Styles of making a living and ecological change on the Fon and Adja plateaux in South Bénin, ca. 1600-1990

This study is a comparative analysis of the joint development of the Fon and Adja styles of making a living as well as the ecological changes between the two adjacent plateaux in South Bénin on which they live. The period of analysis is between ca. 1600 and 1990. The South Béninese plateaux are usually described as a homogeneous category. However, popular opinion also holds that the Fon plateau is ecologically more degraded than the other plateaux, and that the Fon are socially more organised, technologically more advanced, and socio-economically more successful than the Adja. This thesis challenges the popular images about the Fon and Adja, and analyses how and why ecological processes on their plateaux differed between ca. 1600 and 1990.The two plateaux form part of a chain of plateaux in South Bénin and Togo. They have similar soils (Nitisols) with the same bimodal rainfall pattern and precipitation. The Nitisols are regarded as the best tropical soils for arable farming, but their fertility depends strongly on organic matter and clay content. This led in South Bénin, as in many other places, to high population densities on Nitisols. The Fon and Adja plateaux both had about 110 inhabitants per km ² in 1960 and about 300 inhabitants per km ² in 1990 respectively (the Fon plateau 20-30% less if the urban population is excluded). In 2002 the population density was 409 inhabitants per km ² on the Adja plateau and 377 inhabitants per km ² on the Fon plateau. Today, the Fon form the largest ethnic group, while the Adja are the second largest group in Bénin, accounting for 19.9 % and 8.6% of the total population, respectively. Ethno-linguistically they are closely related and their cultures have much in common. The two plateaux have the same distance to the coastal urban markets and since 1900, have been subject to fairly homogeneous government policies. Researchers and development interventionists alike tend to assume that any ecological, agronomic and socio-economic data from one plateau can be extrapolated to the rest. Popular belief, on the other hand, holds that the Fon and Adja differ. For example, since early colonial times the Adja are regarded as economi­cally and technologically backward, socially disorganised because they lack higher-level family structures and chiefs, and lazy because they till their land only superficially and do not sell much to export companies. The Fon are internationally known for their pre-colonial Danhomε kingdom and their predominance in all spheres of public life in the colonial and post-colonial state. Fon houses and compound walls are often made from cement-bricks, while the Adja live primarily in clay houses, often having no compound wall at all. In addition to this, the Fon are generally believed to have a more coherent family organisation and to be technologically more advanced than the Adja. So much so, that agricultural extensionists recommend the Fon's ridge tillage, their oil palm planting densities, and their commercial palm oil production to all South Béninese farmers and to the Adja in particular, who practise flat minimum tillage, plant more oil palms per hectare, and harvest more palm wine than oil from their trees.For several decades, degradation of the plateau soils has been a serious concern of policymakers and agronomists. Many argue that the high population densities inevitably lead to soil degradation. In other words, they believe that the plateaux have an intrinsic carrying capacity and that ecological change depends in a neo-Malthusian way on demography. Popular knowledge, however, holds that the soils of the Fon plateau are much poorer than those of the Adja plateau. It is also easy to see that their semi-spontaneous vegetation differs. The Fon plateau consists mainly of tiny herbs and grasses ( Cyperus esculentus, Digitaria spp., Brachiaria deflexa, Ipomoea involuncrata etc.) and grasses of 2 meter high ( Andropogon gayanus ), which tend to catch afire in the dry season. The Adja plateau has a greater variety of herbs, trees and shrubs ( Albizia zygia,Antiaris africana, Combretum hispidum, Mallotus oppositifolius, Dialium guineense, Dracaena arborea,Di­chrostachys glomerata, Securinega virosa, Zanthoxylum zanthoxyloides etc.), and the principal grass is the medium-sized Imperata cylindrica . Bush fires do not occur on the Adja plateau.This leaves all those who believe in a linear relationship between population density and agro-ecological change, both the (neo) Malthusians like Homer-Dixon (1999) who regard population growth as a threat, and the followers of Boserup (1965) who consider population growth to be an opportunity for agricultural growth, with the mystery of why Fon and Adja plateau soil fertility levels and vegetations differ in spite of similar demographic conditions. They raise various assumptions in order not to abandon their cherished beliefs in population density models. Many think that the Fon plateau is more densely populated than the Adja plateau, but demographic figures reveal that this is not the case. Others hypothesise that the Adja plateau was more fertile and more forested than the Fon plateau in its 'original' state.My thesis argues that the Fon and Adja plateau ecologies were similar in the past but