JUSTICE BRIEFING NOTE 56006 POOR for the June 2010 Volume 5 | Issue 3 The Lay of the Land: Land Access and Dispute Resolution in Timor-Leste1 T his report examines land access, disputes,2 and dis- Some regional variation in responses was also found, pute resolution in Timor-Leste, using findings from with 81 percent of respondents in Region 5 (Oecusse) the justice module included in an extension of the stating that titling would reduce disputes vs. just 58 2007 Timor-Leste Survey of Living Standards (TLSLS2) and percent in Region 2 (Ainaro, Manatutu, Manufahi). a review of relevant social-science literature.3 The extension However, this variation was primarily due to the large survey (TLSLSx) revisited a nationally representative sub- number of respondents in Region 2 who believe that sample of the TLSLS2 between April and October 2008.4 The titling would both resolve and create conflicts.7 respondent for the justice module was randomly selected to The majority of respondents (78 percent) believe that be the household head or his/her spouse, and the TLSLSx a paper land title is a stronger claim in court than a revisited 1,716 respondents across Timor-Leste's thirteen community-recognized traditional land claim. districts.5 Survey topics included: (1) access to information Ten percent of households had experienced land-related and decision making; (2) opinions and knowledge of the law; disputes in the year prior to the survey. While the overall (3) trust and local institutions; and (4) dispute resolution. number of disputes was low, 40 percent of land-related Findings from this survey have been summarized in three short briefing papers focusing on land and conflict, youth per- spectives, and community trust and decision making.6 1 Justice for the Poor (J4P) is a World Bank program that focuses on mainstream- ing justice considerations and conflict management into development processes. In Timor-Leste, J4P began in July of 2008 with the generous support of AusAID. This report presents TLSLSx findings related to land and 2 A note on terminology: throughout this report, terms such as "disputes," "con- property. It begins by presenting a context for land and prop- flict," and "trust" are used. These terms should be understood as referring to relation- erty in Timor-Leste, goes on to discuss East Timorese per- ships between individuals and/or households, and not broader relationships between groups (that is, regional, ethnic, and so on). spectives on land ownership and access, and concludes with 3 TLSLSx is a supplement to the living-standards survey implemented by the gov- a discussion on land disputes and their resolution. It is hoped ernment of Timor-Leste National Statistics Directorate and the World Bank in 2007. For more information and a copy of the TLSLS statistical abstract, please visit http:// that this report will be a valuable resource for civil servants, dne.mof.gov.tl/TLSLS/AboutTLSLS/index.htm. representatives of civil society, and donor agencies working 4 For additional information related to the survey module, including a breakdown of on land and property-related issues in Timor-Leste. respondents by region and district, residence (urban or rural), gender, age, and other categories, please see "Justice Module of the Timor-Leste Survey of Living Standards Extension: An Overview" available at http://go.worldbank.org/ZRKELPETD0. 5 It should be emphasized that this report presents findings related to heads of house- hold and their spouses, and these views are not necessarily representative of the Key Findings population as a whole. 6 All Justice for the Poor Timor-Leste reports can be accessed at www.worldbank. There is widespread support for land titling in Timor- org/justiceforthepoor. 7 The five survey regions are: Region 1: Lautem, Baucau, Viqueque; Region 2: Ain- Leste; overall, 72 percent of survey respondents believe aro, Manufahi, Manatutu; Region 3: Aileu, Dili, Ermera; Region 4: Bobonaro, Cova that land titling would reduce land-related conflict. Lima, Liquica; and Region 5: the enclave of Oecusse. This report was written by Pamela Dale (Consultant, J4P-TL), Kristen Himelein (World Bank, Development Research Group), Denis Nikitin (Consultant, World Bank South Asia Human Development), and Angie Bexley (Anthropologist, Australia National University). Kristen Himelein, Geraldo Moniz da Silva (Consultant, J4P-TL) and the staff of the government of Timor-Leste's National Statistics Directorate assisted extensively with the planning and implementation of the research and this report. Lene Ostergaard (Coordinator, J4P-TL), Matthew Stephens (Task Team Leader, J4P-TL), David Butterworth (Consultant, J4P-TL), and Rod Nixon (Consultant, J4P-TL) provided useful comments on drafts of this report. THE WORLD BANK disputes resulted in property damage and 10 percent significant progress has been made since 2009. The govern- led to injury. ment of Timor-Leste, with the support of the U.S. Agency Respondents expressed a clear preference for local and for International Development (USAID), has launched a