Navigation – Plan du site
L'Afrique équatoriale

Shifting cultivation in the Ituri Forest [Haut-Zaïre]

Colonial intervention, present situation, economic and ecological prospects
Michael Rösler
p. 44-61

Texte intégral

Texte intégral en libre accès disponible depuis le 29 juin 2009.

I am grateful to Werner Petermann, Johannes W. Raum and Polly Wiessner for their critical comments on a preliminary draft.


1The study of shifting cultivation in the Central African rain forest has rarely been made the subject of anthropological research. One can only speculate upon the reasons for this reluctance. Maybe there is a general awareness of the conceptual complications brought about by colonial intervention and its impact on local agrarian systems [Miracle, 1967]. If the colonial context is fully taken into account, long-standing anthropological paradigms are threatened indeed. The Ituri Forest of North Eastern Zaire is a particular case in point. Until now, anthropologists have restricted their efforts to groups of so-called hunter-gatherers; one fancies Pygmies as one of the last exotic peoples of the world apparently representing the earliest stages of human evolution. As a consequence of this rather naive and romantic view, the village-dwelling agricultural people, who represent the majority of the population in the Ituri Forest, have been completely neglected. After Turnbull's (1965, 1983) verdict that there is an insurmountable contrast between “village world” and “forest world”, nobody has dared to take up the “villager's point of view”. The banishment of the agriculturalists, by being characterized as being essentially a non-forest population who immigrated only recently and destroyed the forest by shifting cultivation, created a powerful anthropological myth that led to erronious interpretations not only of the cultivators but also of the socio-economic status of the Pygmies as hunter-gatherers in the overall local economy.

2Such assumptions underlying hunter-gatherer research in the Ituri have been questioned only recently. Thus ecologists doubt that the forest provides the necessary carbohydrate resources to support independent hunter-gatherers [Hart and Hart, 1986, Bailey and Headland, 1991]. For the first time it is fully recognized that Pygmy foragers like the Efe are heavily dependent on their Lese agricultural neighbours, with whom they maintain close economic and social relations [Bailey and Peacock, 1988]. Although agriculture is currently receiving special attention, some misconceptions on its apparent static and unchanging nature still prevail. Going a step further and hypothesizing that today's agriculture in the Ituri forest, far from being traditional, is a product of former colonial development measures, one can conclude that the celebrated opposition of village and forest world represents the functional differentiation of agriculture and hunting-gathering as separate, albeit interdependent, economic systems which came into being only under colonial rule. The current system of shifting cultivation can best be described as neo-traditional, composed of modern (colonial) and traditional (pre-colonial) structural elements. It is a locally adapted version of what was introduced by the Belgian Colonial Administration into the Ituri Forest under the heading of paysannat indigène including resettlement schemes and compulsory cultivation [Jurion and Henry 1967].

  • 1 For a lengthy discussion of the implications for hunter-gatherer research in the Ituri see Rösler ( (...)

3If far-reaching measures like this are taken into account, it becomes clear that it is much less the traditional than the transitional aspects of the Ituri societies that come to the fore. This paper will discuss the major ethnological, economic, and developmental implications of this perspective “from the villager's point of view”1. It attempts to reconstruct pre-colonial agriculture, shows how it was transformed under colonial rule, and it traces recent post-colonial developments, or the lack of them. The main thrust of the argument is that Ituri Forest societies, cultivators as well as hunter-gatherers, cannot be treated as static isolated communities, who have survived in one of the last untouched rain forest corners of the world. Although their pre-colonial culture history remains largely unknown, we can at least reconstruct the social, political and economic changes that occured during the last 100 years of interaction with external powers. If we can understand the dialectics of structural change and continuity in the local production system, we may be able to discern the dynamics of ongoing and future development.

4Empirical data were collected during extended field-research from 1987-89 in the Collectivité des Bombo. The agricultural system of a community of some 200 inhabitants, situated near the junction of the Lenda and Ituri rivers, was studied for two cultivation periods. Although the villages in this area are nowadays enclosed by forest, forty years ago there was a road south of the Ituri connecting them to the main road, where most of the villages are situated today. Trading with the outside world is maintained by frequent intervillage visiting. The general situation south of the Ituri does not differ significantly from that of roadside villages. The data presented here refer primarily to one ethnic group called Bombo (Mho) numbering about 2,400 people. As frequent observation in the neighbourhood clearly showed, they can be regarded as representative of a larger population of some 7,000 people living in villages along the road between Nia Nia and Mambasa (see Fig.1), the latter being the administrative capital of the Ituri Forest.

1: Forest clearance: Bambuti men assist in cutting trees on the field of their village patron.

Stages of colonial and post-colonial transformation

5The Ituri Forest is inhabited by several Bantu and Sudanic-speaking ethnic groups (Bila, Lese, Ndaka, Budu, Mbo), all of them maintain symbiotic, hereditary patron-client relationships with local Pygmy populations known as Bambuti (Mbuti) and Efe. While the latter are generally classified as semi-nomadic hunter-gatherers, the former have been arbitrarily labelled agricultural villagers. Conventional anthropological wisdom on the Ituri Forest is still unaware of the profound impact of colonial rule that shaped current ethnic identities as well as socio-economic systems. Much of what has been labelled the opposition between “forest world” and “village world” can be understood as the direct result of colonial intervention, leaving the forest almost exclusively to the Pygmies for hunting and gathering, and confining the economic activities of the cultivators to the village and its immediate environment.

