The last three chapters focused on the changes that have unfolded in Konso, and the way they have impacted on the institutions that are central to the production of Konso indigenous agriculture and landscape. Together with the earlier chapters, they emphasize the way in which material struggles over land and labour are played out in discursive struggles over the legitimacy of different beliefs and forms of identification (Konso ‘custom’, Protestant Christian, Orthodox Christian). In this conclusion, I return more explicitly to the landscape and examine what can be learnt from the material presented here about how landscapes like Konso are produced and maintained over time. I also explore the significance of this account for development practice and policy.

The principal message is that the terraced landscape of Konso is a cultural landscape; it is produced out of networks of collaborative action which are also networks of meaning. Historically, the indigenous institution of the poqalla has been central to the network of meaning, and has helped to produce the landscape and the indigenous intensive landscape that it supports. The poqallas have been powerful institutions because they were intimately connected to others in Konso, who identified with them personally: the poqallas were their relatives; they represented, structurally, every Konso person's father and their origins. The poqallas also performed multiple functions, providing material, juridical, social and cultural support. In rituals, the poqallas were thought to be able to influence forces beyond the control of ordinary mortals.

The previous two chapters explored the way in which the values and practices of the Derg government and Protestant Christians intersected. The relationship between these two bodies was not straightforward as they did not share the same values, but they shared some of the same goals. They shared a desire to eradicate aspects of culture they found harmful, and they were both modernizing in their own ways. The outcome of this convergence of aims was a ‘discourse coalition’ (Hajer, 1995). The Protestant Christians became some of the main supporters of reforms brought in by the Derg. As a result Konso society became bifurcated. Two groups of people, discourses, and ‘ways of doing things’ emerged, which are described here as corresponding to Bourdieu's orthodoxy and heterodoxy. The orthodoxy was made up of customary practices based in the belief that ‘they are the way things have always been’. They had the legitimacy of ‘tradition’, the power of the belief in ‘das “ewig Gestrigen”’, ‘the eternal yesterday’ (Weber, 1978). The heterodoxy was made up of a combination of appropriated Derg policies and Protestant Christian teachings. Whereas the orthodoxy involved a commitment to the indigenous institutions of the poqalla, to customary tenure and to indigenous networks of labour access, the heterodoxy involved a commitment to state-defined land tenure, and rejected indigenous practices as ‘devil worship’. The heterodoxy saw itself as ‘modern’, the customary as ‘traditional’.

Chapters Two, Three and Four examined the way in which the production and reproduction of the landscape in Konso is part of a social, cultural, ritual and political process of which the poqallas are at the heart. The power of the poqallas rests in their control over and embodiment of forms of social, symbolic and economic capital. The study of the poqalla shows that the institutions for managing land and labour are embedded in other aspects of society to the point that the economic, the ritual, the judicial, and the political cannot be separated. This raises questions about the usefulness of the term ‘embeddedness’, often used to refer to non-economic aspects of production in economics. The term ‘embedded’ implies that the economic (or productive) practices are grounded within existing social networks and cultural beliefs and practices, from which, therefore, it could be separated. This research shows that the economic, ritual, social and political are mutually constitutive. Everyday life in Konso (and elsewhere) is made up of interactions and exchanges between people which are at once social, cultural, ritual, economic and political; it is difficult to prioritize any one dimension of an exchange over another. To separate them would be to impose dichotomies and divisions between different sectors that do not exist in practice.

The last chapter compared data from oral histories, from burial statues, and from the political events from the 1890s, to argue that the role of the poqalla in the production of the landscape is ancient.

In this chapter, the study of the poqallas is developed further to examine more critically the nature of the power relations that exist between the poqallas and others. The previous chapters have demonstrated that the poqallas play a role in turning the stony grounds into fertile ones, which means that they exist as a form of institution central to the intensive agriculture. Development organizations have ‘discovered’ the importance of the institutional dimensions of environmental management in recent years, and indigenous institutions are seen as particularly valuable resources because they are considered to be already existing institutions that can potentially be harnessed for environment and development ends. These approaches to working with indigenous institutions tend to focus on identifying and illustrating the role that they play in environmental management. In arguing for these institutions' positive role, there is little time to think more critically about the nature of the power relations between these institutions and the others in the community. If anything, development organizations normally assume that the institutions are the legitimate representatives of the community and working for the common good.

This chapter argues that development organizations need to think more carefully about the nature of the power of the institutions with which they are working. For example, are they simply representatives of the community among whom they live and institutions that provide a coping mechanism for the poor and sustain the landesque capital?