diverged under the impact of different human management practices. To 'test' the hypothesis that the two plateaux were ecologically dissimilar before their human occupation, I adopt two approaches, namely an oral history approach and a comparison of processes. In Chapter 4, I analyse local myths of origin of villages and local historical narratives about livelihood activities in their socio-political and technological context and compare these with palaeontological evidence from, amongst others, a lake 20 km from the eastern border of the Fon plateau. My compilation of local traditions, stripped of their likely socio-political intentions, portrays the vegetation of both plateaux as a forest-savannah mosaic when they were colonised. According to various myths on their origin, on both plateaux some villages were installed on grassland, others near isolated trees, and others in more forested areas. These local histories therefore do not require us to reject the opinion of ecologists and palaeontologists that all the South Béninese plateaux were covered, since the end of the last wet period, not later than 950 AD, with a savannah-forest mosaic of the type which is still the dominant spontaneous vegetation of the Adja plateau (and the other plateaux in the chain) today, and that only the Fon plateau is degraded.The triangulation of migratory myths from many different sources and localities also indicates that the Fon and Adja plateaux became more densely populated from the 16 ^th century onwards, the Adja plateau mainly by Adja from Tado who brought their own iron tools, and the Fon plateau by various Adja-related peoples (Wemenu, Za, Ayizo, Jinu, etc.) and a small Yoruba-related group (the Gedevi) that was socio-politically dominant over its Adja-related neighbours. This Yoruba-related group and visiting traders from the north-east introduced iron tools from the Yoruba and the Bariba to the Fon plateau, especially hoes which were suitable for ridging. The Adja hoes from Tado however, were only suitable for flat cultivation and for mounding. When more iron became available in the 16 ^th century through the arrival of European traders on the coast, the Yoruba hoe and ridge tillage rapidly spread on the Fon plateau, while the Adja plateau was increasingly colonised by flat-cultivators using Adja hoes. Therefore, the different orientation of the Fon and Adja's socio-political and tool trade networks — their different socio-technical networks, encouraged the development of different tillage styles not later than the 16 ^th century. The ecological impact of these tillage styles is discussed in Chapter 9 (see below). Around 1610 another Adja-related group arrived on the Fon plateau, the Agasuvi, who became accepted by the resident population as their royal family. In Chapter 5 I show that the Fon, under their leadership, formed a kingdom called Danhomε, whose strength resided in the centralisation of weapon production and of military power around a few smithies, the centralisation of religion around a number of State cults, and in the promotion of a warrior ideology. From the latter part of the17 ^th until the late 19 ^th century, the Fon raided neighbouring groups, selling many of them to transatlantic slave traders, and retaining others for their own domestic and agricultural work. Trade in other local and imported commodities also flourished in Danhomε at that time, and every adult - including the more fortunate slaves - could participate in it. Contrary to what Polanyi (1968) wrote about Danhomε to support his substantivist theory, the king did not control trade except for that relating to the court's standing army. My findings also do not support attempts in the literature to characterise Danhomε's economy with a single label. And neither does Elwert's (1977) 'slave raiding mode of production articulated to a subsistence mode of production', nor Coquery-Vidrovitch's (1971) 'tributary feudal system', nor Manning's (1982) 'commodity exchange mode of production joined by a slave labor mode of production' describe it sufficiently. What I do show is that the elites' urbanised styles of making a living, including trade, warfare, forging, religious activities as well as weaving and wearing of prestigious cloth, rose in status in the Fon kingdom, while rural life and agriculture became stigmatised. Nukanmε , literally 'secondary bush', became a Fon synonym for backwardness and was a derogatory label for the countryside, in general. Inhabitants of rural areas who neither traded nor engaged in 'urban' crafts were called nukanmεnu or 'backward people of the bush' . At the same time, for the Adja, bush and countryside were central to their wealth and safety. During the era of the slave raids, the Adja had no other defence than to hide in small villages surrounded by woody vegetation, to engage in agriculture, and to avoid long distance trade in areas scoured by slave raiders. The Adja acquired wealth and prestige by working hard in the fields in small domestic groups.Around the mid-19 ^th century, overseas demand gradually shifted from slaves to palm oil and their kernels, opening, in principle, the same commodity production opportunities for all Fon and Adja plateau farmers, because oil palms grew spontaneously on both plateaux. In Chapter 6, I show that Fon farmers responded by planting oil palms and trading in