6There is growing evidence from hunter-gatherer research in the northern Ituri Forest that shifting cultivators like the Lese can be considered as forest people who “inhabited the Ituri longer than two centuries” [Bailey and De Vore, 1989:464]. The Bombo in the southern Ituri can serve as another example of such long-established forest cultivators [Rösler in press]. Even from the scant ethnographic information available on the Bila, Ndaka and others, one can conclude that this holds true for nearly all agriculturalists of the Ituri Forest [van Geluwe 1956, 1959] (Ref. missing in bibliography !). According to archaeological and linguistic data, agricultural settlement in the Ituri Forest can be traced hundreds of years back [Vansina, 1990], but it is difficult to correlate these data directly with existing populations. For the present purpose it is convenient to focus on the crucial developments of the last hundred years that have obscured the political and cultural tradition of the Ituri cultivators and their relation to local Pygmy groups. Among these we have to consider measures of colonial indirect rule, like circonscription indigène and paysannat indigène, which included political centralization and the reorganization of agriculture [Young 1965].

Fig. 1. Ituri Forest

Fig. 1. Ituri Forest
  • 2 Indigenous expressions given in Kingwana, the regional Swahili dialect, that is widely used as vern (...)

7The Bombo became an ethnic or tribal unit, as they conceive of themselves today, after the colonial administration imposed petty chieftainships in the Ituri Forest during the first decades of this century. At that time, they lived in small temporary settlements dispersed throughout the forest on both sides of the Lenda and Ituri rivers. These isolated villages, fortified by palisades, represented the largest political and economic units, claiming huge tracts of forest territory they utilized for hunting and gathering wild forest resources in addition to the cultivation of yams and bananas. The villages were organized on the basis of segmentary patrilineages combined into exogamic patricians. External relations were based on marriage alliances and military co-operation, sometimes reinforced by initiation confederacies. Local groups of Bambuti Pygmies were associated with certain village-based patrilineages by hereditary patron-client relations granting them the right to hunt and forage in the forest territory of the particular lineage they were associated with. The Bombo often refer to pre-colonial times as wakati ya vita, “the times of war”2. They maintain, as Putnam [1948] has pointed out, that their Bambuti performed important military and logistic functions to help defend the forest territories of their patrons. Today, many local Pygmy groups are still associated with their former Bombo patrons though exchange relations have become more flexible and transactions are confined to economic services.

8Colonial power removed this traditional politico-territorial order with the imposition of chieftainships as a measure of indirect rule. The process of incorporation into the colonial state occured in several steps. The people who are called Bombo today comprised about 30 patricians consisting of some 100 patrilineal descent groups distributed over more than 50 settlements. In 1916-17 these were aggregated into four petty chieftainships (chefferies). In 1930 one of these units received the status of an all-embracing Chefferie des Bombo, incorporating the others as sous-chefferies. At present these divisions represent the Collectivité des Bombo including three groupements. The position of the chief has remained with the same family up to this day and has become integral to the local administrative structure. Chiefs were endowed with administrative powers and economic privileges. Only in the beginning was the new political hierarchy opposed by traditional village authorities. Now it is largely accepted by the population, although disputes within the famille régnante over the succession to the office of the Chef de collectivité continue. The authority of the chiefs depends on the overall supporting political structure. As the Zairian state declined over the last decades and the national economy collapsed [Young and Turner, 1985], basic administrative functions of the Collectivité like tax collecting were suspended.

9Contemporary village communities are virtually autonomous. The heads of local lineages exercise authority by settling disputes and co-ordinating economic activities. In this respect the current situation resembles that of former pre-colonial times.

2: Secondary settlement (malende) of village and Bambuti households.

10Bombo reckon their history in terms of epochs. The current situation is often referred to as wakati ya bazairois, “the time of Zairians”, designating arrogance and incompetence of administrative personel and party cadres as well as the general condition of economic and infra-structural decline. Colonial times, referred to as wakati ya babelges, are remembered by elders with very mixed feelings about the achievements of European (wazungu) culture and the price they had to pay for it. Things look different when the epochal label is changed in favour of wakati ya Kayumba, the time of Chief Kayumba, who ruled the Chefferie des Bombo from 1930-48. This time span is identified as an epoch of glory full of heroic deeds. One of them is the construction of a 60 km roadway south of the Ituri, that was initiated by Kayumba, after the population was forced into villages along the major road that crosses the Ituri Forest from west to east. This saved the southern Bombo communities from being evicted from their ancestral territories on the bank of the Lenda river. Nevertheless, compulsory cultivation of cotton, reaching its zenith in the fifties, could not be averted. The western Ituri Forest became part of the zone cotonniere Nord and was subjected to experimental paysannat politics [Drachoussoff et al. 1992, Vol:2 788 ff]. Agricultural production under the paternalistic supervision of the colonial agronomist came to govern the laborious working-day.

