June 2018 Research Report Promoting equity and managing conflict in development Safety and Security at the Edges of the State: Local regulation in Papua New Guinea’s urban settlements David Craig and Doug Porter  1 Justice for the Poor is a World Bank research and development program aimed at informing, designing and supporting pro-poor approaches to justice reform. It is an approach to justice reform which sees justice from the perspective of the poor and marginalized, is grounded in social and cultural contexts, recognizes the importance of demand in building equitable justice systems, and understands justice as a cross-sectoral issue. Justice for the Poor research reports are aimed at development practitioners, partner governments, researchers, and others interested in justice reform. Research reports are reviewed by at least two external referees who are independent of the program. © 2018 International Bank for Reconstruction and Development / International Development Association or The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Disclaimer This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclu- sions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. For permission to reproduce any part of this work for commercial purposes, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Contact details Justice for the Poor Justice Reform Practice Group Legal Vice Presidency World Bank 1818 H Street, NW Washington, DC 20433 USA Email: j4p@worldbank.org Cover photo: Sunday afternoon after church, the Lae 5 Mile komitis (including 2 women komitis) convene. David Craig/Worldbank 2 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS June 2018 Research Report Safety and Security at the Edges of the State: Local regulation in Papua New Guinea’s urban settlements David Craig and Doug Porter Papua New Guinea Urban Safety and Security ASA Activity Justice for the Poor (J4P) Trust Fund Acknowledgements This report was prepared by David Craig, Senior Government Specialist, and Doug Porter, consultant to Urban Safety and Security for the World Bank’s Advisory Services and Analytics activity. Virginia Horscroft was the task team leader. Caroline Sage and Doug Porter were the previous task team leaders, with support from Debbie Isser and Rob Taliercio. Research in Lae was assisted by Levi Johnston and Don Savi and facilitated by many members of Lae’s urban settlement leadership. This report draws on earlier work in Port Moresby to which Fiona Hukula, Rabura, and Keke Aiga contributed. The work was funded by Australia’s Department of Foreign Affairs and Trade (DFAT) through the Justice for the Poor Trust Fund. None of this work would have been possible without the constant, warm and positive engagement of urban settlement, mediation and law and justice leaders, in both Port Moresby and Lae. These leaders, VC magistrates, women leaders and komitis of both genders welcomed the team, included us in events, talked and discussed openly at all times about what they were doing, kept us safe, and provided no obstacle when we needed to triangulate regarding komiti activities with other sources. Special thanks to Micah Yosman, Philemon Boddeh, Tony Miria, Andrew Kuno, Chris Waima, Doreen Yak, Peter Simbago, Esther Toss, Grace Laman, Helen Illitavalo, Hiob Awagasi, Anthony Guken, Francis Kanasi, James Kinjin, James Mukale, Judith Gidisa, Laura Dawa, Leo Raka, Mark Gigmai, Nalau Agi, Oscar Aiya, Michael Nakut, Sam Temiel, Zuabe Tinning, Mote Zaza, Anna Zerica. Contents Acknowledgements.................................................................................................................................................. II 1. Introduction: safety, security, and investment in Papua New Guinea’s urban future......................1 1.1 CITIES AND URBAN REGULATION ARE KEY TO PNG’S FUTURE......................................................................................................................1 1.2 PNG’S URBAN DEVELOPMENT AND REGULATION CHALLENGES..................................................................................................................2 1.3 THIS REPORT’S INSTITUTIONAL CAPABILITIES APPROACH TO UNDERSTANDING URBAN REGULATION...................................................5 1.4 BRIEF BACKGROUND, SITES, AND METHODS OF THE REPORT.....................................................................................................................6 2. PNG’s urban political economy and local settlement-regulating institutions .................................7 2.1 A BRIEF OVERVIEW OF THE HISTORY OF PNG’S URBAN POLITICAL ECONOMY AND GOVERNANCE ......................................................7 2.2. THE RISE OF INFORMALITY, CONFLICT, AND CRIME, 1980–2005...................................................................................................................9 2.3 BEYOND CRISIS? NEW FORMS OF URBAN LEADERSHIP, AUTHORITY, INVESTMENT, AND INSTITUTION, 2005–17.................................10 Local regulatory capabilities in urban settlements: what exists and how does it work? .............14 3.  3.1 KOMITIS AS SITES FOR COLLECTIVE ACTION ...............................................................................................................................................14 3.2 LOCAL REGULATORY FORM AND FUNCTION: THE KOMITI SYSTEM...........................................................................................................16 3.3 KOMITIS’ EMERGENCE AND LEGITIMACY AS AN ETHNICALLY AND LOCALLY INCLUSIVE INSTITUTION.................................................18 3.4 KOMITIS AS LEADERS: AGENCY, PERFORMANCE, GREATER RECOGNITION..............................................................................................20 3.5 COORDINATION AND REACH: KOMITIS AND THE ASSEMBLING OF FRAGMENTED LOCAL AUTHORITY................................................22 3.6 WEAK COOPERATION: WHERE KOMITIS ARE UNFORMED AND DIVIDED...................................................................................................23 Local regulatory capability: the four principal regulatory functions of komitis and lidas..............25 4.  4.1 FUNCTIONAL CAPABILITY 1: REGULATION OF SETTLEMENT AND TENURE: ESTABLISHING BASIC SECURITY OF TENURE AND PHYSICAL CAPITAL OR PROPERTY ....................................................................................................................................26 4.2 FUNCTIONAL CAPABILITY 2: SOLVING DISPUTES, BUILDING TRUST: PROTECTING SOCIAL AND LEGAL-POLITICAL CAPITAL...............28 4.3 FUNCTIONAL CAPABILITY 3: CREATING LOCAL ECONOMIC SECURITY AND OPPORTUNITY AT MARKETS: PROTECTING ECONOMIC CAPITAL................................................................................................................................................................31 4.4 FUNCTIONAL CAPABILITY 4: POLITICAL AND LEGAL REPRESENTATION AND COORDINATING WITH OTHER LOCAL REGULATION: PROTECTING LEGAL/POLITICAL AND SYMBOLIC CAPITAL ........................................................................36 5. Power asymmetries, outcomes, and the limits of local regulatory reach......................................39 5.1 THE LIMITS TO KOMITI’S AUTHORITY ............................................................................................................................................................39 5.2 ETHNIC ASYMMETRY IN SETTLEMENT REGULATION: HIGHLANDER HEGEMONY ....................................................................................39 5.3 THE EXCLUSION OF YOUTH............................................................................................................................................................................41 5.4 THE KOMITI’S GREATEST FAILING: PERPETUATING WOMEN’S INSECURITY...............................................................................................44 6. Local Regulation: komiti and lida relations with police and subnational authorities ....................49 6.1 WEAK VERTICAL LINKAGES.............................................................................................................................................................................49 6.2 KOMITIS’ DIFFICULT RELATIONS WITH POLICE: MUTUAL ACCUSATIONS OF DISRESPECT, AS WELL AS PRACTICAL DYSFUNCTION AND DANGER.........................................................................................................................................................................49 6.3 EMERGING DISTRICT AND CITY AUTHORITY: NEW OPPORTUNITY FOR LOCAL GOVERNMENT LINKAGES?...........................................53 What is the significance and potential of komiti and lida regulation of settlements? 7.  Conclusions and implications ....................................................................................................... 55 7.1 SUMMARY CONCLUSIONS...............................................................................................................................................................................55 7.2 PRACTICAL IMPLICATIONS FOR URBAN ENGAGEMENTS: “DO NO HARM” ..............................................................................................56 7.3 OBSERVATIONS AND ADVICE FOR URBAN DEVELOPERS: CARRYING “DO NO HARM” INTO PROJECT PRACTICE................................58 7.4 MPLICATIONS FOR THOSE WHO WANT TO STRENGTHEN KOMITI AND LIDA CAPABILITIES AND PERFORMANCE ..............................59 7.5 A FUTURE RESEARCH AGENDA.......................................................................................................................................................................61 Annex 1: Different Komiti Structures in Lae Settlements....................................................................................... 62 Annex 1: Research Methods.................................................................................................................................. 64 Bibliography........................................................................................................................................................... 66 III Safety and Security at the Edges of the State: Local regulation in Papua New Guinea’s urban settlements IV S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Introduction: safety, security, and investment 1.  in Papua New Guinea’s urban future 1.1  CITIES AND URBAN REGULATION ARE KEY wealth in human resources, services, assets, and real TO PNG’S FUTURE estate. Urban growth will also deliver a demographic dividend5 in reduced birth rates and enable the 1.1.1  Cities are key to reducing poverty and “youth bulge” to be absorbed into productive activ- promoting shared prosperity in Papua New Guinea ity. Another benefit will stem from enhanced service (PNG). Cities generally are sites for cultivating delivery as a result of the economies of scale pos- and fostering the accumulation of multiple forms sible through urban concentration. And urban voter of capital.1 Citizens and businesses create physical power may very likely bring a shift from governance or economic capital as they engage in powerful via resource rent patronage to political structures that urban markets and invest in assets, homes, and respond to demands for population-wide services infrastructure. They produce cultural or human capital and institutions to deliver them.6 by investing in education and developing skills in sectors that are largely absent in rural areas. As they 1.1.3 Nurturing all of these forms of capital and build new relationships and urban identities, cities turning them into development outcomes require foster development of social capital in its most prized security and regulation. People need to feel secure and economically valuable “thin” or bridging forms,2 in going about their business: and that they can enabling trust and reciprocity over ethnic difference trade and exchange within clear rules. Security and and kin preference. As city arrivals represent their regulation are needed not just for business transac- interests and resolve conflicts, customary law gets tions, but also within the many diverse formal and translated into new contexts, creating legal and informal7 institutions that regulate the social contexts political capital, which is essential to safety and in which markets are embedded. Rules and institu- rights.3 Finally, reputation, recognition, identity, tions need to govern the intense market and social and civic pride are valuable forms of (positive and contests associated with transactions over land and negative) symbolic capital,4 and they work to either settlement, commercial activity, and political affairs;8 attract or deter further investment. and to do this well they need to be adaptive and inclusive. In PNG, the institutions responsible for 1.1.2  In urban areas, these different kinds of capi- handling the inevitable contest and competition tal can bring economic, social, and political ben- involved in this rural to urban transition will need to efits to national development. In the case of PNG, bridge the divides between the country’s myriad urban growth will help the economy to diversify, ethnic groups (involving 840 languages and 10,000 reducing dependence on subsistence rural produc- clans); between rural village-, ethnic-, and kin-based tion and extractive industries, and also to expand justice practices and modern urban and national 1  World Bank (2017a). See also Bourdieu (1986); Agostini et al. (2007), 6. 2  See Lancee (2010); Putnam (2000). 3  See Bourdieu (1986), 241–58; Agostini et al. (2007), 6. 4  See Bourdieu (1981). 5  Mason (2007). 6  See Barma et al. (2012); Auty (2007). The distinction between formal and informal in urban PNG needs to be seen as both a modest descriptor and also a powerful political tool in 7  characterizing, and governing people and places. An examination of how this works in urban governance beyond the partial review provided in this report is in McFarlane (2012). In this report, governance refers to “the way a society organizes to use power to manage public resources, involving the making and 8  implementing of collective decisions, enforcement of rules and resolution of conflicts” (Kassimir 2001), 98. I ntr o ductio n: s af ety, securit y, and invest ment in Papua New Guinea’s urban f u tu re 1 conceptions of human rights and universal laws; gender and age lines, raising costs for the vulnerable and between patriarchal and often gerontocratic in local marketplaces. It reinforces mistrust, exclusion, customary arrangements and the wider emerging patronage, and corruption and denies the vulnerable roles for urban women and youth. Regulatory mech- access to justice. In settlements and at the wider urban anisms need to deal with the ways even minor dis- and country level, failed regulation creates stigma, putes (for example, between children fighting) can destroying reputations and deterring investment. rapidly escalate into widening ethnic conflict that resonates not just across different city settlements but in distant Highland rural communities also. 1.2  PNG’S URBAN DEVELOPMENT AND REGULATION CHALLENGES 1.1.4  The social and economic regulation of informal urban settlements in PNG needs to be 1.2.1 Urbanization in PNG has been delayed to expressed territorially and spatially9 in residential an exceptional degree, but as elsewhere, it is an neighborhoods, public spaces and amenities, and unstoppable feature of national life. It has a long transport nodes and routes.It especially needs to be way to go, however; currently at around 14 percent,13 present in the places where new urban settlers arrive, PNG’s official urbanization figures place it 202nd in the places Doug Saunders calls “Arrival Cities,”10 the world14 and at less than one-third of the global which are essentially settlements located within or average. Nevertheless, some important drivers of at the growing edge of existing cities, characterized urbanization are strengthening and the transformation by mobility, volatility, and cultural transformation, will inevitably have enormous consequences. If PNG as arrival groups mingle and compete. For PNG’s were to reach only the current Pacific average of 50 forthcoming rural to urban transition to succeed in percent by 2050, the total urban population would these crucial areas, new arrivals will need the same be 9 million, which would mean that the capital Port things that urban dwellers everywhere need: entry Moresby, currently approximately 750,000 people, level housing, economic opportunities, access to would exceed 3 million and Lae, the country’s second- transport, and legal-political representation.11 “Arrival largest city, located in Morobe province (currently City” development should occur in ways that remove 250,000),15 could reach 2 million.16 the stigma from younger newcomers and instead help them become assets to city development rather 1.2.2.  Urbanization works best when the nation’s than scapegoats for urban dysfunction. demographic transition (declining death rates followed by declining birth rates) synchronizes 1.1.5 Regulatory failure, on the other hand, can with the urban transition, and as a result, growing, lead to communal disputes and escalating violence mobile populations of young people are able to at all levels that pervert and destroy capital and access better opportunities in cities. Urbanization threaten national stability.12 In urban areas, poor in PNG will be driven by population increases that quality regulation creates and reproduces division lead to land pressures in rural areas, but the “pull” and inequality between ethnic groups and along and integration factors of dynamic urban markets are 9  See Goodwin, Duncan, and Halford (1993). 10  See Saunders (2011). 11  See Baker and Gadgil (2017). 12  See Slater (2010). 13  This is likely to be a considerable undercount. See Jones and Kep (2012). 14  For urbanization figures, see http://www.indexmundi.com/papua_new_guinea/urbanization.html. See also UN (2014). Jones and Kep (2012). Official census data put Port Moresby’s population at just 318,000, and growth between 2000 and 2022 at just 8,400 15  persons. Estimations of growth related to unplanned settlement suggest 500,000–750,000 persons, with upper estimates at 1 million. Census figures for Lae in 2011 are 119,178, but this excludes large sections of the city located outside Lae District boundaries, including 7 Lae Ahi rural wards. 16  Current UN estimates for PNG are for just 23 percent urbanization by 2050, which would make it a serious outlier. But this figure underestimates the actual current level and assumes rural areas will absorb high (currently 3.1 percent) rural growth. See UN (2014); Montgomery (2008), 761. 2 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS often missing in the country.17 Missing too has been 1.2.4 Local regulation of urban settlement: most urban development planning and the provision of important, least understood? The governance of infrastructure capable of embracing and integrating urban settlements is perhaps the least understood settlers. PNG thus faces the difficult prospect of feature of contemporary life in PNG. Its cities’ sprawl- urbanization without urban development, the kind ing, diverse settlements have been highly stigma- described in Mike Davis’ Planet of Slums,18 where tized, but they have not been entirely ungoverned people arrive in large numbers driven by the impact and unregulated. They are perhaps best understood of wider economic downturns and adjustments that as “areas of limited statehood”25 (ALS), with the pro- also restrict opportunities in urban marketplaces. viso that “ALS are not ungoverned spaces, but ‘dif- ferently’ governed.”26 What and how different that 1.2.3 PNG has already experienced a traumatic governance is, what forms, functions, and processes period of urbanization without corresponding it involves, and with what consequences for different urban development. The prospect of more of sections—gender, age, wealth, and ethnicity—of the the kind of urbanization that occurred in the urban population, is the primary focus of this report. 1980s and 1990s is alarming, since PNG’s cities have consistently been ranked among the world’s 1.2.5 Official public authorities in these spaces, least livable.19 The country’s institutional short- including police and justice services, local govern- comings are conspicuously evident in the failure ments and municipal administrations, and service of formal state institutions to provide municipal delivery sectors, appear to behave with the hallmarks infrastructure and services, and to plan for and of “frontier” overstretch, demonstrating rogue regulate commercial activities to offer opportunities dysfunction and the failure to regulate consistently for the youth, underemployed, and swelling numbers over time and across different agencies and areas of of new arrivals.20 As elsewhere,21 urban settlements the city. In these spaces, for all that state governance have long been represented as “the pariah edge”— appears disorderly, neglectful, unpredictable, and locations of dysfunction and deficit, uniformly unaccountable, it also typically coexists with the degraded sites of illegality and violence22 and equally variable legitimacy and effectiveness of other “no-go zones” for police and government services.23 forms of authority, whether based in custom, religion, “Settlement” has become a stigmatizing shorthand, ethnicity, or the new kinds of authority that are the suggesting places that produce crime and violence product of the new urban experience.27 perpetrated by workless, predatory ”raskols” and youth who destroy the safety and the various forms of capital of other urban dwellers and impose costs on business that impede growth across the economy.24 PNG’s exceptionally low urbanization has historical causes. Neither migration “push” factors (dense rural populations and scarcity or 17  enclosure of land) nor “pull” factors (availability of formal- and informal-sector opportunities in work, housing, transport, and political representation) are yet operating at societal scale. But rural population growth above 3 percent means that the population will double in 20 years, and there will be severely limited opportunities for international labor mobility. 18  Davis (2006). In 2004, Port Moresby was ranked as the least livable city in the Economist Intelligence Unit’s annual list. It improved to the third worst in 19  2014 and to the fifth worst in 2016. For the most recent rankings, see The Economist Intelligence Unit, “The Global Liveability Report 2017,” http://pages.eiu.com/rs/753-RIQ-438/images/Liveability_Free_Summary_2017.pdf. 20  See Saunders (2011). 21  See Davis (2006), 20–49. 22  Goddard (2005a). See, for example, William Kep, “Port Moresby: A Squatter Settlement” (blog), August 2, 2016, http://www.pngblogs.com/2016/08/port- 23  moresby-big-squatter-settlement.html. 24  See Lakhani and Willman (2014a, 2014b). Korf et al. (2018), 167. Korf et al. note that ALS spaces are usually understood as “domains where state capacities to control the means of 25  violence are regarded as either weak, privatized or non-existent.” 26 Ibid. 27  See Rasmussen and Lund (2018); Watts (2018). I ntr o ductio n: s af ety, securit y, and invest ment in Papua New Guinea’s urban f u tu re 3 1.2.6 In these edge-of-state settings, urban resi- gence of a new generation of urban leaders has dents with sufficient means may be able to solicit triggered a much closer review of the quality of insti- the amenities of state governance and augment tutions responsible for regulating social order and this through private security, health, education, settlement and economic activities and for handling and other services. The majority of urban residents administrative functions and political authority.29 lack these means, however, and at the same time Recent legislation30 is one signal of this attention that are most vulnerable to the unreliability and often could potentially support improved urban regulation, predatory behavior of formal state governance.28 administration, and political representation. This They must therefore pact together and invest in mandates that the cities of Kokopo, Lae, and Mount creating and sustaining local mechanisms through Hagen join Port Moresby, which has a more powerful which to protect their ability to accumulate capital. city commission, as stand-alone regulatory authori- As this report will show, people are prepared to invest ties, potentially with clear, urban jurisdictions, service a major, and at times astonishing, quantity of their delivery mandates, and revenue sources. time and capital in these everyday forms, functions, and regulatory processes in order to stretim hevi [in 1.2.8 As this report will describe, the chief insti- Tok Pisin (PNG’s most widely used language): solve tutions of local regulation have taken distinctive problems] and create gutpela sindaun [good, livable forms: local committees [in Tok Pisin and hereafter places]. komitis] and flexibly institutionalized leadership [lidaman/meri] roles, all enacted through media- 1.2.7 Port Moresby and Lae are becoming cities tion and the spatial regulation of settlements and with populations and leaders who want a better markets. Komitis, lidamen/meri, and their regulatory urban future. Among other outcomes, this emer- practices coexist with formal state governance (the Figure 1.  Port Moresby and Lae, Papua New Guinea 28  See Davis (2006), chapter 3. 29  See Hambleton (2017). 30  Lae, Kokopo, and Mount Hagen City Authority Acts. 4 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS police, local-level government [LLG], Village Courts) 1.3.1  Classic and advanced criminological and draw on their authority (though often not their approaches to security focus attention on crime resources) where possible. All of them “co-produce” “hot spots” as defined by various risk factors, authority and regulation by enrolling the support of which may then be addressed through focused residents and by pacting and working together with intervention.31 All too often, these responses fellow recognized ethnic and other leaders. In doing (so-called “targeted/saturated patrols”) highlight so, albeit with highly uneven effectiveness in rela- the deficits and deepen the stigma of already tion to gender and violence, they can help produce, struggling communities or regard them as simply combine, and protect some of the rich but precarious ungovernable and therefore as legitimate targets for capital that is produced and accumulates in urban rough intrusion or tabula rasa redevelopment. Urban settlements. settlements are certainly sites of intense and often conflictual competition, but they tend also to be highly, if unevenly, governed by institutions and actors 1.3  THIS REPORT’S INSTITUTIONAL that perform the basic regulatory functions needed CAPABILITIES APPROACH TO UNDERSTANDING for transactions relating to tenure, business, and URBAN REGULATION social life. Box 2.  Definition Note: Local Regulation of Box 1.  The “System in Place” for Regulating Settlements PNG’s Settlements Local regulation of settlement is defined here Unless you go into an area and know it, you in functional capability terms. It refers to the can’t know what is going on. So, when people ability of residents, groups, and leaders to say, “These areas always looks rough, how can create security and social order through people live in there?” Well, there must be a practices that resolve personal, family, and system in place, and you’ve got to know how communal disputes, to secure and manage to ask the questions to find it, otherwise, tenure and residence, economic opportunity, people don’t realize what is in there and and representation, and to do this for all working. It’s hard to describe unless you see it residents. in specific cases. Mediator, Sabama, Port Moresby 1.3.2  This report focuses on the everyday ins- There’s a song about this: “Kerema yu no savi. titutional arrangements that regulate the safety Yu yet cam na lukim.” “You don’t know Kerema and security of PNG’s urban settlements in relation [a provincial town of mixed reputation]: you to people and places where the reach of formal need to come and see.” Community leader, authorities is limited, dysfunctional, and/or lacks Port Moresby legitimacy. Section 2 sketches out the history and underlying urban political economy cities in PNG, presenting them as the conditions of possibility for local regulation. Section 3 focuses on local regulation itself: what machinery is there and how it works.32 Crime “hot spots” are small areas (typically less than 5 percent of a city’s area) with high crime densities (accounting for on the order of 50 31  percent of all crimes). See Helbich and Leitner (2017); Murray, de Castro Cerqueira, and Kahn (2013); and Weisburd et al. (2017). The analysis in this paper was informed by, and illustrates some of the precepts of, the World Development Report 2017 Governance and 32  the Law (World Bank 2017b). It follows the report’s injunction that analysis should pay attention to: 1. Functions, not forms: what functions are the institutions supposed to achieve? Where are the implementation/effectiveness gaps, and why do they exist? 2. The ability of elites and citizens to bargain and forge agreements or “pacts,” embodying commitment, cooperation, and coordination that effectively reach across economy/society. 3. The policy arenas within which these bargaining processes happen, bringing together the “rules of the game” and including competing/collaborating actors to generate development outcomes. 4. The ways asymmetric power relations (the greater power and resources of some groups and actors) work within the policy arenas, reducing effectiveness and producing (and reproducing) exclusion, capture, corruption, and unequal outcomes. 5. How these processes do and do not adapt, and where careful intervention might make them more inclusive and able to deliver better functional outcomes.  