1. Introduction

Access to land is the basis for almost all rural livelihoods in the rural hinterland of Timor-Leste (East Timor). Despite major challenges, local customary institutions are managing access, through a range of extant relationships to land and natural resources. Since independence from Indonesia in 1999, however, the formal rules governing land access have been the subject of intense debate, community speculation, and proposed legislation. Following scholarship on rural land use and its social relations, we argue that diverse customary institutions can co-exist and work with state institutions to build collective political legitimacy and to fight “exclusion” that can result from inequitable land reforms (Hall et al. 2011, 5).

We explore the importance of customary institutions in Timor-Leste’s proposed land tenure reform legislation, and their relationship with enhanced support to diverse rural livelihoods and better infrastructure. We question if proposed privatisation and state appropriation of land, with some limited demarcation of ‘community lands’ (all part of current draft land laws), are necessary given the range of local customs and traditions available that have served the communities well for centuries. Drawing on empirical research in four villages we assess the extent to which traditional land allocation rules serve rural communities, and we argue that the meaningful recognition of custom and tradition is a prerequisite for relatively conflict-free, equitable and sustainable development for Timor-Leste’s commons.

Timor –Leste’s agrarian livelihoods require security of land access. Theorists suggest caution is warranted as agrarian change occurs, with value captured by key actors including the state, urban based elites and outside interests (Harvey 2004; McCusker et al. 2013). We conclude with a discussion of the ways in which extant local moral economies are continuously cultivated to enable access and sharing of land and resources, and we point to ways to strengthen rural livelihoods that recognize and valorize traditional land and resource management. Land privatisation is not a ‘magic bullet’ for addressing inequalities, or rural poverty (Fitzpatrick 2005). In addition to legal reform, well-targeted government extension and service delivery in rural areas is needed, building on extant local strengths, capacities and values.

2. Timorese challenges

The newly independent Timor-Leste (Figure 1) faces complex social and economic challenges that have occupied its key institutions since independence from Indonesia (Fox 2001; Hill and Saldanha 2001; Hunt 2004; Booth 2005; Philpott 2006). Timor-Leste was one of the five poorest countries in Asia before oil and gas revenues entered the national accounts. Some 41% of the population lives below the national poverty line, and adult literacy is only 50% (UNDP 2009). Timor-Leste lacks major revenue-earning opportunities aside from its offshore oil and gas revenues, which are now the major government income stream. It has a relatively large international aid sector, with Australia, Portugal, and multilateral banks as major donors.

The occupation of the country by Indonesia from 1975, following five centuries of Portuguese colonial rule and very brief independence, saw military actions and opposing guerrilla warfare by Timorese forces (CAVR 2006). Indonesia brought large numbers of civil servants, troops and, in some regions, transmigrants, as well as building new agricultural settlements. Many Timorese opposed to the occupation suffered abuses of human rights, and extensive damage to property (Nevins 2005). Approximately 200,000 Timorese starved or were killed by armed forces and militia groups, during the period of occupation and the final departure of Indonesia in 1999 (Nevins 2005; CAVR 2006; Tanter et al. 2006). United Nations peacekeeping forces stayed for five years (Curtain 2006). In 2006, they returned and stayed for another seven years following dissent in the Army, societal tensions and renewed internal conflict. In recent years, political relations with Indonesia have improved and economic ties are strong.

Most rural Timorese practice indigenous agriculture, often without mechanization and with few inputs. They have a high degree of self-sufficiency, alongside limited, although growing, production for the market. Since independence, economic forces have spurred rural to urban migration for schooling and waged employment: most notably to Dili, the capital city, and some regional towns. Many rural districts remain without modern basic utilities while Dili has high unemployment and poor infrastructure.

Since 1999, there has been constant debate over service delivery and land reform between new political parties and among major stakeholders in land and businesses, including the Catholic Church and private landholders (Pederson and Arneberg 1999; Ospina and Hohe 2002; Kehi 2005; Anderson 2008; Stead 2015).1 Infrastructure improvement has been slow (SDP 2011). In the early years of independence little of the country’s small aid budget was spent on basic rural development and addressing unemployment. Tensions around land tenure, livelihoods and human security have continued (Ho and Spoor 2006; Valters et al. 2015). To explain the importance of these concerns, we first describe customary and modern land management, before discussing rural livelihood challenges of which land access is just one component.

3. Timorese non-market relationships with land and community

The majority of Timor-Leste’s poor occupy a complex, and regionally diverse, rural social landscape. Land remains central to livelihoods, social relations and ecological and spiritual relations – these are inseparable despite the pace of agrarian change. People in Timor articulate their sense of belonging to particular places, and assert claims to land by invoking histories of ancestral origins and movement, recalled in oral histories. Sacred sites mark events in the origin history, and ceremonies held there commemorate spiritual ties and meanings . A sense of place, of social, cultural and ecological identity, and rights and obligations to land, water, forest or other local resources are still defined largely by ‘precedence’, or by relative proximity to the sacred origin place or source of ancestral migration (cf. Fitzpatrick et al. 2012; Palmer 2015).

The ultimate ritual custodianship of a territory is vested in the descendants of the people who first settled the area and had a ritual covenant with the land. Commemorative practices generate a profound sense of emplacement, and a memory of past movement. They also strengthen collective memory (McWilliam and Traube 2011). Traditional ideas about land and territorial entitlements have been renegotiated in response to the tumultuous history of colonisation and incorporation within legal and administrative structures. The significance of secular and Catholic Church institutions is also recognized, given their centuries-old presence.

Many origin narratives recognize and attempt to reconcile conflicting value spheres of spiritual and political power. At the same time they acknowledge that, as individuals with a strategic interest in producing food, people also have a concern for dividing and controlling land for productive uses. Land, therefore, has material and non-material qualities and “cannot be quantified in monetary terms because it is unthinkable to do so” (Peñaflor 2011, abstract). A modernizing discourse exists among some social actors, particularly in government, where customary tenure is now framed as less desirable because of a belief that ‘modern’ private property may act as surety or as a tradable asset. This tension has haunted debates over land since independence, as we will show.

