Regenerating Woodlands: Tanzania's HASHI Project.

UNTIL RECENTLY, THE SHINYANGA REGION JUST SOUTH OF LAKE VICTORIA WAS nick-named the Desert of Tanzania. Its once-abundant woodland had been stripped away over decades, first to eradicate the disease-carrying tsetse fly, then to create cropland and make space for a growing population (Monela et al. 2004:14). Now the acacia and miombo trees are returning, courtesy of the HASHI project, a major restoration effort based on the traditional practice of restoring vegetation in protected enclosures or ngitili.

The region-wide HASHI project, whose success was recognized by the UN Development Programme with an Equator Initiative prize in 2002, is run and mainly funded by the Tanzanian government. But its striking success stems from the rich ecological knowledge and strong traditional institutions of the agro-pastoralist Sukuma people who live in the region.

By 2004, 18 years into the project, at least 350,000 hectares of ngitili (the Sukuma term for enclosures) had been restored or created in 833 villages, encompassing a population of 2.8 million (Barrow and Mlenge 2004:1; Barrow 2005b). Benefits of the restoration include higher household incomes, better diets, and greater livelihood security for families in the region. Nature has benefited too, with a big increase in tree, shrub, grass, and herb varieties, as well as bird and mammal species (Monela et al 2004:3-4). Table 1 summarizes these wide-ranging benefits. It is drawn from an in-depth study of HASHI’s impacts on local livelihoods commissioned by the Tanzanian government and the World Conservation Union (IUCN).

People, Trees, and Livelihoods: A Short History of the HASHI Project

Shinyanga is one of Tanzania’s poorest regions, its low hills and plains characterized by long dry summers with only 700 mm of rainfall a year on average. As its woods were cleared from the 1920s onward, land and soil became over-used and degraded, causing a sharp decline in the natural goods on which the Sukuma people had depended for centuries.Women spent more time collecting formerly plentiful fuel wood; grasses to feed livestock became scarcer, as did traditionally harvested wild fruit and medicinal plants.

The region’s ecological problems were compounded by a booming human population and by the Sukuma’s extensive land-use needs. Nine in ten of Shinyanga’s households live by small-scale farming, with families dependent on cropland and livestock pasture for both subsistence farming and cash crops such as cotton, tobacco, and rice (Monela et al. 2004:21-22). Since cattle are highly valued as a liquid asset, many households also kept livestock herds too large for their land to sustain, and burning of woodland to create pasture was common practice.

By the 1970s Shinyanga was under severe ecological strain, its people feeling the consequences in the form of falling incomes and lost livelihoods (Monela et al. 2004:12-13). Early attempts at reforestation launched by Tanzania’s government, the World Bank, and other agencies largely failed to stem the loss of indigenous woodland and its impact on communities. Top-down, bureaucratic management of projects meant that villagers had little involvement or stake in the success of these efforts. During the 1970s, the socialist government of President Julius Nyerere also adopted laws that increased communal ownership of rural land and encouraged people to live in discrete villages where services could be better provided—a process called “villagization.” Individual ngitili enclosures, which many villagers had carefully sustained for food, fodder, fuelwood, and medicines, were no longer encouraged. Indeed, many ngitili were destroyed during the period, as the villagization process undermined traditional institutions and practices (Monela et al. 2004:102).

In 1986, Tanzania’s government shifted tactics dramatically and launched the peoplecentered, community-based Shinyanga Soil Conservation Programme, known simply as HASHI (from the Swahili “Hifadhi Ardhi Shinyanga”). The impetus came from President Nyerere himself, who declared Shinyanga the “Desert of Tanzania” after touring the region. By 1987, HASHI was operational and by 1989 it had attracted additional, long-term support from the Norwegian Development Assistance Agency.

The Revival of Ngitili

The project’s innovative efforts to improve rural livelihoods are based on reviving “ngitili,” an indigenous natural resource management system (Barrow and Mlenge 2004:1). Traditionally, ngitili were used to provide forage for livestock— especially oxen—at the end of the dry season when villagers plough their land. Vegetation and trees are nurtured on fallow lands during the wet season so that livestock fodder supplies are available for dry months.

There are two types of ngitili: enclosures owned by individuals or families, and communal enclosures owned and managed in common. Both were originally developed by the Sukuma in response to acute animal feed shortages caused by droughts, the loss of grazing land to crops, and declining land productivity (Barrow and Mlenge 2003:6).