I ntr o ductio n: s af ety, securit y, and invest ment in Papua New Guinea’s urban f u tu re 5 1.3.3  The report describes the functions, forms, and violence and the very slow emergence of women processes typical of local regulation. Considering into local komiti and lida roles. Section 6 considers local regulation to be a function frames its role in local regulation in light of recent (but still nascent) enabling capital accumulation through four domains: developments in formal subnational governance, in 1) securing basic settlement and tenure and protecting particular, city authorities. Section 7 summarizes the property (physical capital), 2) solving disputes and arguments, presents conclusions, and points to the safeguarding family and communal relations (social safeguards and other implications for all agencies capital), 3) ensuring economic opportunity (financial whose work will bring them into contact with komitis. and human capital) in relation to markets, finance, and employment, and 4) providing representation and protecting rights (legal and political capital). These 1.4  BRIEF BACKGROUND, SITES, four domains are discussed in Section 4. AND METHODS OF THE REPORT 1.3.4 These functions are realized locally through 1.4.1  This report is the product of a World Bank various local institutional forms (settlement com- PNG Urban Safety and Security Technical Assis- mittees, leader networks, mediation activities) that tance, led by the Justice for the Poor program and have differing capabilities to secure commitment funded by the Australian government. The objec- (and enable contest) and extend regulatory func- tive of this engagement was to improve the qual- tions across settlement populations. These functions ity of the policies and interventions of PNG’s public and forms are enacted through distinctive processes authorities that are aimed at improving citizen secu- through which local regulation adapts, develops rity in Port Moresby and Lae. This report is one of a “rules of the game,” and produces outcomes that series based on extensive field research conducted display various limitations and patterns of inclusion in the two cities from late 2014 to June 2017.34 and exclusion.33 They do this in the wider context of The research endeavored to create new empirical the very uneven and often dysfunctional reach of the knowledge of the ways citizens and a range of state, state, where the elements of state authority cannot be community, business, and other authorities address relied on to produce good outcomes. neighborhood disputes and govern markets and transport nodes, and through this, to support dia- 1.3.5  The limits to local regulatory capability are logue between municipal authorities, business asso- significant. First, there are limits to local regulation’s ciations, community organizations, the police, justice ability to reach beyond its local status to draw in offices, and local government agencies to identify higher authorities, and second, there are limits to practical ways to improve public safety and security. local regulators’ own ability to reach across different Research for this particular report was conducted ethnic, social (gender, age), and other differences. primarily but not exclusively in the five Lae settle- Section 5 shows how these limits reflect the ways ments described in Annex 1, and also, for several local regulation is shaped by and reproduces shorter periods, in Port Moresby (Sabama, Tokarara/ patterns of power and thus acts to reproduce June Valley). exclusion, capture, and corruption. Most significant is the mistrust that women have for komiti structures, A full description of the research process and meth- involving the important issues of regulatory failure odology the report draws on is found in Annex 2. and malfeasance related to family and sexual In relation to these spaces, Korf et al. (2018), 168, refer to the creation of “public authority through shifting networks of relationships that 33  cross-cut socially constructed scales.” 34  See Craig and Porter (2017); Craig, Porter, and Hukula (2016). 6 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS PNG’s urban political economy and local 2.  settlement-regulating institutions 2.1  A BRIEF OVERVIEW OF THE HISTORY dence colonial center, Port Moresby, was dominated OF PNG’S URBAN POLITICAL ECONOMY by Australian administrators. Indigenous and diverse AND GOVERNANCE Papuan and New Guinean arrival populations were subject to curfews; many were corralled into work 2.1.1  Notwithstanding the recent nature of compounds just outside the urban boundaries, and urban-ization in PNG and the country’s distinctive from this place, beyond the immediate purview of ethnic diversity, PNG’s cities exhibit the kind of the colonial administration, contemporary urban features found in similar geopolities elsewhere. settlements grew. Port Moresby residents of the Port Moresby and Lae are both based on an 1960s (population 24,730, one quarter expatriate) underlying colonial planning system and order, upon recall a time before crime, barbed wire, and even which numerous different settlements have emerged locked doors.35 Governance was controlled by expa- either on lusgraun (unoccupied state or other land triate public servants, but indigenous institutions with contested title), on “low-covenant”, “sites- were emerging by the 1950s, fashioning themselves and services” state leasehold land, or through the on colonial templates. At this time, ethnic lead- leasehold of customary land. Although Port Moresby ers known as “councilors” were appointed to liaise is not connected by road to most other major with the population and preserve public order, and population areas, it has a primacy over those cities in when the councilor system failed in the 1960s, some a way similar to the capital cities of other archipelagic migrant groups revived this practice. countries. This dominance seems assured to grow, as patterns of circular and other migration tend to 2.1.3 Committees/komitis emerged from within crowd new settlers mainly into existing settlements. urbanized (and male-dominated) ethnic groups, Like cities elsewhere, Port Moresby and Lae are and early “courts” were urban adaptations by eth- melting pots, containing extraordinary diversity, and nically homogeneous groups. By the early 1970s, their sprawling develop-ment can perhaps be partly Moresby’s first historian recounted that “in nearly all attributed to patterns of resettlement following some [homogeneous] residential areas… individuals distin- conflict. Certainly Lae is somewhat polarized between guished by wisdom and forcefulness of character may the highlands and coastal ethnic groups (“coastals” be formally recognized by their own groups,”36 and are non-highlander settlers). that “the committees of the [heterogeneous] no- covenant settlements at Morata and June Valley are 2.1.2  Urbanization is a relatively recent phenom- now demanding powers to maintain public order in enon in PNG, as are the challenges of combating their settlements.”37 urban violence, providing security, and regulating economic and other public interactions. Since the 2.1.4  Ethnic heterogeneity was always a hallmark earliest colonial times, PNG has been governed as of PNG’s cities. Three quarters of the population an overwhelmingly rural country. The pre-indepen- were from the Central and Gulf provinces, but high- 35  Stuart (1970). 36  Oram (1976), 150. Ibid., 151. “No covenant” refers to the arrangements for distribution of the individual land parcels; no-covenant houses were expected to 37  be self-built and not subject to the formal urban building codes applied in high-covenant subdivisions. Committees emerged from within ethnic groups, and early “courts” were urban adaptations of single ethnic codes by ethnically homogeneous groups. Oram recounted that though “There is a large potential leadership within villages and homogeneous residential groups which as yet can only find inadequate expression… Leadership is lacking in heterogeneous residential areas such as Kaugere…” PNG’s ur ban po lit ical economy and local set t lement-regulat ing inst it uti o ns 7 landers, who first appeared in 1961, were already 2.1.6  Diversity and potential conflict also encour- conspicuous and highly mobile immigrants.38 Urban aged urban sprawl, and migrants in PNG struck a mixing also led to the emergence of other forms of range of settlement deals with traditional land- social organization and solidarity; pidgin English owners across the settlement frontiers of a wide [Tok Pisin] became the urban lingua franca, enabling urban territory. Closely settled sites to the east the formation of labor and ethnic welfare move- of Port Moresby’s city center became the notori- ments, as well as inter-ethnic associations, espe- ous “raskol settlements” of Kaugere and Sabama.41 cially those affiliated with sports.39 There, urban gang membership was ethnically mixed but based on urban rather than rural place 2.1.5  But “wantok” social relations stressed the of origin, and was dominated by Papuans and not transference and adaptation of communal kin highlanders, who arrived in increasing numbers preference into urban settings. In urban settle- in Port Moresby later in the 1970s–1990s. A simi- ments, wantok [“one talk,” that is, linguistic affili- lar pattern emerged in Lae, where ethnic fights in ation based on a shared village language identity] diverse compounds divided that city along Morobe ties meant petty quarrels would quickly escalate “highlander” and “coastal” identity lines, with cen- into communal conflict. Large ethnic fights between tral highlanders coming into dominance in some people from Central province and Morobe high- parts of the city in the 2000s. landers were recorded around contested sites of economic opportunity (in Koki market in Moresby, for example) beginning in 1961.40 Figure 2.  Maps of POM and Lae showing settlements studied 38  Rew (1974), 10–12, in 1966 described the indigenous population as “drawn from every sub-district in the joint Territories.” 39  Ibid., 9–10. 40  Ibid., 9. 41  Ibid., 18. 8 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Figure 3.  Maps of POM and Lae showing settlements studied 2.2. THE RISE OF INFORMALITY, CONFLICT, AND to enable property ownership, many joined the informal CRIME, 1980–2005 or custom land settlement spillover. 2.2.1 Urban population numbers have been increas- 2.2.2 On custom land, new arrivals established rela- ing rapidly, but political and administrative boundar- tions with indigenous landowners. Today, many landlords ies and urban tenure have remained largely unaltered. [papa grauns] remain assertive and evictions are common, Between 1978 and 1988, Port Moresby grew by 8 per- often on the pretext of law and order or public health cent per annum and Lae by 6 percent, but no new formal issues.44 Occupied state land, however, is less secure. suburbs were developed in Port Moresby (or Lae) after Typically, the most violent unilateral evictions, often 1989, despite the fact that its population increased by accompanied by houseburning in Lae [kukim, rausim, 60 percent in the 1980s and added another 30 percent burn their house, kick them out], occur on state land. in the 1990s.42 The cessation of state alienation of land43 led to the growing importance of informal settlement 2.2.3 A deepening sense of crisis emerged in the on state-owned land, usually in insecure or otherwise 1980s at both the grassroots and official levels.45 precarious locations and in pockets along rivers and hill Reports—and popular dismay—continued to focus on lines, spilling over from low-covenant areas into neigh- illegal settlements as particular centers of criminal activ- boring vacant land with a variety of owners. Alienated ity: urban villages, ethnic/migrant groups, and especially land became hugely expensive, and as formal sector groups of young men.46 By 1988, senior police officials employment no longer provided sufficient remuneration were in the media describing the “terrorizing” of Lae by 42  For population statistics, see http://www.citypopulation.de/PapuaNewGuinea.html. During the colonial period, customary land was routinely alienated by the state. This practice effectively ceased during the 1980s, as landowner 43  resistance grew. 44  See “Lae Woman is Homeless Twice at Limki Settlement,” posted on YouTube at https://www.youtube.com/watch?v=OGvuRy98geQ. 45  See Norwood (1984). 46  Clifford, Morauta, and Stuart (1984). PNG’s ur ban po lit ical economy and local set t lement-regulat ing inst it uti o ns 9 armed gangs that were committing robberies, home Moresby) and a range of subcommittees that could be invasions, gang rapes, and armed holdups.47 Informal formed under local discretion. Most wards have had settlements became a political issue, and their resi- some kind or remnant of a ward development com- dents a political base. Prominent leaders (including mittee since then, though these entities have almost then-Prime Minister Bill Skate) gained political sup- never been adequately funded and have fallen into port from settlement youth. In the Miles settlements edge-of-state dysfunctionality. Still, most settlements’ along Lae’s western highway route from the city, the chairmen, presidents, and komitis would claim some Morobean 585 gang found common cause with poli- kind of mandate from these committees or from ward ticians who proclaimed “Morobe for Morobeans” bylaws that have never been made public. and targeted highlander arrivals. Morobe politicians bulldozed Highlands settlements, trying to stem the flow of immigrants. Citizens fortified their houses 2.3  BEYOND CRISIS? NEW FORMS OF URBAN and neighborhoods, and markets became sites of LEADERSHIP, AUTHORITY, INVESTMENT, AND entrenched conflict. INSTITUTION, 2005–17 2.2.4  The peak of the urban crisis occurred in the 2.3.1  Port Moresby emerged as a global symbol of 1990s and early 2000s, as registered in the gangs’ urban violence in the 1980s. In the mid-2010s, even public presence, influence, and violence. But LLG as the city was renewing itself, it continued to be structures were also set up during this time, involv- regarded as one of the world’s least livable cities.48 ing elected ward councilors (but appointed in Port In 2010, Lae and Port Moresby reported homicide Box 3.  Definitional Note: The term “settlement,” as with the distinction between “formal” and “informal” in urban PNG, needs to be seen as both a modest descriptor and also a powerful political tool in characterizing, stigmatizing, and governing people and places. PNG’s 1973 White Paper Self Help Housing and Settlements for Urban Areas related settlements to planned and officially recognized no- or low- covenant subdivisions, with sites and services set up to enable occupier-built housing. Documents from the United Nations Human Settlements Programme (UN Habitat) and the PNG Office of Urbanization refer mainly to “informal settlements,” but also to “unplanned” or “squatter settlements,” as well as to “village settlements,” which they describe as “informal settlements on traditional land on the urban fringes, whose occupation is arranged through informal agreements… involving an annual land rental fee.” Urban development discussions have referred to “cosmopolitan networks of tribal groupings or anarchical sub-cultures, which have been defined by ethnicity and regionalism within an urban context.” Media accounts vary from the recognition of settlements as diverse but established neighborhoods, distinguished by affordable, often informal housing that houses 40–50 percent of urban residents, to dwelling places of unemployed and informal sector workers, to dirty, unhealthy, and “notorious breeding grounds for criminals and violent persons.” In this report, “settlement” is applied to align with popular contemporary usage, referring to residential settlement on land where custom is the basis of tenure and where there is informal or low-covenant tenure on state land (settlements can include mixes of all three). Crucially, these are places where mixed ethnic tenure means that there is a core need for the kinds of regulatory arrangements this report describes. “Blok,” which has an overlapping meaning, here refers to either a small residential area, section, or compound of land occupied by a single clustered ethnic group, or to an entire settlement. 47  Sinclair (1998), 389. See Elena Holodny, “The 11 Least ‘Liveable’ Cities in the World,” Business Insider (Australia), August 21, 2014, http://www.businessinsider. 48  com.au/worst-cities-economist-intelligence-unit-2014-8?op=1. 10 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS rates of 66 and 33 per 100,000 persons, respectively, crime.52 Yet despite the persistent crime problem in nine and three times the global homicide rate.49 Lae’s Lae, across urban PNG, the deep crisis of the late rate places it in the top 15 cities globally, alongside 1980s and 1990s began to pass during the long a cluster of notorious Latin American cases.50 Still, as resource boom of the 2000s and early 2010s. Figure 4 in many urban settings internationally, property crime shows that 33 percent of households in Port Moresby and repeat victimization in fact declined in Moresby had not experienced any crime in 2004, and by 2010, during the 2000s,51 though not in Lae, where in 2005, this figure had increased to 60 percent. 78 percent of all families were victims of at least one Figure 4.  Port Moresby: Reported Victimization, 2004-2010 Selected Household Crimes Multiple Household Crimes 40 70 62 60 35 60 56 30 35 50 43 25 29 39 37 28 26 40 Percent Percent 20 33 30 15 17 20 10 11 5 7 8 10 5 1 3 5 2 1 1 0 0 2004 2006 2008 2009 2010 2004 2005 2006 2007 2008 2009 2010 Stealing property Firearm use Sexual assault None Once 2-4 times 5-9 times More than 10 Source: Papua New Guines (2011a). Figure 5.  Comparative Lae-Port Moresby Victimization Figures, mid-2000s 100 100 90 90 80 80 70 70 60 60 Percent Percent 50 50 40 40 30 30 20 20 10 10 0 0 Port Moresby 2007 Lae 2007 Lae 2008 Port Moresby 2007 Lae 2005 Lae 2008 Individual victim of at least one crime Household victim of repeat crime Household property crime Household sexual assault Household victim of at least one crime Household victim of 5 or more crime Household violent crime Stealing household property Household victim of multiple crime Household firearm use Source: Papua New Guines (2011b). 49  Lakhani and Willman (2014a), 3. 50  “List of Cities by Murder Rate,” Wikipedia, https://en.wikipedia.org/wiki/List_of_cities_by_murder_rate. 51  Papua New Guinea (2011a, 2011b). 52  Guthrie, Laki, and Hukula (2006). PNG’s ur ban po lit ical economy and local set t lement-regulat ing inst it uti o ns 11 2.3.2  There are likely several reasons for the Box 4.  New Forms of Urban Association changing rates of crime and victimization during and Identity: Port Moresby Market this 2004–10 period. Life-course patterns of crime Customers Speak and especially gang engagement may be distinctive in PNG. Unlike notorious, long-lived gangs in other Before we used to go in ethnic lines; now, we countries, PNG’s urban gangs have failed to institu- go by street. But it’s the same thing: you have tionalize in recent decades.53 Gang members have to be in that community to gain that respect. been “growing out of crime”54 in numbers that have You’ve got no status in the community unless outpaced generational renewal, and more members you prove yourself that you are part of there. are now “resting at home” than are being recruited You have the luksawe [recognition] in the for robberies. What has institutionalized in settle- relationships with people that you build all the ments is much more positive and interesting and is time, everyday. And from that you get the the main focus of this report: community investment understanding of leadership in the community. in committee structures, along with churches and pri- That’s how you live here, and grow up here, mary schools. But behind their influence lies the rise and be safe here. Tokarara resident of multi-generational families living in compounds, bound together by economic interdependence We are not wantoks [kin or language mem- through investments that all have made in institutions bers]; we grew up here together, we became to regulate everyday settlement life. brothers in the same place. I was born here. I’m a [Tokarara suburb] man. But so are they. 2.3.3  Urbanization processes in and of themselves When we were young, we played touch, we bring together diverse peoples, creating powerful played sport together. Now, we are older, incentives for everyday investments in safety and higher status. We still know, we trust each security, for example, by creating new forms of asso- other. So, when we are here together in the ciation and new bases for trust. In PNG’s cities, village market, we are relaxed. Tokarara resident and local wantok identities have continued to adapt, being “strategically edited and usually simplified”55 into regional, province, or urban ones (Coastal and formal or informal employment multiplied in [northern coast], Morobe [province], Kaugere mangi PNG’s cities during the long 2003–15 resource [a Kaugere suburb person]). These adaptations in boom. Contemporaneously, beginning in the mid- turn provide new patterns of urban affiliation, trust, 2000s, leadership in urban development emerged and opportunity. Churches, sports, business, and in Port Moresby,56 riding the resource investment, social issues networks all enable these new identi- including spillover from the US$19 billion Exxon ties, relations, leadership, and regulatory capability Mobil/GoPNG liquid natural gas project.57 Boom- to emerge and institutionalize in various ways. time revenues created roads; public buildings; port, airport, and land transport nodes and links; new 2.3.4 At the same time, alternative pathways suburban development;58 and facilities for the 2015 for settlement youth to move into education Pacific Games and the 2018 Asia-Pacific Economic See Hagedorn (2008). Institutionalized gangs everywhere rely on generational recruitment, as older members follow global patterns of 53  growing out of crime in their mid-20s, when domestic arrangements become important. Recruitment involves the branding of identity and commitment through rites of passage involving dependence on illicit drug and extortion economies, cooperation and participation in conspicuous crime, and maximum security incarceration. Jails become major institutional strongholds for protection based on the cementing of gang members’ identities and loyalties, as well as tight hierarchical structures from which imprisoned leaders coordinate gang activities outside. 54  Rutherford (2002). 55  Davis (2000), 17. 56 See https://www.pomcci.com/yumi-lukautim-mosbi/. See Celine Rouzet, “Papua New Guinea: The Gas Boom,” Pulitzer Center, June 20, 2013, http://pulitzercenter.org/reporting/china-japan- 57  papua-new-guinea-Exxon-mobile-gas-oil-local-poor-poverty-foreign-executives-gang-slum. “Plan to Build 40,000 Houses in Port Moresby,” RNZ, February 28, 2014, http://www.radionz.co.nz/international/programmes/datelinepacific/ 58  audio/2587399/plan-to-build-40,000-houses-in-port-moresby. 12 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Lae Miles leaders meet to discuss working with local corporate leaders. Photo: David Craig/Worldbank Cooperation summit.59 Lae too experienced invest- Box 5.  Urban Insecurity: Ongoing Fears ment, and legislation was passed for a Port Moresby- style city commission. Investment in urban youth employment has demonstrated the impact of creating I recently moved to Bundi camp to reside and more opportunities for job and other training for I think it’s terrible. The worst thing is ethnic young people.60 clashes: one time they threw stones at our house and we had to jump the fence. It’s like a 2.3.5 As anticipated above, however, the effec- nightmare to us. We took it to court, but there tive reach of all these local regulation arrange- were too many conflict of interest in the police ments is often limited. On the one hand, the wider and we needed money to appeal so we left it. “enabling environment” of formal urban governance They paid money to the police to hide our and law and order that could recognize and support files. There is no control of noises— youths or local regulation is frequently either disabled or itself people make noises till day break so you don’t predatory. And on the other hand, emerging local really have a peaceful life or privacy. When regulatory arrangements themselves struggle to be youths are taking in steam (homebrew) and effective and maintain trust across settlement popu- drugs, they have no control over their minds. lations. As elsewhere where customary, traditional, Female settlement resident, Ward 1 Lae Urban and hybrid legal practices predominate,61 insecurity typically remains pronounced for women, youth, marginal ethnic groups, many informal sector traders such as market vendors, and residents on customary and informal state land. Rowan Callick, “Coup for PNG as ‘Least Liveable’ City to Host 2018 APEC Summit,” The Australian, October 10, 2013, http://www.theaustralian. 59  com.au/national-affairs/foreign-affairs/coup-for-png-as-least-liveable-city-to-host-2018-apec-summit/story-fn59nm2j-1226735720143. 60  Ivaschenko et al. (2016). 61  See Amnesty International (2005); Castillo Diaz (2006); and Lee and Haider (2012). PNG’s ur ban po lit ical economy and local set t lement-regulat ing inst it uti o ns 13 Local regulatory capabilities in urban settlements: 3.  what exists and how does it work? 3.1  KOMITIS AS SITES FOR COLLECTIVE ACTION 3.1.2  In international comparative perspective, komitis in PNG represent a relatively mature (but 3.1.1. This section considers how everyday insti- also somewhat disempowered) level of urban set- tutional mechanisms in urban settlements fulfill tlement leadership development. Internationally, their basic regulatory function by becoming local many urban settlements’ leadership (and, not coin- arenas wherein leaders and common people com- cidentally, much of the literature on settlements) is bine efforts to solve disputes, execute judgments, more focused on struggles to secure land access and manage settlements, make local markets safe, and resist removal, the participation and roles of local create opportunities for youth. It focuses on komiti leaders in upgrading or delivering basic services for systems and emergent local leadership and their abil- residents, or settlement politics as a stepping stone ity to create consensus and cooperation through per- to higher office.65 In contrast, in Port Moresby and forming public roles and mediation, forging horizontal Lae, of greater concern and preoccupation to settle- and vertical links and working pacts, and collectively ment leaders are managing ethnic relationships and framing and enforcing rules. As will be described, issues arising from social disorder and economic komitis are assemblages distinctive to PNG. But in competition. As will be shown, komitis are not sub- many ways, they resemble some of the forms of local stantially in conflict with formal urban governance or collective efficacy that international research has policing agencies, and they have fairly good relations shown to have considerable influence on community with local quasi-state law and justice agencies (espe- crime and safety.62 The literature on collective efficacy cially Village Courts). suggests that “a community’s ability to exercise infor- mal control and its collective capacity for action were 3.1.3  Thus, local settlement regulation performed the more proximate mediators of structural disadvan- by komitis and leaders does not happen in a reg- tage on crime.”63 Although the distinction between ulatory vacuum. Rather, as will be developed, it formality and informality is difficult to sustain in depends to a significant extent on an ability to “co- practice, “informal control” and “collective capac- produce” regulation66 by reaching both vertically and ity for action” are very much what komitis manage horizontally to locally recognized forms of authority. to construct from the diverse social groupings in In the diagram below, the bold lines represent regular PNG’s settlements. In terms of violence and social and important links to such authority, and the dotted order,64 a komiti can be seen as a site for and mode of lines represent nominal or weak links. pacting between leaders that functions to regulate and suppress violence between group members. See Mazerolle, Wickes, and McBroom (2010), 3: “Recent research, however, indicates that highly specific, collective processes are more 62  important and proximate factors in differentiating community crime outcomes than are social relationships ….. Using the term collective efficacy to capture the task-specific nature of these community processes, Sampson and others suggest that the level of collective efficacy in a community explains spatial variation in crime problems better than systemic theories of crime.” 63  Ibid., 5–6. See also Sampson (2004). 64  See North, Wallis, and Weingast (2012). 65  See, for example, Sakaya, Sanonib, and DEng (2011); and Koster and de Vries (2012). 66  Bovaird and Loeffler (2012); Joshi and Moore (2003). 14 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Figure 6.  Structure of Local Regulatory Arrangements in PNG Settlements (Key required). Governance Level District Court DDA Police District DISTRICT LLG LLG WARDS Village Courts Ward Ward Committees Law & Justice WARD Chair Law & Justice Chair Papa Graun Blok Commitee Chair BLOK K O M I T I Local (Women’s Youth) SECTIONS Organizations Areas Ethnic COMPOUNDS Market Churches Sections Groups Box 6.  Disclaimers: There is a range of issues related to urban safety and security that this report does not deal with directly This report does not address the traditional Motu Koitabu or Ahi landowning villages of Port Moresby and Lae, which generally have hybrid rather than komiti governing arrangements.67 The ways Lae’s traditional landowners regulate their bloks is seen through the lens of komiti arrangements (which are present even with the most capable landlords), so that landowning is not treated as a separate function or institutional form. Although komitis in some places and sectors are mandated to function under LLG and ward bylaws, in other places they operate in parallel or supplementary to, or in the practical absence of, local state forms and recognition. This report covers all komitis operating in the areas of limited statehood constituted by urban settlements, many of which would want to or could claim that they are the mandated expression of LLG or ward levels of the state. But by nature of their formation and operation, they blur the boundaries of the state and operate along powerfully diverse lines rarely governed by bylaw or bureaucratic procedure. Nor does this report separately treat the quasi-state Village Courts, which are already the subject of good research.68 However, in all the research areas, the Village Courts functioned alongside and overlapped with the komiti system, sharing members, cases, and recognition. This report also does not examine the functions, ways of operating, and impacts associated with the Royal Papua New Guinea Constabulary (RPNGC), magistrates and district courts, and district, city, or LLG agencies. The report authors are aware of the issue, but for reasons of space and purpose, do not elaborate on the different ways in which notions of 67  tradition and traditional authority, are politicized and constructed in areas of limited statehood. See, for example, Forster and Koechlin (2018). 68  See, for example, Goddard (2009); Demian (2016), 13-30; and Evans, Goddard, and Paterson (2010). Local r egulatory capabilit ies in urban set t lement s: what exis t s and how does it wo rk ? 