Land allocation is traditionally based around ancestral clan houses (Tetum [T]: uma lulik), marriage relations, and ritual prohibition ceremonies (T: tara bandu).2 There are regional differences in “operational rules” (Ostrom 1990, 51–52); inter-generational transfer mechanisms for symbolic and material resources, matrilineal and patrilineal systems, socially defined rights and obligations relating to land, and in the terminology used to describe local relations (Fitzpatrick and Barnes 2010). Nonetheless, these different local cultural systems have much in common historically, as comparative ethnology of Austronesian land tenure shows (Fox 1997; Reuter 2006).

Land scarcity, regarded by some analysts as a key driver of the eventual formalisation of land rights, is becoming a pressing issue in and around towns and cities but is not widespread across the country (Fitzpatrick and Barnes 2010). Land pressures are more visible, and this was observed during the appropriation of land for conservation, potential agribusiness and industrial development projects in the late 2000s (Cullen 2012; Palmer 2015). Community disputes occur, especially where collectively owned land has been legally reclassified as alienated to the state; as individual private property that is divided and sold off; or due to forced displacement during the periods of colonial occupation.

4. Rural livelihoods and development

Access to land is so vital in Timor-Leste because it is central to socio-political relations in this agricultural society (cf. Palmer 2015). Despite the growth of Dili and other towns, as noted above, at least 75% of Timor’s population practice subsistence and semi-subsistence agriculture in rural areas, often distant from infrastructure, services and towns (UNDP 2009). Farming systems are of four main types. Rain-dependent dryland farming (crops include maize, rice, cassava, taro, sweet potato, different types of legumes); irrigated rice production; plantation cash crops such as coffee, coconut, rubber, candle nut; and lastly hunting and foraging in forests and coastal areas for a range of resources (game, fish, seaweed, honey, medicinal plants, construction materials) (Meitzner Yoder 2007, 2011). Corn and rice are the dominant staples, gleaning and hunting are important depending on the region, and coffee is the dominant international cash crop, largely grown in the western districts (Gunn 2004). Timor’s unique geography – particularly its mountainous terrain and pronounced seasonality – render a large percentage of the population food-insecure, particularly before the main harvest time in February/March. Food shortages are exacerbated by transport and communication difficulties, periodic climatic events and natural disasters, and crop failures are common (Barnett et al. 2007). During food shortages, residents rely on hunting, wild legumes, taro, cassava, yams, sago, and other leafy plants. Attention is now being paid to better local market systems and infrastructure, sustainable agricultural extension, rural employment programs, and modernization of traditional farming systems (Shepherd 2013).

Hundreds of years ago, Portuguese outposts were built for security and taxation, and some plantation agriculture was developed, primarily in the west (Shepherd 2013). There were also large land holdings, some still in the hands of Timorese mestizo families (Aditjondro 1994, 9). Plantation workers were at times heavily taxed and exploited for their labour. Under later Indonesian control, many settlements were relocated close to roads to enable better surveillance in the face of a hostile independence movement (Thu 2012).

Rural ‘development’ and support to the rural poor has an interrupted history (Shepherd 2013). The Tetum term for ‘liberation’ or ‘independence’ is ukun rasik an. This can be translated as ‘self governing’ and can be interpreted to mean that each individual in society has power to govern themselves. Ukun rasik an was boosted by the leaders of the Timorese revolutionary front (FRETILIN) in the early 1970s. Resistance leaders promoted principles of community self-reliance. Many Portuguese educated Timorese students formed brigades and fanned out across rural villages to raise political awareness and to mobilize communities for different causes (Hill 2002). Some started cooperative programs in agriculture, introducing literacy programs and prompted community health through nutrition and sanitation management (Hill 2002). Some of these initiatives were continued during the Indonesian occupation, especially from 1975 to 1983. Farming cooperatives shared labor, materials and seeds to secretly produce food for front line forces.

Self-reliance principles are less visible in today’s agricultural policies. These prioritise market-oriented farm production, often without sufficient training or guidance on the use and maintenance of new mechanized equipment and hybrid seeds (Shepherd 2013). Despite important national and local efforts, basic infrastructure including tarred roads, functional buildings, water supply and communications are still scarce. While the government and aid agencies distribute seed, machinery and other aid, our surveys in the late 2000s showed extremely low levels of livelihood support, and significant logistical and economic challenges. Currently the government infrastructure fund, through the Millennium Development Goals Suco program, is providing new houses, solar power, water, sanitation and roads. The Decentralized Development Program (PDD) and Decentralization Development Package (DDP) fund local business development and capital works (UNMIT 2011). Funding is still short at the village level, creating governance tensions (Maia et al. 2012). Meanwhile, Timorese communities are busy diversifying livelihoods to include new income sources and activities, as we describe below.

5. The contemporary land debate

The customary system of land tenure in Timor-Leste has evolved and adapted over time, demonstrating remarkable resilience, but there has been a consistent resistance by elites and foreign advisors to full recognition of communal and customary land management. Legal pluralism prevails, despite several land tenure reform efforts and draft legislation.

In the Portuguese era some land became private property, but few land titles were issued. By contrast over 44,000 were issued during the occupation (1975–1999), as efforts were made to apply Indonesian laws and norms to the territory (Pederson and Arneberg 1999; Fitzpatrick 2002; Stead 2015).3 This has resulted in a patchwork of approximately 47,000 formally titled and hence ‘privately owned’ landholdings, predominantly in the country’s urban centres. Overlapping titles and claims are still common. There have been unlawful occupations of properties in times of conflict and displacement, resulting in social tension and sometimes violence (Cryan 2012).

Expatriate advisors first arrived in Dili in 1999 to assist in developing new constitutional and legal systems. Land reform was strongly recommended to the new government, on the grounds that clarity, and law, were necessary. Some advisors were favourable to legal pluralism empowering communities, but others were more committed to rolling out stronger private property rights, or wider state ownership (Elderton 2002; Fitzpatrick 2002; Hill 2002; D’Andrea 2003; Wallace 2007).