traditional authorities in resolving land- and property- land registration program in district centers and peri-urban related disputes. However, substantial regional varia- areas, which has led to the registration of over 10,000 claims tions emerged, as did urban-rural divides. in seven districts.14 A transitional land law15 has been social- Respondents were divided in their understandings of ized throughout Timor-Leste, and was recently approved by marital property. Approximately half of respondents the Council of Ministers; consideration by Parliament is ex- believed that all marital property was under the con- pected in mid-2010. This law aims to establish a regime for trol of the head of household, while 39 percent believed the identification and regulation of land ownership, including that land in the possession of one party at the start of the ordering of competing claims. It also provides a start- a marriage remained personal property throughout the ing point for the legal recognition and protection of commu- marriage. nity land rights. In recognition of the importance of ongoing discussions on land and property in Timor-Leste, this report is intended to provide the government and other actors in The Context of Land Access and Use in Timor-Leste's land sector with data on Timorese perspectives Timor-Leste on land ownership and the value of titling, as well as infor- mation on the prevalence and resolution of disputes over land Land is an important feature in the daily lives of Timorese, and property. representing both an ancestral legacy and the primary source of income for the majority of the population.8 Land owner- ship in Timor-Leste is dominated by customary norms, par- Perspectives on Land Ownership and ticularly outside of urban and peri-urban centers.9 Customary Access ownership takes a variety of forms, and can differ between communities. Communal or common property ownership ex- Land Titling ists, particularly in areas used for food-crop cultivation and As noted above, the Timor-Leste government is currently un- around water points.10 However, most customary groups in dertaking a land registration program (known as Ita Nia Rai) in Timor-Leste also have heritable individual- or family-use district centers and peri-urban areas of Timor-Leste. Through rights to land such as residential, garden, and plantation this program, as a precursor to land titling, the government is plots.11 Finally, ritualized prohibitions to access and use (re- identifying and registering areas of uncontested land owner- ferred to as tara bandu) are maintained on some areas of land. 8 According to 2007 TLSLS data, approximately 73 percent of Timorese, and 80 During the final century of Portuguese colonization, the percent of poor Timorese, depend on agriculture for their livelihood. Timor-Leste, period of Japanese occupation during World War II, and es- Directorate of National Statistics and World Bank, "Timor-Leste: Poverty in a pecially the quarter century of Indonesian occupation, large Young Nation" (Dili, Timor-Leste: Directorate of National Statistics, 2008), http:// pascal.iseg.utl.pt/~cesa/TL-Poverty-in-a-young-nation-25-Nov-2008.pdf. Accessed numbers of East Timorese experienced displacement from June 2010. their ancestral lands as a result of state co-optation or strategic 9 Rod Nixon, "Challenges for Managing State Agricultural Land and Promoting Post-Subsistence Primary Industry Development in East Timor," in Crisis in Timor- relocation.12 Following Timor-Leste's historic vote for inde- Leste: Understanding the Past, Imagining the Future, ed. Dennis Shoesmith (Dar- pendence in 1999, an estimated 83 percent of the population win: CDU Press, 2007). Note that according to 2005 data, only about 3 percent of was displaced as a result of militia-perpetrated violence, and rural land in Timor-Leste appears to have been alienated from customary tenure, meaning that customary systems of authority predominate in approximately 97 the majority of land and property records were destroyed.13 An percent of the land. See Rod Nixon, "Non-Customary Primary Industry Land Sur- outbreak of violence in 2006­07, stemming in part from ex- vey: Landholdings and Management Considerations," USAID/ARD Inc. Research Report (Washington, DC: USAID, 2005), http://pdf.usaid.gov/pdf_docs/Pnade790. isting land and property disputes, led to the displacement (or pdf. Accessed June 2010. redisplacement) of an additional 10 percent of the population. 10 Daniel Fitzpatrick, Andrew McWilliam, and Susana Barnes, "Policy Notes on Customary Land in Timor-Leste," East Timor Law Journal (2008). 