11Although the colonial state had earlier tapped the indigenous labor force through notorious rubber campaigns, the imposition of cultures obligatoires quickly set new standards in terms of demands for labour force, agronomic innovations, economic organization and productivity. Daily life became almost completely centered upon agriculture. Ancient sacred institutions of leadership like the imbaa, who functioned among the Bombo as a judge and priest, were abolished, while other social institutions were reinforced as long as they served paternalistic tutelage. The exigencies of cash-crop production had their profound impact on the creation of a neo-traditional order within the village. The European agronomist had his indigenous counterpart represented by the moniteur agricole who supervised the details of cultivation. The head of the village, along with the village elders, was responsible for complying with the production quota of each household. Whipping and fines became an everyday occurence. We can assume that these “educative” measures failed among the Bombo as in the rest of Congo Beige [Jewsiewicky, 1983]. Nevertheless, some agronomic innovations should be kept in mind. They have survived and were modified by the local population resulting in a less intensive but nevertheless potentially effective form of shifting cultivation.

Colonial innovation and post-colonial adaptation in shifting cultivation

  • 3 These are mostly plantains designated by the Bombo with the generic term ibugu, of which they still (...)

12Pre-colonial agriculture was quite different from that practiced today. For the cultivation of the traditional staples bananas and yams only moderate clearings were needed, which were cut into the forest in the vicinity of the villages (see Fig. 2.1). These gardens were enlarged every year while old stands of bananas could remain up to five years. Shifting cultivation implied shifting the fields annually and moving the villages after 5-10 years. This became necessary when soil fertility declined, or the fields were too far away, or hygienic as well as strategic military options caused people to move. However, the cultivation of bananas3, the primary supplier of caloric food, was not at the center of economic activities. Reconstruction of the traditional economy of the Bombo revealed that hunting and collecting wild forest food was at least as important. Hunting methods of the Bombo, which included a wide array of trapping techniques designed to kill large game like elephants and buffalos and net hunting for antelopes, were in no way inferior to those of the Bambuti. Since the oil palm had not yet been introduced into their territory the Bombo were heavily dependent on the collecting of wild forest resources like nuts and insects to fill the fat gap in the subsistence system. These foraging activities rather than the cultivation of bananas determined the annual cycle [Rosier, 1997]. Colonial powers knew how to tap indigenous forest skills when they imposed the rubber campaign and demanded tributes in ivory tusks.

Fig. 2.1. Pre-colonial fiel shifting

Fig. 2.1. Pre-colonial fiel shifting

Each household shifts his field within the limits of the territory of his chut/lineage.
Note: Number of households/fields is reduced, households 1-4 representing Clan A, households 5-8 representing Clan B.

Fig. 2.2. Colonial and post-colonial corridor shifting

Fig. 2.2. Colonial and post-colonial corridor shifting

Fields are grouped in two corridors shifted separately according to the division of land into clan territories.
Vegetation types
primary forest: black
secondary forest: grey
field: white

  • 4 Drachoussoff et al. (1992, Vol. 2: 799 ff), contains more detailed information.

13Resettlement and compulsory cotton cultivation introduced a new dimension of economic differentiation. Agricultural production became the mainstay of the newly established villages. After some negative experience with the introduction of European methods of permanent cultivation, shifting cultivation termed the “Bantu system” came to be recognized as “perfectly adapted to heavy rain forest” [De Coene, 1956]. However, resettlement in permanent villages posed new problems of soil conservation and rational management of long forest fallow. The practical solution to this was the field corridor (couloir), which required that all fields be grouped in a single large clearing to assure collective shifting of all plots of a village. Although there were different systems of field-forest rotation developed by the Institut National pour l'Étude Agronomique du Congo Belge (INÉAC) on research stations in the 1940s, it is difficult to determine which system was applied in the Ituri Forest4. It is questionable, to what extent suggestions of research stations, such as alternating strips of field and forest to ensure rapid forest regrowth [Tondeur, 1957], were put into practice. In most cases there was learning by doing. Over-exploitation of soil may have occured since length of fallow periods was unduly shortened. From local informants, who were functioning as moniteurs agricoles in the fifties, and current cultivation practice, one can conclude that, at least in the last colonial decade, crop rotation was as follows: first three years — Banana, rice banana/cassava; next two years — groundnuts, cotton.

14Since cotton, groundnuts and rice were short-term crops that required 3-4 months of growth they could be easily integrated into existing practices of intercropping and field shifting. There was no need to introduce individual private ownership of land, as it was usually propagated in paysannat politics especially to cultivate perennials like coffee. The corridor system respected communal land tenure, as land was redistributed to every household each year. This corresponded to traditional notions that land was the property of the patrician or lineage, which distributed individual usufruct to its members. Only crops, not land, were individually owned. Resettlement nevertheless presented the problem that kinship groups became settled on the territory of others. In the Bombo community studied, which was founded in the early thirties by the association of two patricians, one clan, whose territory was situated in the west of the village, came to settle on the territory of its neighbouring clan. This clan occupied the western part of the village and established their field corridor west of the village on their former territory. Until today each clan operates its own corridor (see Fig. 2.2).