The resulting land legislation has an elaborate history (see Fitzpatrick et al. 2012, 93 and Rede Ba Rai 2013). The government seemed keen to support elements of Western land management principles, thereby diminishing the extent of customary land (Batterbury and Longbottom 2007). From 2003 onward, a temporary arrangement saw all ‘real estate assets’ without clearly identified owners become the property of the state until such time and to the extent to which legitimate and successful claimants could be found and issued with a title (Law No. 1/2003; Fitzpatrick et al. 2012, 100). This was only one year after the government of Timor-Leste took over from the United Nations transitional regime, and was justified given the considerable governance challenges and conflicts present in the country.

The East Timor Land Law Program prepared a report in October 2005 to help guide government policies on land reform, with funding provided by The United States Agency for International Development (USAID). The report asserted that:

“…the development of an effective land administration organizational structure in East Timor will be an important component of establishing the proper institutional base for economic growth and the reduction of poverty in that country. Among other things, effective land administration provides land title guarantee and land tenure security, supports the process of land taxation, provides bank loan security, develops and guides land transactions, protects public and private land, reduces the severity and frequency of land disputes, facilitates land reform broadly, and improves infrastructure planning and development.” (ETLLP 2005, 16).

Customary land management and dispute resolution mechanisms did not feature prominently in these early statements. The discourse was that of tenure formalization, through individual title deeds for poor families, issued to enable them to cultivate, or to sell or mortgage land in order to lift themselves from poverty; a transition promoted by De Soto (2000, see also Perz et al. 2014).4 In an interview conducted with a senior member of the National Directorate for Land, Property and Cadastral Services (DNTPSC) in 2006, we learned of plans to advance a cadastral survey of the whole country, and then to issue property deeds to those with sufficient historical claim to parcels of land. The process was to be conducted centrally, with uncertain recognition of communal property and its management mechanisms. Our research began from the realisation that this was an ambitious plan with uncertain outcomes.

The agency USAID then funded a program later called the Ita Nia Rai (T: ‘our land’) pilot project from 2007 (Ita Nia Rai 2008), implemented by ARD (Associates in Rural Development), Land Equity International and other contractors. Its second phase ended in 2012. In the initial stages, cadastral surveys of prior property rights were conducted in pilot areas, along with social investigations of claims. The first phase claimed success in finding single owners for parcels of land, with 648 claims in two pilot areas, Liquica and Manatuto, with only 35 cases or 5% disputed (Ita Nia Rai 2009).5

Tim Anderson noted two threats from government desires to institute privatisation of land, and the allowance of de facto state control over land it allocates to itself or to other interests (Anderson 2008). Both could alienate rural people from land, favouring agribusiness and commercial or state uses. But USAID, rejecting this and other criticisms, argued private deeds would simply permit individuals to have their “active” land holdings recognized; a legal recognition of use claims (USAID 2008). Nonetheless “accumulation by dispossession” (Harvey 2004) is a real possibility in Timor-Leste, as Anderson suggests. Issuance of land titles does permit poorer people to trade and sell them. This permits land “accumulation” by richer individuals purchasing those titles, and dispossession of former title holders (Hall et al. 2011). The failures of the World Bank-supported Cambodian titling program is often cited to prove this point (Rede Ba Rai 2013, 20). Fitzpatrick et al. (2008) recommended very clear demarcation of state land (T: rai estado) since any parcels appropriated by government would probably encompass land believed by communities to be under their control.

In June 2009, the Minister of Justice launched the ‘Draft Law on the Special Regime for Attributing Property Rights’. The proposals came before Parliament after only two months of public consultation. After revisions, the fifth draft was issued on 10 March 2010. A land law was finally presented to the national Parliament of Timor-Leste in February 2012 and passed, but in March 2012 it was vetoed by the President of the Republic, Jose Ramos Horta, because amongst other concerns, common property and traditional tenure were insufficiently protected and recognized.6 The consensus is that the draft Law “…does not provide for explicit recognition of customary forms of tenure or authority relating to land,” this absence being blamed in part on government players expressing some hostility to customary ownership, but secondly to the state desiring authority over land earmarked for national development interests (Fitzpatrick et al. 2012, 91). Revisions to the Law were approved by Ministers in 2013, but have still not been ratified by the current President. They do include zones in which communal land will be protected. But Rede ba Rai, an NGO land network that represents the rights of the poor and those excluded from the debate, finds this protection much too weak (Rede Ba Rai 2013). Whether the state may make ‘national interest’ land claims has emerged as a key battleground between competing interests. There are plans for large scale industrial zones and land expropriation is already underway on the south coast in Suai, for a proposed oil and gas mega-project.7

This complex history of land reform raises a significant question: Do traditional land allocation rules actually lead, or not lead, to persistent vulnerabilities and poverty? Do they require change, or is the modernisation of tenure driven by other interests? This is crucial, and of great interest to scholars of common property represented by this journal (Armitage 2007). The two case studies below suggest well-developed traditional processes do deal with transfers of property and conflictual situations, for both titled and untitled land.

Our argument is based on interviews with key Timorese policymakers and NGOs in 2006, 2008 and 2009, and fieldwork in four villages at remote ends of the country’s poor road networks and many hours travel from Dili. The four detailed studies were first carried out in 2008 in field visits over several days, with livelihood questionnaires administered in the same communities later that year (c.20 individual respondents per village, n=80). We worked in Tetun, Bahasa, Fataluku, Tokodeden and English. We made further field visits in 2011–2012, and testing and confirmation of key findings has been possible during related research by team members until early 2015. The case studies are paired. In each we outline the particularities of the communities, and how land tenure decisions and livelihoods are locally manifested within the context of the Timorese commons. The reason to discuss livelihoods as well as land access is because they are mutually interdependent (see Grenfell et al. 2008; Fitzpatrick and Barnes 2010; Fitzpatrick et al. 2012; Thu 2012).

6. Land and livelihoods in two Lautem communities

In the first case, traditional land tenure management has been challenged by a territorial decision to institute a national park in which communities were already living and working. The eastern Lautem region (Figure 1) is primarily a subsistence-based economy with around 55,000 people. It has monsoonal rainfall between December and February, particularly along the drier north coast, which includes Tutuala and Mehara villages in the Tutuala sub-district. Local livelihoods in both villages are based on subsistence farming supplemented with tree plantations, hunting, forest and coastal foraging and fishing (with some sale of livestock, agricultural products, and in some cases receipt of overseas remittances) (McWilliam 2002). Food shortages are noticeable from December up until harvest time towards the end of the rainy season.