The 1974 revolution heralded the beginning of a new period in Ethiopia. The Derg government penetrated the structures of government more deeply into the grassroots communities of southern Ethiopia than any regime that had gone before. In this chapter, the general history of the revolution is reviewed, setting the context for the discussion of the experience in Konso. As well as exploring the way the revolution impacted on Konso, the way in which the ideas and practices associated with the revolution combined with those of Protestant Christians, described in the last chapter, is examined. The combination of the modernizing forces of Protestant Christianity and scientific socialism had severe impacts on the institution of the poqalla. As the modern project failed, however, some Konso began to seek alternative forms of meaning and organization, and returned to some of their older institutions, including the poqalla. Here the orthodoxy of customary practices and the institution of poqalla show their resilience.

The impact of the revolution and the Derg period in Ethiopia

The revolutionary government was known as the Derg, from the Amharic word for ‘committee’. The events surrounding and following the 1974 revolution in Ethiopia have been described in some detail by Clapham (1988), Dessalegn Rahmato (1984), Ottaway and Ottaway (1978), Pankhurst (1992), Donham (1999) and James et al. (2002). The revolution brought some dramatic transformations, including changes in state administrative structures at the local level, and a land reform programme implemented in 1975, which has been described as one of the most radical agricultural reforms that have been set in motion in sub-Saharan Africa (Dessalegn, 1984).

The poqallas' control over larger amounts of land gives them access to more labour, which allows them to invest in the construction of the landesque capital. As institutions, the poqallas are structures of power, and the material presented in the previous chapter suggests that their power is derived mainly from their economic control, and that a poqalla can be perceived as a kind of landlord. The revolutionary Derg government certainly took this view when it assumed power in 1974, and accused many poqallas of being landlords who exploited the people. The extent to which the poqallas can be justifiably viewed as landlords is examined in more detail in Chapter Four; Chapter Six looks at the way in which these government assumptions impacted upon the position and role of the poqalla after the revolution. In this chapter, I examine the nature of the different exchanges between the poqallas and others in the community and the way in which they go far beyond the realm of the economic. In the literature, the poqallas have first and foremost been understood as sacred ritual leaders or priests, and it is this aspect I consider here. This chapter argues that the ritual role of the poqalla is as important to the production of Konso landscapes and livelihoods as the economic role, and that the ritual and the economic cannot be separated.

The customary agricultural practices and institutions that are described in this Part of the book are those practices and power relations whose legitimacy is grounded in the claims to tradition, to the ways things have always been and should be. Using Bourdieu's terms, this represents the orthodoxy in Konso. The title of this specific chapter is borrowed and developed from Appadurai's edited book, The Social Life of Things (1986). Others have taken up his approach before now, for example Nyerges (1997) uses his ideas to talk about ‘the social life of resources’, and Longley (2001) uses his ideas to talk about ‘the social life of seeds’. For these authors, Appadurai's approach to things is useful because it sees them not as inert, but as objects with ‘careers’, ‘life histories’ and ‘biographies’, which are ‘fully part of social life’ (Nyerges, 1997: 12). Thus the environment is presented not merely as a backdrop to social life, but as deeply inter-connected with, and part of, that social life. This approach has therefore much in common with the approach to agriculture as a social process discussed in the introduction. In the next chapter, the approach is taken further to show that not only are the practices and institutions part of social life, they are also part of ritual life. The title the ‘Ritual Life of Agriculture’ is used to show that these social processes are also cultural processes too, replete with ritual actions, relationships and meanings.

The Konso people live on and around a small range of mountains, some 600 km south of Addis Ababa, in the Rift Valley of Ethiopia (see Map 1.1). These mountains rise to a height of 2,500 m, and from above 700 m they are scored with neat, dry stone terraces (Amborn, 1989). To the outsider, the area is distinctive in many ways: the terraced landscape marks it out as different from the hot lowlands that surround the mountains, and from the other cultivated mountains in the area. In the area marked on the map, they also speak their own language, Afa Xonso, an Eastern Cushitic language. It has some links to Oromiffa spoken by the Boran and the Guji to the east, and other links to the Dirashe to the north-west, but the languages are not mutually comprehensible. The Konso are also recognizable by their style of dress: women often wear thick home-spun cotton skirts which are gathered and folded into two layers; the men sometimes wear home-spun and woven cotton shorts, often striped and brightly coloured.

Konso villages are nucleated, walled and extremely dense and compact. The density of the settlements and the way in which they are criss-crossed by many narrow and labyrinthine paths led Hallpike (1972) and others, Sutton (1989), to describe these settlements as towns. Though in many ways this word is appropriate, it is less confusing to describe them as villages, distinguishing them from the more urban recent settlements of, for example, Fasha and Karate, both of which have grown up around market places. Karate is the main town, and the location of government offices.