15From an agronomical point of view the corridor has many advantages. Pooling the fields of a village ensures a rational field-forest rotation with planned long fallows of 15 years and more (see Fig. 2.3). Agronomic supervision is relatively easy. The annual allotment of plots to the households is flexible and can be adjusted to actual labor capacity. Kinship-based patterns of land use and co-operation can be integrated. Annual movement of the corridor can be adjusted to variations in soil quality and the landscape [De Coene, 1956, Tondeur, 1957]. From the cultivator's point of view, however, things look different. A common field corridor is not flexible enough in terms of both labor allocation and social relations. At times the priority of long forest fallow causes unreasonably long walking distances from and to the villages. Although this is generally dealt with by building secondary huts in the fields, where the family can stay for longer periods of time, a certain portion of disfunctional households (see below) fail to allocate necessary labor to cut mature secondary forest or manage long transport distances. Lineage and patrician solidarity set limits to individual decisions. Such group loyalty becomes a menace when conflicts arise within the larger community of the village.

16The shape of the corridor today is different from its colonial forerunner in that it is shorter and more compact (see Fig. 2.4). Certain rules like the east-west orientation to use sun light more efficiently have been dropped. Arranging the fields in one long strip has been modified in favour of a double-row of fields, which shortens the corridor and saves labor time when measuring and distributing the fields, but the standard size of 50 x 100m (0.5 ha) per household has been retained. Field neighbourhood reflects household neighbourhood in the village. When a new corridor is cleared, fields are arranged in a double-line. In the next year, fields are shifted to the head-end of the old field. In the third year fields shift a step further. The corridor is split up into sections corresponding to units of households belonging to one lineage. There may be rearrangements of new common corridors after some years, but the general dynamic of isolating lineage-based corridors persists. The corridor directly reproduces processes of fission and fusion of lineages in the village. In other words, the corridor expresses the local politics of temporary houshold alliances.

Fig. 2.3. Colonial allotment of fields within the corridor

Fig. 2.3. Colonial allotment of fields within the corridor

Situation of a corridor in the third year. Fields are grouped in a single strip of land and shifted collectively in one direction. Plots are arranged according to lineage membership of households.
field: white
field of preceding year: light-grey
secondary vegetation: dark-grey
primary forest/secondary vegetation: black

Fig. 2.4. Pre-colonial allotment of fields within the corridor

Fig. 2.4. Pre-colonial allotment of fields within the corridor

Situation of a corridor in the third year. The double-sided corridor of the first year is split up into smaller sections, each moving in different directions.
field: white
field of preceding year: light-grey
secondary vegetation: dark-grey
primary forest/secondary vegetation: black

17Indeed, there are many advantages of a flexible corridor. It produces a certain chaos which is functional in dealing with external as well as internal enemies. As the lessons of the Simba rebellion 1964-68 taught, the single corridor is vulnerable to looting. So has plunder by uncontrolled bands of soldiers in recent years. Even administrative officials are quick in confiscating crops for so-called official reasons. In fact, authorities of Collectivité and Groupement often take up the initiative to build new corridors, that may only be abandonned afterwards. Distrust and suspicion of envious neighbours is another incentive to hide one's field from others. Moreover, the small corridor is more adaptable to the specific demands of the household. If someone is ill, one can cut young secondary forest and plant at least some cassava. The effect on forest regeneration is not necessarily negative, for secondary growth comes up rapidly, as fields are generally planted for one year only. Clearings are small and encircled by mature trees which seed secondary growth. If marketing opportunities should become better, cultivation of rice as cash crop could be intensified with little effort. The single corridor can be restored, if economic incentives override petty village affairs. For the present, however, priorities in organizing the field corridor have switched from economic and ecological requirements to social needs. The single corridor presupposes a degree of social harmony or consent the larger village community rarely can provide and it can be used as a source of social discrimination including witchcraft accusations, if crops fail on some fields while others are spared. The corridor system, as it exists today, reflects these social constraints.

Cultivation cycle

18Just like the corridor system, the cultivation cycle has been adapted to local conditions. Following the colonial innovations, crop rotation represents a mix of the non-seasonal perennials, bananas and cassava, and the seasonal annuals, groundnuts and rice. The former can be planted and harvested at any time of the year. The latter deserve proper timing to make optimal use of seasonal rainfall (see Fig. 3). Together with cotton, groundnuts and rice have come to determine the cultivation cycle under colonial supervision. Although their importance as cash crops is continually declining (cotton cultivation has been abandoned completely), they still govern the agricultural work calendar (see Fig. 4).

19Men begin forest clearance in October/November. After measuring the fields, the undergrowth is removed with the machete, work which is usually done cooperatively by neighbours. Then trees are felled with axes, arduous work that continues for weeks with breaks of some days. Vegetation is not removed completely. Particular hard wood species are left standing. Clearing should be finished in January, so that felled vegetation will be dry enough to be burned in the middle of February at the peak of the dry season. Further cutting and burning of logs and trunks is carried out later, as pieces of land are prepared for planting. The first bananas and cassava can be planted in the new field as soon as the underbrush has been cut. Subsequent burning does not harm shoots or larger plants. Groundnuts are sown March/April and harvested June/July. Land is weeded and burned before rice is planted in June to August, to be harvested from November to February. High amounts of rainfall over nine months of the year generally provide considerable latitude in the timing of activities, and there are variations of up to three months in planting and harvesting dates of groundnuts and rice. Nevertheless, yields can vary according to rainfall, damage inflicted by game, and labor input. In a sample of 4 particularity productive fields the yields of three staple crops could be calculated: medium yields of rice were 1,640 kg/ha, of bananas 6,150 kg/ha and of cassava 5,100 kg/ha [Rösler, 1997].