oil and kernels, first only on communal lineage land, and later in the 19 ^th century, also on individually-owned land. It became a sacrilege to 'kill' a Fon oil palm by felling it. Contrary to what the literature asserts, there was no compulsory palm oil tax for all Fon farmers. Farmers were mainly motivated by trade opportunities. Because the fallow vegetation of the Fon plateau consisted by the 19 ^th century primarily in the fire-prone grass Andropogon gayanus and because fire endangers but tillage benefits oil palms, Fon farmers developed various strategies to keep their palm groves free of weeds during the dry season — permanent cultivation being the preferred technique. This enhanced palm fruit yields in the short-run but depleted the soils in the long-run. Not later than 1850, the Fon plateau started importing food from the Adja, and between the mid-19 ^th and the mid-20 ^th century, some Fon settled on the north-eastern Adja plateau to produce food and palm oil, sometimes with the help of slaves. Now many central Fon plateau soils are too poor to produce staples, and only the oil palms yield fruit, which is then sold for food.The Adja also had oil palm groves, but these were usually so densely grown with palms and semi-spontaneous bush that they produced little fruit. The Adja felled these palms at the age of 20-25 years, tapped their trunk, and sold the obtained wine at the local markets. The new palm oil export opportunities did not motivate the Adja men to plant their palms less densely and to sell much more oil, though some Adja women independently gleaned windfall palm kernels and sold these on their own account. Rather, when a distilling technique was introduced around 1920, Adja men gradually increased their oil palm planting densities from 600-1000 palms per hectare in the earlier to 1000-1600 palms per hectare in the later 20 ^th century, distilling more and more sodabi for sale. During the first 6-8 years, annual crops are grown between young oil palms; then oil palms and bush are allowed to cover the land during ca. 10-15 years. Grasses disappear during this 'oil palm fallow' period and the soil restores its fertility to some degree. This 'wine palm' management style allows the Adja to plant much more palms per hectare than the 200 which Zeven (1967) and Hartley (1988) regard to be the maximum. These two oil palm 'experts' also think that oil palm densities drop below 200 per hectare if the human population density increases beyond 250 inhabitants per hectare. However, this population density was reached on the Adja plateau around 1986, and aerial photographs indicate that the average palm density of the plateau was already around 500 palms per hectare (land without palms included) in the same year. Some Fon in the frontier zone experiment with intermediate oil-wine palm management styles (densities and ages which hold the middle between those of the Adja and the Fon plateau) but they do not introduce these to their native Fon plateau villages, where farmers remain reluctant to 'kill' the palms which enable them to eat.In 1900, the French colonisers exiled the Fon king Agoli-Agbo and the Adja's chief of the land Kpoyizun and replaced them with several chefs de canton . Chapter 7 shows that from then onwards, the colonial and post-colonial governments applied fairly homogeneous policies on both plateaux. Most policies encouraged the production of the same agricultural commodities (i.e. cotton, coffee, tobacco, palm oil, and until the 1960s groundnuts and castor bean) by means of the same farming techniques in the whole plateau zone (ridge tillage, palm oil production from hybrid oil palms, and from the 1960s onwards fertilising cotton and ploughing). The commercial production of maize was discouraged in most years, and felling oil palms was first forbidden and then subjected to a fee. Contrary to the common assumption that Africans either cling to autarky or respond positively to commercialisation policies and programmes by producing the commodities which the government demands, the Fon and Adja developed diverse market oriented styles of making a living, but most of these diverged from those encouraged officially. Until the 1930s the Fon pleased the administrators by selling groundnuts, cotton and palm oil to export companies, by paying their taxes promptly, and by collaborating in a general sense with the colonial government. Then they dropped the cultivation of cotton on the plateau, and increasingly sold their palm oil and groundnuts through private traders to West African consumers, and abandoned plateau agriculture more and more. The Adja sold, besides castor and more recently cotton, large amounts of food to West African consumers - especially maize, gari , tomatoes, chilli peppers and sodabi - which went quite unnoticed by the official economic statistics.Chapter 8 presents the livelihood activities of members of some Fon and Adja lineages since about 1900. These family histories illuminate how individual actors motivate and evaluate their own practices. They illustrate in general a persistent Adja esteem for agriculture and pride for working hard on the land, and a Fon preference for trade, crafts, white collar work, and spiritual income-generating activities — in the Fon lineage studied in more detail the trade in magic charms and 'medicine'. They show how most Fon