15 3.2  LOCAL REGULATORY FORM AND terms express a similar objective: mekim bel isi stap FUNCTION: THE KOMITI SYSTEM insait lo blok [establish peace within the settlement]. Concerns about security are at the core of the komiti’s 3.1.1. PNG’s urban settlements are not usually function, as well as the ways different ethnic groups places where outsiders simply arrive, or even can pact together to coproduce protection and access, without local recognition or connections. manage peace and good order locally. The komitis And as the quote in box 1 suggested, “These areas represent their ethnic group or blok and perform always look rough,” leading to questions about this pacting, which provides protection that includes “how can people live in there?” But across urban personal safety but also communal well-being and PNG, settlements and bloks (subward local areas) on the various forms of capital. both state and custom land all have people who are recognized as a form of local authority, in addition 3.2.3  Daunim hevi, gutpela sindaun gives komitis a to the landowners’ customary authority. Universally, wide functional mandate involving both proactive research shows that this authority—and the and reactive regulation: creating the conditions for individuals exercising it—is now known as a komiti. peace and good order and sorting out the challenges to it as quickly as possible (see Section 4, box 13). 3.2.2 The function of the komiti, local leadership, Interestingly, the komiti’s four functional areas and regulation is expressed most neatly in the Tok line up reasonably well with the factors that have Pisin phrases daunim (or stretim) hevi, gutpela been identified as the most important elements in sindaun: to resolve or straighten out problems and successful urban arrival locations: entry level shelter, create a good way of life for people here. Other work, transport, and political-legal representation.69 Box 7.  Basic Elements of PNG Urban Komitis • Self-professed function: solving problems and creating stable living conditions • Form: multiethnic committee with widely recognized members ranging in number from one to around a dozen. Some individual members are designated to be “senior komitis.” • Membership: flexible but always composed of locally resident and recognized (almost entirely male) ethnic leaders from across all local ethnic groups • Geographic reach: site specific—a settlement can consist of a number of locally identified bloks or zones, which together can be a subset of a wider urban ward. A single blok can comprise formal and informal or kastom lands. Blok size varies from 1 to 3 or 4 hectares. Zones or settlements can include up to 20,000 residents (see Annex 1 for a variation in actual Lae blok/zone composition) • Regulatory capabilities (see Section 4, box 13): manage the multiethnic settlement, carry out dispute resolution and mediation, manage local marketplaces, represent the settlement to the wider authorities • Common strengths: recognition, ethnic representation, accessibility, experience, skills in mediation • Common limitations: gender and age representation, management of gender-based violence, uneven relations with police and formal urban government 69  See Saunders (2011). 16 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Above: Lae Law and Justice leaders, June 2018. Below: 4 Mile Lae Law and Justice leader Philemon Boddeh, with family Local r egulatory capabilit ies in urban set t lement s: what exis t s and how does it wo rk ? 17 3.2.4  The are multiple forms of local (komiti, lida) is indeed an individual, but he (and there are very regulation, carrying out local functions. Enacting few female komitis) is an individual selected and the wide scope of that common function, forms of recognized by his own and other groups. He also komiti are diverse and adaptive. Komitis as collective necessarily acts as part of a collective construction of entities vary in name, definition, appointment basis, authority involving several different individual, group- and scope of operation. Some law and justice representing komitis, who, by definition, should be komitis are simply known as “the komiti”; others as working together at various points. the hevi [problem] komiti, by the prefix “blok,” or “line” (compound or group of houses).70 There are 3.2.7  Within these basic outline and function land management or blok komitis that collect rent parameters, komitis exist in plural forms. Invest- for indigenous landowners and manage settlement igations in Lae and Port Moresby have identified locations, law and justice komitis that hold regular considerable overlap in the usage of terms (komitis, courts [cots] and mediate disputes throughout the bloks, zones), but no two identical komiti structures week, and market komitis or “business reps” who (see Annex 1 for an overview summary of five different collect fees and enforce local rules in relation to local Lae configurations). This diversity relates in part to markets and businesses. the fact that these systems are not framed in a written statute and in part to the different types of tenure, 3.2.5  A komiti (understood as a collective group) ethnic mixes, and strength and level of engagement will usually have a chairman, who is often elected of the traditional landowner. by the komiti members themselves and who takes a more or less dominant role in komiti business, including appointing other members of the komiti. 3.3  KOMITIS’ EMERGENCE AND LEGITIMACY The blok and/or law and justice chairman has a AS AN ETHNICALLY AND LOCALLY INCLUSIVE representative, quasi-political, or territorial authority INSTITUTION role, especially in relation to police, customary landowner, or ward authority. Most but not all law 3.3.1 Komitis emerged very early in PNG’s and justice or blok chairmen and komitis would urbanization process in both homogeneous and claim (or want) some kind of recognition or mandate heterogeneous ethnic settlements.71 Describing from the local ward-level community development the early emergence of endogenous settlement committee, as well as from local residents and their governance in pre-independence Port Moresby, ethnic groupings. Relations with ward councilors, Oram recounted that the komiti appeared first within who are supposed to have some funding for these homogeneous residential areas, but that “leadership kinds of committees, are typically limited, though the [was] lacking in heterogeneous residential areas LLG/ward will fund local komitis to run community such as Kaugere…”72 After the failure of the colonial activities, including by providing forms of diversion government–appointed councilors system, it was during the crime-heavy buildup to Christmas (and committee (i.e., komiti) members, each individually sometimes on PNG’s Independence Day). Election recognized as leaders by their groups, who emerged and/or appointment procedures to local komiti status to manage local social relations in ethnically vary widely and seem to be rather flexible. heterogeneous areas. 3.2.6. The term komiti is also commonly used 3.3.2 Both the individual and collective senses in relation to an individual (“he’s my committee/ of the word komiti have long been in usage. komiti,” “he’s the market komiti”). This does not Individually, ethnic settlement leaders were called necessarily cause confusion, as it is an important way komitis (as in, “he’s a komiti”), with each one of understanding the komiti role and its individual appointed by the group and leading procedures and collective authority and workings. A komiti through which the group might, as outlined below, 70  Hukula (2013). There is no extensive international comparative literature on the historical development of collective efficacy in urban informal settlements. 71  But see Perlman (2006). 72  Oram (1976), 150. 18 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS “mind their own business.” Together, these individual (through mediation) or pacts (through recognized komitis comprised a wider, collective komiti that, membership in a group of local leaders or komitis). when in strong and expressed agreement with each Insofar as he can do this, the agreements and pacts other, was not easily ignored. As noted, by the early he makes resolve a fundamental challenge to basic 1970s, komitis in the towns of Morata and June Valley security in urban PNG: the threat of inter-communal were “demanding powers to maintain public order in violence. These are, in this sense, protection pacts their settlements.”73 Both of these communities were that give the komiti his power and underpins his distinctively heterogeneous, and the committees that authority.74 emerged combined homogeneous legitimacy with heterogeneous inclusion and reach. 3.3.4  Dealing fairly and inclusively with heteroge- neity is a crucial challenge for komitis and lidas. Striking the right balance between representing the Box 8.  Community-Appointed Komitis? group’s interests and negotiating settlements fairly across the locality is not easy work, but it is essential to In the settlements, ethnic groups live together, komiti effectiveness. Komitis are subject to frequent but they also live alongside other groups. But accusations of being wansait [one-sided along ethnic, to do that you need to have someone who will gender, or other lines] or working baksait [being cor- speak on your behalf, someone you and rupt, subject to behind-the-scenes deals]. With the others can go to. So the community says, “this accusations come threats, and komitis emphasize is the guy, we give him our blessing, he will be the ways they put their own lives on the line despite the one to mediate, to negotiate, to settle.” little more than the rewards of luksawe [reputation].75 Moresby mediator. Again, performance of these roles and aspirations varies, and frequent accusations of komiti weakness Without a group of people behind us, we and one-sidedness are part of everyday settlement won’t be leaders. A leader comes out from a life. As is demonstrated below, capacity to deal with group of people; back inside your communities, ethnic diversity does not necessarily translate into the your streets, there are a group of people who ability to manage problems involving other kinds of recognize you and push you up. People are difference, such as gender, age, or class. happy if you continue to stand up. If you ignore problems they will stay, and you will Box 9.  Diversity, Cooperation, and damage [bagarap] everyone who stands Effectiveness in Komiti Work behind you. You’ll turn around and see everyone, they’ll look at you, you’ll hang your IPeople come with different languages, but if head. Lae settlement senior komiti we can all work together in Tok Pisin we can make it work. If we get five people from five different places, and we are all covered by 3.3.3. Komiti legitimacy and regulatory authority unity, we can all contribute ideas. We depend on people’s need for representation in compromise, we work together. We don’t just multiethnic contests (see box 8 above, “…you need go back to our languages. Then we stick in to have someone who will speak on your behalf”). this together: it’s the right way. If we include In these terms, above all else, the komiti needs to people from different places we can solve the be able do two things: 1. represent his people as a problem. But if we stand by just one particular personally invested, local authority figure, and 2. solve way, we won’t solve it. Pita Blok chairman problems with other groups by forging agreements 73  Ibid., 151 See Slater (2010), 10-13. Protection pacts involve elites pacting together to prevent urban and communal violence, which can fundamentally 74  | destabilize national unity and economic development. Protection pacts can be performed locally, where, as in wider political settlements, they impose costs on local elites and locals alike. See also Khan (2010); and Craig and Porter (2014). Fees are a thorny matter in relation to komiti mediation. See section 4.2.7, on Mediation Fees and Leadership Recognition. 75  Local r egulatory capabilit ies in urban set t lement s: what exis t s and how does it wo rk ? 19 3.4  KOMITIS AS LEADERS: AGENCY, grounded in and able to “go back” to and respect- PERFORMANCE, GREATER RECOGNITION fully adapt custom; he should also be active, able to talk “like a man,” and extract clarity and consensus. 3.4.1 Although komitis depend heavily on local legitimacy, there is also a wider competitive and 3.4.2  There is also considerable agency and aspi- aspirational element to becoming a respected ration on the part of komiti chairmen. Being chair- komiti, reflected in popular expectations about man can be a powerful, widely respected (though leadership and performance. A strong or “senior” not often lucrative) role, and becoming chairman komiti will embody and channel the reputation of his means achieving and performing a kind of urban ethnic group as people who “know how to act” in patronage status, expressed in terms of (Highlands) an urban environment and do not maintain backward kastom recognition of leadership. Chairmen regularly rural practices [pasin blo bus], and as people who and proudly refer to the other komitis in hierarchical, know how to “mind their own business” and coop- cliental terms as their senior komitis, or sometimes erate with other groups to solve problems (or keep their “little boys.” Both characterizations express out of their problems when they are not their busi- hierarchy and a level of trust: “He is my komiti, I ness). He (again, it is most commonly a he) should be chose and appointed him.” Box 10.  Choosing and Mentoring Komitis in Kamkumung, Lae I look for the active ones, whoever knows how to fight against corruption and those who can handle such corrupt activities, and I get them on board. They handle cases themselves, but when they see that the case is fairly serious, they call me in to handle the case, and then I get and apply an overall view. If there are those [komitis] who want to raise concerns or any verbal attack [about the judgment], I put all views together and call everyone to sort it out. We must all know how to make good judgments, and go back to custom. A blok chairman Pita Blok settlement, Lae. Photo: David Craig/Worldbank 20 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS 3.4.3  In recent years, komiti or chairman status has 3.4.4  What distinguishes a lidaman from the mass become a stepping stone to emerging urban lead- of komitis is their ability to both transcend his ership [lidaman/meri] roles. Such an aspirant leader ethnic origins and reach well beyond the territory must perform like a “super komiti.” S/he needs to of the blok komiti “vertically” to sources of power be available and active at any significant event, be and authority beyond it. A well-regarded lida will among the first to turn up when there is trouble, and be heavily in demand by komitis elsewhere for a stick to the business until it is resolved. The person range of activities and semi-official advisory, con- must be seen to reach above ethnic or local wantok sultation, and witnessing roles, including business allegiance to resolve problems and build consensus, and other deals, politics, urban development, court cooperation, and peaceful relations [bel isi]. This con- cases, youth issues, and event organizing. Thus, lida sensus must be achieved through public speaking status is earned, and once earned may provide the but also needs to be pragmatic; the lidaman should basis for creating wider, issue-resolving networks; be able to pull together diverse perspectives into becoming a ward or council of women councilor, city- short-term action. wide ethnic group representative, or part of a district member of parliament’s support team; or contesting an election himself. Box 11.  Lae Residents Talk about the Way of a Leader A leader in the community is not elected based on his appearance or wealth. We see those who can talk and his words can remove the bad and bring good to the community. His words will bring a sense of peace in people’s lives, regardless of his wealth, possession, and appearance. Whenever the leader speaks in any given scenario, people display great respect. Regardless of unfavorable weather conditions or proper sitting arrangements, people patiently pay attention, without unnecessary distraction, to what their leader conveys out of respect for him. If he’s a good leader, he could talk forever, always bringing the people along with him, them agreeing with him. 4 Mile resident If there is a situation within a particular community, it is the leaders’ responsibility to immediately attend to it. Any matter that arises, the leaders are already there to address the situation. But if the leaders are not present, they are negligent; the problem will continue and bring destruction to lives and properties. Nawaeb Blok leader You can assess a leader’s skill in this by the state of their community. If there’s problem in a community, leaders are not working or there is no cooperation between the community, and the leaders are not working together due to disrespect, conflict of interest, etc. When trying to address a situation in a particular community, first look at the leader. How does the leader conduct himself in the community? How do the leaders formulate solutions when disputes arise? Lae City Law and Justice leader He has the reputation that he’s always there for the people. He stands for the people, he represents the people of his community in both good times and bad times, he is always committed. He is always on the front line. He has time to listen to everyone, from teenagers, women, men, youths, and persons with disabilities. He gives openly to the people in his community who are in need in times of sorrow or disaster or compensation demands. He’s got the care attitude, unselfish. He has the network of his boys who take orders from him and carries out without arguing or complaining because he takes care of them. But the most important thing of all is whether they are playing a big part in the community, in maintaining peace and order. Ahi Ward Law and Justice leader Local r egulatory capabilit ies in urban set t lement s: what exis t s and how does it wo rk ? 21 3.4.5  All of these elements can be summed together as the “pasin” (fashion/passion, usage or practice) blo lidaman. This comprises a set of cultural forms (rules, roles) that are currently expanding and becom- ing institutionalized in urban PNG. 3.5  COORDINATION AND REACH: KOMITIS AND THE ASSEMBLING OF FRAGMENTED LOCAL AUTHORITY 3.5.1  State authority might be remote or fractious, but settlements have an endowment of various kinds of authority in their several generations of mixed residency. It is this fragmented, quasi-formal, lightly resourced, and individual reputation–based authority that the komiti system picks up on, recog- Lae Law and Justice leaders hear a dispute under umbrellas.. nizes, grasps, and pacts together, and from which it Photo: David Craig/Worldbank also builds new forms of credible local authority. Table 1.  Grasping Authority at Kamkumung: the Multiple Hats of Komitis and Mediators Land ownership Ahi/ Church/ Ethnic Village or formal ward/ Police network leader Court role title Mediator politics Business links leader Jacob √ √ √ √ √ √ Peter √ √ √ √ Julian √ √ √ √ √ √ √ Ruth √ √ √ √ √ √ Mote √ √ √ Banjo √ √ √ √ Peter √ √ √ √ Nick √ √ √ 3.5.2 This local grasping together of available near the market, and all the different sectors (youth, forms of authority happens in two dimensions: church, women, business) and brings them “inside vertical and horizontal. Vertically, it draws in individ- the komiti.” uals with formal state or police connections, experi- ence, and respect, as well as a reputation. It does not 3.5.3 Thus, with these diverse leaders come a matter that the policeman is now retired and running diverse but collectively powerful base of the vari- a local business, or that the sportsman’s glory days ous forms of capital described in the introduction: on the national team are over; both still have respect, human capital of experience and mediation skills along with powerful connections and ways of operat- or knowledge of the legal system; cultural capital ing that can support local authority. Horizontally, it of ethnic expertise and knowledge of pasin or the draws together all the ethnic and church groups pres- village language [tok ples]; legal/political capital in ent, all the potential leaders living inside the blok or terms of representation and official or quasi-official 22 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS 3.6  WEAK COOPERATION: WHERE KOMITIS ARE UNFORMED AND DIVIDED 3.6.1  Komitis perform a complex and multifaceted regulatory role, reflecting and working with both homogeneous and heterogeneous local interests and processes, and with various elements of state and quasi-state agencies. Where the reach of the state is limited, komitis co-produce state-like local authority by borrowing and mimicking some of the paraphernalia of the state (providing cots services, holding office, adopting bylaws, taxing local busi- Multi-hat wearing lida, 4 Mile. nesses, and so on). This is a complex performance Photo: David Craig/Worldbank that is common in governance situations that are beyond the actual reach of the state or in areas of limited statehood. Overall, it produces institutions office-holding within the blok or zone; and symbolic that are neither state nor society but that work across capital in terms of titles held, places afforded at each and are recognizable and intelligible from both meetings, demonstrated links to formal hierarchies, sides.76 and the recognition [luksawe] of powerful outsiders. The physical capital of money is, on the other hand, 3.6.2  The komiti needs to be able to legitimately not usually a significant advantage or, crucially, even represent and work horizontally across local inter- necessity. A strong individual komiti will personally ests as well as vertically, connecting the settle- embody many of these forms of capital, emerging ment to other regulatory forms, including the Vil- as a “multi-hat wearing” actor with multiple inter- lage Courts, police, and formal government. Many connected roles. individual komitis have overlapping or divergent roles, each of which gives them some kind of author- 3.5.4  These forms of capital and the roles of the ity and horizontal or vertical connection. In disputes people who possess them often expand to reach as in other local regulatory activities, it is precisely the across wider urban territories. To the extent that a ability to combine individual standing and authority komiti performs in a fair and balanced way, his/ her with plural, communal recognition that makes the reputation will expand and with it, the geographical system work or fail. scope of his/ her engagement. In Tok Pisin idiom, a komiti “whose reputation is not destroyed” [nem 3.6.2  Not all komitis are created equal, however. blo em no kisim bagarup] will be invited into other Some do not function, and their failure to achieve settlements to help solve disputes, and as noted commitment and cooperation is evident in a pro- above, he will be looked upon as a potential leader liferation of local divisions and internal fights. The in the broader sense, as someone able to bring wider lack of long-established and recognized leaders within communities together in consensus and action. In ethnic groups can make problem solving harder or all of this, komitis and their system demonstrate a subject to errors of judgment and poor timeliness. A kind of adaptive capability, able to adjust, depend- weak and inactive indigenous landlord [papa graun], ing on which groups are involved, and to reach out with multiple, factionalized komitis (see Annex 1), or to and include new ethnic or other arrivals. That is a powerfully ethnically divided community (as seen at not to say that youth or women are often included in Lae’s Sialum compound, site of a major ethnic fight in komiti roles, but it is to recognize this possibility 2015 involving hundreds of burned houses) seems to going forward. undermine functioning on several levels, especially in such core operations as solving interethnic disputes. This performance is arguably a contemporary version in PNG of a wider phenomenon often referred to in recent scholarly discussions as 76  “new governance arrangements [which] rearticulate the [wider] state-civil society relationship … in a two way, ‘Janus faced’ relationship.” See Swyngedouw (2005); and Brenner (2004.) Local r egulatory capabilit ies in urban set t lement s: what exis t s and how does it wo rk ? 23 Emerging settlement areas, such as Ass Mambu on are generally not well aligned or linked with verti- Lae’s notorious Back Road, may not yet have had time cal, city, or nationwide governance or policing. This to produce strong komiti leadership; in its absence, impacts directly on komiti effectiveness; for instance, youth associations or nascent gangs can cause seri- komitis are simply out of their depth in relation to ous safety and security issues. In general, areas with larger issues of social inequality or disputes arising in small highlander populations seem more vulnerable conjunction with large-scale urban development. This to these concerns. feature also reinforces the fact that komitis reproduce local power relations (for example, between men and 3.6.3  On the other hand, a weak papa graun can women or highlanders and coastals, as described be supported by a powerful komiti, and a strong below), and this affects the perceived fairness and papa graun (as at Niu Rocks and Awagasi market, see legitimacy of their activities and decisions (see Sec- below) will rely on komitis only for a limited range of tion 5 below). But when a komiti brings locally avail- tasks that he does not want to do himself. Hybrid or able authority together to achieve collective efficacy even competing komiti arrangements, as at Pita Blok, in settlements, its intrinsic pivotal powers and adapt- can also be functional and effective. ability become patent. Such authority maintains a credible threat of sanction against those creating 3.6.4  There are also implicit structural weaknesses problems, particularly in relation to the availability of in the komitis’ roles. As the report shows in Section economic opportunities. 6, a feature of the local nature of komitis is that they Box 12.  “Some Komitis but No Leaders” at 4 Mile Stage 1 Stage 1 area at 4 Mile settlement has had a long history of local conflict and notoriety as the place where armed holdups of vehicles on the Highlands Highway frequently occur. Until recently, it was a stronghold of gang organization, and the predominantly ethnically Morobean 585 gang used it as a base. Streets were named after gang allegiances. Highlander settlement was informally restricted, and fighting across territorial lines with highlanders in Stage 2 was endemic. But a series of deaths and retirements from the gang leadership did not actually lead to a violent leadership contest, as is often the case elsewhere. Rather, it left local youth acting autonomously, committing crimes on a much more episodic basis. These youth, known locally as “sortimobs” (short, i.e., youthful gangsters), have not institutionalized a new generation of gang practices, and as a result, ethnic Morobeans are underrepresented in overall 4 Mile zone komiti arrangements, and 4 Mile komitis struggle to exert authority in Stage 1. Meanwhile, the sortimobs continue to commit robberies but are so loosely organized that 4 Mile komitis and leaders find it hard to respond systematically. Former gang-associated leaders who are settling down have in recent years adopted recognized komiti roles in Stage 1. But no one—and especially no Morobean—has yet emerged in Stage 1 as a blok-wide law and justice leader capable of regulating the sortimobs. Thus 4 Mile Stage 1 was characterized in interviews with Miles leadership as having “some komitis but no lidas.” 24 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Local regulatory capability: the four principal 4.  regulatory functions of komitis and lidas This section details the four sets of regulatory capabilities of the komiti/lida system. These are summarized immediately below in box 13 and explained further in the text following. Box 13.  Settlement Problems to be Resolved: The Four Core Komiti/ Lida Regulatory Functions 1. Regulating settlement and tenure: establishing basic security of tenure and physical capital or property • Rental/tenancy/occupancy, allowing and taxing certain usages, including disputes, infringement on another’s property, damage, boundary issues, downstream waste or sewerage impacts, unlawful sale of property; property crime (theft, burglary); eviction of malfeasants (or others) from the settlement; impact of safety and security in relation to infrastructure and commercial development • Peace and good order: overseeing the blok and/or its local markets and bus stops personally via patrols or networks of komitis; regulating youth and children’s public behavior; restricting homebrew, public drunkenness, and damage to property (especially around public holidays and Christmas, when actual funding for some activities from the LLG ward has usually been available); reducing disruptions to neighbors and public spaces; dealing with drunk driving • Personal injury: occurring by accident, from dog bites or as a result of children’s activities, or involving vehicles • Diversion of youths: organizing sporting activities around public holidays 2. Solving disputes, building trust: protecting social and legal-political capital • Mediation around fights, ethnic conflict: fights between individuals and groups at schools, markets, and local sports events, and fights and ongoing conflicts between larger groups, all while recognizing each group’s customs and leadership • Family, marriage and relationship disputes, adultery, violence: extra-marital affairs, disputes between marital partners/family members over money (including inheritance, shares of windfalls); negotiations over bride price repayment; access to children (after awarding of custody); treatment of different wives in polygamous relationships; daughter’s loss of virginity (and thus bride price); family violence, sexual violence, including abuse and rape • Swearing and slander: public or private defamation, talking baksait [behind another’s back], spreading gossip, including possible curses, sorcery accusations • Developer impact: disputes with outside actors, including developers, businesses, infrastructure contractors L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 25 3. Creating local economic security and opportunity at markets: protecting economic and human capital • Local marketplace regulation: what is sold or done and by whom, and managing disputes related to access or trading • Loans and money: disputes at the local market or over street selling, loans and repayment, irregularities around sports or other funds • Diversion of and employment for youths: organizing sporting activities around public holidays, finding work (local contracting, formal employment selection) • Sourcing money for local groups: especially from the ward and around public holidays 4. Ensuring political and legal representation and coordinating with other local regulations: protecting legal-political and symbolic capital • Representing the settlement: in relation to the ward, police, schools, and higher government, and with politicians, though mainly at election times • Defending residents’ rights: in relation to police and legal representation and with regard to the impact of infrastructure and other development • Reinforcing place-based and law/rights-based identity and authority alongside ethnic or wantok-based authority: encouraging locals to defend these in relation to outside conflict • Having cases referred by police: back to the settlement • Networking with/providing cross support: for other komitis and in cases within wider law and justice concerns (police, Village Court, ward, or city domain) 4.1.  