Fig. 3. Rainfall Ituri Forest

Fig. 3. Rainfall Ituri Forest

Fig. 4. Agricultural Work Calendar

Fig. 4. Agricultural Work Calendar

20In terms of food preferences, bananas are often declared by Bombo to be more favoured than rice. Cassava is considered a food reserve, a substitute for bananas on poor soils. Since bananas demand fertile soils, their high prefential rank requires long forest fallow. In the agricultural season of 1986/87 two new corridors were established for this reason three and five kilometers away from the village, in mature 20-year-old secondary forest with stands of primary forest. Groundnuts, rice and cassava are planted by some households on soils that have lain fallow for five years only. Since it is not marketing opportunities but subsistence needs that determine decisions as to what is cultivated, production levels of groundnuts and rice have fallen considerably. Nevertheless, even the subsistence level cannot be attained by many households. There is a seasonal food shortage in April/May every year. It does not only occur in years of irregular distribution of rainfall resulting in bad rice harvests [Bailey and Peacock, 1988], but rather, as my findings suggest, is mainly correlated with the intricacies of household production and the general lack of incentive to raise production [Rösler, 1997].

Institutional constraints of domestic production

  • 5 Sample size was 50 households with a total of 190 individuals. Household productivity was calculate (...)

21In the agricultural season of 1986/87 62 % of households in the village studied were not self-sufficient and in 1987/88, 72 %, though they were dependent upon agricultural production and had no significant alternative source of income. In 1986/87 30 % of housholds were self-sufficient and only 8 % were surplus-producers. In the following year, 28 % were self-provisioning and there were no surplus-producers at all5. The primary reason for such low agricultural productivity is that a considerable proportion of housholds are disfunctional. As a rule households are monogamous nuclear families, with a medium size of 3.8 persons. They are thus particularity fragile production units vulnerable to shortage of personel. In 1986/87 30 % and in 1987/88 44 % of the households were underproductive due to death or illness of one of their members or because the couple divorced or separated (the latter being the case particularly with young couples, who live together before bridewealth is paid). Any household can experience such a situation from time to time. If cases of domestic underproduction accumulate they can result in chronic local food shortage.

22The majority of households, however, can be considered as consolidated and functional, as they are based on marriages sanctioned by payment of bridewealth and have at least two able-bodied adults at their disposal. Nevertheless about 45 % of these households are underproductive as well (accounting for 32 % of the total number of households in 1986/87 and for 28 % in 1987/88). The reasons for this vary. All households, including underproductive as well as self-suppliers and surplus-producers, face one or more of the following internal and external constraints on agricultural production.

  1. Absence of women from husband's household. This is a function of the inferior social status of women in the patri-virilocal context. In the village of their husband women are treated as “strangers”, who are not allowed to participate in decisions of the communal hut of the local patrilineage. Emotional bonds of women to their own natal lineage remain strong, resulting in frequent visits to the village where they were born (see Fig. 5). These can reduce the wife's labor input in the fields of the husband's household. Long absence of wives sometimes results in complete abandonment of a field.

    • 6 Time allocation data were gathered in 36 village households over 12 months in 1988 using the random (...)

    Conflicts over the sexual division of labor. Although agricultural labor input is moderate and differences between the sexes are not strongly pronounced (see Fig. 5)6, the latter give rise to conflicts about the responsibility for certain time-consuming tasks like weeding. Whereas these were performed under colonial supervision by men and women in co-operation, today they are left almost completely to the women which leads to a bottleneck in the case of rice planting.

  2. Lack of cash economy. Only a small fraction of the rice harvest of 1986/87 was sold to official byers. In the next year there was a complete breakdown of agricultural marketing due to the restrictive price policy of the government. Most of the surplus-producers refused to sell their crop. Moreover, the status of rice as potential cash crop has resulted in different opinions within the household about who is the owner of the crop, Women should receive their shares when the crop is sold, but men often pocket the money. Women express their discontent by refusing to cultivate and visiting their native villages.

  3. Disfunctional administration and increasing political insecurity. The administration has crumbled and regular taxation has been abandoned. Military personal occasionally raids fields and villages. Compulsory collective labor, called salongo, forces the population into time consuming activities (see Fig. 5) that are nevertheless highly ineffective. In 1988 one such salongo-campaign kept the population busy for two months to re-open the former roadway south of the Ituri in order that trucks could pass to collect rice, but there was neither a barque to cross the river nor any crop to be sold.