Tutuala village is at the far eastern extremity of Timor Island, and has a population of approximately 1400 (3700 in the administrative district). There are 28 Fatuluku-speaking ratu (clans or agnatic groups) in the vicinity who hold ritual authority over land, although some lack customary power. Tutuala’s proximity to the coast, and a commanding geographical location lends it customary power and importance.

Every household cultivates agricultural crops, with the only non-farm incomes being some artisanal fishing and hunting, lodgings at the Portuguese-built posada, and work carried out as a part of a small community tourism cooperative at the nearby Valu Sere beach. Members of some households have left the village to work or study in Dili and further afield, and remit limited cash and goods.

Mehara village is situated further inland, close to an upland lake, Lake Iralalaru. It is larger at around 2400 people (2368 people comprising 564 households in 2010). Formalised as a settlement during Portuguese times, the sub-communities it comprises are connected to a longer history of ratu taking up residence in the area after shifting from other nearby locations, often under colonial pressures. Mehara has better agricultural soils and lake access, where some fishing and recessional rice cultivation can be practiced and livestock grazed (Plan Timor Leste 2008). Compared to Tutuala it has a larger number of family members living overseas and returning remittances.

Both villages are located within a tropical dry forest area. This is one of the finest contiguous blocks of dense lowland tropical and monsoon forest on the island (McWilliam 2007b). The area has high conservation status, including the heavily forested Paichao Mountain Range rising to 925m. The forest was traditionally home to a number of shifting populations, (although some defensible stone dwellings, now abandoned, signify permanent settlement). During the Portuguese era these communities were encouraged to move closer to the newly colonial administrative centres. Some, however, retreated deeper into the forest to evade forced labour and taxation (Pannell 2011). During the Indonesian occupation from 1975, the region was a resistance stronghold and fighters used their knowledge to live off the land (Collins et al. 2007; Pannell 2011). Further centralization occurred for control and surveillance. When the Indonesian military departed in 1999, houses, infrastructure, livestock, property and lives were lost (Tanter et al. 2006).8 Many people never returned to their ancestral lands. Finally, the National Park was decreed by the Minister of Agriculture, Forestry and Fishery (MAFP) in 2007 and declared in 2008, named after the national resistance leader Nino Conis Santana. Some fields now lie inside the park (O’Connor et al. 2013).

6.1. Land access and control

The area encompassing Tutuala and Mehara was, according to local clan origins stories, settled from the coast with some clan groups moving progressively inland over time. The location of Tutuala sacred houses (Fatuluku [F]: lé téin) and settlements altered due to water shortage, as well as later forced relocations (Pannell 2006). Tutuala’s origin narratives identify a ratu said to emerge from the ground, and three ratu arriving by boat (and see O’Connor et al. 2013, 212). While they are said to come from nearby islands, the names of these places are closely guarded clan secrets. Other ratu followed (also by boat), and land and reef areas were divided between the older and newer arrivals according to sacred oath. This process also set in train an ongoing division of clans into older sibling – younger sibling relationships (F: kaka – noko). From the outset primary leadership and decision making relating to land and resources was by the mua cau (- [F]; the keeper of the word; T: lia nain) advising the liurai (- [T]; local king; F: malai) of the various clans. Conflicts over land ownership and use were referred to the mua cau, with some involvement from political leaders. Today land is still obtained by sacred oath with reference to ancestral markings, graves and past settlement or garden sites (see also Fitzpatrick et al. 2012, 155). An assessment is made of the historical and extant social relationships between land owners and users and land uses. The oath gives usufruct rights only, although these can often be transferred intergenerationally. The critical issue is that permission is sought from the malai or the land owners (F: mua’a hocava).

Land requests were based on expressed need for access, including land obtained by in-marriage (rather than by evoking principles of ‘ownership’). Interviewees in Tutuala said that there is a need to document these local land allocation decisions better, in order that they have a basis on which to negotiate with the government over proposed land reforms. People openly acknowledge that there are land conflicts from time to time. In such cases the mua cau have ceremonies to bring the parties and the relevant sacred houses together to seek resolution. Some of the larger ceremonies were said by respondents to be so spiritually powerful that illness, deaths and other disasters can result, so they are avoided in all but the most intractable disputes.9

There is now active involvement of local political leadership in the form of the chefe de suco (village head) or chefe de aldeia (sub village head). The chefe de suco of Tutuala told us the central government asked to see evidence of their customary laws in the late 2000s. Present and proposed land use changes such as the National Park and a previously proposed hydroelectricity project mean that land and sea tenure security is at the forefront of peoples’ minds, and a source of tension (McWilliam 2007b; Palmer and Carvalho 2008; Pannell 2011; Carvalho and Palmer 2012; Cullen 2012). Government compensation to land owning ratu and the six communities affected by the National Park (Mehara, Tutuala, Balrol, Com, Maupitine and Loro 1) is unlikely. There is also a boundary dispute across two different sub districts.

Despite considerable debate at the national level, there was still no formal land titling in the Lautem region and at the time of fieldwork, and the DNTPSC had not been involved in adjudicating land conflicts (Cullen 2012). The sub-district administrator of Tutuala had only been asked to intervene in one case.10 Local people alluded to the need for some sort of government intervention in such cases but did not specify exactly what form this intervention should take, other than it should recognise customary practice for land and resource allocation, particularly their own processes for resolving conflicts.

6.2. Livelihoods

Agro-climatic conditions vary between the coast and the high plateau on which Mehara sits. Mehara lake access is governed by defined ratu owners who permit grazing as the lake level recedes. Aldeia residents practice communal and individual rice production and there is government encouragement to grow crops, supported by projects.