11 Ibid. Note that these individual or familial use rights often do not translate to a This history has contributed to continuing confusion over right to sell the land to those outside of the community. land and property ownership in Timor-Leste, as customary 12 Daniel Fitzpatrick, "Mediating Land Conflict in East Timor," in Making Land Work, vol. 2 (Canberra: AusAID, 2008), http://www.ausaid.gov.au/publications/pdf/ rights, colonial Portuguese land titles, Indonesian titles, and MLW_VolumeTwo_CaseStudy_9.pdf. Accessed June 2010. nontraditional long-term land occupation compete for pri- 13 Ibid. 14 macy. Since independence in May 2002, there has been a sig- The project was piloted in Manatutu and Liquica districts. Data collection in Manatutu has been completed, with collection in Liquica expected to follow shortly. nificant delay in establishing a formal land administration in The program is now active in Aileu, Baucau, Bobonaro, Oecusse, and Dili. A public the nation, partly due to uncertainties in land ownership and information effort is ongoing in advance of a launch in Lautem district. For more information on the Ita Nia Rai program, see www.itaniarai.tl. the time taken for state actors to understand and respond to 15 Known formally as the Special Regime for Determination of Ownership of Im- the complexities of customary ownership systems. However, movable Property. ship (either by individuals, groups, or legal entities), as well Chart 1: Potential Impact of Land Titling in as registering and attempting to resolve disputed ownership Timor-Leste and boundary claims. With this in mind, the TLSLSx survey 100% posed several questions on the perceived value of land titling. 80% 60% The survey found strong support for a land titling program 40% in Timor-Leste. There was widespread belief that an official land title was more valuable than a traditional land claim 20% when pursuing a claim in court.16, 17 Additionally, 72 percent18 0% National Region 1 Region 2 Region 3 Region 4 Region 5 of respondents indicated that land titling would help to reduce (Baucau, (Ainaro, (Aileu, (Bobonaro, (Oecusse) disputes over land and property.19 There was little variation in Lautem, Manufahi, Dili, Cova Lima, Viqueque) Manatutu) Ermera) Liquica) support for titling across gender, education level, type of em- ployment, or marital status. Furthermore, though one would Reduce Conflict Create Conflict Neither Both predict stronger support for land titling in urban areas, where tenure is generally seen as less secure, there was surprisingly little variation between urban and rural respondents.20 There and most property rights are thus subject to the authority of does, however, appear to be a trend of younger households be- customary systems. Timor-Leste is home to both patrilineal ing more in favor of the land titling process than older popula- and matrilineal customary land-tenure systems, with matri- tions, with 79 percent of youth believing that land titling could lineal systems common in Bobonaro, Cova Lima, Manatutu, potentially reduce disputes, compared with only 70 percent of and Manufahi districts, and patrilineal systems predominat- respondents over the age of 60.21 Also, while only 16 percent ing in the remaining districts. While generally, land inheri- of youth thought titling could lead to an increase in conflict, tance patterns in matrilineal areas favor women and girls and a substantial minority over 60, 28 percent, believed that ad- those in patrilineal areas favor men and boys, land inheritance ditional conflict might emerge with a titling regime.22 Some and use are subject to various exceptions and negotiations.28 regional variation in responses was also found, with 81 per- For example, unmarried women in patrilineal societies may cent of respondents in Region 5 (Oecusse) stating that titling be permitted to inherit or to hold a life interest in property, would reduce disputes vs. just 58 percent in Region 2 (Ainaro, Manatutu, Manufahi).23 However, as shown in the chart below, 16 The two options were: "When pursuing a land claim in the court system, the this variation was primarily due to the higher degree of am- strongest claim is an official paper land title," and "When pursuing a land claim bivalence in Region 2, where more respondents believed that in the court system, an official paper land title is not regarded as a stronger claim than a community-recognized traditional land claim." Respondents could also select titling had the potential to both resolve and create conflicts. "neither" or "both," though these responses were discouraged. Overall, the reasons behind both the regional disparities and 17 A 2004 survey by the East Timor Land Law Project found that just 15 percent of the lack of urban-rural variation require further exploration. rural and 33 percent of urban respondents were aware of a land claims process. De- spite the generally low awareness, the same survey found that respondents generally saw official state claims as valuable. See East Timor Land Law Program "Report While this data shows support for titling, and though stud- on Research Findings, Policy Options and Recommendations for a Law on Land Rights and Title Restitution" (Burlington, VT: ARD, Inc., July 2004), http://www. ies conducted in Timor-Leste have concluded that in some laohamutuk.org/Oil/LNG/Refs/020ARDLandRights.pdf. Accessed June 2010. instances (particularly in rural areas), land titling could be 18 Ninety-five percent confidence interval (C.I.) (68.1­75.0). 