FUNCTIONAL CAPABILITY 1: REGULATION as described below). Almost all komiti members are OF SETTLEMENT AND TENURE: ESTABLISHING highlanders, although appointments are based on BASIC SECURITY OF TENURE AND PHYSICAL community respect and performance, not on ethnic CAPITAL OR PROPERTY allegiance or identity. 4.1.1  There are several case examples of regula- 4.1.1.1  Regulatory authority over Lae settle- tory function 1, including the crisis and emerging ments has emerged through numerous strug- regulation at 4 Mile, Lae. 4 Mile, introduced in box gles, as the rise and fall of local gangs has been 12 above, the most notorious of the Lae Miles settle- a significant factor in generating public order in ment areas, is a hybrid tenured settlement, compris- the settlement. Until the mid-2000s, 4 Mile settle- ing areas of formal leasehold on state land, within ment was “regulated” (more precisely, terrorized) Ward 4, Lae Urban LLG. There are three zones: 4 by gangs extorting money on paydays and raping Mile Stages One and Two are on low-covenant women in and out of their homes. Conflict between land,77 99-year leaseholds (with a fringe of infor- gangs and youths raged across informal boundaries, mal settlement on swamp land toward the river), and the centrally located 4 Mile market became the and a third, called Mountain Zone, is an informal site of regular turf fights. In 2007, the severed head settlement on state land. Each zone has five komi- of an Eastern Highlands market trader was placed tis (individuals) appointed by the community for in the center of the market by the perpetrators as life (although each komiti receives some guidance a trophy and intimidating message. The rejoinder from the law and justice chairman/blok president, to the challenge, however, came not just from the 77  No-covenant settlement allocated sections on a self-help basis, with no requirement to build according to housing standards. Low- covenant areas demanded a certain basic standard. 26 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Eastern Highlands group and its former policeman Ahi LLG, its customary titled land over the past two komiti/leader, but from across the 4 Mile community decades has become a highly significant arrival and in the reformation of the local komiti into a wider settlement point for chain migrating families and informal network able to physically respond to any villages from Enga province. Regulation inside Pita particular threat in the settlement. Alerts were gen- Blok is a story of partnership, growing (highlander) erated by community members banging pipes on ethnic hegemony (see Section 5.2 below), and com- steel power poles. Police support was enlisted, mitment to diverse, inclusive settlement. A strong, and summary justice, including extrajudicial pun- long-term relationship between the papa graun ishment, was enacted. Offenders were jailed, their and his appointed Engan blok and law and order families were expelled from the community, and chairman has given the settlement a reputation for their houses burned to the ground, even when they safety that has attracted other ethnic groups hoping had housing corporation leaseholds. In the 10 years to settle. As remarked by an ethnically Engan local since these measures were put into action, violence resident, “They like to live here, because Engans are has not returned to 4 Mile Stage 2. The leader of the strong. For the papa grauns, we are looking after counter fight was the 4 Mile law and justice chair- their land, we make it safe.” man, combining this role with the blok president and chair of the (not gazetted [officially registered]) 4.1.3 The Pita Blok landlord-appointed komiti HaiCoss 2 Village Court. faces some minor competition from another, ward-related committee structure and chair with 4.1.1.2  4 Mile komiti functions include daily a Sepik support base (see Annex 1). There is active mediation and cot events, many conducted at territorial and security management of the blok. The the local police station, with the cooperation of komitis move around the community, especially at komitis and non-confirmed magistrates. Due to night, when they conduct foot patrols. As outlined the relative stability, there have been few evictions below, the local police also perform foot patrols in since the 2007 crisis, and so far, informal tenure in the afternoon on the invitation of the komitis, who the mountain and swamp areas has not been chal- ensure their safety. But if problems come up at lenged. Active youth leaders collaborate with komi- night, “… they don’t call the police, they call us. tis to manage public holiday and Christmas peak We live close together, it’s dangerous… we move pressures, although in recent years, funding for this quick as possible to kill this fire. Other-wise people from the LLG ward councilor level has been scarce, are aggressive and use weapons. We don’t wait for as allegations of misappropriation of funds circulate. daybreak.” As explained in the section on market regulation below, komitis’ influence extends to 4.1.1.3 Notwithstanding 4 Mile komiti capabili- local markets: according to a Pita Blok resident, ties, the extension of effective komiti regulation “Normally we have the same leaders surrounding to engage 4 Mile Stage 1’s youthful Morobean the markets, living together there, working, so it’s sortimobs [young gangsters] remains a challenge. ok, fine.” But the komiti has established his author- Like all other komitis, 4 Mile faces real limitations ity through recent elections. There are no women in terms of its ability to coordinate the links across among the blok or law and order komitis, and wom- the city and to regulate the numbers and possible en’s issues are not properly dealt with by these enti- residences of new arrivals. A 4 Mile komiti related ties (see Section 5). that “now that the population has increased and we have influx of people, and the rent houses which are 4.1.4  Pita Blok settlement regulation extends to not controlled, we can’t identify which people live in housing arrangements for new arrivals. Pita Blok which blok. The population in this blok used to be has more than 20 “renthouses,” multi-unit rental around 17,000. Now it must have gone up to around accommodations, usually popular with younger 23,000-plus population.” recent arrivals to the city, both singletons and cou- ples. These, like residential compounds that have 4.1.2  Another case study of regulatory function 1 young arrivals staying with wantoks, can be prob- deals with the consolidation of order in a chang- lematic sites to regulate, especially on weekend ing arrival community at Pita Blok, Lae. Pita Blok is afternoons, when drunkenness can be widespread. named after its papa graun, Pita Tiba. Situated in Lae The komitis are active in their regulation of youth L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 27 and homemade alcohol; they are active too in the regulation of rent houses, demanding registration and fees and the establishment of house bylaws, including curfews and visiting hours. Rent house owners impose their own conditions, including the careful selecting of ethnic group makeup and a common landlord preference for formally employed residents. Rent house regulation is where the kinds of close local management of settlements that are achievable on custom land become visible. As noted below, there is also spillover of security from Law and Justice leader Chris Waimo at a Rent house, the residential komiti areas into the local market. Kamkumung, Lae. Photo: David Craig/Worldbank Box 14.  Regulating Rent Houses/Entry-Level Housing and Local Businesses in Pita Blok, Kamkumung I tell the residential blok owners ok, you want to earn money, and run a rent house: so. You need to set up bylaws, and everyone has to submit to them. No beer, no drugs, no fighting, you come home by this time, etc. Anyone wanting to run a rent house who doesn’t like these rules won’t get permission from the papa graun. And if any problems come up, the man running the rent house will be responsible: “You, you are harboring these people and now trouble has come up. Since they are staying here, and there are problems, you as the blok owner need to take ownership.” Because the tenant, he’s going to sleep then go, but you accommodate him. We charge different amounts (of fees). If a rent house has 30 rooms, okay, every fortnight each one is 100 kina, so if it’s more than k1,000 you are going to pay a different charge. So the rent house pays k500 or 250, and we make a payment schedule for them. A proportion of the money goes to the papa graun, and a separate proportion comes to the komiti. We have a list, everyone follows the list, and when we say to pay k250, that’s what they pay each month or fortnight. Blok law and justice chairman 4.1.5  Pita Blok’s highly functioning komiti to us. Ee don’t get permanent living, we don’t get arrangements have not resulted in guarantees land titles, we settle only.” that settlers’ tenure arrangements are on a legal footing. The Engan komiti chairman has strong incentives to enable his and other ethnic groups to 4.2  FUNCTIONAL CAPABILITY 2: SOLVING achieve long-term security of tenure. But though DISPUTES, BUILDING TRUST: PROTECTING he has acted to clarify issues of generational SOCIAL AND LEGAL-POLITICAL CAPITAL succession by signing a personal memorandum of understanding with the aging papa graun, the tenure 4.2.1  Komitis function most conspicuously in of settlers is insecure. Land values are rising rapidly, solving disputes [stretim hevi], enabling com- and legally there is nothing whatsoever standing munity life [gutpela sindaun] to go on. Achieving between the papa graun (or his successor) and the safety and security in settlements is heavily shaped sale of the land, followed closely by mass eviction. by leaders’ ability to come together, bargain, and That this happened just down the road at Speedway commit to collective action.78 At mediation events, Blok in 2015 is not lost on the komitis; according to komitis commonly work as a team, together estab- an Engan Village Court official, “We make it safe, we lishing a local arena where authority and its “rules are looking after their land, but they never turn back of the game” are reassembled and recombined to 78  See World Bank (2017b), especially chapter 4 on security and governance. 28 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS produce outcomes that people can agree to and live Box 15.  Adaptation, Leadership, and with. Mediation events perform and enact—in fact, Cross-Cultural Communication within the dramatize—the komitis’ authority. As described in a Komiti System previous research report, “Come and See the System in Place,”79 neighborhood mediation in urban PNG is It’s been going on for a long time; we make an established, locally recognized practice in which a community leader or organization brings together decisions and we bring peace together, even different parties in an openly and explicitly neighbor- in cases like the [Coastal or Morobe Highlands] hood venue to negotiate and agree on an arrange- Sialums and the Kabwums cases, where they ment acceptable to the people living there and to don’t usually solve issues in the open. We others immediately affected. Such “voice” is joined [Highlands] Chimbus, we have a different to local loyalties;80 in the terms famously developed culture but we value the leadership, how you by Albert Hirschman, the ensuing settlement is then play a different role in the community. reinforced over time by the vocal, loyal, and commit- ted leadership of these same participating actors.81 [In many cases] it’s the leaders only. [After a recent fight] I had to go and bring the leaders 4.2.2  As described in a previous World Bank from up there to come down and we talk Urban Safety and Security activity research report, together. They go back to their community, mediation is a widely adopted, closely adapted they talk to their community and get their modality for dispute resolution.82 Although subject views. Then we leaders, we come and sit down to enormous local variation, local capabilities in dis- and discuss the solutions. And we give the pute resolution have been a part of every traditional mediation dates, organize the location, and society in PNG’s long human settlement. In recent call all parties to come to the location and we years, however, mediation practices have coalesced mediate. to the point where they are highly recognizable across PNG’s major cities. What is coalescing has So it depends on the leadership and all the three constituent elements: (1) a translation of village communication: how you network with each or rural ethnic-based tradition [pasin blo ples], (2) a other and you brainstorm together. People distinctly urban hybrid kastom law based on exten- will respect you and come to an understanding sive local but networked practice, and (3) an adap- to make peace and it depends on the leaders. tation of globally recognized mediation practices, It’s not the community but the leaders who as promoted across PNG by the Peace Foundation makes the decision. But if another ethnic Melanesia, during the 2000s. group is involved, it’s very hard to communicate. That’s the reason why we have 4.2.3  These processes involve the significant trans- set up the Haikos [Highlands-Coastal] leaders’ lation of ethnic custom and also its adaptation into group here. If there’s a fight between the modern contexts. It is not just the content (or institu- Highlands and Coastal, we have leaders tional “rules”) that gets translated in these processes. working together already so we have a good Institutionalized leadership roles are rearticulated and understanding of each other and respect of renegotiated within these relations, especially as (as each other. Lae 3 Mile law and justice chairman developed further below) Highlands-based “men’s house” [haus man] roles and relationships interface with Coastal leadership structures related to chiefly authority. In these processes, komitis are not simply 79  Craig, Porter, and Hukula (2016). 80  Hirschman (1970). This definition aligns broadly with international perspectives on local mediation, also known as “insider” or local, community, or 81  neighborhood mediation. See, for example, Mason (2009); UNDP and EC (2014); Suurmond and Sharma (2012); Lederach and Thapa (2012); and AFSC (2013). 82  Craig, Porter, and Hukula (2016). L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 29 domains of consensus, cooperation, and adaptive magistrates) and aspirant komitis often join the pro- harmony; rather, they embody and reproduce power cess, jumping in to lead the questioning or discussion relations between ethnic groups, (mostly male) lead- or withdrawing while another takes the lead. Indi- ers, and women and youths, and between competing vidual komitis are seen as representing their ethnic forms of local authority (police, Village Courts, ward group in these situations, leading some to recuse governance). themselves in some cases and others to be told to stand aside. Conversely, they might be asked to stay, 4.2.4  PNG’s settlement residents approach media- because their participation in the consensus adds tion not just as individuals, but inevitably as mem- authority. After a fight, procedures start rapidly, with bers of a communal or ethnic group. Each group initial meetings within hours of the bout or event, fol- brings with it beliefs about justice in general and lowed quickly by “gut cooling” [bel col] compensa- how these relate to the particular case at hand. In the tion payments. A formal mediation hearing is sched- course of mediation, these beliefs and perspectives uled, sometimes within days and other times after are actively aired and frequently explicitly recognized the initial bel col payment has been given time to as reflecting the custom or pasin [fashion or legal calm passions. After the mediation event, the formal usage] of the group that reaches back to the rural handover of the compensation payment is scheduled setting of the original migrants: pasin bilong Sepik, for a date one to two weeks into the future to allow or “the Sepik way.”83 In everyday practice, this rec- for the communal collection of contributions. If it is ognition is broad brushed, expansive, and flexible; it clear that initial or negotiated resolution amounts might relate to the acceptance of polygamy or to the will not be reached, the mediator needs to go to the conviction that some groups (for example, Papuans) groups and try to negotiate a lower payment. The are not familiar with large compensation payments. formal handover itself may involve final negotiations, Each of these traditions has its own gender framings; but once compensation is agreed, serious efforts are many are highly patriarchal, but significant matrilin- made to express goodwill and an end to the conflict, earity also exists in parts of PNG. again underlined by the presence of wider commu- nity and urban leadership. 4.2.5  Participation in mediation events consti- tutes a significant investment of a family’s time resources. Extended family and kin/ethnic groups Box 16.  Mediation Fees and Leadership are expected to attend three or four sessions, and Recognition dozens—if not hundreds—of them do. This invest- A lot of komitis have been there a long time, ment involves a combining of various forms of capi- others are more recent. But they all speak the tal: the human capital of time, skill, knowledge; the (thick) social capital of trust in intragroup relations, heart [Tok Pisin bel] of all and oversee law and being translated into (thin, bridging) intergroup rela- order and community issues, and they keep tions; the cultural capital of knowledge of custom and on doing it. They all work for free, without the basic national law, and how these might be applied recognition of the government. When they in urban, intercultural settings; and the symbolic cap- hear a case, they charge 50 kina, but that only ital of a formal event and process, presided over by buys them cold water or betelnut. recognized leaders and turned up to and respected Women’s committee member, Kamkumung by different groups and outside witnesses. 4.2.6 Mediation involves commitment and facili- 4.2.7 Mediation is resourced by table fees that tated consensus at several steps over time. At are often calculated as a percentage of the Lae’s 3 and 4 Mile and Nawaeb bloks, komiti media- agreed compensation payments. Together, these tions are held in open, public locations (under a local combined costs are usually substantial enough tree, in a field, at the local police station, beside the to require further extended family commitment. market). There, local komitis (some of whom have Mediation or table fees are usually arranged as a part been identified but not gazetted as Village Court of the agreement and paid before the settlement is 83  See Demian (2003); and Filer (2006). For international comparatives, see Wolfe and Yang (1996); and Anleu (2009). 30 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS announced. Fees are often calculated as a percent- such as the police, local komitis, Village Court age of the overall settlement and might be as much magistrates working outside court hours, or other as 10 percent of that amount, but divided between public figures. the often plural mediators and yielding a typically low hourly rate (not including, however, any side pay- ments made to mediators). For violent crime involv- 4.3  FUNCTIONAL CAPABILITY 3: CREATING ing serious offenses, however, including family and LOCAL ECONOMIC SECURITY AND sexual violence, the fees are large enough to be an OPPORTUNITY AT MARKETS: PROTECTING incentive for mediators to hear cases and ask for fees ECONOMIC CAPITAL that are well beyond the formal limits imposed, for example, in Village Courts, and to claim customary 4.3.1  PNG’s local urban markets are basic to urban legitimacy for doing so (see below). But in general, life and livelihood. The country’s urban produce the relatively small fees (physical capital) charged by markets range from central markets, which generally the mediator and his or her team are only one form of combine wholesale with retail trade, through to much capital being exchanged; reward and value are seen smaller local or neighborhood markets. The latter in the recognition, security, and trust gained. The tend to cater to more diverse needs and market func- compensation payments themselves, however, are tions, including retailing fresh fruit and vegetables, a substantial and credible commitment, as well as a but also hot food snacks, cigarettes and betelnut, mechanism of punishment and reward. old and new clothing, crafts, and the cheap everyday electrical, plastic, and other household items known 4.2.8  Group commitment and cooperation around as “minigoods.” In addition to being a functioning the mediation are needed to make agreements site of livelihood and economic opportunity, local stick. Participants are thus highly motivated at markets are also a center of social and recreational mediation events, as basic personal and family safety activity and a forum where different groups with dif- are at stake, as well as time and physical capital. ferent power and resource bases are represented.85 An agreement needs to be not just negotiated and agreed, but adhered to as a genuine resolution of the issue, embodied in a judgment or pact approved across family or ethnic lines. Ultimately, the authority to make the deal stick through commitment over time84 comes from all sides. It depends only partly on the mediators themselves. Much more important is the agreed commitment of the families and clans involved, and the sealing of the agreement through mutual compensation payments. In the words of a Moresby mediator, “It’s because everyone there [at the mediation] has been a part of it, that’s why it will last. The family, the neighbors, the police, the komitis; everyone wants it to be over, for the thing to be sorted out. They want other people to know that is the case; people who know both sides. Mediation belongs in public, with witnesses. It’s not something small you can walk away from, and no one will say anything.” This commitment is further reinforced by some sort of ceremony and the witnessing of the Snack vendors, Awagasi market, Lae. Photo: David Craig/Worldbank agreements by local and outside authority figures, 84  See World Bank (2017b), 54. 85  See Craig and Porter (2017). L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 31 4.3.2  Self-regulation and ethnic diversity are core features of local markets. Local markets in Port Moresby and Lae are mostly regulated by vendors themselves, backed up by immediately available/ appointed market komitis (individuals given komiti status in the market). In Lae as elsewhere in PNG, different areas and specialties within the market have evolved as the domains of different groups: snack, single betelnut, and hot food sellers to the front, bulk betelnut sellers in another zone, fruits, vegetables, used clothing, minigoods, and so on, elsewhere. Dispute resolution, Tokarara Market, Port Moresby. Photo: David Craig/Worldbank Table 2.  Mixed Marketing: Specialization and Diversity at Awagasi Market, Lae Total Usages/Ethnicities Enga EHP SHP WHP Chimbu Coastal vendors Wholesale buai 7 7 14 Wholesale cigarettes 4 4 8 Retail buai/cigarettes 20 10 30 Minigoods, household items 7 4 2 11 Cooked food 15 1 16 Drinks 17 9 26 1. Fresh vegetables 9 14 4 3 30 2. Used clothes 2 1 3 3. Dried tobacco 1 1 4. Firewood 2 1 3 5. Fresh fruits 2 1 1 4 6. Moneylending 1 1 7. Frozen meat 3 3 8. Baked goods 2 2 9. Eggs 5 1 6 10. Bilum/craft 1 1 11. Store clothes 4 4 12. Live poultry 1 1 13. Phone/accessories 3 3 Note: EHP refers to Eastern Highlands province, SHP to Southern Highlands province, and WHP to Western Highlands province. Source: Authors’ field notes. 32 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Map 4.  Awagasi Market, with Different Usages and Ethnic Groups 4.3.3 Many local market sections or usages are ethnically dominated, but overall, ethnic plurality commonly predominates. Plural ethnic groups and communal affinities work as both a risk and a “col- lective action” resolution. Individual disputes always threaten to escalate rapidly into ethnic fights, but any dispute also faces communal scrutiny and collective interest. In practice, this means that few disputes will be allowed to escalate to the point where the integ- rity of the market as a whole might be impaired. Quick de-escalation of problems is important but also pos- sible—and highly incentivized, as a bad reputation among vendors or buyers will diminish patronage, and everyone loses. Rogue traders prepared to “big- head against the market” are routinely sanctioned at Vendors, Awagasi market. Photo: David Craig/Worldbank several levels and commonly excluded. L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 33 Box 17.  Self-Help/Communal Regulation at Awagasi Market I feel safe here at Awagasi market because there are so many Engans selling at the market, so we support each other. Table Market vendor from Enga province I come here to do my marketing because this is an Engan market and we control and look after the market. The Engans look after the market properly with no trouble or problems, so everybody likes to come here to do their markets. Betelnut customer from Enga If you disturb me and my market, you disturb all the Kaguas doing market in Awagasi market. Table market vendor from Kagua in Southern Highlands province Everyone at the market or the community contributes. They are making a living in there! So they take ownership of the place. There is competition between rival ethnic groups, but they all ensure that law and order is maintained. The papa graun, VC, komitis work together, and the community supports these leaders to make the community safe for all to live. Eastern highlander customer 4.3.4  By definition, it is more difficult to regulate 4.3.5  Market traders themselves are unlikely to be market usages that routinely introduce problems able to regulate the entry of problematic usages (and their regular perpetrators) into the wider into the market or their containment in separate market. Betelnut and cigarette selling, both staples sections. There are immediate gains to be had from of local markets, have this reputation, as they com- having males with discretionary resources in the monly attract young males and aggressive High- market buying food, snacks, and vegetables. More- lands groups. Illegal or quasi-legal activities (mari- over, advocating or prosecuting exclusion is dan- juana sales, gambling games [cards and darts], and gerous. Here, papa grauns and komitis supported especially the sale of homebrew) also attract these by clear bylaws can have enormous influence. The vendors, as well as buyers who will socialize around papa graun in Awagasi market is very much hands- the points of sale, variously intoxicated and variously on, prepared to evict vendors and buyers alike, and prone to displays of money or other male dominance. has closed the whole market for months at a time Harassment and robbery of vendors and customers after trouble. With his young, fee-collecting komiti, (and passing traffic) are regularly ascribed to those he has effectively enforced a ban on several high-risk involved with or hanging around these activities. activities. Similar hands-on traditional leadership on Back Road has also simply closed down the develop- ing Centa Market due to insecurity related to youths Box 18.  Betelnut [buai] Trading and Ethnic and homebrew. Market Violence, Tokarara Market It’s always about buai. The fighting. There’s always a group who want to control the buai market. When they get drunk, they like to complain about the other people who come in. So a Wabag man is from here, but then he will go outside and get their family from Wabag to come in and sell too. And the Wabags are selling cheaper, and it brings the price down—they want to get all the customers. So the others don’t want the outsiders to steal their customers. And they challenge [the newcomers], physically. They kick the stall. The family come and join in the fight. Female market vendor Buai (betelnut) vendors, Awagasi market, Lae. Photo: David Craig/Worldbank 34 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Box 19.  Safety and Security Outcomes at 4.3.7 Making the market an ethnic- and gender- Kamkumung, Lae inclusive arena of opportunity—and a financial success—requires a diversity of uses in the market. Awagasi market is much safer and accessible In achieving market security, the diversity of usages than Kamkumung market because there are and the different ethnic groups specializing in them no drug bodies and steam (homebrew) bodies constitute both an opportunity and a challenge. In around the market. We feel safer to have our regulatory terms, there is diversity and competition lunch here. Students from a nearby girls (the essence of a functioning market),86 potentially vocational school. forming a stable ecology of opportunities in which each group/ethnic identity feels that it is “their” I do vegetable shopping at the Kamkumung space. This enables ongoing commitment and invest- market early in the day, because in the ment. There are incentives for all to work to make afternoon it’s all crowded and there are this happen, but the local authority, whether it is a possibilities of drunkards around the market, papa graun and/or komiti, is crucial. Achieving this which disturbs the market. Awagasi market is equilibrium requires that a papa graun or komiti not safe, with the papa graun keeping the market ignore ethnic differences and power asymmetries but under control, with the help of the police instead actively work with them to create a function- presence in the afternoon. The market vendors ing, complex, and multi-use environment in which mind their own business by selling their stuffs ethnic and other dominance might exist but does not and you are free to do marketing without become the basis for exclusion. intimidation, so if anyone wants to disturb the market, he is asking for trouble from these vendors because that’s where they earn their Box 20.  