Political sociology has from its very inception had an overriding concern with the nature of political order and stability, and the threats to that stability. Ever since ‘the entry of the masses on to the stage of history’, at the time of the French Revolution, one source of that threat has regularly been seen as the industrial working class. That has been so, whether the threat was perceived by the liberal centre and conservative right; or whether is was converted, by the left, into a definite promise to overthrow ‘bourgeois’ stability. In both cases, in the anxious speculations of Mill and Tocqueville as much as the triumphant predictions of Marx and Engels, a key role was marked out for the developing working class of nineteenthcentury Europe.

I shall deal in this paper with the Jagannath temple of Puri which nowadays constitutes one of the biggest remaining North Indian temple and pilgrimage centers. This town of 60,000 inhabitants still draws approximately one million pilgrims annually from all over India. Puri is situated on the seashore of the Bay of Bengal and is the greatest temple town of Orissa, an Indian province lying adjacent in a westerly direction to the province of Bengal. The present temple was constructed in the eleventh century, and the God's name, Jagannath, which translates as ‘ruler of the world’, spells out well the intent of the founders of the temple, the Ganga, the main dynasty of medieval Orissa. The name was appropriate, because the history of the Gangas shows that they dedicated their kingdom to Jagannath and they only ruled as regents or ministers on behalf of their deity. In the context of the overall Indian high tradition, Jagannath is regarded as the ninth incarnation of Vishnu. The sculpture that is actually worshipped in the temple is singular in the Indian setting. It is the only wooden idol which receives veneration, and in addition, this sculpture is, compared to the conventional blue-stone images of Vishnu, of a particularly crude, quasitotemistic style. This strange, iconographic artifact has never deterred the Oriyas from lavishing a splendid ritual in honour of their Overlord. In the course of thirteen daily main rituals, the God receives three meals, the main lunch consisting of fifty-six different dishes (the temple kitchen houses 270 hearths and ovens); he is dressed and undressed four times a day; and some 10,000 priests organized into 108 different service-groups attend to his needs and desires.

Much has been written in the last two decades on clientelism and patron-client relations. From a concept of relatively marginal concern in the social sciences, clientelism has gained analytical clarity and has been applied with greater precision to political and social relationships found in many contemporary societies.

Since Hobsbawm published his Primitive Rebels (1959) and Bandits (1969), the study of organized banditry is slowly moving out of the dusky fringe of history and social science. In the 1970s some interesting regional and biographical studies have come to light. Apart from the encouraging example set by Hobsbawm, this has largely been the result of the increasing rapprochement between history and social science. Ordinary, rural and marginal people are being set on the stage of history. However, detailed comparative local and regional studies are still needed to test Hobsbawm's frame of analysis. Moreover, as he himself admitted, the evidence on which Bandits relies—poems and ballads—is rather ‘tricky’ and needs to be supplemented with other sources.

If proverbs are sociologically significant the one quoted encapsulates the theme of this paper: the extreme fluidity and uncertainty of post-Independence Greek Cypriot politics; the tension between individual strategies and collective action in Greek culture; secrecy, and the use of violence for political purposes (what Aristotle had called stasis). In this paper I want to examine the patterns of covert politics in Cyprus by tracing the evolution of the cultural phenomenon of omadhes (small secret groups of armed men activated for political purposes), and examining their relationship to political parties. Although there has been some literature on omadhes (Loizos, 1975; Markides 1977) I shall offer a different interpretation to Loizos who sees them merely as ‘action sets’ by showing that they became something more substantial and were strongly influenced by the overall political framework within which they operated; I shall also take a different line to Markides who treats them as an expression of discontent.

An enormous amount of scholarly attention has been devoted to the phenomenon of patron-client relations both in the form of conceptual elaboration and to the application of the patronage model to a wide variety of empirical situations. However despite the prodigious amount written on the relationship its analytical status remains equivocal: no one, for example, has been able to say with any degree of precision what patron-clientage is, and especially where patron-clientage ends and the reciprocity which pervades all social relations begins. But a certain lack of clarity has not deterred social scientists from resorting to the patron-client model and few studies of social and political change in the Third World manage to get by entirely without it.