19 beneficial,24 experience from nations with similarly strong The two options were: "A process of land titling would help to draw definite boundaries, and reduce conflict in the community," and "A land titling process systems of customary land management suggest caution in would cause more problems than it solves by stirring up old problems and offer- extending a land titling system into rural areas, where bound- ing the opportunity to steal land with fake claims." Respondents could also select "neither" or "both," though these responses were discouraged. aries between villages are strongly contested and wider polit- 20 Sixty-nine percent of urban and 73 percent of rural respondents believed that ical areas tend to demarcate sites of past conflict and linger- titles could reduce conflict, while 27 percent of urban and 18 percent of rural re- spondents saw the potential for increased conflict with titles. ing, possibly unresolved enmity.25 Additionally, titling has the 21 This difference, however, is not statistically significant (t-stat = 1.39), likely due potential to disenfranchise weaker members of the commu- to the relatively small sample size. Please see the statistical abstract for more details. 22 nity, particularly women, and there is also great uncertainty This difference is statistically significant at the 5 percent level (t-stat = 2.32). 23 This difference is statistically significant at the 95 percent confidence level about the capacity of the Timor-Leste state to administer a (t-stat = 3.07). titling program covering the entire country. 24 Fitzpatrick et al., "Policy Notes on Customary Land in Timor-Leste." 25 Andrew McWilliam, "Meto Disputes and Peacemaking: Cultural Notes on Con- flict and its Resolution in West Timor," The Asia Pacific Journal of Anthropology 8 Gender and Access to Land no. 1 (2007): 75­91. 26 Section 54 of the Constitution of Timor-Leste guarantees Government of Timor-Leste, Constitution of the Democratic Republic of Timor- Leste. all citizens the right to property,26 and the draft Civil Code 27 Timor-Leste, Secretary of State for the Promotion of Equality, "Goodbye Conflict includes provisions protecting women's inheritance rights.27 and Welcome Development" (Written Statement, Development Partners Meeting, April 7, 2010). However, the reach of the formal justice system, and by ex- 28 Pyone MyatThu, Steffanie Scott, and Kimberly P. Van Niel, "Gendered Access to tension the protections it offers, is still limited in Timor-Leste, Customary Land in East Timor," GeoJournal, 69 no. 4 (2007): 239­255. and societies with active marriage-exchange practices often Chart 2: Land Ownership within Marriage, link inheritance to the completion of barlaque29 payments.30 by Region and Residence Thus, gender aspects of property rights under customary law 100% cannot accurately be determined solely by examining pat- 80% terns of matrilineal or patrilineal descent. 60% 40% The substantial variations in land inheritance and marital property throughout the country are reflected in the survey 20% data. When asked about control of property within a marriage 0% R1 U R1 R R2 U R2 R R3 U R3 R R4 U R4 R R5 U R5 R according to the law, approximately half of respondents indi- cated that property was controlled by the (usually male) head HH Head Original Owner Neither Both of household, while 39 percent believed it to be retained by the Source: Regions are as follows: Region 1 (Baucau, Lautem, Viqueque), Region 2 (Ainaro, Manatutu, Manufahi), Region 3 (Aileu, Dili, Ermera), Region 4 (Bobonaro, Cova Lima, Liqui- original owner.31 There were also considerable differences ac- ca), and Region 5 (Oecusse). cording to residence (urban vs. rural) and region. The majority Note: R1U = Region 1, Urban; R1R = Region 1, Rural, and so forth. (54 percent) of respondents in urban areas believed that land remained under the control of the original owner during a mar- riage; just 35 percent of rural residents believed this, though vey in early 2008.36 This is consistent with findings from the an additional 10 percent responded "both."32 The starkest dif- Ita Nia Rai program--as of March 8, 2010, the program had ferences emerged between regions, perhaps reflecting local- collected over 10,000 claims, approximately 8 percent of ized patterns of land management and control. In Region 2 which were disputed.37 While the low overall number of land (Ainaro, Manatutu, Manufahi), just 31 percent of respondents disputes prevents region-by-region comparisons, there was indicated that marital property was controlled solely by the a statistically significant difference in the percentage of the household head; in contrast, 68 percent of respondents from population experiencing a land dispute in the lowest-dispute Region 4 (Bobonaro, Cova Lima, Liquica) believed this.33 The area (Region 2 [Ainaro, Manatutu, Manufahi]--5 percent chart below presents beliefs about control of marital property, experienced disputes) and the highest (Region 4 [Bobonaro, disaggregated by both region and residence. Cova Lima, Liquica]--16 percent experienced disputes).38 While it is difficult to identify the contributing factors to the higher relative number of disputes in Region 2 vs. Region 4, Land Disputes and Dispute Resolution there are interesting parallels to the findings of a 2004 Land Law Program study that found that several sucos in Bobon- Land Disputes in Timor-Leste aro, Cova Lima, and Liquica were subject to a higher-than- In recent years, numerous books and articles have discussed the complexity and volume of land disputes in Timor-Leste.34 A history of displacement, overlapping titles, and lack of le- 29 The practice of gift exchange upon marriage between wife-giver and wife-taker gal clarity contribute to land-related disputes. Conflict may groups is often referred to as "bride price." 