All households in Tutuala cultivate maize, with cassava, yams and tree crops including numerous fruits, particularly bananas and papaya. Fertilizers and compost are scarce. Commercial crops are limited but include sales of copra (first cultivated commercially by the Portuguese), and livestock to the urban market (McCord 2011). Small numbers of goats and pigs are reared by most households, with some cattle and buffalo. All livestock have customary importance in marriage exchange relations between clan houses. The sale of firewood is common along the road in the western part of the village, and distilled palm spirits are sold.

Food shortages are seasonal and uncertain. Government surveys report food security suffers from occasional flooding, mudslides, and locust outbreaks (Simpson and Sun 2002). In 2007 for example there was a locust outbreak and erratic early season rainfall, reducing harvests by up to 30%. In 2008, just before we carried out surveys, the maize harvest failed due to a drought spell. In early 2011 villagers were again exposed to increased food insecurity due to crop damage from continual rains throughout 2010. Both villages suffer irregular and low rainfall and are reliant on springs far from the village centre. A lack of piped water, now partially redressed, compounds the difficulties facing farmers and householders.

Market sales are reported as being small. For the people of Tutuala, trading takes place as far away as Los Palos, which is usually too distant and expensive for regular trips. Very small amounts of candlenut, lucrative for its oil, are cultivated in Mehara. Coconut oil is sometimes used domestically for cooking, or sold, particularly in Mehara.

There are few other income opportunities in the region. Plan Timor Leste (2008) reported that half of the households in Lautem rely on a combination of crop sales, livestock sales and remittances for subsistence. Some 28% just sell crops, 23% just sell livestock and only 4.5% rely on incoming cash remittances (p. 28). This is a different pattern to many parts of the developing world where non-farm work is almost universal, and diversification of household economy is seen as ‘adaptive’ to uncertainty and changing trends (Batterbury 2001). Diversification has been slow, perhaps because of the region’s unique and troubled history, high transport costs (domestic and international) and present difficulty trading and working with Timor’s nearest neighbor, Indonesia. Despite the presence of Indonesian islands closer than Dili, trade with them is limited (McWilliam 2007a).

In Tutuala, our surveys reported sales of livestock and chickens during times of hardship. A few mentioned sales of sandalwood (now illegal), one mentioned selling turtle shell (illegal) and one woman opened a seasonal kiosk. However in general the range of asset sales was extremely limited. In Mehara, two respondents worked as carpenters concurrent with farming, and one ran a local NGO. Many families often shift to hunting (now illegal) to supplement income and protein deficiency as well as fishing.

Across the eastern region as a whole, Plan’s surveys showed 24.5% are female headed households (Plan Timor Leste 2008, 5). Women encounter greater challenges generating sufficient income outside the home given their major household responsibilities. They are also often involved in animal husbandry and water collection, which is a major task, particularly in Tutuala where steep climbs are involved. Weaving is a subsidiary activity but is not lucrative.

About one fifth of households in our survey had members who moved away for education. Almost all children receive formal education, studying as far away as Los Palos and Dili. In Mehara, there were relatives of the respondents in Cuba, on Java, in West Timor, and several work in Portugal, Ireland, and the UK. In Tutuala, there were migrants in all these countries except Cuba.

In recent years the Haburas Foundation, Timor-Leste’s leading environmental NGO, began engaging local people in a number of alternative community focused economic development projects including the Valu Sere community tourism enterprise. This small venture hosts and caters for visitors in simple beach side bungalows overlooking Jaco Island. Villagers themselves invested money and labour to improve a road from the village to the beach. Any profits from the venture are shared between participating clan groups. Not only does it provide an appropriate way for some Tutuala residents to gain revenue, but it strengthens social capital for those households and participating ratu. It has more recently been joined by a second private enterprise nearby.

The legality of Valu Sere has been disputed by the Forestry Department, since it lies just inside the National Park. From a community standpoint the development has been constructed on community land. For this reason they see it as an appropriate community oriented protected area under IUCN category IV principles, but this remains unresolved. Local communities said they would welcome the opportunity to mesh together customary land ownership and resource use principles with a guiding plan for National Park management, meeting the twin aims of local development and conservation. As O’Connor et al. (2013) write, promoting biodiversity and conservation values in the park at the expense of local peoples’ involvement is counter-intuitive, since the park has several thousand years of occupation, and biodiversity has only been preserved to its current extent because of population relocations in the Portuguese and Indonesian eras. As this example demonstrates, land tenure and livelihoods are mutually reinforcing and both are affected by current government policies. It also shows that alienation of land by the state, a form of “exclusion” (Hall et al. 2011), affects rural livelihoods, unless other options are proposed or if traditional activities can be maintained.

7. Land and livelihoods in Leotela and Guiçu districts

The western case studies are located in the district of Liquica in the villages of Guico in Maubara sub-district, and Leotela (Faulara/Lepa) in the Liquica sub-district (Figure 1). The landscape transitions sharply between steep slopes, with high erosion risks, and floodplains. The two major rivers, Laueli (branches Riti, Lowes, Mantala) and Lilaho, are subject to seasonal flooding and present a major impediment to transport around the communities, especially since there was very little road maintenance after 1999. Both lowland villages have people with ancestral lands in the mountains to their north, but who currently live on the flood plain. These areas have experienced a greater exposure to population movement and resettlement and a greater abundance of western-style land ownership can be observed. Tokodeden is the dominant local language spoken in the region.

In the Portuguese era, plantation crops were widely introduced in Liquica. These included coffee, rubber, cocoa and vanilla, particularly in the highlands. There was also sandalwood exploitation and logging. Landholdings were created for a local political class whose fathers and grandfathers were political exiles from Portugal. The emerging mestizo elite were given land by the Portuguese administration on the condition they refrained from involving themselves in anti-colonial activities. In the Indonesian era from 1975, transmigrants from within Timor and the archipelago were moved into the region’s lowland areas. This was accompanied by a general population shift from the highlands to the lowlands, and a shift from highland coffee to lowland rice production. Nevertheless, all the sacred origin houses (T: uma lulik) of the four traditional owner clans – Laueli, Dilaho, Selelebu and Lilebu – remain in the highlands.