Fig. 5. Time Allocation

Fig. 5. Time Allocation

23It is no wonder that under such conditions surplus-production is exceptional and confined to a few privileged households from strong, politically dominating lineages whose male heads have the status of powerful patriarchs, or are his sons or brothers. Moreover, some members of these lineages occupy administrative offices, their traditional as well as modern status allowing the recruitment of additional labor for cultivation. While administrative officials can demand communal labor like salongo, patriarchs can count on younger relatives and associated Bambuti. Underproduction, by contrast, is correlated with fragile households from shallow lineages, sometimes represented by less than three males who rarely can support members in need of help. These units fail to pool either labor force or agricultural produce.

Bambuti and co-operation in agriculture

24The Bambuti associated with the Bombo cultivate small gardens situated side by side with the fields of their patrons. Nevertheless, labor input and yields are far from being sufficient to secure their own subsistence. Bambuti mainly rely on their patron's harvest or on that of other associated lineage members. Assistance in agricultural labor has been reported as a significant element in exchange relations between Bambuti and village cultivators [cf. Bailey and Peacock, 1988, Wæhle, 1986], but helpers in the garden are a mixed blessing for the owner of the garden. Bambuti do not behave differently from village neighbours or relatives in this respect for they want to be compensated for their efforts. Helping on the field of others is sometimes more rewarding than working one's own. Harvest tasks are easily preferred to clearing and weeding, as the former are compensated immediately with certain shares of the crop. Cultivation of groundnuts has been reduced to a minimum, since, due to too many selfish helpers, the owner's share has become unacceptable. Only the rice harvest still presents a good opportunity to obtain food by assisting others. As such strategies of individual provisioning undermine co-operation and solidarity, “generalized reciprocity” [Sahlins, 1974] as a mode of transaction between relatives becomes obsolete. This applies to patron-client relations between Bambuti and their patrons as well as to villagers in general. Only good personal relations and frequent contacts can prevent conflicts resulting from different opinions about the actual state of debts and credits. A patron has to offer something, at least some palm wine or marihuana, to have “his” Bambuti clear his field. Bambuti should at least offer a prospect of supplying honey or some other forest resources when demanding food. Quite a lot of spontaneous exchange goes on between Bambuti and village household's, and transactions are not confined to the patron-client relationship. Nonetheless the latter offer the opportunity for long-standing stable relations between larger corporate groups on both sides allowing for different arrangements according to circumstances. In the village sudied there were three such combinations:

3: Forest clearance: a field corridor cut into a mixed stand of primary and mature secondary forest.

  1. a weak village patrilineage associated with a strong Bambuti residence group. This large local group of Bambuti which temporarily comprises up to 25 members was “confiscated” by the Chef de Collectivité for some weeks to work on his own field. Under colonial rule this had become a habit of the powerful chiefs, and ever since that time there has been contention as to who is the proper patron of these Bambuti. The latter show indifference about this matter, but nevertheless keep up good relations with their actual patron who decided to leave the village and join “his” Bambuti in their base camp situated five kilometers away on an old field corridor of his lineage. The patron, being the head of a shallow lineage of only three households, is married to two Bambuti women and has a child with one of them. The camp subsists onamixed economy, including agriculture, fishing and hunting. The three households of the village lineage join the Bambuti camp regularly for some months. Economically they are more integrated in the camp than in the village;

  2. a strong village patrilineage associated with a fragmented Bambuti patrilineage. The village patrilineage is headed by the most influental, charismatic patriarch of the village. He “gave” his Bambuti to his younger brother and his two sons to provide food for them. In 1986/87 and 1987/88 the fields of the younger brother produced extraordinary surplus furnishing the Bambuti camp (comprising up to 14 members), which was situated in the midst of them, with food almost all the time. The Bambuti in turn performed some work on the field of the patriarch (not on that of his brother) and became heavily engaged in organizing the boy's initiation ceremony, the women supplying firewood to be piled up in the village and the men attending to the initiation camp where some of their own boys were initiated. In terms of reciprocity, transactions between members of the Bambuti camp and the patrilineage constitute a mutual support network rather than a sequence of one-for-one exchanges. The patron-client relationship is based on mutual agreement and the undisputead authority of the patriarch;

  3. segments of village patrilineages associated with segments of Bambuti lineages. Families of villagers as well as Bambuti can become fragments of formerly larger descent groups by demographic decline either through death or outward-migration. Segmentation of lineages can occur simultaneously, if a patron moves into another village together with his client. Bambuti camps can afford to split into one or two households of nuclear families, if they are in good terms with their patron who must in turn rely on them, for his own lineage is a small one. Thus, a close co-operation in economic activities can develop. The most convenient arrangement is living together within a secondary settlement in the field corridor. There were at least four village households co-existing with single Bambuti households on these terms. Again there is much co-operation in agriculture, fishing and hunting to the mutual benefit of all.

4: Wanande immigrants panning gold at the bank of the Ituri river.

25As these examples show: a regular economic exchange of goods and services is connected with an intimate relationship between Bambuti and villagers. This is not to say that relations are never strained. The patron-client relationship provides the institutional framework for dealing with conflicts just as if relatives were concerned [Rösler, 1992]. That this relationship is hereditary indicates long-standing bonds over generations in the male line on both sides [Terashima, 1985, 1986], with the Bambuti having acquired the one-sided right to be provisioned with agricultural produce. This is the reason why they do not practice agriculture. If the Bombo are still on good terms with “their” Bambuti, this possibly has to do with their marginal economy. Since there is no commercial agriculture and no meat trading as in other parts of the Ituri [Hart, 1978], no conflicts arise about the distribution of the earnings — the Bambuti being part of the larger community of relatives who all claim their share. On the other hand, where commerce arises, Bambuti are apt to adapt more rapidly [Ichikawa, 1991]. Low agricultural performance of their former patrons is possibly related to their readiness to look for other exchange partners.