30 Thu et al., "Gendered Access to Customary Land." arise when certain groups claim overarching stewardship of 31 The question asked that respondents choose the response closest to their own rights to agricultural farm lands; their claim may be contested view: "According to national law, all household property is under the control of the by other groups who "opened" or "tamed" the land by clear- head of the household," or "According to national law, property acquired by the man or woman prior to marriage remains under their personal control throughout the ing the site (and undertaking the necessary rituals associated marriage." "Neither" and "both" were available, but discouraged, options. In this with taming the land). Areas of agricultural resettlement pro- case, 2 percent of respondents answered "neither," while 9 percent selected "both." 32 The difference between urban and rural percentages is statistically significant at grams are more prone to land and natural-resource conflict. the 1 percent level (t-stat = 3.07). Additionally, ethnic tensions, land and property disputes, and 33 In both regions, a high number of respondents--11 percent in Region 2 and 9 arson during the 2006­07 crisis resulted in the displacement percent in Region 4--answered "both." The difference between the two regions in respondents indicating solely the household head is statistically significant of over 100,000 people and shifted the demographics of some (t-stat = 6.58). 34 urban neighborhoods. While most of the displaced have now See for example: Daniel Fitzpatrick, Land Claims in East Timor (Marricksville, NSW: Southwood Press, 2002); Integrated Regional Information Networks (IRIN), returned to their communities or been resettled, many of the "Timor-Leste: Settling Age-Old Land Disputes," IRIN, October 31, 2008, http:// property disputes stemming from or aggravated by the crisis www.unhcr.org/refworld/docid/4917f25cc.html. Accessed June 2010; and Andrew Harrington, "Ethnicity, Violence and Land Disputes in Timor-Leste," East Timor Law remain unresolved. Journal (2007), http://www.eastimorlawjournal.org/ARTICLES/2007/ethnicity_vio- lence_land_property_disputes_timor_leste_harrington.html. Accessed June 2010. 35 Interestingly, despite the attention given to land and Ninety-five percent C.I. (7.7­13.0). 36 The question was: "Have you or anyone in your household experienced a dispute property disputes in Timor-Leste, the TLSLSx survey found with land in the past year?" a relatively low incidence of land-related disputes: just 10 37 Note, however, that Ita Nia Rai is being implemented only in urban and peri- percent35 of respondents reported experiencing a land-related urban areas, whereas the TLSLSx survey was also fielded in rural areas. 38 Ninety-five percent C.I.s: Region 1 (1.1­13.9), Region 2 (1.5­8.2), Region 3 dispute between January 2007 and the fielding of the sur- (7.6­16.5), Region 4 (9.6­21.7), Region 5 (3.0­13.0). average number of land claims.39 In contrast, no sucos in Ain- Chart 3: Authorities Resolving Land Disputes aro, Manatutu, or Manufahi experienced higher-than-average 100% 2.8% 1.0% 2.1% 4.7% claims. However, because the TLSLSx data cannot be exam- 90% 11.5% 6.1% 80% 35.3% 23.3% ined with confidence at the subdistrict or suco level, it is not 20.5% 70% possible with this data to determine if there is a link between 60% high claim and high conflict sucos. 50% 33.1% 40.0% 40% 45.8% 30% 3.4% One would expect that urban and peri-urban areas, where 20% 3.8% 0.4% 29.3% 22.1% customary systems are weaker, recent conflict-related dis- 10% 14.7% 0% placement was higher, and groups from across Timor-Leste With Another Household With Another Village Conflicting Ownership coexist, would experience a higher rate of land-related con- National Sub-District / Chefe de aldeia / Traditional Leader / flict. Surprisingly, however, urban and rural residents re- Police District Authority Chefe de suco Adat ported approximately similar numbers of disputes, though District / Appellate Courts Other / None there was some variation within districts.40 The reasons for Source: Respondents were asked where they would take hypothetical disputes where (a) this finding are unclear and would merit further study. Also, your household is experiencing a property boundary dispute with another household; (b) though the reported number