The Papa Graun, Komitis, and living. Market customer, Kamkumung Cooperation at Awagasi Market Hiob [Awagasi, the papa graun] keeps control of the Awagasi market close to himself 4.3.6 Horizontal authority achieved through the because of constant competition between representation of ethnic groups in market komiti rival factions of the Engans, Eastern Highlands, management seems not to be as important as the and the Southern Highlands of who would sense within each group that it is included and control the market. He makes each and every secure, under whichever komiti and papa graun vendor or ethnic group be involved in be- leadership is in place. Komiti success in regulating coming part of the system. Whenever there is nearby settlements helps; according to an Engan a problem, everybody takes part in solving it market vendor, “The market here is good compared because there isn’t a lot of komitis available… to Kamkumung market, because in here fighting but Hiob gives the ownership and respons- never spills over to residential areas but is contained ibilities to the vendors themselves. So, there is where it started.” The fact that a Village Court or a cooperation between ethnic rivals, but there is active mediation practice is nearby also helps, as the also a quest among who could be best at solving market is continually generating some kind of issue, hevi/problem. Market Vendor, Kamkumung especially around debt and its repayment. The prox- imity of the police station also matters; at Awagasi market, this relationship is cultivated and at the large, multiethnic, betelnut-dominated Bumbu market, the head komiti has the local Chinatown police on speed dial. But he has also fenced off a part of the market that is restricted for the exclusive use of Sepik buai sellers. 86  See Craig and Porter (2017). L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 35 4.4  FUNCTIONAL CAPABILITY 4: POLITICAL AND 4.4.4  But Village Court expansion into settle- LEGAL REPRESENTATION AND COORDINATING ment areas that are dominated by komiti and lida WITH OTHER LOCAL REGULATION: PROTECTING arrangements has been slow and uneven. In part, LEGAL/POLITICAL AND SYMBOLIC CAPITAL this may be due to the collapse and ongoing vola- tility in global commodity prices, which resulted in 4.4.1  Core representational and coordinative reg- budget cuts to the second round of expansion. This ulatory functions involve the komiti and lidas in co- meant that Village Courts in areas including 4 Mile production, symbiosis, and overlap. All of the func- were not officially gazetted, nor were the additional tional capabilities of komitis are co-produced with magistrates that were appointed, including most of other forms of local authority. Section 6 of this report the women magistrates, or they were gazetted but will consider the more formal elements of these rela- then unpaid and ungazetted. Many of these persons tions, particularly with the police, LLG and district are still active in local law and justice activities, and government, and emerging city authorities. Here, the some claim Village Court magistrate standing in the focus is on how komitis function as mechanisms of interactions with komiti colleagues. Hopes that the legal and political representation in relation to agen- province would use some of its law and justice func- cies of the state—in this case, the Village Courts — tion grant to fund the second phase of expansion on and local corporates and infrastructure developers. Village Court salaries have so far not been realized. 4.4.2 Village Courts, where active and respected, 4.4.5 In many settlements and neighborhoods, represent a vital and important part of the komitis’ hybrid law and justice arrangements are emerg- regulatory environment. Individual komitis may well ing that involve Village Courts and komitis work- be court magistrates or “peace officers” (bailiffs), and ing side by side, cooperatively or competitively. Village Court magistrates will participate in (and some- The boom in Village Court investment happened times compete for cases with) komiti mediations, often after years of neglect, when the urban boundaries on as senior members of a wider team of local komitis which court jurisdictions are based had been vastly and leaders. Village Courts can be very flexible in their expanded. Of Lae’s two LLG areas, the urban zone forms; they are as likely to comprise a few magistrates with its six wards received all seven of the new Village and komitis as they are to be a bench sitting of the Courts. Lae Ahi, which has the anomalous distinction five gazetted magistrates. In Lae, various ward, blok, of having seven of its 17 wards outside Lae District, and section cots model themselves on Village Court received no increase to its four Village Courts. But practices, issuing summonses and orders, collecting local law and justice will not wait for formal appoint- fines and fees, and imposing compensation. ments, and all of Lae Ahi’s wards have active law and justice chairmen and komitis, working at ward, zone, 4.4.3 The formal Village Court has gained local section, area, and blok levels and holding regular significance in recent years, particularly in urban mediation cots at all of them. contexts where it was not expected to succeed. Expectations that Village Courts would not work in 4.4.6  Although there is some competition urban areas have been overtaken by actual experi- between komitis and Village Courts for mediation ence, which mirrors the success of komitis as arrange- work, in general, the presence of both is advan- ments able to combine different kinds of ethnic tageous. Komitis are a proving ground for aspiring custom and authority. As a result of the 2012 Alotau Village Court magistrates. This collaborative compe- Accord pact of the government of Prime Minister tition often gives complainants a choice and limits Peter O’Neill and revenue from PNG’s long resource opportunities for rent seeking from Village Court boom, Village Courts have received increased fund- magistrates. Cooperation with the Village Court adds ing, including for magistrates’ stipends and an authority and a mandate to issue orders. Being able increase in the number of courts, especially in Lae to morph instantly into mediation means that Village Urban LLG. The number of magistrates was increased Court magistrates can hear all sorts of cases as soon to seven, one of whom was necessarily a woman. The as they arise, beyond the formal limits of the Village selection of Village Court magistrates in Lae seems Court mandate, especially in relation to statutory to have gone well; quality komitis were selected compensation limits. As needed, they can fall back and, in many cases, formally gazetted and placed on on the full bench, as the Village Court offers a point of the payroll. appeal or a higher court when komiti mediation fails. 36 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Box 21.  Cooperation with Village Courts 4.4.7  Komitis and lidas have a crucial role in repre- and a Dual Role for Village Court Magistrates senting the blok and mediating relations with cor- porates and developers. Urban settlements along as Komitis major infrastructure routes face significant challenges in safety and security as well as law and justice. The full [Village] court. That is the rule of law. Informal settlements, mixed into dense residential What we do down in mediation is different: it’s “arrival” neighborhoods, spring up alongside these to assist them to come up with a decision. We routes, only to face problems as young populations allow them to talk, and when they get stuck, lack opportunities, property values rise, evictions and you give them some ideas. They get tied up related conflicts start, or the sites are bypassed or into a situation where they can’t get out; as marginalized by route upgrades. Communities, and mediators, we offer ideas, solutions to the especially their youth, frequently become identi- problem, and they decide whether to buy [the fied with and stigmatized by their relationship with solutions]. If it can’t be solved there, we take it the road (“He’s a 4 Mile boy”). Petty crime at bus to the full bench. Village Court magistrate stops and street markets, as well as (literally) high- way robbery, attracts extra attention from media, police, citizen, and corporate interests. At the same time, infrastructure development and corporate busi- ness activities in these localities can provide gainful employment for local youths (see the case in box 22 below). Box 22.  A Corporate and Lae Miles Youth Leaders Try to Engage Komitis to Regulate the Highway 4 Mile on the Lae section of the Highlands Highway has long had a reputation for youth crime, especially armed holdups on the highway. The recent expansion of the highway to four lanes represented an opportunity for change, but design elements meant that level crossings for pedestrians could be used to stop and hold up traffic. Following the very public armed holdup of a corporate vehicle at 4 Mile, a group of youth were sponsored by the corporation to try to improve safety and security at the notorious hotspot. Youth were employed on casual community work and with the business sought to engage law and justice leaders and komitis. The engagement was a kind of regulatory challenge for the komitis: could they and the youths work with the company and others to create diversion and employment for youth and safety and security on the highway? Komiti leaders were recognized as the only viable form of local leadership, and they dominated elections to the new Lae Miles Community Safety Network. The leaders and komitis had achieved some success in getting compensation from the Chinese road company for drainage issues. As a community safety pact involving a diversity of recognized local authority, the network held some initial promise. But despite some high-profile awareness raising success, the network was subject to very uneven engagement from komitis. Komiti capabilities, roles, and routines were aligned to set up mediation processes around grievances and not to organize a network that could reach out to marginalized youth. As a result, youth were disillusioned and sidelined from decision making by leaders and komitis, who rarely showed up to meetings. The company’s contribution focused on informal employment for the youth; leaders and komitis were not compensated for their time commitments, even in token ways (e.g., a phone card, for example). L ocal r egulatory capabili ty: t he four principal regulat ory func t ions of komit is and lid a s 37 The youth leaders themselves struggled to engage beyond highlander kin networks, especially into 4 Mile Stage 1, where most of the holdups were committed by youths unconnected to family and highlander kin groups. Getting commitment and coordination along the highway proved especially difficult for youths, whose local leaders lacked both horizontal links as well as links to vertical authority, and levels of mistrust between the different Miles youth groupings were high. A Lae youth training organization offered special courses for Miles youth, but funding secured from a donor law and justice partner for expanding the network and its training was not expended, as youth struggled with community group formalization and with maintaining connections to the training agencies, the donor program, and the Miles senior komitis. The outcomes point to a number of factors working against the ability of komiti and youth organizations to cooperate and coordinate. Komiti functions and capability were very much related to the dispute case, and komiti members granted little time or even communication resources (such as phone credits) to the community development leadership (they rarely even attended monthly network meetings). Neither youth nor corporate and nongovernmental organizations (NGOs) involved with the network were able to secure serious komiti commitment or cooperation. In the end, a komiti leader leveraged his position on the Miles Community Safety Network to create a wider employment network involving a different company under his own control. He lost the youths’ confidence, especially when he appointed only his local youth to the employment positions available. He was evicted from his position on the network. The two employment programs carried on, with the original still relating to the original corporate, which has gone on to employ several youth leaders on a full-time basis. The leader who started the alternative program lost status in the komiti network; his involvement with politics, as well as questions about his handling of grants from the ward for seasonal neighborhood security, left him sidelined for a time. Meanwhile, expanding the highway to four lanes led to an increase in vehicle speed (and many fewer holdups) along the route. Locals, however, counted 16 fatalities in the three months after the road’s opening, as level crossings painted onto the highway were rubbed out because they enabled youth to hold up slowing vehicles. Komiti protests have not changed the situation. 38 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Power asymmetries, outcomes, 5.  and the limits of local regulatory reach 5.1  THE LIMITS TO KOMITI’S AUTHORITY domination by older men offers few openings. Arguably the most salient area of power asymmetries Section 4 described four broad functions of local related to komitis is gender, however, a significance regulatory mechanisms: 1. Regulating settlement made more acute by the prevalence of gender- and tenure, 2. Resolving disputes, 3. Regulating local based violence and the importance of the komiti as markets, and 4. Representing the settlement itself a site for women to seek redress and security. Here, to the wider authorities. It showed how komiti and the persistence of patriarchy means that komitis are lidas clasp together different kinds of local authority exclusionary and implicitly one-sided; women’s trust and illustrated the kinds of regulatory capability this in the komiti system is therefore greatly reduced, creates. Section 5 considers the ways local regulation and outcomes, in terms of the absence of protection interacts with the very unequal power relations that against prevalent family violence, can be fatal. exist within and around settlements to produce uneven security and capital outcomes for different 5.1.3 As a consequence, men who were inter- groups. These include power asymmetries between: viewed often expressed more confidence in local women and men, ethnic groups (for example, regulation and, as in many similar places across the highlanders and coastals), age groups (including globe, felt safer in neighborhoods and settlements youth), and social and economic classes. than women and youth. These asymmetrical capabilities mean that komitis put at risk the various 5.1.1 To regulate settlements, komitis and lidas forms of urban capital—social, economic, political/ need to be able to pact, commit, and coordinate. legal, and so on—of all residents, but especially Part of this involves coming to consensus, though the most vulnerable. Failures in legal and political consensus is not some kind of state of natural representation, for example, can leave everyone equilibrium. Rather, very real power asymmetries exposed to the loss of rights and potential violence, are produced and reproduced within the process as well as to trust-denying collusion and corruption. of regulating local settlements. Coordination is fragile, effective commitments are constrained, and outcomes from these processes often reflect 5.2  ETHNIC ASYMMETRY IN SETTLEMENT fragmented, exclusive, or captured power relations. REGULATION: HIGHLANDER HEGEMONY But even in the best circumstances, where consensus is achieved and shared commitment is generated, 5.2.1  Komiti procedures are not simply about the situation cannot be separated from the power collaboration and consensus building along PNG’s or hegemony of particular groups. Creating order familiar words, bung wantaim, wok wantaim [meet demands some kind of authority and credible together, work together]. Some traditions come to commitment or enforcement, and this itself derives dominate and others are excluded. And in practice in part from hierarchies of leadership and respect and effect, this domination in itself becomes part of for education or mediating skills. But it also derives the institution’s power to forge local commitment from patriarchy, ethnic hegemony, gerontocracy, title and resolve disputes. holding, and recognition from the state. 5.2.2  It is common for PNG’s central highlanders 5.1.2 This has serious implications for legitimacy to emphasize their distinctive leadership role and and effectiveness, and also for the outcomes that influence in practice on komitis. As described in produce asymmetries by ethnicity, age, and gender. box 23 below, they stress the relevance of traditional Ethnic—in these cases, highlander—hegemony can Highlands male modes of dispute resolution, win- be inclusive, but it can also marginalize people and win justice, and mutual compensation payments. leave them vulnerable. For youth, komiti structures’ But other groups also recognize (and some resent) Pow er asy mmet ries, out comes, and t he limit s of local regulat ory re a ch 39 the ways that both traditional dispute-resolution Box 23.  Highlander Hegemony in approaches brought from the Highlands and Settlements and the End of Raskol Rule Highlanders’ own raw power in community disputes have provided an important basis for mediation, the Before the highlanders came in 80s, 90s, komiti system, and, as outlined below, the emergence things were worse, much worse. They came of wider lidaman roles. and changed this community. Before, the policemen, servicemen tried to control this 5.2.3  At Lae’s Pita Blok, the situation is, in many area, but there were armed gangs, the gangs minds, an unexpectedly positive example of could march into your house and do anything: highlander arrival and even dominance. As in car theft, robberies. They came armed into chain migration patterns elsewhere, Pita Blok has the community, they were known to us but we become a preferred arrival destination for young couldn’t stop them. The highlanders tried to people and families from a particular part of Enga fight against those raskols, to stop them from province. In urban areas, the reputation of Engans doing that. They would go straight to the is generally unfavorable; they are seen as among the most assertive and aggressive settlers, who raskols’ house, show the police the suspects, “fight first and talk later.” But the strong and stable and then burn the house. Nowadays, these leadership exerted by the Engan komiti chair—and are not the raskols from the past: they are new the traditional landowner’s high trust in him—has ones, more into steam and marijuana. They led other ethnic groups to feel confident that they just young people, not organized gangs. The will be treated fairly in disputes and settlement highlanders cleared the real raskols out. They questions, and in access to local markets. Thus, the drove them out, and there’s no raskols from Pita Blok komiti in effect coopts the ethnic power of a the highlands here. Papuan church pastor, dominant highlander group but also constrains it by Horse Camp, Port Moresby bringing it into a komiti-based policy arena where it produces peace and good order locally.87 Lae law and Justice leader James Kinjin at a mediation event, Nawaeb Blok. Photo: David Craig/Worldbank 87  See World Bank (2017b), 120. 40 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Box 24.  Comparing Highlands and Coastal Dispute Resolution: Popular Discourse A Highlands view: In my view the Coastal [people, i.e., Morobeans, including upland Morobeans] are not very good in solving problems, because in their village, they are scattered all over the place and don’t sleep like a community. You will find two or three families living near its other [ethnic group] members. Whereas in the Highlands we have this hausman [male-exclusive meeting house]; when we have a problem we go to the hausman and we sit down and listen to the leaders talking and come to an understanding. So we solve problems. But in Coastal areas you don’t see this sort of practice or gathering. The communication and understanding between the different leaders are not there. That’s why you see people are having more problems, you know, like the [Morobe uplands] Kabwums and the Nawaebs. They had a big fight last month, they got their own leader but he could not solve the problem because he does not have a good relationship with the [wider] community. So they had to get leaders from outside to solve the problem. Lae settlement dweller A Morobe/ Coastal view: In the highland areas you have all these men wanting to become leaders… they will talk, that’s how they get on top of each other and gain respect. Whereas in the Coastal areas, only one person talks, and that is the chief. And people respect that decision from him, he has the final say in everything. That’s why in every blok, highlanders all want to become the leader, they want everyone to listen to them and acknowledge their superior views. But no Coastal person wants to become [one of] the settlement leaders. We are used to just one leader, the chief. Why should we talk? We have a leader there everyone respects, and he’s the only one. Lae settlement dweller 5.2.4  Thus, Highlands’ style of dispute resolution 5.3  THE EXCLUSION OF YOUTH has become hegemonic in both the practice of mediation and the wider komiti structure and pro- Major asymmetries in settlement regulation persist, cess. In doing so, highlanders have created possibili- including with regard to class, age, and especially ties for other Highland elites and ethnic group mem- gender. But there are other major asymmetries that bers, bringing them into a process of local problem pose challenges for komitis and lidas. Understanding solving. At best, highlander dominance can locate how well the komitis function and what their disputes within an extended, carefully managed and limitations are also depends very much on who you staged time frame. ask, and how those groups have been affected by the asymmetric power relations komitis work from. 5.2.5  But non-highlander settlers, known in Komitis themselves have narratives of their own Morobe province as Coastals, get left out in a range success, albeit tempered by fears for their own personal of different ways. They are less likely to become safety and a lack of recognition, especially from police komitis and to adopt Highlands dispute modalities and civic leaders, but also increasing strains as settler in the first place. Coastals notoriously resent the numbers increase. However, interviews with middle- exorbitant compensation payments highlanders tend class residents (known as working class, or formally to impose. Enforcing decisions can be problematic employed), youth, and especially women produced for them, especially where their sorcery practices are a much less positive view about the komitis: a great rumored to be brought into play as a covert, spoiling deal of mistrust and narratives of ineffectiveness, low tactic not easily covered by mediation. Adaptation is capability, overstepping their mandate, and collusion progressing, though, as the roles of mediation and with the perpetrator. komitis deepen. Skilled leaders, Highlands or Coastal, learn and will try to incorporate all kinds of ways to frame agreements within mediation approaches. Pow er asy mm et ries, out comes, and t he limit s of local regulat ory re a ch 41 Box 25.  Middle [“Working”] Class Views of Komitis The other thing is that most of the Komitis are not working; they are unemployed and the fees they collect is how they make their living to buy food. They get money from the victim’s husband so they won’t take the case to court, won’t take it too seriously. Mediators are not skilled properly to mediate, really—what background to they have? Who has really trained them? They just look at the surface results of family violence and not the root causes. And the root cause is males dominating females. So they smooth it over on the surface side. So people, women draw back. You are unprotected, not recognized. If you think you can you can take it on and refer it to the police. Different groups of people take it on themselves to enforce the law or deal with trouble. Here at the market or at home with neighbors, working class [formal sector employed] people are offended by the drinking and what it leads to. They take it on themselves to enforce good behavior: “If you drink, you behave!” Well, the illiterates, if they are confronted, they will make an issue out of it too. They accuse the working class people, and take it to mediation. The working [middle] class, though, they will just chase or belt the illiterates. Moresby mediator 5.3.1  Are komitis and youth subordinating a threat Box 26.  Lae Mothers Talk about their Children to public order? The decline of urban youth gangs and Other Youth and their failure to become institutionalized were described in Part 1 of this report. A number of factors There are lots of problems with youth and were involved: the emergence of multi-generational children, sitting around at the blok. They get household structures, the lack of a major illicit econ- drugs and make homebrew… lots of them omy and a high-security prison, and the development have got grade 10 or 12, but they are still of other institutions (church, schools, sports) and, in unemployed. They ask their families for particular, highlander hegemony. Komitis now say that money, and fight with their mothers for money they are not dealing so much with gang or organized crime and robbery/burglary threats as with “rats and when they are drunk. When they are involved mice” issues, involving (mainly male) youth getting with sports, or get a contract [from the papa in fights when drunk (especially under the influence graun] to clean up the cemetery, or if they can of homebrew and marijuana). Male youths are also sell mobile phone cards at the local market, involved in violence within families, particularly fight- and they have some money, it’s OK. ing with mothers over money. The kids that are doing the holdups along the highway are us widow mothers’ kids. My teenage kid does not want to go to school; he is doing these things. But I am a good friend of the community, and I tell him ‘you will not turn your frustration on the community. He doesn’t beat me, I usually beat him. When they don’t eat they go and cause more trouble. So my teenage boy usually brings them over to our house and asks me to cook rice for them. I usually give them some money to go buy greens, kaukau , rice, tin fish to cook and eat. My teenage kid will not bash me because I control him; if does not listen I belt Women including leaders from the notorious 4 Mile settlement. him too. Photo: David Craig/Worldbank 42 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Kumkumung Leader Chris Waimo with a group of local youths. Photo: David Craig/Worldbank 5.3.2  The settlement youth problem is a matter of might broker scarce formal-sector jobs or facilitate wider disadvantage and stigma that komitis cannot local trading. In Niu Rocks, “leaders here find work realistically deal with. A young man will be a youth for us at the local canning factories, or contracting well into his 40s, especially if he lacks either formal work, making fences or cutting grass. They opened employment or still lives with extended family, even if the local market and let us trade there, and gave us he is married. Young women, on the other hand, do work patrolling the bus stop. But the leaders closed not seem to be categorized as youth at all; they are the market—too much fighting between sellers.” even more ensconced (and often invisible to outsid- ers) in domestic economies and petty trading, craft 5.3.3 Youths’ relation to komitis is therefore not manufacture, or some other contribution to subsis- simply oppositional or criminal. Yet, very few are tence. A young settlement man is much more likely to actively involved in law and justice either, though be underemployed and working to support the family many would like to be. Nor do youth generally in petty or table marketing to boost the overall family have ways (such as business links beyond kin busi- income than to be involved in gang-related crimes. nesses or other forms of thin or bridging social Those involved in the holdups at 4 Mile, a number capital) to link with other youth beyond their local of whom were met in the course of this study, were area, or (outside of sports and church) to establish likely to be from broken families where, by their own relations with leaders at ward or other levels who frank admission, actual food shortages and the inabil- might help them organize. Rather, interviews with ity to coerce money from their mothers drove them to youth revealed a mix of some respect for komitis and “stand up on the road.” Others’ “sitting ‘round” [stap local leaders, with much distance and distrust. They nating] behavior is likely to involve everyday social- also showed how asymmetric power relations affect ization related to selling or recreation: playing guitar, not just access to resources or justice, but physical maybe selling cigarettes and/or marijuana, waiting for safety and vulnerability to violence. Here, as box 27 casual work. Much youth employment is informal or shows, the komitis candidly admitted to exercising short term and contractual. They might have access control through hierarchies of simple violence, both to a leader who, as a local patron, organizes or pro- within their patronage networks and in their wider vides petty, perhaps security- or construction-related work arena. employment for them as “his boys.” Other leaders Pow er asy mm et ries, out comes, and t he limit s of local regulat ory re a ch 43 Box 27.  