CAP aligns more closely with local aspirations for socially embedded policing. It constitutes a hybrid model, evident in its alignment with community leadership and governance structures and the different sources of authority and legitimacy on which it is able to draw. Although they are sworn members of the BPS, and bound to uphold and enforce state law, CAP officers are nominated by village chiefs and their candidacy has to be approved by the local CoE. In this regard, they are potentially critical intermediaries between different social orders, with a foot in both camps. ⁴⁸ This hybrid quality also means that the CAP is, in theory at any rate, subject to different layers of accountability. At the local level, this entails that exercised by the communities in which these officers live and work, including the relevant CoE. Accountability is also exercised through the CAP itself, notably, by the international advisers who appear to act as the final arbiter of individual performance and discipline within the CAP. ⁴⁹ Evidence from the recent evaluation undertaken by the present authors indicates that a much higher level of accountability applies in the CAP than in the BPS, where few of the serious allegations made against officers are ever followed up. ⁵⁰ CAP’s important policing role is also demonstrated by preliminary research, which indicates that it is the source of most of the incidents (around 85%) reported to the BPS. ⁵¹

While it is important to acknowledge the innovative qualities of the CAP initiative, it is also clearly necessary not to exaggerate its potential, particularly given the many challenges facing Bougainville as it moves forward. Conflict stresses are re-emerging and these are likely to increase in light of population growth, urbanisation and rising popular expectations that the weak autonomous government simply cannot fulfil. The economic realities of running an autonomous region, let alone an independent state, provide the backdrop to current discussions around the resumption of mining. This prospect, along with the looming window for holding the referendum on Bougainville’s future political status, is already generating high levels of contestation and potential for conflict. The availability of weapons left over from the conflict adds to the risks. An effective security capability to manage this situation is clearly needed. The main strength of the CAP, as we have argued elsewhere, ⁵² is its potential role in the longer-term process of state formation and the contribution it might make to helping build the kind of hybrid political order envisaged by Bougainville leaders.

Conclusion

In contrast to much of the literature on police development, which tends to focus on the failings of international approaches, this article has sought to illuminate relatively rare examples of attempts to do this work differently. Our claims in these cases are modest and provisional. They do not present definitive solutions to complex problems of security and development, nor are they necessarily scalable and replicable. In so far as they evince success, it is because they are context-specific rather than being drawn from an international ‘toolkit’.

That said, we believe that these cases of experimentation and innovation are significant and merit closer study by scholars and practitioners of conflict and fragility than can be gleaned from a one-off evaluation. These programmes have sought to engage directly with the plural realities of these socially complex and fragmented environments, characterised by the entanglement of myriad forms of order and authority. Rather than simply trying to transplant familiar institutional solutions and hoping for the best, the entry point in each case has been an acknowledgement of the messy and entwined reality of plural policing. In this respect, the implications of our study go well beyond these three places and have resonance for the fragile and conflicted states that have become such a high priority for international interventions in recent times.

In each of our cases, we have shown a disconnection between the administrative and political centres and the rural peripheries where the majority of the population live. Donor programming nevertheless continues to privilege the centre, thereby engaging in partial, limited and, sometimes, counterproductive ways. The focus on state authority to the exclusion of other sources of power and legitimacy that continue to prevail in places like Timor-Leste, Bougainville and Solomon Islands means that a large part of the realities of everyday security governance are simply not seen. An important strength of each of the three projects examined is their potential for linking along the vertical and horizontal axes, which can help break down the pronounced urban–rural divide. Similarly, these endeavours can also contribute to the fostering of productive relations across the multiple social orders found in such places. In contributing to the extension of the functional authority of the state, these community policing initiatives can become catalysts for state formation and consolidation in environments where limited statehood is increasingly recognised as a major source of conflict and instability. Perhaps the most significant contribution of these modest community policing experiments ultimately lies beyond the realm of policing in relation to the larger process of political and social change.