of property disputes is quite low, your village is experiencing a boundary dispute with another village, which involves part of the land on which you farm; and (c) someone is claiming ownership over your land. the survey showed the potential for violence and property damage associated with these clashes. Nationally, 40 percent of reported land disputes resulted in property damage, and 10 complicated land disputes.46 Communities may be seeking a percent involved injuries.41 neutral mediator, who comes from outside the disputing com- munities, in order to resolve the dispute fairly. It may also re- Land-Dispute Resolution flect the potential for land disputes to lead to property destruc- Land disputes in Timor-Leste are entwined with local political tion and violence, as described in the preceding subsection. and cultural contexts, as well as livelihoods, and are thus resis- tant to the win-lose models often used in formal legal adjudica- While respondents were most likely to take interhouse- tion.42 Timor-Leste's complex history of colonization and forced hold boundary disputes to nonstate authorities,47 consider- resettlement has also created a situation where determining the able differences within and between regions emerged. For validity and primacy of competing land claims is extremely dif- example, the percentage of respondents indicating that they ficult. TLSLSx respondents showed a clear preference for local would approach a nonstate actor for assistance in resolv- and traditional authorities in resolving land-related disputes, though substantial minorities believed that disputes were best resolved through the police or subdistrict authorities.43 This is 39 Based on an evaluation of the number of land claims submitted to the Directorate consistent with the findings of previous studies on land- and of Land and Property by early 2004. East Timor Land Law Program, "Land Rights and Title Restitution." property-dispute resolution in Timor-Leste.44 40 Given the small sample of reported disputes, however, these differences are not statistically reliable. 41 Ninety-five percent C.I.s: Property Damage (27.8­52.0) and Injury (0.0­20.5). As depicted in the chart above, when asked about three 42 Fitzpatrick, "Mediating Land Conflict in East Timor." hypothetical dispute types related to land (property-boundary 43 Respondents were able to identify multiple fora for dispute resolution. Through- disputes between individuals, property-boundary disputes out this section, the data depicts the first forum identified by the respondent. 44 East Timor Land Law Program, "Land Rights and Title Restitution." between villages, and conflicting claims over land owner- 45 Fitzpatrick, "Policy Notes on Customary Land." 46 ship), the majority of respondents preferred to seek help from While land boundaries, inheritance practices, and dispute-resolution rituals are often known and well accepted within communities, variations in practice exist village and traditional leaders. However, respondents would between communities. Dispute resolution thus becomes less effective as disputes engage formal institutions (particularly police and subdistrict move beyond village borders. For more on this, see, for example, Dionisio Babo- authorities) in more serious or difficult-to-resolve property Soares, "Nahe Biti: The Philosophy and Process of Grassroots Reconciliation (and Justice) in East Timor," The Asia-Pacific Journal of Anthropology 5 no. 1 (2004): disputes. It is worth noting that subdistrict authorities, while 15­33; and Rod Nixon, "The Crisis of Governance in New Subsistence States," civil servants, have commonly been ritual leaders in Timor- Journal of Contemporary Asia 36 (2006): 75­101. 47 Nonstate authorities refers to chefes aldeia/suco, traditional leaders/adat, direct Leste, and thus it is difficult to discern whether communities negotiations, and a very limited number of "nothing" (where respondents indicated are approaching them due to their ritual or state roles. that they would not take the dispute to any forum). State/formal authorities include all types of police, courts, lawyer/paralegal, judges, the Provedor, subdistrict and district authorities, and the Land and Property Directorate. Note that this distinc- The role of the police in land-dispute resolution is perhaps tion is the authors' and does not indicate that communities themselves view these surprising; however, other research has found that the police officials as state or nonstate authorities. In the survey, respondents were asked to indicate the actor specifically, not state or nonstate actors generically. For more on often take on adjudication and mediation functions that would this, please see Timor-Leste Land Law Program, "Report on Research Findings normally rest with courts or conflict-mediation entities.45 This and Policy Recommendations for a Legal Framework for Land Dispute Mediation" could reflect respondents' past experience that local and tradi- (Burlington, VT: ARD, Inc., Feb. 2004), http://www.jsmp.minihub.org/Tradition- al%20Justice/Reports/LLP_Mediation%20Report/LLPMediationReportEnglish. tional authorities are underequipped to resolve intervillage or pdf. Accessed June 2010. ing an interhousehold boundary dispute ranged from a low Chart 4: Regional variation in Nonstate of 60 percent in rural parts of Region 1 (Baucau, Lautem, Authorities and Boundary Dispute Viqueque) to a high of 95 percent in rural areas of Region 100% 2 (Ainaro, Manatutu, Manufahi). The chart below provides 90% additional detail. 