Perspectives on Berber Politics: On Gellner and Masqueray, or Durkheim's Mistake
Roberts, Hugh, Journal of the Royal Anthropological Institute
The fact that the scientific pretensions of political science have driven many of its practitioners to ape their counterparts in more exact disciplines, notably economics and econometrics - with the result that many studies in political science today are abstruse exercises in logic or algebra as much as anything else -- may not matter in the developed democracies of Europe and North America, where the long-established discipline of political history is available to compensate for the way in which much political science now approaches or retreats from its duties. But it is liable to matter a great deal in the countries of the South and the East, where the discipline of history is less firmly established. It is not for nothing that anthropology has tended to step into the breach, and that the works of anthropologists often furnish the best introductions to the politics of much of Asia and Africa, including the Middle East and North Africa.
But how does contemporary anthropology conceive politics and approach its study? Is its conception fundamentally different from that of sociology? Or is the difference confined to secondary matters - for example the substitution of kinship or caste for class - in an otherwise substantively sociological theory of politics which assumes the social structure to be the locus of the main action, with room for argument about the relative importance of process' and 'agency', but no recognition of 'the political' as constituting a distinct sphere that can be properly apprehended only in its own terms?
Two particular concerns have prompted me to raise these questions. I am a student of North African, and particularly Algerian, politics. I am therefore concerned about what has become of political studies, and I am even more concerned about what has become of Algeria. Over 100,000 people have reportedly been killed in the violence which has ravaged the country since 1992, but for much of the time it has been far from clear who has been killing whom or why; as numerous observers have regularly complained, Algerian politics are unusually 'opaque'. While there are undoubtedly objective reasons for this, to do with the character of the Algerian polity, part of the problem is located in the social sciences themselves and has its roots in the way in which anthropologists have approached the study of the political organization of what one might call the spinal column of Maghribi society, the Berbers of the Atlas mountains. My thesis is that over the last hundred years the political anthropology of North Africa lost the plot, with consequences which are still being felt in political studies in North Africa and beyond.
Berber political studies: the problem stated
Algeria and Morocco form the core and greater part of the single field of Berber studies. The study of Berber politics in the modern era began in Algeria in the nineteenth century, but has throughout the twentieth century been dominated by work done in Morocco. This article outlines the point of view which I developed, while carrying out fieldwork in the Kabylia region of Algeria in the 1970s and 1980s, (1) on a field of study in which Moroccanists have long held the theoretical initiative. But what is at issue here is not a particular, let alone intrinsic, Opposition between Moroccanists and Algerianists, such as might be of no interest to anyone else. The opposition in question is of a different and more significant character. It is also to be distinguished from that between contemporary critics and defenders of the theory of segmentarity. While I take issue with the champion of this theory, the late Ernest Geilner, on one cardinal point, my difference with him has nothing to do with the objections which ha ve been levelled against his theory by the disciples of Clifford Geertz in the 'interpretivist' or 'hermeneutic-relativist' school in Moroccan studies (Gellner 1991), and is also distinct from those made on empirical grounds by Henry Munson (1989; 1993). …