The HASHI project’s approach to ngitili revival was to work with local people, first to identify areas requiring urgent land restoration, and then to restore them according to customary practice. Field officers, employed by the Division of Forestry and Beekeeping in the Ministry of Natural Resources and Tourism, worked closely with both district government staff and village government authorities—the lowest accountable bodies in Tanzania’s government (Barrow 2005b).

Technical guidance and information was also provided by the Nairobi-based International Center for Research in Agro- Forestry (ICRAF), which had researched ngitili restoration. ICRAF studies documented appropriate vegetation and management practices, and noted the important role played by traditional knowledge and local institutions in successful land management (Barrow 2005e).

In many villages, HASHI field officers used residual natural seed and root stock to restore ngitili enclosures. In others, active tree planting (first of exotic species, later of the indigenous tree species preferred by local people) was carried out, especially around homesteads. Some of the restored ngitili dated back to pre-villagization days. Others were newly created by farmers and villages. In addition to restoring ngitili, villagers were encouraged to plant trees around homesteads (particularly fruit and shade trees), field boundaries, and farm perimeters. This helped improve soil fertility and provide firewood, and had the side benefit of helping farmers to stake out and formalize their land rights within villages (Barrow 2005c).

A range of tools were used to educate and empower villagers. These included video, theater, newsletters, and workshops to demonstrate firsthand the links between soil conservation, forest restoration, and livelihood security. Participatory rural appraisal methods helped villagers to identify local natural resource problems and agree on solutions (Kaale et al. 2003:13-14). Farmers and villagers received training in how to get the most out of their ngitili. For example, they learned which indigenous species were best suited to enrich farms soils or create dense boundary plantings.

Armed with this powerful combination of traditional and scientific knowledge, villages across Shinyanga gradually revitalized the institution of ngitili and broadened its use from simple soil and fodder conservation to production of a wide range of woodland goods and services. Products such as timber, fodder, fuelwood, medicinal herbs, wild fruits, honey, and edible insects enhanced livelihoods and provided a vital safety net during dry seasons and droughts (Barrow and Mlenge 2003:1).

In the early years, restoration efforts proceeded gradually as cautious farmers and communities assessed the benefits and rights which ngitili regeneration produced. By the early 1990s, with the project’s effectiveness beyond doubt, restoration efforts spread rapidly through the region. In 1986, about 600 hectares of documented ngitili enclosures existed in Shinyanga. A survey of 172 sample villages in the late 1990s revealed 18,607 ngitili (284 communal, the rest owned by households) covering roughly 78,122 hectares (Kaale et al. 2003:8, Barrow and Mlenge 2004:1). Extrapolating from these figures, project managers estimate that more than 350,000 hectares of land in Shinyanga were in use as ngitili, with nine in ten inhabitants of Shinyanga’s 833 villages enjoying access to ngitili goods and services (Barrow 2005b).

Wendelen Mlenge, longtime manager of the HASHI project (recently renamed the Natural Forest Resources and Agroforestry Center) has closely observed its success. The enthusiasm and commitment with which communities have embraced ngitili restoration demonstrates, she says, how “a traditional natural resource management system can [be adapted to] meet contemporary needs” (Barrow and Mlenge 2003:10).

Making It Work: Traditional and Local Institutions

HASHI’s empowering approach was unusual among 1980s rural development programs, but critical to its success. Promoting ngitili as the vehicle for land restoration increased local people’s ownership over natural resources and their capacity and will to manage them. Likewise, allowing traditional Sukuma institutions and village governments to oversee restoration efforts helped to ensure their region-wide success. While elected village governments officially manage communal ngitili, and also decide disputes regarding individually owned ngitili, in practice traditional institutions have played an equally important role in most villages (Kaale et al. 2003:14-16; Monela et al. 2004:98).

For example, while each village sets its own rules on ngitili restoration and management, most use traditional community guards known as Sungusungu and community assemblies known as Dagashida to enforce them. The Dagashida is led by the Council of Elders which decides what sanctions to impose on individuals caught breaking ngitili management rules, for example by grazing livestock on land set aside for regeneration (Monela et al. 2004:98-99).

HASHI field officers have worked to build the capacity and effectiveness of both official and traditional governance institutions. Elected village governments, for example, are increasingly using their powers to approve by-laws that legally enshrine the conservation of local ngitili. Such by-laws, once ratified at the district level, are recognized as legitimate by the national government (Barrow and Mlenge 2003:9, Barrow 2005c).