Lae Youth Talk about Trust in the Komitis Most of the community leaders in law and order are biased in their decision making, and a lot of them get bribe money to make decisions. So most of them are not trusted in decision making any more these days, and there’s no trust either in what they are doing and where they would want to lead us. They are also greedy and creative in finding ways to use the youth’s name to get money from the government or NGOs. But they are not so creative in finding ways to get the youth involved. They use youth’s name, but don’t include us; youth hear about it and are angry. A lot of youth would like to get involved in the law and justice; we have made a request for that, proposed that, but nothing. Also the community leaders along the Miles are not so strong when it comes to problems between different Miles communities. They can solve problems within communities, but if it’s outside, you are better to go and try to get the police involved. If you can take a respected leader like x or y with you, and the police, something might happen. Youth leader, Lae Miles Box 28.  Leaders and Youth Regulating Settlements: Hierarchy and Violence in Lae The current youths listen to me when I give out instructions because I supply them with betelnuts and smokes, and so they respect me when I speak. My youths are crime stoppers; we disturb the other youths who try to do holdups on the highway. The youths only listen to the right lidaman who communicates, respects, and has time for these youths. I am a short-tempered guy but I have changed. But when I am being provoked by the youths, I lose my temper and beat them up and they realize that there is law and order in the community, and us komitis are here to implement it. Senior Komiti, Lae We go around the community, and we know that kids like this usually make trouble. We stay close to the community, and we educate all. We talk to their parents, and we tell the kids, you see you only make trouble, you bring trouble home. Your parents are not like you, they are good law-abiding citizens staying in this community. You bring trouble to them. Do you feed [your parents]? No, you don’t feed them, you don’t bring a packet of rice to the table. All you do is to find trouble. Where is your money to pay your fee, penalty, or charge? So we go hard on them, we sometimes swore at them, too. Because they don’t understand, they don’t know where they are. Every time we see them drunk on the street, abusive language in the public. But when time comes and they run into trouble, we go into their houses and deal with them good and proper. Female Komiti, Lae 5.3.4 Thus the authority of dominant groups in that both youth—and, as described below, women— settlements, as elsewhere, is backed by a credible are much less likely to feel secure in settlements. threat of violence, and komitis are a part of that situation. There is violence directed at perpetrators and violence reflecting basic class, age, gender, and 5.4  THE KOMITI’S GREATEST FAILING: other power differences. Regulation of settlements PERPETUATING WOMEN’S INSECURITY happens within these everyday relations of hierarchy that are expressed within families, between ethnic The biggest challenge for komitis is clearly the and class groups, and within institutionalized prac- asymmetric gender power and their failure to regu- tices, including mediation. Functioning komitis might late family violence. Arguably most important, how- suppress violence, but there remains a kind of vio- ever, is that for all the kinds of capability described lence within komiti authority itself. The outcome is above, women generally have little trust in komiti 44 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS structures, despite the fact that a large proportion of better access to komitis, and this leads to the influ- cases heard in mediation relate to family matters and encing of komiti processes—and strong distrust from that komiti structures are often the only accessible women seeking justice. Even in cases where a komiti form of dispute resolution available to women.88 leader will extol the virtues of human rights and the new domestic law, there are still numerous features 5.4.1 Komitis everywhere are male dominated of komiti regulation of family issues that women find and traditionally patriarchal on several levels. unsatisfactory. Men’s community networks may well involve much Male Lae komitis hear a case involving a marriage dissolution. Photo: David Craig/Worldbank Box 29.  Women, Komitis, and Family Cases, Including Gender-Based Violence: Lae Women We don’t trust the man, especially the male komiti leader in the community, because they always go around the back of the woman. Man they don’t keep it a secret, don’t keep confidential information to themselves. They come back and inform their husband saying “Hey your wife came to us and told us this and that and she wanted to put you in court.” And he goes back to the house and bashes her. That’s the fear that woman have, so they are not opening up to go to the komitis. GBV survivor Man think they own the woman, they see us as their properties, and so when they are drunk they demand food or water, and if the woman talks back, that’s when they start to beat the woman even in the middle of the night. So [in komiti hearings] it’s always hard to get the woman’s side because man has that mentality of thinking that they are always right and they are man. They mix traditional culture with governance: they believe in men. Women don’t make decisions. That’s the Highlands patrilineal way. But I’m a strong individual coming from a matrilineal society. So in my case when man does that—I mean the komitis and leaders—I don’t have the second thought of bringing my case to the komitis. I bring it to the police station or courts. Emerging female lida 88  See Goddard (2005b), 247–67. Pow er asy mm et ries, out comes, and t he limit s of local regulat ory re a ch 45 5.4.2 Komiti processes of gender and legal rep- Unit. In practice, seeking justice can begin at any resentation lead to poor outcomes for women, one of these points, and referrals (or at least refer- especially in relation to referrals to higher forums ral upward in the imagined hierarchy) between them and courts. Gender power asymmetries are force- are rare, even when women komitis or Village Court fully (and dangerously) reproduced in the ways that magistrates are involved.89 Police, however, do con- komitis manage cases, and this leads repeatedly to sistently refer cases, but they often refer back to the unsatisfactory outcomes, in particular to the entrap- komiti or family, effectively meaning back to the per- ment of the victim in local structures that perpetuate petrator.90 The low levels of resourcing and/or com- the problem. Some women reported much better mitment of police FSV Units, along with the minimal outcomes from going directly to the local or cen- trust and respect between law and justice leaders, tral police; others talked of the desirability of having translate into an environment where responsibility for komitis or Village Court magistrates from outside the serious crimes is passed back down to komitis (see community sitting in on mediations and hearings. Still part 4 below). Non-resolution, or perhaps temporary others despaired over gender relations, especially relief, is a very common outcome. the lesser educated among the younger generation. In theory in gender-based violence cases, as in all of 5.4.3  The slow emergence of women komitis and PNG’s local justice arrangements, there exists a hier- lidameris [women leaders] is an uneven but prom- archy of resort (or jurisdiction) and referral, wherein ising development. In response to komiti patriarchy, women might go to the komiti first and then the many women would like a better representation of police station, and/or the Village Court or the cen- women on komitis. But women leaders taking up tral police station’s Family and Sexual Violence (FSV) roles within komiti or court structures face a consider- Box 30.  Outcomes from Referral and Resort outside the Community There are cases where the komitis have helped solve problems, and many are referred by the komiti to the [local] police. And when women go to the police, they get the case “cooled down,” no action, referred back to community, because men are friends with the police. Mostly if it’s referred back to them they will say it’s a family matter, go get a live chicken, some bananas and food, get the wider family together, cook, let’s have a big feast, nothing more said. Woman, Lae How can you trust the komiti: they might be our own relatives or people who just reside here. It’s all too close by, and there are too many connections. If the komiti make a decision, the man who lives locally might get drunk and fight the komiti… then they will delay. They don’t solve it. If there’s a domestic problem they should get outside komitis to come in [from other communities]. Woman, Lae The komitis will refer family cases to [the woman Village Court [VC] magistrate] and the VC. They refer via a complaint form, that they can fill in at my home at night. Then the VC peace officer (bailiff) gets the form and takes the form to the husband. Then the man will try to solve the problem; the males call in their mates, and try to solve it in their way. And the men on the VC don’t work with us [women], they work alone. When she goes to the komiti she won’t explain what has happened, she’s scared to explain to a male komiti. And then the senior VC magistrate can use his discretion and relationship with the police to get the man let off… and then the man comes out and goes drinking with his mates, and they are all boys together, swearing at the komitis and at [the woman VC magistrate]. There must be female komitis… the men won’t get all the problems straight. Woman lida, Lae 89 Ibid. 90  Medecins sans Frontieres (2015). 46 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS able set of challenges, not least from the expecta- tions and ways of working their male colleagues have developed. Lidamen might achieve their status both on their own merits and with the privilege of free- dom and resources to actively engage in civic action across a wide geographical and social terrain. Entry and early stages of involvement commonly require a patron, and very few existing lidameris means very few potential patrons. But churches, commu- nity organizations, and women’s business and other networks are providing komiti and lidameri develop- ment possibilities. Box 31.  Being a Lidameri I would call myself a lidameri. The role is not Lae Law and Justice/ Community Lidameris. so different; if you are a woman you still have Photo: David Craig/Worldbank to serve. Once you are labeled as a leader, you need to listen well and look in with people, you need to stand with them to help. Lidameris difficult relationships, including with higher courts don’t really have law and justice roles, that’s and the police. Referrals have to be done on a per- the komiti system, though we are always sonalized basis or there is likely to be little response, involved with peace-making, with finding the and women magistrates and komitis are not always solution, with sorting out worries. We have taken seriously. But failure to provide access to higher our lidameris who we follow, who look after courts or the police can reinforce repeat offend- ing or result in unevenly effective measures, such us, but most of the support comes from as repeated verbal warnings and escalating fines underneath, and we support each other. But imposed on repeat perpetrators. If these processes the men whisper behind our backs, “what sort drag out, it means a loss of confidence on all sides: of woman is that?” Emerging lida, Lae victim, perpetrator, and magistrate/komiti. 5.4.5  On the other hand, women komitis and mag- 5.4.4  Women komitis, lidameri, and Village Court istrates tend to get much more closely involved magistrates are drawn immediately into difficult with families and their struggles. This is partly questions and conflict over local cot/court author- because they recognize that issues are embedded ity jurisdiction and the referral of cases. They are in family structures from which women cannot always also drawn into relations with their fellow court offi- escape, requiring them to be creative about seeking cials wherein they face pressure to publicly assuage relief. Speaking of a Miles woman komiti, a neighbor doubts about the capability of the komiti or court. related that “She deals with all these youths. I have At the same time, women in these roles face pres- witnessed. I come to her all the time, her children sures to try to regulate male magistrates and komi- when they march around in the street, they always tis and ensure that they are acting fairly and appro- bring in kids who have no food to the house and priately, both in court and in the backroom or after she is always feeding them in her house.” Another hours negotiations around a case, when bribes are woman Village Court magistrate runs choirs, march- often rumored to be paid. In practice, this leads to ing groups, and writing workshops for gender-based conflict with the male komitis and magistrates—and violence survivors, in the course of which women to a level of avoidance, wherein a woman komiti or might write down the story of their abuse and then magistrate can be sidelined, left out of mediations, have it turned into an affidavit used to procure a and/or not informed of cases and hearings. Whether protection order. This woman also got two women a female komiti or magistrate gets heard and con- elected (for the first time) to blok komiti roles where sulted depends heavily on her ability to maintain she lives. Pow er asy mm et ries, out comes, and t he limit s of local regulat ory re a ch 47 Box 32.  A Woman Komiti Speaks about Community and Komiti Work I work with the community leaders; I am a komiti. The male komitis asked me three times and I rejected their request to become a komiti, but then I thought about it and I said “Oh, we widow mothers are trying to create this association to help stop law and order problem in 4 Mile in the ward four area.” So here I am. Those kids that create problems are the kids of widow mothers and single mothers. They are our kids so I think it’s wise for me to work with the leaders so when our kids see me they will listen when I call them. If other leaders get on them, they will fight back at them, but when I get on them they listen to me, they will say, “Yes mum.” And I am happy to work with the community. The women, when their husband bashes them, up they come to me for help and I take them to the komitis to solve the problem. But if I see that the problem is too big I refer them to the police. I am from Sepik. All these criminals, I like them. I do not get cross to them like them and I treat them with respect. I give them food, when I have money I give them. I am doing that to try and help them to leave crime. When they don’t eat they go and cause more trouble, so my teenage boy usually brings them over to our house and asks me to cook rice for them. I usually give them some money to go buy greens, kaukau, rice, tin fish to cook and eat. 48 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Local Regulation: komiti and lida relations 6.  with police and subnational authorities Urbanization everywhere intensifies contests over crucial features of these forms of local state authority the creation and control of resources, including and how they impact the outcomes of regulation in the economic, social, and other types of capital urban settlements. referred to in the introduction. The arrangements described in this report are distinctively Papua New Guinean, but they are also recognizable elsewhere 6.2  KOMITIS’ DIFFICULT RELATIONS as urban versions of the sort of institutions that WITH POLICE: MUTUAL ACCUSATIONS arise in all areas of limited statehood. Thus in urban OF DISRESPECT, AS WELL AS PRACTICAL PNG, as elsewhere,91 komitis and lidas intersect, DYSFUNCTION AND DANGER overlap, align, and compete with other edge-of-state institutions that also assert regulatory authority, most 6.2.1 Komitis often feel that they perform the importantly the police and local government councils bulk of frontline law and order, especially in situ- (although as described in box 22, corporate interests ations requiring quick response policing to avoid also play a role in these territories). The nature of escalation and serious violence. For this, komitis these other edge-of-state institutions, and how these long for respect, recognition, and resourcing. “We relationships with komitis play out, is crucial to how are putting our own lives at risk daily, but the state local mechanisms perform in the four functional doesn’t recognize or reward us.” The police, it is often domains outlined in Section 4. noted, are frequently absent. Komitis feel neglected and maligned by official law and order/law and jus- tice agencies and also that their claims for material 6.1  WEAK VERTICAL LINKAGES resources and symbolic recognition are too easily dismissed as entrepreneurial rent seeking. 6.1.1  Komitis are typically able to include and com- bine forms of authority horizontally, for instance, across different ethnic groups, and adapt their membership and procedures to include almost everyone. But as Section 5 shows, there are limits and asymmetries in komitis’ institutional reach, especially up vertical hierarchies of official power, such as the police. Komitis typically reflect the interests of one ethnic, gender, or age group more strongly than others. The extent to which some interests prevail, however, is not simply a product of the horizontal balance of powerful local players or traditions. In some regulatory contests, it is vertical linkages with state governance (in particular, the police and local government) that determine the capability of komitis. Although it was not the purpose of this research to appraise the nature and 4 mile leaders holding home made weapons captured performance of these agencies, through the lens of by local police. the komiti and lidas, it is important to unpack some Photo: David Craig/Worldbank 91  Korf et al. (2018), 171. Local Regulation : k omit i and lida relat ions wit h police and subnat ional aut ho ri ti e s 49 6.2.2  Certainly it is true that komitis’ unmet in Section 6, in crucial ways, komiti and state insti- demands for official recognition reflect a perverse tutions are fundamentally different in character. In division of labor. Local komitis and lidas, and the consequence, although komiti and lida crave fiscal people they represent, are able to pact together to and symbolic attention, as this section will show, they protect and secure their settlements and capital, but are particularly vulnerable to disruption and disarray they produce order and security that is of equal ben- when such wishes are granted. efit to economic and political elites and to middle classes and urban businesses. But lidas and their 6.2.3  Resourcing issues are an important aspect of people can impose the costs of doing this work only the problem—and a potential source of improve- on themselves. Except for around the Christmas ment in komiti-police relations. Under the O’Neill period, they are unable to form stable, remunerated government’s 2012 Alotau Accord political pact, pacts with the incumbents of formal state offices, police and law, and justice agencies have received the police, or local governments. There are several significant reinvestment. No published review is avail- reasons for this, including the obvious point that able of the effects of these funds, however. Despite the latter are competitors with komitis for authority regular media reports that even new police are still and resources and are much better oriented to the regularly involved in excessive violence, positive new vertical lines of power through which budget and alignments between komitis and police are possible patronage is channeled. As will be further developed and can certainly add to komitis’ authority. Lae Miles police and law and justice leaders. Photo: David Craig/Worldbank 50 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS 6.2.4  But for komitis, both collaboration and com- opportunistically. Komitis and local residents protest petition with police can be ruinous, bringing coer- that different police commands compete with each cion into mediation and raising costs and fees to other and with private security or ad hoc auxiliary participants. In part, this is due to the fact that the police to forage and provide for themselves. police in urban settlements, no less than other gov- ernment agencies, are operating where the state’s 6.2.5 Having the police and their independent presence is most brittle, underfunded, and poorly vertical authority involved in komiti mediation can connected to central accountability and discipline. be literally fatal, where susceptibility to corruption Here, as state representatives, the police do not act skews procedures in the interests of the party who in relation to komitis as unitary authorities subject is most able to bribe or otherwise influence whom- to hierarchical sanction; rather, local residents per- ever is involved (see box 33). ceive them to be untethered, reaching into their lives Box 33.  Local Perceptions of What the Police Can Do to Komiti Work We do not involve the police, they are the last resort. If we go to the police, they will send us back and tell us to sort it out within our community. We only go to the police when someone has been killed or if it is a very serious matter such as rape. … The police come around here and they muck it up; they play around with the law. .... I can say in all honesty, you can’t rely on the police. The moment you are in trouble, if you want the police involved, whether it’s to mediate or for safety in the mediation or whatever, you need a whole bundle of money. Every time. Always. They will detain you, charge you or the other guy, depending on who gets to them first. Even if it’s just a small thing. If they have got you, just get out of there, get out of the police station whatever it takes. Moresby Village Court official There’s a motto in law and justice: “Community and Police working together to reduce crime.” But this motto isn’t making way; the police come and butt heads against the komitis and community leaders. Some of you are victims of police brutality. You are the volunteer people, the hard-working people, they are hard-working people from the ground but never recognized. Lae District law and justice leader, talking to 350 Lae komitis at the election of officers 6.2.6 Typically, relationships with the police are community regulation by komitis have helped the highly idiosyncratic and dependent on good per- police to feel safe enough to do this. At 3 Mile sonal connections between komitis and senior Police Station, the community-oriented station com- police officers. There are exceptional places in both mander allows all of the hybrid (un-gazetted) Village Port Moresby and Lae where official PNG police pres- Court/komiti hearings to be held on the police sta- ence is being re-established according to the com- tion site. The fence and police presence provide munity policing model. At Sabama in Port Moresby, security and an official backup if issues get heated. police from the community policing–oriented Badili On the other hand, the police do not usually partici- Police Station were temporarily relocated to Sabama pate in mediations (though a previous police station during the rehabilitation of their station. More than 18 commander [PSC] would often cherry pick cases to months later, they remained heavily involved in local hear). The current PSC believes that this separation of mediation, using the police outpost built beside the powers between police and courts/komitis is impor- market and mediating alongside community leaders tant to maintain. He, like many of PNG’s police offi- who are mostly happy to have the added authority of cers, also believes in sending disputants back to the a uniform present. komiti to be dealt with. 6.2.7 In Lae, at the invitation of the papa graun 6.2.8  The komiti-police axis can be a major obsta- and his chairman, Pita Blok has recently had regu- cle in cases of serious family and sexual violence, lar afternoon RPNGC foot patrols because active however. Komiti, Village Court, and police arrange- Local Regulation : k omit i and lida relat ions wit h police and subnat ional aut ho ri ti e s 51 ments frequently reproduce and reinforce gender pensation to each other and publicly share responsi- dominance, confirming women’s subordination within bility for acts that were highly one-sided. Conflict and the family unit and within local law and justice institu- fighting involve two sides, and sexual violence gets tions. In dispute mediation, protocols for referral and a similar treatment, with the rights of the perpetrator prosecution of serious gender-related cases (rape, and his family emphasized as a way to get beyond sexual abuse) are routinely set aside. Families (and individual shame and to communal reconciliation. often male family leaders) determine the future of Once the case has been heard in a kastom setting using approaches that they can collectively live with, and an arrangement agreed to, recourse to formal which are usually under the authority of traditional and criminal courts is rare; in practice, what media- leaders and gender roles, bride-price kastom, and tors claim are complementary systems become in notions of “win-win” or restorative justice.92 These fact alternative systems. often insist that both victim and perpetrator pay com- Box 34.  Police Referring Cases Back to Community and Family: Lae Settlement Dwellers Rape and murder get referred to the police; we can’t [legally] deal with that. But even serious cases get pushed back to the komiti. Maybe the complainants themselves or the [victims’ and in-laws’] family ask to have it taken back to the community. And there it stays. Families are linked together, connected through marriage, their customs are bound together. That means disputes or violence is only taken to a certain level outside, a level that still involves the family. The family have to think, if this goes on then this will affect all the family ties, they might be cut off, and we need to consider them. So the brothers will speak for the family, for the woman, and if there is no male in the family, the older sisters will speak. Senior settlement woman, Lae Really the komitis are only there to mediate, to deal with things in the customary way, to mediate right down with the people, and make peace and normalcy as the bottom line. So how do they come up with a solution? They join up and with the family they make the woman the focus of the situation, they and the family try to put them into a solution. But some of them are her in-laws, and they are often the ones backing up the violence against her. Woman, Back Road They say they will be fair, “Our decision must be 50/50,” so they give equal weight to both sides. If we want to bring evidence to tip the balance, well, they are always right and they are men. They won’t have a second thought about it. So that’s why in the end, even in the city, the woman can’t speak, and then she’s rather go back to the police. Woman, Nawae Blok Another case regarding another young woman occurred in which the komitis took the case up to the police station. The police on duty at the duty counter told them to go and see the FSV [Family and Sexual Violence] unit at the central police station. The FSV unit sent them back to the general duty counter, then they were told to come back to Serious Offences in the morning. Always excuses, too busy, no vehicle to investigate, so they came back to the settlement. So they decided to solve the case in the settlement with the komitis. I told them this is a rape case and it shouldn’t be dealt with at the community level, it should be dealt with by Serious Offences. But it was heard in the settlement, and the komiti charged k500.00 fine to the accuser! Emerging woman leader, Lae The komitis are always trying to get the cases referred back to them. And they are getting the cases like rape referred back to them, and they entertain them illegally for just one reason: money. One case, the komitis offered the mother of a daughter who had been abused k1,000 in settlement money if they could hear the case. But she refused. They are looking for business. They prolong the case even if they are not wanted. Woman, Lae 92  Papua New Guinea (1999). 