In the highlands, much of the coffee land continues to be owned by the SAPT (Sociedade Agricola, Patria e Trabalho) —a part-owned Portuguese company of great longevity. Along with the other large land holdings owned by well known Timorese families (Carrasacalo, approx 100 ha; Oliveria, approx 100 ha), there are also many small coffee holdings.11 In 2008 small-scale coffee producers earned only US$0.25/kg when selling to the Cooperativa Café Timor (CCT) and on to VCBA, a US company, although this price has fluctuated.

The SAPT’s production was halted by the Indonesian invasion and the region was declared a military operational zone in the late 1970s. With intense fighting with the Timorese resistance, the area was depopulated and the coffee left untended.12 Indigenous shade trees (ai samutuku) were cut down, and vanilla trees planted. These were less hardy and soon died. The plantations were subsequently taken over by PT Denok Hernandes Indonesia, a company set up by Indonesian military officers (Aditjondro 1994, 17; CAVR 2006; ICG 2010). From 1999, local people began to re-occupy plantation land to which they had ancestral claims. Leotela experienced an influx of substantial numbers of displaced peoples in the 1999 conflict but, during the same period, Guico, an Indonesian-built roadside town, was virtually razed to the ground by the departing Indonesian military and militia, and later rebuilt.13 The land itself and plantations are in a degraded state, affected by severe gullying and erosion. Farming involves burning off and planting subsistence crops such as cassava and maize.

In recent years the once small, lowland village (suco) of Leotela has become a larger and more permanent settlement in the lowland region of Faulara. The suco contains 5 sub-village units or aldeia. Informants noted that the lowland region had been part of their land from ancestral times but had been forested until the 1990s, when villagers began to move to the lowlands to start cultivating irrigated rice (harvested twice a year), maize and fruits.

In addition, transmigrants arrived in 1994. In Leotela these were predominantly Timorese. Around 100 plots were allocated to as many households, but many were vacated in 1999 and then re-occupied by local people. In Leotela each Timorese transmigrant household was given 1 ha to farm, or less in some cases. In Guico where transmigrants were predominantly Balinese, new arrivals were allocated 2 ha per household.

7.1. Land access and control

In Leotela there are four house clans (Tokodeden: sua lulik): Laueli, who have a special status as custodians of the water, irrigation and field opening/harvest rituals; and Dilaho, Selelebu, and Lilebu, who have ritual responsibilities. According to custom (T: lisan) this reflects a division of authority between the Uma Mutin (white house clan) and the Uma Metan (black house clan). The ‘keeper of the word’ or ritual leader (T: lia nain, Tokodeden: rebu) comes from the latter. The local chefe de suco is usually drawn from the Laulei house clan. The uma lisan and their branches are connected through a range of social relationships, predominantly configured through inter-marriage or sibling relationships. Land is normally passed down the male line (except in matrilocal marriages), but if a family member ‘asks’ to use vacant land it is usually granted for free, with no firm obligations attached other than to participate in agricultural rituals. In cases of intermarriage with outsiders a newcomer husband (referred to as a ‘leaf’) is allocated land to use, but the ownership remains with the wife’s local land owning clan (referred to as the ‘root’). When asked why they would give away land, informants stressed that “people need to live,” explaining that a refusal of land to the needy would be immoral on principle. This moral sympathy and solidarity goes against the neoliberal spirit of privatised ownership.

Land conflict appeared remarkably absent in the lowlands area. Even remaining transmigrants generally respect the rights and responsibilities of the customary land owning groups and their privileged role in maintaining the ritual order. There has been some conflict between transmigrants themselves, possibly on account of their rather meager land allocation.

Boundary markers made of stones (T: fatu kladik) are sometimes used, both between plots and between clan territories, but informants also stated: “people simply know what belongs to them.” However, beyond the issue of demarcating individual plots, the major way in which the overall authority of customary owners is maintained is through ongoing observation and community participation in annual agricultural rituals, as well as the periodic ritual renewal of customary resource prohibition markers or tara bandu. Transmigrant households are active participants in these rituals and usually also pay small tributes in kind to the ancestors of the landowning groups at harvest time. They also consult local elders to resolve disputes relating to land, even between themselves.

When conflicts do arise, the nahe biti (- T; Tokodeden: lo’e biti) ritual of mediation is used to settle a dispute (lit. ‘unrolling the mat’) (Soares 2001). The ancestors are invited as witnesses to the dispute resolution meeting, meaning that if someone speaks a falsehood they risk spiritual retribution, through illness, accidents or death. The complainant must provide food, tobacco and betel nut in woven offering baskets. Respectful conduct must be observed. Allocating fault is called tadan hene we in Tokodeden – ‘where lies the weight’ – that is, the heavier fault, but both parties in a conflict are assumed to be somewhat responsible because allowing a difference to become a conflict is seen as upsetting social, ecological and spiritual harmony. The aim of dispute resolution is not only to establish right and wrong but to restore balance between the opposite camps (known as loho mane loho feto, balancing male and female). The party at fault must pay for another and bigger feast of reconciliation called lo’e biti-hi’it biti.14

Respondents differed in their acceptance of the idea of government-issued land certificates. Some said they would not want a land formalization process to elevate usufruct rights to ownership, but more recent transmigrants said they would welcome this. Significant conflicts from such intervention were noted, and concern about the levying of a government land tax which could, they thought, accompany any land certification process. According to local community members the government is welcome to negotiate with the community for the beneficial use of currently vacant land, but they stressed it had local owners or custodians.

In Guico most village residents we asked indicated that they inherited their land from their ancestors, and the village itself traces its origins to the arrival of the Portuguese, with rice cultivation established much later by the grandparents of the present generation. Customary law and customary resource prohibition operates in similar ways to Leotela. Our interviews revealed some Portuguese formal land titling, and then some acquisition of land by the state, with compensation. After most Balinese transmigrants left the irrigated lowland plots, their land reverted to the government. The DNTPSC later surveyed for a Timorese agri-business firm that was interested in fruit production. Few local residents had seen or understood this survey and there was a generalized concern in both villages that the state was attempting to appropriate customary lands under the banner of ‘vacant’ or state land. The rights of longer-term residents are generally accepted as taking precedence over those of newer arrivals. As in Leotela, selling land is rare – rather, it is held for children and relatives.