A 2003 study funded by the World Conservation Union concluded that this twin-track approach had paid off. “Traditional groupings, such as Dagashida and Sungusungu have complemented, rather than conflicted with village government. The blending of the traditional and modern has clearly been an important factor in the success of the restoration” (Kaale et al. 2003:21).

Despite popular support, however, decisions over where to situate ngitili and what rules should govern them are not always democratic. While many communities establish communal enclosures through the village assembly—in which every registered adult can vote—others are chosen arbitrarily by village governments without public consultation (Monela et al. 2004:8). “There is no single way of establishing ngitili and some are more democratic than others,” explains Professor Gerald Monela of the Department of Forest Economics at Tanzania’s Sokoine University of Agriculture. In general, he says, devolution of decision-making to village institutions has clearly increased local responsibility for natural resource management and promoted the success of ngitili conservation in Shinyanga (Monela 2005).

This success has not been lost on Tanzania’s other regions, two of which, Mwanza and Tabora, are now adapting and replicating HASHI’s empowerment methods (Barrow and Mlenge 2004:2).

Paying Dividends to People

Of the more than 350,000 hectares of land now occupied by restored or newly established ngitili, roughly half is owned by groups and half by individuals. Communal enclosures average 164 hectares in size, while individual plots average 2.3 hectares (Kaale et al. 2003:9; Barrow and Mlenge 2004:1).

While the impressive speed of ngitili-based reforestation has been apparent for several years, its impact on people’s livelihoods and income has only recently been quantified. A major study by a ten-person task force, launched by the Tanzanian government and IUCN in 2004 and directed by Prof. Monela, combined detailed field research among 240 households in 12 villages with market surveys and other data analysis to quantify the HASHI project’s benefits (Monela 2005).

The task force estimated the cash value of benefits from ngitili in Shinyanga at US$14 per person per month—significantly higher than the average monthly spending per person in rural Tanzania, of US$8.50 (Monela et al 2004:6). Of the 16 natural products commonly harvested from ngitili, fuelwood, timber, and medicinal plants were found to be of greatest economic value to households. Other valuable outputs included fodder, thatch-grass for roofing, and wild foods such as bush meat, fruit, vegetables, and honey (Monela et al. 2004:54-56). (See Table 2.)

In surveyed villages, up to 64 percent of households reported that they were better off due to the benefits derived from ngitili. The task force, headed by Professor Monela, concluded that ngitili restoration “demonstrates the importance of tree-based natural resources to the economies of local people” and offers “a significant income source to supplement agriculture to diversify livelihoods in Shinyanga region” (Monela et al. 2004:7,16).

The study also documented the ripple effect of these economic benefits in people’s lives. Maintaining ngitili has enabled some villagers—mainly through sales of timber and other wood products—to pay school fees, purchase new farm equipment, and hire agricultural labor. Income generated by communal ngitili has been used to build classrooms, village offices, and healthcare centers. One farmer, ‘Jim’ of Seseko village, reported how he had been able to send his son to secondary school and his daughter to university in Dar es Salaam. “My ngitili assists me …I fatten my cattle there and therefore they fetch a good price. Then I use the money to educate my children” (Monela et al. 2004:91).

The new abundance of fruits, vegetables, and edible insects has also improved local health, while easy access to thatched grass has improved housing. Raised water tables due to soil conservation have increased water supplies within villages.

The study also confirms that villagers, particularly women, are saving considerable time by no longer having to walk long distances for fuelwood, fodder, and thatch. (See Table 1.) This frees men and women to concentrate on other income-generating activities while also fostering improved child care and school attendance (Monela et al 2004:108). “I now only spend 20 minutes collecting fuel wood. In the past I spent 2-4 hours,” reported one Sukuma woman who harvests branches from the family ngitili (Barrow and Mlenge 2004:2).

According to Edmund Barrow, Coordinator of Forest and Dryland Conservation and Social Policy at IUCN’s Eastern Africa office, the task force findings “demonstrate that natural resource assets are significantly more important in terms of livelihood security and economic benefits than is generally assumed.” There are useful lessons to be drawn, he argues, both by Tanzania’s government and other comparable countries. “At a time when conservation is increasingly being asked to justify itself in the context of the Millennium Development Goals, the HASHI experience offers detailed insights into the reasons for considering biodiversity conservation as a key component of livelihood security and poverty reduction” (Barrow 2005b; Barrow and Mlenge 2004:1).