The local agriculture in perspective: adaptation to the informal sector

26Agricultural production in the Ituri forest is on the retreat. The ongoing Zairian crisis and disorder creates considerable insecurity and intimidation among the rural population. As in other parts of the country [Schatzberg, 1988], the people of the Ituri Forest react by opting for exit: in order to reduce losses due to governmental and military seizures, they limit cultivation to mere subsistence necessities, at times leaving their villages and hiding in the forest, occupying temporary forest camps they traditionally used for hunting and fishing, and which today are suitable for gold-digging. By doing this they renew ancient customs, which fell into disuse only in colonial times. Since the semi-nomadic way of life is not restricted to Mbuti and other Pygmies alone, and since the overall political and economic situation of the country does not offer an integration into a broader cash economy, it is worth considering some local institutions in terms of their adaptive function within the context of political insecurity, economic decline and the possible ecological degradation of the forest that is yet to come.

27The Ituri has one of the lowest population densities in Central Africa. It borders on nearby savannah and mountainous areas, with the highest population densities in the country [Wiese, 1980: Map 7]. The pace-makers of future development will stem from these areas, and the indigenous people of the Ituri, either villagers or Pygmies, will have no say in this matter. Traders of Beni and Butembo already hold key positions in commerce. As the state structure rapidly declined in the last decade, informal economy has taken over. The northern Kivu, ethnically dominated by Wanande, became a booming center of more or less illegal commercial activities, including smuggling from and to Uganda, Kenya, and the Sudan [Vwakyanakazi, 1991]. The Ituri Forest constitutes the hinterland of this informal sector. It functions as a reserve of forest resources like ivory, hides, gold and timber. Information on how far local people have already become involved in these activitites is scarce.

28At least some current studies of Mbuti Pygmies suggest that the commercial meat trade to serve the Kivu town markets has resulted in depletion of game and endangers Mbuti subsistence needs [Hart, 1978]. In the northwestern corner around Wamba, Mbuti are supposed to have "hunted out" the forest [Wæhle, 1986]. As Mbuti become involved in a cash economy they are not used to, they are ridiculed and treated as fools by village people [Ichikawa, 1991] - At the same time low agricultural performance seems to be causing Mbuti to dissolve former patron-client relations in favor of new more spontaneous commercial relations. The opposition of “village world” and “forest world”, as reported by Turnbull (1965, 1983), was confined to a few commercial centers like Epulu in the sixties. It is now becoming the scenario for future development under frontier conditions.

29The Bombo, situated in the inner Ituri Forest, as well as the Bambuti associated with them, are not yet involved in either frontier or informal activities. For them the situation is different and poses other problems. The expansion of informal commercial activities is confined to the gold sector. The carrière is integrated in their normal seasonal activities including semi-nomadic camping in the forest either for fishing or hunting and/or gold-digging. Bambuti may serve carrières and meat traders occasionally showing up with smoked antelope meat. Bombo, if they do not dig for gold themselves, supply carrières mith fish and manufactured produces like pirogues. In general this trade is moderate. Along the banks of the western Ituri Lokele fishermen from Kisangani have settled and some disputes over fishing rights and competition in trade occur, but interethnic relations are not strained. Occasionally, poachers, who are said to be deserted soldiers armed with army guns, present dangerous foes in the forest and some remote villages. There is a rumor among Bombo that Pygmies of Lombi, who live farther off towards Kisangani, defend their hunting grounds against intruders with poisoned arrows, though this sounds like wishful thinking. When soldiers show up in the villages, the population usually does not dare to raise its voice, let alone to take any action.

  • 7 For example, Alur cultivators coming far from the upper Blue Mountains and settling at the central (...)

30Although the carrière does not offer great prospects, it will be the economic focus of the years to come. As alluvial gold deposits do not suffice to make a man rich, the golddigger's camp serves as a decent source of income, either by the direct revenue from washing gold or by selling small amounts of local produce. The informal sector does not demand agricultural products from the Ituri, for they can be more easily supplied by the fertile areas of the Kivu. The agricultural system of the forest, though based on agronomically improved field-forest rotation and intercropping, is currently underused. It can serve local subsistence demands, but it is questionable whether it will ever significantly contribute to the national or regional economy. While corresponding to current notions of sustainable development in terms of marketable surplus production and forest conservation granted by long-fallow shifting cultivation [Hunter and Ntiri 1978], it cannot compete with the more productive agriculture of the mountains. Moreover, it is questionable whether it will be accepted by new settlers, who often cling to accustomed food preferences and agricultural practices7.