52 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS 6.2.9 Thus, with regard to gender relations and funding. This resulted in a large wish-list of propos- violence, the proud claim that “we people know als sent to the district administrator, but for which how to look after our own business and not put our no funds were available. The 300 graduates of the heads into others, we solve disputes down here” course were issued with identity badges and T-shirts, can be an obstacle to women’s ability to obtain however, confirming their authority and that of the more effective help than is available locally—or magistrates and komitis. Returning to their settle- even a trap in which resolution is denied by komi- ments, newly trained komitis set up alternative komiti tis seeking rents. But resort to komitis is also shaped networks, some in active competition with estab- heavily by day-to-day failings in the wider pathways lished Village Courts and komitis. of FSV resort and referral. In these circumstances, even for serious criminal offenses, komitis become A further round of DSIP funding was the subject places of resort for family matters—preferred, acces- of dispute and irregularity, sending the elected sible, affordable, and timely resort, but also highly committee chairman into hiding for a time. Another unsatisfactory and even dangerous. incident involving the misappropriation of funds from the district and LLG led to the disciplinary side-lining of a senior Miles law and justice chairman by his 6.3  EMERGING DISTRICT AND CITY AUTHORITY: peers. In all these cases, formal rules and procedures NEW OPPORTUNITY FOR LOCAL GOVERNMENT around the Lae District law and justice structure were LINKAGES? either not in place or were able to be manipulated to grant funds to people lacking a local mandate or 6.3.1 The problems inherent in komiti regulation expenditure control. The irregular arrival of significant (uneven outcomes, subordination and entrapment amounts of money led to a culture of rapid and of the vulnerable) are often magnified by their bulk dissemination within the District Development relations with local governments. At face value, Authority. the advent of the District Development Authority, along with the prodigious funding reputedly avail- able under the highly discretionary District Services Improvement Program (DSIP) (10 percent of k10 mil- lion per year nominally earmarked for district-wide law and justice activities) would appear positive. The reality, however, is that in urban settlements, local governments are subject to volatile, unreliable fund- ing under current arrangements. The effect is the reproduction of local political and patronage rela- tionships and the creation of incentives for leaders to divert available discretionary funding into low-quality, one-off, and low-transparency investments. Indeed, the instances identified during this research in which local government monies were channeled into local law and justice activities proved to be more disrup- tive than enabling. Port Moresby coordinator of city wide mediation Micah Yosman with Sammy Temiel, Law and Justice Chair, Ward 17, Lae. 6.3.2 District funding sources that are not well Photo: David Craig/Worldbank aligned with local komiti or Village Court law and justice needs and development can have perversely negative impacts. In Lae District, an Ensuing disputes, infighting, and further allegations elected district law and justice committee received of mishandling funds between Lae’s senior law and DSIP money to fund training for roughly 300 of Lae’s justice komitis led to a petition and threats of a law- komitis. A nongovernmental organization (NGO) with suit against the district by the newly and duly elected no law and justice experience was contracted to carry district law and justice chairman and his deputy. this out, drawing on some preexisting training mate- Demands from these leaders for access to funds rial to teach participants how to write proposals for became heated, and the chairman and deputy, whose Local Regulation : k omit i and lida relat ions wit h police and subnat ional aut ho ri ti e s 53 election was overseen by 350 of their peers and the munity policing, renovation of secondary markets, Electoral Commission (the election was financed by and improvement of services (or perhaps negotiated the district from DSIP money), were told that they security) to settlers on custom land. In addition to had no formal or statutory standing and could not the complex issues involved in achieving a satisfac- receive further funding. tory political arrangement with the Morobe provincial government—which has the technical and adminis- 6.3.4  Nonetheless, the potential promise of urban trative capacity needed by the city authority—the authorities in relation to komitis is considerable. experience of Port Moresby, PNG’s largest and most Beyond the District Development Authority and the functional city authority,94 shows that the Lae City DSIP, the growing national political significance of Authority’s ability to capture sufficient tax revenue urban areas is leading to new legislation that elevates from goods and services and, for capital spending, the governance status of PNG’s cities. These reforms to be included in the national Public Investment Pro- excise the cities’ jurisdictions from the existing dis- gram will be crucial to its ability to function. tricts and province and reconciles fragmented urban areas under new urban authorities somewhat akin to 6.3.6 With these statutory, fiscal, and political Port Moresby’s National Capital Development Dis- provisos, it may be anticipated that city authori- trict. In Lae, this manifests in the Lae City Authority ties could act to formalize and fund their own law Act enacted in 2015. Lae’s current city council bound- and justice committee arrangements. The Service aries and governing arrangements are an anachro- Delivery Partnership Agreements envisaged by the nism since they do not reflect the growth that has Department of Provincial and Local Government occurred since the late 1980s, and this deprives large Affairs between city authorities and other levels of areas of the city of access to municipal services. Much government may provide umbrella arrangements to of the Lae Ahi Rural LLG area is dependent on neigh- support this, but it remains to be seen whether this boring Nawaeb District for support, but Nawaeb will be able to superintend the advantages flowing DSIP money is spent on the mainly rural electorate. to ward-level procedures where relationships with Ahi settlement dwellers see themselves as “kicked komiti and lida would be needed in practice. around like a football between local governments.” “We have boundary problem, so Nawaeb thinks that we are in Lae Open, Lae Open thinks that we are in Nawaeb. This confusion deprives us of rights, so we get none of the development nor services others do. This boundary system breaks us up and we are in the middle, confused.” 6.3.5  Potentially, the City Authority Act, enacted also for Kokopo and Mount Hagen, gives city authorities remarkable scope and autonomy as service delivery agents able to contract or set up their own executive and commercial entities at will as determined by their board and member of par- liament.93 The ability of urban authorities to bring together different agencies is unproven, but in view of the powers granted by the board and the minister, they will likely have a broad but poorly defined man- date to address crosscutting policy issues, including Lae Law and Justice leaders congratulate James Kinjin on his gender-based violence and community safety. Port election to District law and Justice Chairman. Moresby’s experience suggests that possible activi- Photo: David Craig/Worldbank ties include the re-establishment of suburban com- 93  See Papua New Guinea (2015). 94  Port Moresby’s National Capital District Commission is a city commission, with greater powers than a city authority. 54 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS What is the significance and potential of komiti 7.  and lida regulation of settlements? Conclusions and implications 7.1  SUMMARY CONCLUSIONS reform ambitions, and brutality by law and justice agencies. For many urban residents, these condi- 7.1.1  Cities are key to reducing poverty, promoting tions will continue to cast an unfavorable and threat- shared prosperity, and improving governance in ening shadow on their ability to securely create and PNG. Urban growth will help to encourage economic accumulate capital. These conditions will also mean diversification, to expand wealth in human resources, that komitis and associated forms of leadership will services, and public and private assets, and to reduce continue to play core regulatory functions, including economic reliance on subsistence production and in mediating everyday disputes, securing tenure and extractive industries. A shift from governance via shelter, regulating market and business activity, and resource rent patronage toward a politics that res- representing the rights and interests of settlement ponds to demands for population-wide services—and dwellers. There is no reason to suggest that they for institutions to deliver them—may well accompany will not continue to do this flexibly, in real time, and an increase in urban voter electoral power. sustainably, adapting, co-opting, and refashioning ethnic and place identity, kastom, and social relations 7.1.2  Realizing the potential economic, social, and beyond kin preference and thereby enabling local political dividends of urbanization hinges on the reciprocal altruism. quality of urban regulation. Yet it remains one of the least understood areas of governance in PNG. Cur- 7.1.4 Komitis and lidas will continue to play a rently, the reach of formal authorities into settlements crucial role, but the forms of regulation they pro- is uneven and fractious, and there is a sizable gap vide will inevitably reflect and reproduce local between formal policy commitments and outcomes. patterns of power and opportunity. Urban settings Recently, however, the profile of urban areas such as are, by definition, intensely competitive zones where Port Moresby and Lae has increased, partly as a result the reach of state regulatory agencies—chiefly the of the emergence of new styles of political leader- police, courts, and local governments—is limited, ship and partly due to the growing significance fragmented, and only incidentally aligned with the of urban areas as locations for investment by the interests of everyday citizens. Regulation by komiti public sector and by the beneficiaries of national and lida can be deeply intertwined with state admin- energy projects. Assuming that the current national istrative as well as law and justice agencies, but this fiscal challenges are addressed, the growing politi- will not reliably protect the tenure security of long- cal significance of urban areas will prompt demands term holders of customary leases, for instance, or for, and may lead to improvements in, regulation. enable recently arrived settlers to access municipal Budget outlays may increase for law and justice services or the police. Local regulation by a komiti agencies, including for community policing and Vil- and lida will remain vulnerable to elite capture (by lage Courts; jurisdictional conflicts between city, dis- older males), exclusion (of women and youths), and trict, and provincial authorities that hamper service corruption and clientelism. delivery and accountability may be resolved; funding could become reliable and matched with responsi- 7.1.5 Local settlement regulation therefore will bilities; and increased political engagement may lead continue to be both an indispensable and trou- to more positive forms of supervision of the conduct blesome feature of PNG’s urban governance. In of police and local governments. reproducing power asymmetries based on gender, age, ethnicity, or class, decisions by komitis and Improvements in urban regulation will 7.1.3  lidas can seriously harm the realization of rights and need to overcome the formidable legacy of frag- due process in law. Although women continue to mentation and disorderly practices, unrealized access komitis, and most things komitis deal with W h at is th e signific a nc e and potential of komiti and lida regulat ion of set t lement s? Conclusions and implic ations 55 Lae women’s network. Photo: David Craig/Worldbank have important family implications affecting women, is, dispute mediation as it pertains to social order, women’s interests are not represented at all reliably. shelter and security of tenure, market and business The result is that gender-based violence victims are activity, and political representation. Each audience diverted from formal justice structures and women is addressed below. are denied meaningful access to local justice and basic security. 7.2.2 It would be ill-advised to recommend that the role of komitis and lidas be elevated and ex- panded, formalized, or adjusted. Clearly, it makes 7.2  PRACTICAL IMPLICATIONS FOR URBAN no sense to stigmatize or disparage local mecha- ENGAGEMENTS: “DO NO HARM” nisms as merely “transitional arrangements” to be tolerated while formal state agencies catch up to 7.2.1  It can be difficult to draw practical implica- reduce the distance between their statutory obliga- tions and policy actions from these conclusions, tions and their actual performance. Komitis and lidas but there are two audiences for whom the find- are not simply transitional; they play crucial roles in ings are crucial. This report’s findings have implica- urban regulation that, if anything is predicted, are tions for what may be termed “urban developers,” likely to increase in scale and scope. But much less which here includes public officials responsible for straightforward is the need to devise practical impli- urban infrastructure and public facilities—for exam- cations or recommendations for either of the above ple, the creation or redevelopment of markets, water two audiences. However, before proceeding, there and sanitation, lighting, roads, and buildings. Urban are three caveats on the scope and limitations of this developers are also the private sector actors who research, the inherent nature of komiti and lida as undertake these works or the commercial develop- forms of regulation, and their relations with formal ment of areas or specific sites, including for commer- state agencies. cial retail, housing, factories, and warehousing opera- tions. A second audience includes officials (donors, First, recommendations require a comparative knowl- government) who engage in urban settlements with edge of the alternatives to the kinds of urban regula- the aim of improving justice and governance out- tion provided by the komiti and lida. Most immedi- comes in the various ways explored in this report ately, they need to be understood and compared to relating to the functions of the komiti and lida, that the actual local work and reach of statutory local gov- 56 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS ernments, the police, courts, prosecution and referral 7.2.3  “Do No Harm” should be the guiding prin- agencies, and their higher echelons: what they are, ciple of all engagement. The komiti and associated the kinds of disputes and issues they are mandated forms of leadership are not the only regulatory author- to handle, and how do they this and with what dis- ities at work in settlements, since komitis do nothing tributional consequences, both intended and unin- that is not also the formally mandated responsibility tended. These data could enable a comparative of local governments and so-called law and justice review of the opportunity costs—the likely returns to sector agencies. In practice, however, no one else and risks of public sector and/or donor investment in can or will do what they do in providing local regula- the available range of options, including the institu- tion, as they are crucial to, and currently perhaps the tions examined in this report. backbone of, urban safety and security. Contained in the principle of “Do No Harm” are three points. Second, even if there are merits to giving further rec- ognition, stature, credentials, funding, or capacity First, experience shows that local mechanisms are building to the komiti and lida, the burden of con- easily undermined and disrupted by engagements veying such support would most likely be handled with external agencies, with the effect that these by the host of agencies with which there is at best forms of local regulation can lose their intelligibility an uncertain relationship and limited flexibility—and and legitimacy. They can themselves become the at worst, competitive hostility: the police, the formal focus of grievance, causing existing conflicts to re- justice sector, and land and other urban authorities. ignite or new cleavages to emerge within kastom or As shown in this report, regulation in Lae and Port komiti leadership while at the same time reinforcing Moresby in fact occurs through shifting, contradic- the marginality of already excluded women, ethnic tory, fragmented, and overlapping systems in which groups, and youth. the state, kastom, and private sector governance act in parallel and independently, and variously com- Second, urban developers—governments or private pete, collide, and collude. Komitis need to be able sector investors—create the greatest potential for to flexibly adapt to such contexts. harm when they are driven by narrow interests and short time frames and lack clarity and consistency. Third, efforts to alter the credentials, ways of operat- Grievances will be disruptive if some legitimate fac- ing, or outcomes of regulation by komitis and lidas tions within kastom arrangements perceive that they will inevitably involve some formalization, which will have been excluded. Undoubtedly, some urban risk destroying their current capabilities. Formal busi- developers will perceive advantage in “gaming the ness or state actors value stability and continuity over system” and knowingly act in ways that are divisive flexibility and iterative change. But komitis need to and disruptive. Others will know that their interests be respected as urban kastom constructions that are not well served if they engage with local mecha- must deliberately mix and bridge between two very nisms only sporadically and opportunistically, that is, different forms of public authority, that is, elements of with the “fly in-fly out” fickleness typical of project- formal state authority on the one hand, and diverse based commitments. customary and localized forms of authority on the other. To a degree much greater than public sector Third, the potential for positive engagements is institutions, local regulatory mechanisms enact and increased where realism prevails, namely, where it is bring together rules, roles, and resources in ways recognized that anything to do with money, land and that combine and blur formality and informality, as pecuniary opportunity, or honorific entitlements will well as different social mores and politico-legal ways become subject to contest and rival claims by kastom of ordering life. They seldom create the kinds of leaders. By coupling a positive outlook with realism, durable, accountable relationships sought after and urban developers would shift from a conventional valued, even if rarely achieved, in public sector ways outlook, “How do I avoid disputes and grievances?” of doing business. Their local arrangements must to another that asks “How can I preempt disputes always be plural, flexible, personalized, and depen- and grievances from escalating, turning to conflict, dent on who is there now. and resulting in unacceptable costs?”—regardless of whether these are borne by the urban developer or W h at is th e signific a nc e and potential of k omiti and lida regulat ion of set t lement s? Conclusions and implic ations 57 the “communities of interest” represented by komiti, komitis and lidas will be unavoidable, but it is unlikely lidas, and kindred other local mechanisms.95 to occur on terms defined by the urban developer. 7.3.2 Do not assume that it will be possible to 7.3  OBSERVATIONS AND ADVICE FOR URBAN instrumentalize or draw on the authority of the DEVELOPERS: CARRYING “DO NO HARM” INTO local komiti or to scale up or re-engineer its activi- PROJECT PRACTICE ties for project purposes. As box 35 indicates, despite their apparent local authority and efficiency, While emphasizing the overarching concern that “Do komitis and lidas will not provide the kind of author- No Harm” should be at the heart of any engagement ity that is universally applicable beyond the indi- that impinges on komitis and local lidas, there are a viduals involved, durable over time, representative range of caveats and issues related to safeguarding of all groups (including women and ethnic groups urban development activities, especially around the evenly), and enforceable according to formal rules instrumentalization of komitis in project contexts and of accountability. Urban developers are unlikely to around grievance. These are included here because find that komitis’ authority in resolving disputes can of their potential to “go with the grain” of effective be applied, for instance, to managing temporary komiti arrangements. labor contracts or community in-kind contributions for infrastructure projects. Neither is it likely that a 7.3.1. Anticipate that your activities will trigger komiti will successfully handle a contract to manage disputes and that engaging with local komitis and security around a major asset or provide a stable and leadership will not be optional. State, donor, or gender- or age-equitable platform to impartially rep- commercial engagements to upgrade water, light- resent community interests during an assessment of ing, or roads, renovate markets, or establish com- damages and a determination of compensation to mercial centers will always trigger disputes and griev- people impacted by the urban developer’s activity. ances about who benefits from, or bears the adverse Nor is the kind of authority enacted by a komiti at a impacts of, these activities and who has the author- particular time or place readily able to be scaled up ity to make such decisions. Grievances will almost or taken from one part of the city to another, such as always be articulated by groups and individuals close when a developer needs a “community interface” or in character and capability to the komitis and lidas a “focal point.” discussed in this report. Engaging with settlement Box 35.  Komiti and Lidas are Not Easily Able to Give State, Commercial, and Development Projects the Kinds of Stable Certainty of Relationship They Want and Need The corporate player involved in the Miles road project (Section 2) worked hard to include multiethnic leaders, youths, men, and women, but secured highlanders and males at the cost of women and Morobean sortimobs [juvenile groups]. The latter actively opposed the project and its work, including by reputedly targeting a corporate vehicle and discouraging Morobe youth from participating in edu- cational outreach. As often occurs elsewhere, the komitis and leaders in this case were approached as if they held some kind of comprehensive authority in their bloks. But their actual reach into youth groups, gangs and their remnants, and local patronage proved frail. Disappointment in leaders followed, along with the unraveling of the pact between them, the corporation, and the youth leadership. In terms of the safeguards and grievance management arrangements typical of donor agencies, including the World Bank, current 95  Grievance Redress Mechanisms could be reframed as Grievance Pre-emption and Management Mechanisms. A short, accessible “Kastom Guide” to dealing with local kastom mechanisms and leadership is proposed to guide these decisions. 58 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS 7.3.3. Practical, rewarding, and safer engage- cive authority will not be able to predictably ments with local lida and komiti members are discipline its involvement, and that it will be dif- more likely where urban developers adopt the ficult to reliably instrumentalize its actions through following principles: agreements or the exchange of favors, sinecures, or material benefits. 1. Proactively consider ways in which komitis and lidas can be engaged in the proposed activities, and ensure that they are consistently informed 7.4  IMPLICATIONS FOR THOSE WHO WANT TO about them. This principle should apply especially STRENGTHEN KOMITI AND LIDA CAPABILITIES to activities involving land and shelter security, AND PERFORMANCE employment and contracts, and the assignment of responsibilities for management and oversight 7.4.1  Given the considerable vulnerability of to officials from LLGs, districts, and provincial komiti and lida structures to change and interven- administrations. tion, “Do No Harm” applies equally to interven- tions designed to recognize and strengthen local 2. Consider who is likely to benefit and be regulation, starting from initial contact. “Do No excluded. Consider how contracts and regulatory Harm” remains the bottom line in any engagement requirements created around the project may, or with komitis and lidas, including those seeking to may be perceived to, exacerbate existing local develop their own regulatory capability. This prin- cleavages or even create new ones, or could be cipal needs to be applied at each point, especially more gender, age, and ethnicity inclusive. where there is the likelihood that engagement will bring komitis into increased recognition and contact 3. Actively “triangulate,” that is, verify through with other regulatory agencies, from the police to multiple sources who local leaders are and their city authorities. Again, suggestions are included here competing interests and sources of respect and because of their potential to “go with the grain” of legitimacy, and how their fields of capability, such effective komiti arrangements. as disputes relating to land, housing, youth, and so on, are locally perceived. 7.4.2  Recognize and support komitis and leaders with care. Locals are certainly keen to receive rec- 4. Ensure representation in agreements. Ensure ognition and resourcing by state agencies. Ideally, that project agreements make it possible to recognition needs to be provided clearly and con- engage a range of different forms of authority. sistently by stable and legitimate forms of local gov- Project agreements must not simply rest on the ernment (including in PNG, Village Courts, LLGs, and putative power of the bearers of formal govern- urban authorities) or by those holding formal land ment office; they should also ensure that agree- title, including papa grauns. Training, for instance, is ments with local komitis and lidas are constructed one way to give headline recognition to komitis and and endorsed in a public way in relation to the lidas while at the same time enhancing komiti media- project. tion skills and knowledge of formal procedures where disputes involve state agencies, and also reminding 5. Frequently revisit agreements. Anticipate that them of their jurisdictional limits over such matters as agreements will be renegotiated and require roles serious sexual assault. and responsibilities, and the distribution of costs and benefits, to be adjusted, and again witnessed 7.4.3 If enrolling PNG’s police, justice, and local through public rituals. government agencies in delivering support to komiti and lidas, be aware of the competition, 6. Anticipate that as disputes and contests are stigma, and antagonism in these relationships. The activated, all local agencies and actors, whether experiences examined during this research show that variously described as formal, official, kastom, local regulatory mechanisms are often undermined or otherwise, will seek to capitalize on the when they interact with ward or district governments opportunities presented through such contests. or with the police; indeed, the authors encountered instances in which state officials reached into komiti 7. Assume that the formal administrative or coer- activities in abusive ways to extract rents, exploit W h at is th e signific a nc e and potential of k omiti and lida regulat ion of set t lement s? Conclusions and implic ations 59 authority, or grab resources. This is because these cally commit to and cooperate around urban set- agencies are competitors in the local regulatory (for tlement regulation. Dialogue should include the example, mediation) market. But it is also because ways that local law and justice issues are included when the formal machinery of the state recognizes in the mandate of the district, LLG, and/or city other institutions, along with this will come efforts to authority and how these functions are resourced “freeze” local komitis and networks into singular— either as part of—or supplementary to—the DSIP and subordinate—roles and relationships. This is on a predictable function and formula basis. This partly why the normal arsenal of “capacity building,” kind of closer assignment of function and funding such as providing training, formalizing pathways of is currently being piloted by the Department of response with the police, courts, or special purpose Provincial and Local Government Affairs in some crisis response agencies, or introducing codes of of PNG’s provinces and districts, though not spe- conduct and other measures of credentializing local cifically in urban or law and justice settings. mechanisms and leaders, can easily disrupt the flexi- bility and mixing of norms, rules, and forms of author- 3. Formalize the district (or city authority) law and ity and legitimacy that keep these local arrangements justice committee, enable it to function to solve working. Even where intentions are positive, the larger disputes across the territory, and bring kinds of financial and human resources that the PNG wider women’s network–based lidameri into government is able to provide to back these relation- family cases. Provided city and/or district authori- ships are typically uncertain, unpredictable, and so ties receive a clear legal mandate with respect to closely tied to personalized patronage arrangements law and justice issues (as advocated above), city that they will usually disrupt local mechanisms and authority or district leadership could then be sup- forms of leadership. ported to identify areas and places within the city that could bolster komiti development. This may 7.4.4  If harm is to be avoided, the following gen- include formalizing district or city authority law eral principles should guide any such engagement, and justice leadership, elected, as is current prac- whether the aim is to strengthen or to merely work tice in Lae, by the wider body of komitis and Vil- effectively with local regulation. lage Court magistrates practicing in various bloks, zones, and sections. District law and justice komi- 1. Do not attempt the heavy regulation of komitis, tis could become the sites from which to test the leaders, and mediation. Ensure that the flexible efficacy of networks of komitis (male and female) and adaptable qualities of komitis are protected, who could hear family cases across the city, rather but work to address the egregious gender biases than having them heard within close communi- in both procedures and outcomes. This is most ties and by local leaders. As in Port Moresby, a likely to be supported indirectly by improving citywide mediation capability or team could be gender representation in settlement and komiti established and, with due care and attention to governance and in wider community law and jus- the unintended consequences, be supported with tice. At the settlement level, work with women’s, a budget and dedicated vehicle and logistic sup- church, youth, and business groups to identify port. Overall, this seems likely to bolster, and not emerging women leaders and potential komitis undermine, local komiti capabilities. and mediators. Create local expertise among these groups and leaders around gender-based 4. Expand and extend urban Village Courts into violence issues and provide settlement-level train- kastom land tenure settlements. Priorities are of ing in mediation, basic human rights, and the legal course contingent on the fiscal means but could procedures associated with gender-based violence include establishing at least one Village Court per to women, men, youths, komitis, magistrates, and ward or on a consistent locality and population ward officials. Work down from the city authority to basis; enacting and funding provisions in Village raise expectations and advocate urban bylaws with Court legislation requiring the appointment of provisions about the representation of women in women magistrates; and subjecting the courts ward or zone law and justice komitis. to the regular performance reviews that the Vil- lage Court act provides for to avoid capture, pro- 2. Create a policy dialogue with emerging district mote female magistracy, and ensure turnover and and city authorities about how they can realisti- the access of emerging local law and justice 60 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS leaders to Village Court positions. Village Courts tis, women commonly present their family-related are arguably the agency most compatible with cases in komiti settings. Serious cases continue to be komiti regulation and least likely to damage it. heard in inappropriate and illegal venues. The con- sequences of this practice for the women affected 5. Find ways to address the worst abuses of FSV are not at all understood and the remedies still less case handling related to mediation, komitis, so, as well as the ways komiti processes could be and policing. Work with the Constitution and Law improved to make them more equitable forums in Reform Commission and Family and the Sexual which women could seek and administer justice. Violence Action Committee to identify ways to stop the police and families from referring cases 1. A needs and costing assessment of local Village back to komitis, and ways to penalize komitis, Courts and the capability of existing komitis to mediators, and rogue police officers involved in evolve into full-fledged Village Courts, with suf- circumscribing FSV survivors’ access to formal ficient resourcing to expand across large areas of justice . Review legislation based on close eth- urban settlements (such as Lae’s Ward 17, which nographic research to clarify the wider roles of has some 30,000 people) that are currently almost komiti, kastom, and formal law in relation to FSV. entirely serviced by komitis. A lack of confidence Find enforceable regulatory solutions where pat- that the Village Court system could translate into terns of regular dysfunction related to customary urban contexts has hindered this development up uses or patterns of case referral are clear. to this point. 