The Conservation Dividend

Not only are the restored woodlands important economic assets but, as Table 1 highlights, they are also fostering richer habitats and the recovery of a variety of species. The task force found 152 species of trees, shrubs, and climbers in restored ngitili, where recently scrubby wasteland had stood. Small- and medium-sized mammals such as hyenas, wild pigs, deer, hare, and rabbits are also returning, and the task force recorded 145 bird species that had become locally rare or extinct (Monela et al. 2004:3-5).

The returning wildlife has also created problems, with some villages suffering considerable crop damage. Growing hyena populations, for example, are taking a toll on livestock. However, the costs of wildlife damage, which average US$63 per family per year, are greatly outweighed by the economic gains from ngitili in most villages (Monela et al. 2004:58-61, 67; Barrow 2005c).

Unequal Distribution of Benefits

Not everyone is benefiting equally from ngitili restoration, however. Land use patterns in the region are strongly influenced by Sukuma traditions, with women controlling low-income crops while men control higher-earning livestock and cash crops. The task force found this culture persisting with ngitili restoration, with married women rarely owning individual ngitili or having a meaningful say in their management (Monela et al 2004: 92). On the other hand, all women have access to communal ngitili, a right and resource which has helped them acquire essential household needs such as fuelwood, thatch, and food, and to save time on chores. “Women are better off as a result of ngitili revival, despite patriarchal systems, due to their increased access to forest products,” argues Professor Monela, the task force chairman (Monela 2005).

Better-off households are also capturing a bigger slice of benefits from reforestation measures than poorer families. The task force reported that differences in land and cattle ownership were the most obvious indicators regarding the scale of benefits reaped, and noted that well-off people were buying additional land from poorer households, thus exacerbating local inequity (Monela et al. 2004:92-93). At the other end of the scale, the poorest households cannot afford individual ngitili, although they are entitled to harvest products from communal enclosures, sometimes for a fee.

One impoverished woman, from Mwamnemha village, explained her predicament to a task force researcher: “I do not have a ngitili because I do not have money, nor cattle to allow me to buy land. I therefore purchase some of my needs from ngitili. If I want to purchase grass for thatching I have to pay 200 shillings [US$ 0.20] per bundle. If I want land for cultivation, I have to rent a piece for 12,000 shillings per acre. I am sometimes given these products free of charge, but this is very rare” (Monela et al. 2004:92).

Despite such problems, there have also been improvements for the poorest. The task force found that ngitili were being “used as one of the strategies through which some communities indirectly cushion the vulnerability of households classified as poor…those of the elderly, widows, and households with no assets.” Most communities surveyed included families with no cattle as those in need of help, even if they had some land. The task force reported that each village they visited either lent oxen to plough the fields of cattle-less households, or allowed these households free use of products from communal ngitili. In the village of Seseko, poor households were required to reciprocate by feeding the neighbors who plowed their fields (Monela et al. 2004:95).

Acknowledging the benefits gap between richer and poorer households, the task force warned that additional strategies would be required to prevent social conflicts from erupting and to ensure the long-term sustainability of ngitili. In particular, its report concludes, local institutions should make every effort to “enable people to hold on to land resources so that they can maintain ngitili and enjoy its products” (Monela et al. 2004:110).

A Fragile Future?

The HASHI project is clearly a success story, drawing attention far beyond Shinyanga’s borders. Yet several demographic and land-use trends threaten the continued expansion of ngitili as a cornerstone of natural resource management in Tanzania. These include (Monela et al. 2004:103-4,107):

• Scarcity of land and insecurity of tenure;
• Rapidly growing human and livestock populations, which are driving a surge in demand for resources from the still-recovering landscape;
• Damage to livestock and crops caused by growing wildlife populations; in some areas, this threatens to outweigh the benefits gained from ngitili;
• Growing, unregulated sales of individually owned ngitili.

The government-commissioned task force identified population increase as a particular concern, pointing out that so far “there are not clear indications that the restoration [of ngitili] is sustainable” (Monela et al. 2004:107). Shinyanga’s population rose from 1.77 million in 1988 to 2.8 million in 2002, and continues to grow by 2.9 percent a year (Monela et al. 2004: 21). As a result, fathers are increasingly dividing their ngitili plots between sons, reducing the size and productivity of the plots. Farmers in Maswa district, for example, reported in 2004 that the shrinking size of their individually owned ngitili had forced them to graze only the neediest animals during the critical dry season.