hardly objected to acquiring food on the market, and how the Adja's primary goal remained self-sufficiency in maize. They also describe how actors worked with or for each other within social networks. Adja school children and teenagers tended to work more on their parents' land without payment than most Fon children in the same categories; the Fon encouraged their children to develop non-agrarian skills. Members of the same kinship, village and religious networks often cooperated in crafts and trade, acquiring skills in this way; they migrated to the same destinations, or helped each other to find employment in the same company. This led in many cases to family and village specialisation in livelihood activities, illustrating how networks may encourage style formation.Chapter 9 analyses the interaction between Fon and Adja tillage and manuring styles and the ecological environment. It shows that the Fon's ridge tillage eliminates vegetation, especially trees, shrubs and grasses with rhizomes, more effectively than the Adja's superficial flat tillage. Therefore the Adja have to weed their crops more frequently, and Adja fallows produce more rapidly a large and woody biomass than is the case among most Fon. The Fon's clean weeded ridges also encourage the erosion of clay from the cultivated layer. In response to the savannisation of their plateau, Fon blacksmiths around 1940 invented a scythe which is suitable for clearing savannah grasses. This new tool spread through local markets within one decade to all Fon farmers on the plateau, illustrating how technology may travel through indigenous trade or socio-technical networks. The Adja reacted amongst others by planting, on land infested by Imperata cylindrica, tomatoes and chilli peppers on mounds for urban markets, in rotation with dense plantations of oil palms, cassava, pigeon pea, or Mucuna pruriens , in order to uproot and out-shade this grass with rhizomes. Fon and Adja women increasingly manured fields near the village with household waste and crop residues. Since 1980, Adja men and women have also purchased fairly large amounts of mineral fertiliser for their tomatoes, cotton, maize and occasionally cowpeas, for which they pay cash unless they apply it to cotton. Their fertilisation of local food crops is exceptional for Africa.Chapters 6 to 9 describe how under similar population densities, the Fon plateau degraded as expected by Homer-Dixon and other (neo)-Malthusians, and many of its inhabitants withdrew from farming there, while the Adja reacted rather as Boserup predicts, by indigenous agricultural innovation and by increased labour inputs per unit of land in order to obtain higher returns from it. Chapter 9 shows that the Adja devote between 1.5 and 5 times more labour to one hectare non-irrigated annual crops than the Fon, and that irrigated Adja tomato culture is about 8 to 12 times more labour intensive than Fon maize cultivation. The popular image of Adja laziness therefore clearly does not hold. These different labour needs are the result of their different tillage styles and crop choices.My comparative historical analysis, based on a variety of research methods including ethnographic ones, provides insight into the roles played by local actors and their socio-technical networks, and explains why developments on the Fon and Adja plateaux diverged in spite of similar external and demographic conditions. This shows that none of the systems models of Malthus, Boserup, Homer-Dixon, Zeven and Hartley is intrinsically right; this cannot be because they all neglect the role of human agency. My diachronic comparative study reveals that neither population growth as assumed by Boserup, nor integration into large-scale political, financial, educational and research institutions as assumed by Homer-Dixon, are sufficient conditions for environmental ingenuity and sustainable agro-technological innovation. The thesis shows how clusters of similar practices emerged in the historical process and how, on several occasions, these clusters overlapped with regional vertical and horizontal social relations, including kinship ties and trade networks, and that these practices had socio-cultural value and meaning for the Fon and Adja people. The concept of styles was used to designate both these clusters of meaningful practices and their description in sometimes ideal typical terms. The historical analysis shows that social actors sometimes aspired to the lifestyles of those members of society who already enjoyed the esteem of the population, as also observed by Bourdieu (1979) and Hofstee (1985). Other style elements travelled more horizontally along socio-technical networks. Kinship and neighbourhood ties were however no guarantee for style and knowledge dissemination. Fon and Adja knew of each others' styles, especially in the frontier region, and some individuals experimented with some elements of their neighbours styles, but there was no general trend of style diffusion. The combination of network and historical analysis in this study was therefore necessary to understand the emergence of styles, while the holistic comparison drew attention to crucial factors and points where processes diverged. Moreover, socio-cultural valuation of agrarian versus other types of labour appeared pivotal for explaining preferences for certain livelihood activities.