7.2. Livelihoods

Both communities are adversely affected by poor roads, infrastructure, a lack of agricultural services as well as localized flooding and drought. Rice and maize dominate in the more productive and more reliable lowland farming system. Coffee production in the degraded uplands still yields poor prices from CCT and VCBA, and meager returns only help to meet household and education expenses. Services are usually accessed by a long walk. The great majority of people feel that they were better off economically during the Indonesian era, and in both villages there was a real bitterness expressed about the government and its lackluster performance. People felt that they have been forgotten, and the only real benefits from independence have been the re-established school system, the end to violence, and restoration of personal freedoms. There are several local primary schools, and senior high schools further away in Liquica and Maliana.

Guico families said they were 50% self-sufficient on average. Now that human security threats have diminished, the main risk to food self sufficiency is flooding, or occasional drought. Despite the small size of irrigated plots, some households are farming up to 9 ha of lowland rice. A tractor or buffalo is usually required to seed and weed at this scale. Maize is produced by all households, as are drought-resistant cassava and sweet potato. Fruits grow well in the moist lowland conditions, and included banana, jackfruit, coconuts, papaya and oranges. One farmer interviewed utilized reciprocal or hired labour to cultivate areas untenable by himself and his wife across their 5 ha of land. Reimbursement occurred through partial offerings of the harvest. Occasionally food shortage necessitates wild food collection – wild taro, beans, yams, cassava and sago. Animals may also be sold in cases of hardship. Only one household head, who possesses a Portuguese pension, said his Guico household does not go hungry.

It is important to note that in this part of Timor-Leste, there is social and economic interchange with Indonesian West Timor, specifically Atambua. Extended family live across the border and there are economic possibilities there. The most distant relatives are in Dili, but one family in Guico has relatives who were able to obtain Portuguese passports and resided in Ireland, which in 2008 was a common destination for jobs in poultry farms and factories. Nonetheless, few individuals were said to be wealthy. This is exemplified by one modest household that made just $25 from crop sales in 2008. Maliana and Maubara markets are used, both a distance away, and one farmer took his crops to Dili. Vegetables and fruit are sold in the local market, for a few cents each. Animal ownership was less prominent than in the eastern case studies. Nonetheless animals provide a ‘bank’ of saleable assets, with many households having 10–20 pigs and cattle. One household had 5 horses and 24 cattle, plus a large rice harvest and some coffee.

Some household credit is available from micro-finance organizations, but repayment terms are punitive. In Guico, the international NGO Caritas is active, and they assist with farm clearance, vegetable production and they operate a small plantation. Tuba Rai Metin (standing together firmly) works as a cross-household organization to repair irrigation and carry out agricultural tasks. We were informed that, as in the eastern cases, sacred houses should work together with the government administration to better local livelihoods. In sum, this partially commoditized agrarian region suffers from its remoteness and inadequate market linkages and infrastructure. Land conflicts are not the major driver of rural poverty.

8. The impulse to formalize, the livelihoods challenge, and ethical decision making

“Customary land tenure systems in Timor-Leste are not a significant constraint on agricultural productivity. … Outside areas of chronic conflict, there is little evidence of under-investment in land due to perceptions of tenure insecurity.” (Fitzpatrick et al. 2008, 5).

Our assessment is that Fitzpatrick et al. are correct. There has been further study on this question (Cryan 2012; Fitzpatrick et al. 2012; Thu 2012). In Cryan (2012) over 1200 rural people were surveyed in an NGO study of land access, and their main fears for the future were found to be about state appropriation of land, unclear status of land parcels, and loss of community access (Cryan, 24). Only 7% of participants identi?ed the problem of ‘land disputes within the family’, which had been one of the main problems identified by the (largely urban focused) Ita Nia Rai program. Respondents wanted the government to formalize its recognition and support for customary land owning and resource management practices (see also Rede Ba Rai 2013). We found drought, prompted by El Niño events, and flood, by La Nina, were crucial natural hazards. Given good rains, we found people could feed themselves with reasonable workloads and occasional use of remittances and cash, leaving time for other activities. But as Timor’s situation changes rapidly, with modernization and inflation, money is increasingly needed. Schooling was universally cited as a high priority for households.

The term ‘ita nia rai’ reflects traditional communal land tenure (‘our land’ in an inclusive sense, not as exclusive and meaning ‘our land vs their land’). But we contend that the Ita Nia Rai project was part of a program to privatise some land, encourage a broader community acceptance of the inevitability of land titling, and facilitate state control of other areas, which has emerged as a strong goal of government (Rede Ba Rai 2013). Ita Nia Rai staff, and their counterparts in the Department of Land and Property surveying rural lands, had little trouble with demarcating large areas of so-called ‘vacant land’ (actually used for swidden farming and livestock grazing) as ‘state-owned land’ for future development purposes (Anderson 2008; Fitzpatrick et al. 2012). This also included sacred land (rai lulik) which often is traditionally protected by tara bandu rituals and other ceremonies and norms that do not allow, for example, trees to be cut either by machetes, axes or saws (Miyazawa 2013). Context-specific agreement-making will be very important if there are pressures to utilize ‘vacant’ land for commercial purposes.

Thus far the consequence of the material intervention of the Ita Nia Rai process in urban areas and the discursive intervention of the draft property laws across the country since then, is the emergence of a widespread perception of rapidly increasing land monetization. People are concerned about the power of the state to ‘purchase’ long term leases or expropriate lands for development (Rede Ba Rai 2013). Stead (2015) recounts the example of Aitarak Laran in Dili, a neighborhood razed to make way for a Library and Cultural Centre. In the twinned context of ‘the national interest’ and of weak and unclear processes of community involvement, understanding, consultations, and capacity to give full and informed consent, government sanctioned interventions have established a powerfully discursive set of practices, with new tensions, which overlays existing relational processes. In separate fieldwork carried out by one of the authors in rural and urban Baucau, the result of these faltering material and discursive interventions “...has been disquiet, simmering anxiety, and even violence between local residents, neighbours and families. Sub-village heads say they are now called to intervene and mediate in an unprecedented number of local property disputes” (Palmer 2015 in press).