In addition, there are no constraints on landowners wishing to sell their individually owned ngitili, although, because of the village land title system, it is very difficult to sell private land to someone from outside your community. New owners are free to fell the trees and develop the land as they see fit.

The somewhat ambiguous tenure situation of ngitili is also a significant concern. Despite popular enthusiasm, the establishment of new ngitili is often limited by tenure insecurity—or the perception of insecurity. Although ngitili are formally recorded and registered by village governments, their tenure status remains unclear under Tanzanian law. Villages commonly hold a village title deed to all the land within village borders, while households receive a subsidiary title to their privately owned farmland with the village assembly’s approval. The remaining land is designated as communal village land, under the management of the village government (Barrow 2005c, d).

These communal lands can be used for communal ngitili, but it is not always clear what basis the designation of a village ngitili has in law, and therefore what property rights pertain. For example, village governments and assemblies are sometimes wary of officially designating ngitili as “protected areas,” because they fear the state may appropriate these lands and manage them as public lands at the district or national levels (Barrow 2005d).

Tenure issues can interfere with establishing ngitili on private land as well. Private landowners who don’t have secure rights to their land are sometimes reluctant to establish or expand ngitili for fear of triggering disputes within the community. In some cases, concerted efforts by villagers and local government institutions have overcome tenure problems, with boundary surveys made in order to obtain legally watertight communal and individual land title deeds (Kaale et al. 2003:16). Nevertheless, as pressure on land grows due to rising human and livestock populations, land tenure disputes, trespassing on ngitili, and conflicts over grazing rights are all likely to increase.

Designating in law the specific ownership and use-rights that pertain to communal ngitili within the overall system of village-owned land could help address the tenure problem, according to Edmund Barrow. Formally recognizing individual and family-owned ngitili under Tanzanian law as a separate land management category would also help. Closing these loopholes would help ensure that ngitili continue to play a significant and expanding role in villagers’ livelihood strategies and income (Barrow 2005c).

Despite these challenges, the multiple benefits of forest restoration are increasingly recognized by Tanzania’s government. Since the HASHI project began, new legislation— including the National Land Policy of 1997, the Land Act of 1999, and Village Act of 1999—has supported the formal establishment of ngitili and has begun to address the thorny issue of land tenure (Kaale et al. 2003:16). In 1998 Tanzania revised its forest policy, which now emphasizes participatory management of and decentralized control over woodlands, and strongly supports ngitili.

Enriching the Benefits Stream

According to Professor Monela’s task force, the Tanzanian government can take several additional steps to improve the economic benefits from ngitili and thus their anti-poverty impact (Monela et al. 2004:10). These include:

Support Better Ngitili Management
The state can provide technical help and targeted research specifically aimed at raising ngitili productivity. For example, it could help improve fodder productivity by introducing more nutritive and productive tree, shrub, and grass species. And it can research the best methods and timing of cutting and pruning ngitili trees to maximize production.

Monitor Ngitili Trends and Facilitate Lesson-Sharing
The state is in a unique position to offer certain kinds of support that require a national rather than local perspective. For example, using satellite imagery the state could track nationwide changes in land use and biodiversity related to ngitili restoration to help HASHI officials understand the macroscale impact of their activities and better target their aid. The state can also mount a national effort to document ngitili-related benefits and innovations, helping communities to share their successes and learn from others through public education campaigns and knowledge networks.

Expand Markets for Ngitili Products
Increasing the income stream from ngitilis will help sustain Shinyanga’s land-use renaissance by making ngitilis even more essential to local livelihoods. One of the most effective ways to do this is to expand the markets for ngitili products. The state can help by supporting small-scale processing plants to diversify and add value to ngitili products (by making timber into furniture, for example); by removing burdensome regulations and other barriers to ngitili expansion and the establishment of local enterprises based on ngitili products; and by helping households access local and regional markets for their ngitili products by providing relevant and timely market information.

How Tanzania’s government responds to these and other challenges facing the ngitili restoration movement, remains to be seen. What is not in dispute is a strong national commitment to consolidate the successes of ngitili restoration and the benefits it has brought in Shinyanga, and to replicate these, wherever possible, across Tanzania’s drylands