6. Support the re-emergence of community polic- 2. A study of how young and emerging (individ- ing96 at suburban stations by re-establishing ual) komitis are trained and developed through community policing posts and developing the patronage of older members. This would shed career paths and a community of practice light on possible ways to increase the number of within the RPNGC. District law and justice lead- youth and women entering komiti work. ership could complement this RPNGC initiative by establishing a program of regular interac- 3. A study of best practices for police and other tion with citywide community policing, including agencies that engage with komitis. Police can attendance at mediation, training, elections, and certainly disrupt komiti mediation and other other events. work, but there are strong cases where police sta- tion commanders and others work very well with komitis to resolve conflicts and build a system of 7.5  A FUTURE RESEARCH AGENDA local justice that is better respected and trusted by locals. More could be understood about suc- This study has brought programmatic research back cesses here, and their replicability. into PNG’s urban settlements for the first time in the country’s urban history. And yet, as noted above, its 4. A study of relations between komitis and local practical applications are quite particular and neces- government, especially in the new urban city sarily cautious, as these are contexts in which harm authorities, showing existing good practice and can easily be done, including by further research. looking for windows of opportunity for construc- Nevertheless, a number of possible areas for further tive engagement between komitis and the most investigation present themselves: local branches of government (ward, LLG, and district). This study could investigate ways that 7.5.1 There should be a close study of the fam- komitis can receive some of the recognition they ily-related and gender-based violence cases as crave without disrupting the flexible and inclusive they are currently handled by komitis and Village aspects of komiti function. Courts. For all their distrust of male-dominated komi- Although there is fair agreement in Western democracies about the basic elements of community policing, it should be noted that 96  comparative developing country research makes clear that there is no single uniform model. Successful community policing depends crucially on how models are adapted to the local context and history (see Davis, Hendersen, and Merrick 2003), and efforts to export Western models can be fraught with unintended and perverse consequences (see Fruhling 2007). W h at is th e signific a nc e and potential of k omiti and lida regulat ion of set t lement s? Conclusions and implic ations 61 Annex 1: Different Komiti Structures in Lae Settlements Box 36.  Variety in Komiti Systems in Five Lae Settlements 4 Mile Settlement, Ward 4, Lae Urban local-level government (LLG): Settled in the early 1980s, this settlement is led by the elected president of the 4 Mile subward area, who is also chairman of the overall 4 Mile/Ward 4 Law and Order Komiti and the 4 Mile Haicoss Village Court (11 officially selected officials, as yet ungazetted and unremunerated). There are three zones: two on low-covenant self-help settlements, and one informal one on state land. Each zone has five komitis (individuals) “appointed by the community” for life (though with guidance from the chairman). Appointments are based on community respect and performance and not ethnic allegiance, but almost all are highlanders. There are two practicing woman komitis. Mediations and cots are held regularly across 4 Mile at different sites, especially at the 3 Mile Police Station, presided over by a mix of Village Court magistrates in waiting, komitis, and other local leaders who attend in the hope of becoming komitis. Nawaeb Blok, Ward 1, Lae Urban LLG. Settled in early 1980s by mixed Morobe, Nawaeb, and Chimbu settlers, it is one of seven zones in Ward 1. It has an elected blok president, but no blok law and order chairman (the local Village Court chair fills that role). The zone includes both alienated state land and over 30 years of settlement on what was previously state land (but now returned to the landowner) on a river floodplain. The blok is divided into nine ethnically mixed and numbered sections (with Sepik, Chimbu, or Nawaeb dominant in some sections, reflected in komiti appointments). Each section is represented by one–three komitis, chosen by section residents along local authority lines according to both ethnicity and competence. There have been two women komitis in the past, but currently there are none. There is turnover of komitis, with two of the nine “thrown out” in the recent past. Market fees are collected by the blok chairman and his market komiti. A local police outpost was removed in 2013, but there is the occasional liaison with the Central Lae Community Policing Unit over homebrew. Pita Blok, Kamkumung, Ward 16, Lae rural (Ahi) LLG. Pita Blok has two different, overlapping komiti systems, each with its own chairman: one appointed by the papa graun and one claiming the authorization of Ward 16. The papa graun’s chairman is a member of the dominant Engan ethnicity, who has a memorandum of understanding with the papa graun underpinning his long-term tenure. His komiti is divided into two halves: land management and law and order, both working across the two zones of Pita Blok, with 10 administrative/land (blok) and 10 law and order komitis in each zone. All are selected by the chairman who looks for ability and representation for “all 21 of PNG’s provinces.” The other ward-related komiti structure is chaired by a Sepik man, whose komitis are appointed by residents to represent their ethnic groups. The papa graun land management or blok komiti collect rents and actively manages the settlement, evicting unsociable people. It permits and collects a tax on rent houses. The papa graun komiti has four “business reps” who impose a local tax on commercial activity, including rent houses, and cooperate with local papa grauns to regulate local markets. The papa graun’s komiti chair is also a peace officer or bailiff at the Kamkumung Village Court. There are no women komitis. Niu Rocks. One of nine zones within Ward 14, Lae Rural (Ahi) LLG. Traditional landowning, residential. The papa graun has appointed four komitis, led by a long-term resident who is a Morobe Highlands male to oversee the mixed settlement (which has just five Highland families). This komiti 62 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS allocates lots and collects rents on land, physically separating housing bloks allocated to different ethnic groups, and also adjudicates disputes. The papa graun maintains strong links with the local police. Tenants are regularly expelled from the blok for their or their children’s malfeasance, sometimes including the burning of houses and entire properties. Recent moves by the papa graun to create street-level komitis resulted in the appointment of two women. Abong: one of nine zones within Ward 14, Lae Rural (Ahi) LLG. Traditional landowning, residential. The papa graun is not active in law and justice. Abong Zone has several komitis that are recognized by and supported in different parts of the blok, often competing. One faction of the zone komiti is seen as aligned to a specific ethnic group. The community did elect a woman komiti, “but they left her behind” when hearing cases. Annex 1: Different Komit i St ruc t ures in Lae Set t le m e n ts 63 Annex 1: Research Methods This exploratory and formative piece of research used primary ethnographic and qualitative methods to examine neighborhood mediation practices: observation, participation, case studies, and group and individual interviews. Building on prior work in Port Moresby, and over the course of a six-month period in Lae, this research developed and validated initial findings through triangulation and qualitative sampling strategies, including snowball sampling and sampling for difference and redundancy. At the same time, it applied theoretical and analytic approaches (drawn largely from international/comparative political science, urban political economy, and historical institutional analysis) to predict and explore patterns of institutional development and capability. These investigations enabled a wealth of empirical materials to be assembled. Only the most illustrative of these, however, could be used in the body of the report itself. • Given the almost complete absence of studies in this area within Papua New Guinea (PNG), mainly due to access and safety issues but also stigma, as well as the access and safety issues for this study itself, it was crucial that formative work, framing analyses, and empirical exploration be conducted flexibly. Researchers needed to be able to secure access and at the same time respond to new, presenting perspectives where these came to light. They needed to make clear that they were not looking in the first instance to criticize or stigmatize local mediation processes, nor to enable others to discipline them from formal law and justice positions. • Research began with an extended series of conversations with local people in public spaces (markets, churches, shops, bus stops) and private homes, and moved from these contexts into a series of more structured individual and group interviews and the development of case studies in public sites, homes, workplaces, and businesses. It continued with discussions with komitis and lidas and those around them, and with observations at komiti-run mediation, election planning, meetings with officials, and community events, at which discussions were had with the main actors and various participants who were present. Triangulation to validate emerging findings was actively pursued. • Access to settlements was facilitated through discussions with various local leaders (men and women) and, in a few situations, with the police. Referrals and introductions from networks of mediators, lidas, women, and youth offered a further extension of the research into different settlements. • Relevant international literature and works from PNG were consulted, and the researchers’ considerable experience of community-level organization in relation to law and justice drawn on. But descriptive and analytic framings were also approached flexibly and developed through well-known exploratory and formative investigative techniques. • These techniques included both triangulation and snowball-, difference-, and redundancy-based sampling, as well as ongoing dialogue and iterative analysis conducted by the team. Snowball sampling, where a member of one group with whom rapport and a level of communication and trust has been established is asked to introduce researchers to others who might inform the research, has the advantage for researchers entering the field for the first time of being able to draw on the trust established with one group or interviewee in order to engage others. It works particularly well in intense social contexts such as markets and settlements in PNG, providing researchers with the basic safety of being known and under the care of known local actors. Short of entering under full armed official escort (which would impose constraints of its own on research reliability), there is simply no alternative to this kind of relational approach during entry and to some extent ongoing research in and around settlements. 64 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS • For obvious reasons, snowball sampling needs to be complemented by other sampling techniques, including, in this case: i) applying triangulation (a technique that validates data by deliberately seeking different perspectives from multiple sources to cross verify, with the differences based on accepted social science parameters), ii) using multiple points of entry and research team mixes (not becoming dependent on or captured by one person, group, or site), iii) sampling for difference (ensuring that the qualitative sample includes the full range of points of view), and iv) sampling to redundancy (pursuing these techniques until little or no new material and perspectives come to light). In urban PNG’s mixed settlements, differences that matter include, at a minimum, ethnicity, gender, class, age, and duration of residence, and this research actively sampled for these differences at every point. To some extent, sampling and questions were based on analytic framings found to be significant elsewhere. • Common issues in this kind of research include the predominance of perspectives from (older, male, educated, dominant ethnic) leaders and the difficulties in accessing those most vulnerable or peripheral to power. This research indeed focused on understanding in the first instance the institutional mechanisms themselves and focused its attention on those intimately involved in komiti work: mediators, komiti members, local leaders, and Village Court officials, as well as those participating as parties to mediation events. Around these parameters, which certainly favored older males, initial triangulation and sampling for ethnic, gender, age, and class difference as described above occurred, with interviews and focus groups at and around each site deliberately involving only women, youth, or ethnic minorities. As there was more limited participation in this initial research from those not immediately involved in (but often powerfully affected by) mediation processes, the need for further research, especially involving those vulnerable and peripheral to mediation processes, is strongly stressed. • Single interviews with strangers in these kinds of contexts produce results with very low reliability. Within both neighborhoods and the research team, this research was informed by ongoing iterative practices of reflecting together on emerging findings (in groups, individually, in review of writing and conversation). Clarifications were often sought and different perspectives explored, and time for reflection enabled all sides to sharpen their sense of what was happening and what needed further, wider investigation. Following these protocols respectfully within local settlement contexts meant that interviews needed to be flexible in terms of attendees, since family members would want to be in the conversation and some would want to dominate. Multiple and iterative conversations involving different individuals and groups, and careful use of group and researcher splitting to enable different kinds of voices to emerge, were used throughout. • Findings were discussed and validated during a series of presentations to different groups, including mediators, komitis, and law and justice sector and urban safety program staff, both indigenous and expatriate. • Altogether, more than 150 separate individual conversations, as well as semi-structured individual and group interviews, of this kind were performed in Lae, building on a previous series of 200 similar interviews in the earlier Port Moresby work. Conversations occurred in the familiar mix of Tok Pisin and English, within which much of this “local public” business is pursued. Although the research leader developed a certain facility in at least hearing Tok Pisin, translation was locally and immediately available on both sides. Overall, researchers were pleasantly surprised and impressed by both the eagerness of locals to talk and the ways their discourse reflected the obvious local development of critical perspectives and widely shared understanding of what is happening in neighborhood processes. Variously or even randomly encountered people at very diverse sites would use highly similar language to describe what local leaders are doing. Annex 1: Research Me thods 65 Bibliography Agostini, Giulia, Francesca Chianese, William French, and Amita Sandu. 2007. “Understanding the Processes of Urban Violence: an Analytical Framework.” Crisis States Research Centre, Development Studies Institute, London School of Economics, London. AFSC (American Friends Service Committee). 2013. “The Local Peace Network Handbook: A Reflection and Action Guide for Local Violence Reduction.” AFSC, Tegucigalpa. Amnesty International. 2005. “No One to Turn To: Women’s Lack of Access to Justice in Rural Sierra Leone.” Briefing Paper, Amnesty International, London. Anleu, Sharyn L. Roach. 2009. Law and Social Change, 2nd ed. Thousand Oaks, CA: Sage Publications. Auty, Richard M. 2007. “Patterns of Rent Extraction and Deployment in Developing Countries: Implications for Governance, Economic Policy and Performance.” In Advancing Development: Core Themes in Global Economics, edited by G. Mavrotas and A. Shorrocks, 555–77. London: Palgrave Macmillan. Baker, Judy L., and Gauri U. Gadgil, eds. 2017. East Asia and Pacific Cities: Expanding Opportunities for the Urban Poor. Washington, DC: World Bank. Barma, Naazneen, Kai Kaiser, Tuan Minh Le, and Lorena Viñuela. 2012. Rents to Riches? The Political Economy of Natural Resource-Led Development. Washington, DC: World Bank. Bourdieu, Pierre. 1981. Language and Symbolic Power. Edited by John B. Thompson. Translated by Gino Raymond and Matthew Adamson. Cambridge, UK: Polity Press. ———. 1986. “The Forms of Capital.” In Handbook of Theory and Research for the Sociology of Education, edited by J. G. Richardson, 241–58. New York: Greenwood. Bovaird, Tony, and Elke Loeffler. 2012. “From Engagement to Co-Production: How Users and Communities Contribute to Public Services.” In New Public Governance, the Third Sector and Co-Production, edited by Taco Brandsen and Victor Pestoff. London: Routledge. Brenner, Neil. 2004. New State Spaces: Urban Governance and the Rescaling of Statehood. Oxford: Oxford University Press. Castillo Diaz, Pablo. 2006. Customary Law Reform in Sub-Saharan Africa: The Status of Women Between Customs and Statutes. Unifem Briefing Note, Feb 2006. New York: United Nations. Clifford, William, Louise Morauta, and Barry Stuart. 1984. Law and Order in Papua New Guinea, vols. 1 and 2. Port Moresby: Institute of National Affairs. Craig, David, and Doug Porter. 2014. “Post-Conflict Pacts and Inclusive Political Settlements: Institutional Perspectives from Solomon Islands.” ESID Working Paper 39, Effective States and Inclusive Development Research Centre, University of Manchester, Manchester. 66 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS ———. 2017. “‘There is Security from this Place’: Promoting the Safety and Economic Vitality of Port Moresby’s Local Markets.” Policy Note, Justice for the Poor, World Bank, Washington, DC. Craig, David, Doug Porter, and Fiona Hukula. 2016. “‘Come and See the System in Place’: Mediation Capabilities in Papua New Guinea’s Urban Settlements.” Research Report, Justice for the Poor, World Bank, Washington, DC. Davis, Robert, Nicole Henderson, and Cybele Merrick. 2003. “Community Policing: Variations on the Western Model in the Developing World.” Police Practice and Research 4 (3): 285–300. Davis, Mike. 2000. Magical Urbanism: Latinos Reinvent the U.S. Big City. London: Verso. ———. 2006. Planet of Slums. London: Verso. Demian, Melissa. 2003. “Custom in the Courtroom, Law in the Village: Legal Transformations in Papua New Guinea.” Journal of the Royal Anthropological Institute 9 (1): 97–115. ———. 2016. “Court in Between: The Spaces of Relational Justice in Papua New Guinea.” Australian Feminist Law Journal 42 (1): 13–30. Evans, Daniel, Michael Goddard, and Don Paterson. 2010. “The Hybrid Courts of Melanesia: A Comparative Analysis of Village Courts of Papua New Guinea, Island Courts of Vanuatu, and Local Courts of Solomon Islands.” Justice and Development Working Paper13/2011, World Bank, Washington, DC. Filer, Colin. 2006. “Custom, Law and Ideology in Papua New Guinea.” Asia Pacific Journal of Anthropology 7 (1): 65–84. Forster, Till, and Lucy Koechlin. 2018. “‘Traditional’ Authorities.” In The Oxford Handbook of Governance and Limited Statehood, edited by T. Risse, T. Borzel, and A. Draude. Oxford: Oxford University Press. Fruhling, Hugo. 2007. “The Impact of International Models of Policing in Latin America: the Case of Community Policing.” Police Practice and Research 8 (2): 125–44. Goddard, Michael. 2005a. The Unseen City: Anthropological Perspectives on Port Moresby, Papua New Guinea. Honolulu: University of Hawaii Press. ———. 2005b. “Research and Rhetoric on Women in Papua New Guinea’s Village Courts.” Oceania 75 (3): 247–67. ———. 2009. Substantial Justice: An Anthropology of Village Courts in Papua New Guinea. New York: Berghahn. Goodwin, Mark, Simon Duncan, and Susan Halford. 1993. “Regulation Theory, the Local State, and the Transition of Urban Politics.” Environment and Planning D: Society and Space 11 (1): 67–88. Guthrie, Gerard, James Laki, and Fiona Hukula. 2006. “Lae Community Crime Survey 2005.” National Research Institute, Port Moresby. Hagedorn, John M. 2008. A World of Gangs: Armed Young Men and Gangsta Culture. Minneapolis, MN: University of Minnesota Press. Bibliog ra ph y 67 Hambleton, Robin. 2017. “Inclusive City Governance: International Lesson Drawing from Progressive Urban Leadership.” In DSA 2017: Sustainability Interrogated: Societies, Growth and Social Justice. Bristol: University of the West of England. Harvey, David. 2008. “The Right to the City.” New Left Review 53: 23–40. Helbich, Marco, and Michael Leitner. 2017. “Frontiers in Spatial and Spatiotemporal Crime Analytics—An Editorial.” ISPRS International Journal of Geo-Information 6 (3): 73. Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press. Hukula, Fiona. 2013. “Gender-Based Violence and Masculinity in an Urban PNG Settlement.” PhD diss., University of St. Andrews. Ivaschenko, Oleksiy, Darian Naidoo, David Newhouse, and Sonya Sultan. 2016. “Can Public Works Programs Reduce Youth Crime? Evidence from Papua New Guinea’s Urban Youth Employment Project.” Policy Research Working Paper 8032, World Bank, Washington, DC. Jones, Paul, and M. Kep. 2012. “Understanding Urbanisation in the PNG Context.” In PNG Symposium 2012: PNG Securing a Prosperous Future. Melbourne: Crawford House Publishing. Joshi, Anuradha, and Mick Moore. 2003. “Institutionalised Co-production: Unorthodox Public Service Delivery in Challenging Environments.” Journal of Development Studies 40 (4): 31–49. Kassimir, Ronald. 2001. “Producing Local Politics: Governance, Representation and Non-State Organizations in Africa.” In Intervention and Transnationalism in Africa. Global-Local Networks of Power, edited by T. Callaghy, R. Kassimir, and R. Latham, 93–114. Cambridge, UK: Cambridge University Press. Khan, Mushtaq. 2010. “Political Settlements and the Governance of Growth-Enhancing Institutions.” Available at http://eprints.soas.ac.uk/9968/. Korf, Benedikt, Timothy Raeymaekers, Conrad Schetter, and Michael J. Watts. 2018. “Geographies of Limited Statehood.” In Oxford Handbook of Governance and Limited Statehood, edited by T. Risse, T. Borzel, and A. Draude. Oxford: Oxford University Press. Koster, Martijn, and Pieter A. de Vries. 2012. “Slum Politics: Community Leaders, Everyday Needs, and Utopian Aspirations in Recife, Brazil.” Focaal—Journal of Global and Historical Anthropology 62: 83–98. Lakhani, Sadaf, and Alys M. Willman. 2014a. “Trends in Crime and Violence in Papua New Guinea.” Research and Dialogue Series, Socio-economic Costs of Crime and Violence in Papua New Guinea 1, World Bank, Washington, DC. ———. 2014b. “Gates, Hired Guns and Mistrust - Business Unusual: The Cost of Crime and Violence to Businesses in Papua New Guinea.” Research and Dialogue Series, Socio-economic Costs of Crime and Violence in Papua New Guinea 4, World Bank, Washington, DC. Lancee, Bram. 2010. “The Economic Returns of Immigrants’ Bonding and Bridging Social Capital: the Case of the Netherlands.” International Migration Review 44 (1): 202–26. 68 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Lederach, John Paul, and Preeti Thapa. 2012. “Staying True in Nepal: Understanding Community Mediation through Action Research.” Occasional Paper 10, Asia Foundation, Washington, DC. Lee, Maggy, and Syed Jahangeer Haider. 2012. “A Letter from Bangladesh – Developing Gender-Responsive Community Policing in Bangladesh.” Crime Prevention and Community Safety 14 (1): 69–77. Mason, Andrew. 2007. “Demographic Dividends: the Past, the Present, and the Future.” In Population Change, Labor Markets and Sustainable Growth: Towards a New Economic Paradigm, edited by A. Mason and M. Yamaguchi, 75–98. Bingley, UK: Emerald Publishing. Mason, Simon. 2009. “Insider Mediators: Exploring their Key Role in Informal Peace Processes.” Berghof Foundation for Peace Support, Berlin. Mazerolle, Lorraine, Rebecca Wickes, and James McBroom. 2010. “Community Variations in Violence: The Role of Social Ties and Collective Efficacy in Comparative Context.” Journal of Research in Crime and Delinquency 47 (1): 3–30. McFarlane, Colin. 2012. “Rethinking Informality: Politics, Crisis and the City.” Planning Theory and Practice 13 (1): 89–108. Medecins sans Frontieres. 2015. “Return to Abuser: Gaps in Services and a Failure to Protect Survivors of Family and Sexual Violence in Papua New Guinea.” Medecins sans Frontieres, Amsterdam. Montgomery, Mark R. 2008. “The Urban Transformation of the Developing World.” Science 319 (5864): 761–64. Murray, Joseph, Daniel Ricardo de Castro Cerqueira, and Tulio Kahn. 2013. “Crime and Violence in Brazil: Systematic Review of Time Trends, Prevalence Rates and Risk Factors.” Aggression and Violent Behavior 18 (5): 471–83. North, Douglass C., John Joseph Wallis, and Barry R. Weingast. 2012. Violence and Social Orders: a Conceptual Framework for Interpreting Recorded Human History. Cambridge, UK: Cambridge University Press. Norwood, Hugh. 1984. Port Moresby: Urban Villages and Squatter Areas. Port Moresby: University of Papua New Guinea Press. Oram, Nigel Denis. 1976. Colonial Town to Melanesian City: Port Moresby, 1884–1974. Port Moresby: University of Papua New Guinea Press. Papua New Guinea. 2015. “Lae City Authority Act 2015.” Port Moresby. ———. Law and Justice Sector. 2011a. “Community Crimes Surveys 2010: Port Moresby.” Port Moresby. ———. Law and Justice Sector. 2011b. “Community Crimes Surveys: Lae.” Port Moresby. ———. Law and Justice Sector Working Group. 1999. The National Law and Justice Policy and Plan of Action Toward Restorative Justice. Law and Justice Sector Working Group, Port Moresby. Bibliog ra ph y 69 Perlman, Janice E. 2006. “The Metamorphosis of Marginality: Four Generations in the Favelas of Rio de Janeiro.” Annals of the American Academy of Political and Social Science 606 (1): 154–77. Putnam, Robert D. 2000. Bowling Alone: The Collapse and Revival of American Community. New York: Touchstone Books. Rasmussen, Mattias Borg, and Christian Lund. 2018. “Reconfiguring Frontier Spaces: The Territorialization of Resource Control.” World Development 101: 388–99. Rew, Alan. 1974. Social Images and Process in Urban New Guinea: a Study of Port Moresby. St. Paul, MN: West Publishing. Rutherford, Andrew. 2002. Growing Out of Crime: the New Era, 2nd ed. London: Waterside Press. Sakaya, Claudia, Paola Sanonib, and Toshihiro Hanazato DEng. 2011. “Rural to Urban Squatter Settlements: The Micro Model of Generational Self-Help Housing in Lima-Peru.” Procedia Engineering 21: 473–80. Sampson, Robert. 2004. “Neighbourhood and Community: Collective Efficacy and Community.” Safety. New Economy 11 (2): 106–13. Saunders, Doug. 2011. Arrival City: How the Largest Migration in History is Reshaping Our World. London: William Heinemann, Random House. Sinclair, James Patrick. 1998. Golden Gateway: Lae and the Province of Morobe. Goolwa, Australia: Crawford House Publishing. Slater, Dan. 2010. Ordering Power: Contentious Politics and Authoritarian Leviathans in Southeast Asia. Cambridge Studies in Comparative Politics. Cambridge, UK: Cambridge University Press. Stuart, Ian. 1970. Port Moresby, Yesterday and Today. Los Angeles: Pacific Publications. Suurmond, Jeannine, and Prakash Mani Sharma. 2012. “Like Yeast that Leavens the Dough? Community Mediation as Local Infrastructure for Peace in Nepal.” Journal of Peacebuilding & Development 7 (3): 81–86. Swyngedouw, Erik. 2005. “Governance Innovation and the Citizen: The Janus Face of Governance-beyond- the-State.” Urban Studies 42 (11). UNDP (United Nations Development Programme) and EC (European Commission). 2014. “Supporting Insider Mediation: Strengthening Resilience to Conflict and Turbulence.” UNDP, New York. UN (United Nations) 2014. World Urbanisation Prospects. 2014 revision. United Nations, New York. Watts, Michael J. 2018. “Frontiers: Authority, Precarity, and Insurgency at the Edge of the State.” World Development 101: 477–88. Weisburd, David, Anthony Braga, Elizabeth R. Groff, and Alese Wooditch. 2017. “Can Hot Spots Policing Reduce Crime in Urban Areas? An Agent-Based Simulation.” Criminology 55 (1): 137–73. 70 S A F E T Y A N D S ECUR I TY AT THE EDGES O F THE STAT E: LOCAL REGULAT ION IN PAPUA NEW GUINEA’S URBAN SET TL E ME NTS Wolfe, Alvin, and Honggang Yang, eds. 1996. Anthropological Contributions to Conflict Resolution. Athens, GA: University of Georgia Press. World Bank. 2017a. “Papua New Guinea Systematic Country Diagnostic.” World Bank, Washington, DC and Port Moresby. ———. 2017b. Governance and the Law. World Development Report 2017. Washington, DC: World Bank. Bibliog ra ph y 71 Bibliog ra ph y 73