While the state wished to recognise some customary ownership in the 2012 draft law, how this might be achieved is still unclear. There are now provisions to designate some localities as ‘community’ land, but under government regulation and ‘protection’. Resurgent customary resource management practices such as tara bandu are important for communities continuing to exercise de facto governance and ownership rights over areas of ‘social tenure’ (Wallace 2007). Our findings suggest a fruitful approach to recognizing a ‘legal space’ for custom is to acknowledge, in resource management regulations, the role of broadly interpreted customary governance mechanisms in the regulation and sharing of land and resources. Fitzpatrick et al. (2008) also write

[w]hile the law should provide legal space for custom in rural areas, it is not necessary to define either the nature and rules of “customary law”, or to identify specific types of customary authority. There are few benefits and too much diversity across Timorese customary systems to justify this approach (2008, 8).15

Following the views of former UNTAET Property Rights Adviser, Warren Wright (2010), it seems appropriate to enact legislative processes (i.e. Constitutional Rules; see Ostrom 1990) that give recognition to locally relevant institutions of resource governance, whilst refraining from the specific definition and ‘capture’ of such practices or alienation from communal lands. In this way extant land management institutions can become tools for land regulation, also enabling a formal recognition of the fact that the notion of ‘community’ goes well beyond simply defining a customary land owning group or their territory (cf. Jackson and Palmer 2012). Furthermore it must be recognized that each area will have its own issues and specific needs in relation to an appropriate institutional response.

Such creative local governance responses would seem the minimum requirement for assisting sustainable, equitable and less-conflictual approaches to rural development, providing that full and informed consultation is not only mandatory, but that the interests of the people have precedence over the interests of the state and/or the corporations. We argue, however, that there is an urgent need to recognise the diverse economies in rural localities, including the intricate and multiple customary exchange relationships and labour sharing, as well as the void created by the loss of the rural market economy at the end of the Indonesian era.

Local people in both case study sites rely on the range of ethical relationships they have with each other that guide local decision-making and conflict resolution over land (Soares 1999). Theirs is a moral economy, not just an economy (Palmer 2010, 2015; Peñaflor 2011). Our fieldwork showed that they expect the state to recognize and exercise its ethical obligation to care for the people, in this case through providing access to opportunities, services, and security. The beneficial role of government is in assisting local systems to manage difficult issues – sometimes traditions do need to be overruled, as when there is a lack of local agreement or a serious conflict (cf. Rede ba Rai 2013). But it is local diverse practices of economy and polity that need support. Some community rights supporters have suggested creating a national conselho de lianian (council of elders) to negotiate fair outcomes in intractable land and property conflicts.

To do this the state needs to recognize and make visible the range of complex processes of change, interdependence and ethical decision making practices that are relevant to creating viable rural livelihoods, including local governance, inter-group exchange arrangements and historical relationships (Palmer 2010, 2015; Jackson and Palmer 2012). Moral economies are not intrinsic to traditional economies any more than they are to modern ones. They require continuous mindfulness based on an appropriate social and symbolic process of reflection and negotiation (Palmer and Carvalho 2008; Jacob and Le Meur 2010). If the customary land tenure systems in Timor-Leste have managed to create viable moral economies, as we have argued, then the state – if it is serious about improving people’s lives and livelihoods – can afford to do no less.

9. Conclusion

We have focused on two issues facing rural development in Timor-Leste: 1) poverty, food production and the actual conditions of rural livelihoods systems, and 2) whether the ‘magic bullet’ of land tenure reform to provide private titles and state access to strategic land, at the expense of communal property, will provide equitable and workable solutions to rural poverty and lack of security.

The specific land debate in Timor-Leste is, as we noted, unique due to a turbulent history of settlement and colonisation. But it has classic elements: it divides between those who wish to see customary practices protected fully in law, and those who wish to expunge customary rights to be largely replaced by ‘ownership’ of different forms, including state control and private property (Sikor and Lund 2009; McCusker et al. 2013; Agrawal 2014). But rather than waiting for law, local people are already exercising a range of socially accepted norms, rights and obligations traditionally practised by different communities for centuries.

The land policy debate sidesteps, to a great extent, what actually works – on the ground – for the maintenance of adequate livelihoods (Batterbury 2001; Unruh 2004). From our own fieldwork, and other such surveys, it appears the question facing policymakers is how to support local diverse livelihood systems, for food sufficiency, and sovereignty, redress for endemic rural poverty, and provision of far better income and services, using national revenues derived from oil and gas. Timor-Leste has ongoing development programs to tackle infrastructure and service deficiencies but these are working very slowly. This has led to dissatisfaction. Livelihood diversification opportunities are weak compared to other parts of Southeast Asia and beyond, and have even retreated since independence in 1999.

Sharing of land and resources will be part of this. Precedence and other connections developed over time and through inter-generational familial relationships have endured:

For all the disruption and reconfiguring of residential arrangements that have occurred in Timor-Leste, the social institutions of customary authority continue to provide a locally legitimate basis for organising emplaced social relations and asserting seniority and authority over defined jurisdictions (McWilliam and Traube 2011, 15).

Government and community expectations of outcomes from formal land titling and have been at odds since the mid 2000s. In the absence of a rural land market, artificially creating one through issuance of individual property deeds is hardly a solution, especially if Lautem communities regard placing monetary value on land as “unthinkable” (Peñaflor 2011) and evidence of rural land conflict is far less evident than in towns and cities. The worldwide experience with tenure formalization is not positive, and ushers in the potential for ‘land grabs’ by foreign and corporate concerns (Anderson 2008; Sikor and Lund 2009). Unfortunately, Timor-Leste is not immune from their interest.

After all, across the developing world “it is …important that social actors gain and maintain access to resources in many ways that do not amount to property” (Sikor and Lund 2009, 4). In Timor-Leste, it is access to land and the socio-political relationships this underpins, that counts for more than its conversion into tradable property (Hall et al. 2011; Thu 2012). Culturally specific ideals of fair and equitable access to land have been the basis of customary laws and norms for centuries. Limited land titling over the centuries has not detracted from this reality in the rural hinterland where the majority of Timorese base their livelihoods.