31Commercial agriculture, coffee and oil plantations, are rare in the Ituri Forest. As far as they are remnants of colonial enterprises, they are of minor importance. Immigration from the Kivu is accompanied by setting up plantations of beer bananas. The ecological effect is not yet clear. Agricultural intensification has been going on for decades around semi-urban centers, and forest cover has been removed over vast areas. Loca practices of shifting cultivation are surely not adapted to this situation. It has never been examined, whether agronomic improvements have been developed spontaneously by the local urban population. Agricultural produce coming from the villages like rice could surely serve these urban markets, but a noteworthy market system did not develop under colonial rule and was not supported in post-colonial times. Towns are largely supplied by Kivu products. Local agriculture is once again not in demand.

32Summing up one can state a paradox: although from a technical point of view there is a locally and ecologically well adapted agricultural system, based on agronomically improved shifting cultivation that could potentially integrate forest conservation and agricultural production, the system is badly underused and on the verge of disappearing. What developmental experts may still conceptualize is already there, but nobody practices it, since the developmental issue from the peasant's point of view has become a question of hiding or concealing their crop from the state and the military, and at times even from their nearest neighbours.

Haut de page


1 For a lengthy discussion of the implications for hunter-gatherer research in the Ituri see Rösler (1992).

2 Indigenous expressions given in Kingwana, the regional Swahili dialect, that is widely used as vernacular in interethnic discourse.

3 These are mostly plantains designated by the Bombo with the generic term ibugu, of which they still cultivate 15 varieties.

4 Drachoussoff et al. (1992, Vol. 2: 799 ff), contains more detailed information.

5 Sample size was 50 households with a total of 190 individuals. Household productivity was calculated in terms of caloric needs of household members. Surplus is defined as output exceeding household needs by more than 25 % (Rösler, 1997),

6 Time allocation data were gathered in 36 village households over 12 months in 1988 using the random visit method of Johnson (1975). The sample includes data on the daily activities only of adults (Rösler, 1997).

7 For example, Alur cultivators coming far from the upper Blue Mountains and settling at the central road could be observed cultivating cassava with short fallow of five years and extending the cultivation period of the field to three years. This practice was taken over from the mountains, where soils are much more fertile. In the forest this will deplete the capacity of the soils to recover and ultimately remove secondary forest in favour of bush vegetation and stands of Imperata. This may be instructive of how food preferences (not population density or land scarcity) determine cultivation intensity.

Haut de page

Table des illustrations

Légende 1: Forest clearance: Bambuti men assist in cutting trees on the field of their village patron.
Fichier image/jpeg, 116k
Titre Fig. 1. Ituri Forest
Fichier image/png, 321k
Légende 2: Secondary settlement (malende) of village and Bambuti households.
Fichier image/jpeg, 100k
Titre Fig. 2.1. Pre-colonial fiel shifting
Légende Each household shifts his field within the limits of the territory of his chut/lineage.Note: Number of households/fields is reduced, households 1-4 representing Clan A, households 5-8 representing Clan B.
Fichier image/jpeg, 84k
Titre Fig. 2.2. Colonial and post-colonial corridor shifting
Légende Fields are grouped in two corridors shifted separately according to the division of land into clan territories.Vegetation typesprimary forest: blacksecondary forest: greyfield: white
Fichier image/jpeg, 80k
Titre Fig. 2.3. Colonial allotment of fields within the corridor
Légende Situation of a corridor in the third year. Fields are grouped in a single strip of land and shifted collectively in one direction. Plots are arranged according to lineage membership of households.field: whitefield of preceding year: light-greysecondary vegetation: dark-greyprimary forest/secondary vegetation: black
Fichier image/jpeg, 68k
Titre Fig. 2.4. Pre-colonial allotment of fields within the corridor
Légende Situation of a corridor in the third year. The double-sided corridor of the first year is split up into smaller sections, each moving in different directions.field: whitefield of preceding year: light-greysecondary vegetation: dark-greyprimary forest/secondary vegetation: black
Fichier image/jpeg, 104k
Titre Fig. 3. Rainfall Ituri Forest
Fichier image/png, 176k
Titre Fig. 4. Agricultural Work Calendar
Fichier image/png, 280k
Titre Fig. 5. Time Allocation
Fichier image/png, 318k
Légende 3: Forest clearance: a field corridor cut into a mixed stand of primary and mature secondary forest.
Fichier image/jpeg, 104k
Légende 4: Wanande immigrants panning gold at the bank of the Ituri river.
Fichier image/jpeg, 92k
Haut de page

Pour citer cet article

Référence papier

Michael Rösler, « Shifting cultivation in the Ituri Forest [Haut-Zaïre] », Civilisations, 44 | 1997, 44-61.

Référence électronique

Michael Rösler, « Shifting cultivation in the Ituri Forest [Haut-Zaïre] », Civilisations [En ligne], 44 | 1997, mis en ligne le 29 juin 2009, consulté le 20 avril 2014. URL : ; DOI : 10.4000/civilisations.1608

Haut de page


Michael Rösler

Ethnologue, Docteur en Ethnologie, Chargé de cours à l'Institut für Völkerkunde und Afrikanistik de l'Université de Munich ; auteur d'un PhD sur l'agriculture des Mbo, dans la forêt de l'Ituri, à l'est du Zaïre, et de plusieurs publications sur ce thème.

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page