80% 70% 60% While regional differences were most pronounced for 50% 40% boundary disputes within villages, some differences also 30% emerged in responses regarding resolution of intervillage 20% boundary disputes and conflicting ownership claims. For 10% 0% example, while there was little urban-rural difference on National Urban National Rural R1U R1R R2U R2R R3U R3R R4U R4R R5U R5R resolution fora for intervillage disputes overall (55 percent and 54 percent identified nonstate authorities, respectively), there was a 29 percentage point difference between urban and rural respondents in Region 1 (with 65 percent of urban Non-State Authorities State Authorities respondents and 36 percent of rural respondents identifying Note: R1U = Region 1 Urban; R1R = Region 1 Rural, and so forth. nonstate actors).48 An urban-rural difference was also evident on responses regarding dispute-resolution fora for conflict- Conclusions ing ownership claims; nationally, 69 percent of urban respon- dents indicated that they would seek help through nonstate Solidifying property rights and resolving land and proper- actors, compared with 50 percent of rural respondents.49 ty-related disputes continue to be significant challenges in Across all dispute categories, urban respondents in Region 3, Timor-Leste. As large numbers of youth migrate to over- the region that includes Dili, were among the most likely to crowded cities, investors seek land for economic activities, report that they would bring land-related disputes to formal and the country's population continues to expand, putting authorities.50 The reasons for these findings are unclear, and strain on communities reliant on subsistence agriculture, it in some cases, such as the stronger preference for nonstate will become increasingly important to clarify and protect authorities among urban respondents in Region 1, counterin- both individual- and community-ownership claims. How- tuitive. At a regional level of disaggregation, it is difficult to ever, substantial variations both within and between regions determine precisely why these differences emerge, and fur- on issues such as land titling, gender, and land-dispute reso- ther study may be warranted. lution reveal the difficulty of establishing appropriate sys- tems that are relevant and accepted throughout the country, without resorting to microregulation of the various systems. What is Justice for the Poor? While Timorese by and large seek the assistance of local and traditional authorities in resolving land and property dis- Justice for the Poor (J4P) is a global research and putes, and consider these authorities to be fair and effective development program aimed at informing, designing and arbiters, there is considerable support for additional protec- supporting pro-poor approaches to justice reform. It is an tions from the formal legal system. Despite a weak and often approach to justice reform which: physically distant court system, respondents recognize a role Sees justice from the perspective of the poor/ for external authorities (most often the police, and occasion- marginalized Is grounded in social and cultural contexts ally subdistrict and district authorities) in helping to resolve Recognizes the importance of demand in building competing claims. The effectiveness of the government's cur- equitable justice systems rent efforts to register peri-urban and urban land and pass a Understands justice as a cross-sectoral issue transitional land law, and the extent to which these steps are Justice for the Poor in Timor-Leste is part of the AusAID- taken in cooperation with communities, will help to shape the future of land and property access, private sector investment, Justice for the Poor Initiative. This Initiative includes work and stability in Timor-Leste. in Solomon Islands, Vanuatu, Papua New Guinea, Timor- Leste and Indonesia, as well as regional thematic activities. 48 This is statistically significant at the 1 percent level (t-stat = 3.08). Contact us at j4p@worldbank.org and visit our web- 49 This is statistically significant at the 1 percent level (t-stat = 3.53). site www.worldbank.org/justiceforthepoor for further 50 Thirty-one percent (95 percent C.I. [22.9­38.2]) identified state actors for in- information. terhousehold boundary disputes, 52 percent (43.8­60.5) for intervillage boundary disputes, and 59 percent (50.9­67.4) for conflicting ownership claims. Justice for the Poor Briefing Notes provide up-to-date information on current topics, findings, and concerns of J4P's multi-country research. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank.