The horizon is generally conspicuous, whereas the surrounding landscape, close to us, is taken for granted. Yet, in the social and historical world (and perhaps in some measure in the physical world also), the horizon we perceive depends on our more immediate environment and its general features. A forest or a savannah, a local hillock or a hollow, make a great difference to the kind of skyline that is seen. But it is in the nature of things that what is close and familiar should also be treated with familiarity and contempt, and that its importance should normally be ignored. It is its ordinariness, obviousness, which causes us to take it for granted: but the hold it has over us is immeasurably strengthened precisely by the fact that we do take it for granted. What is noticed can be queried, but that which seems utterly obvious eludes questioning. The horizon, by contrast, errs in the opposite direction. It is often quite spuriously dramatic. If you walk to the point on the distant skyline, you may well find, when you reach it, that it is just as ordinary a place as your starting point. But as long as it remains on the skyline, it occupies the dramatic point at which the sky meets land or water, the point where the sun sets or rises. It has a striking suggestiveness, and symbolises our deeper or more ecstatic aspirations—quite unlike the close and dusty immediate locality, which tends instead to remind us of our compromises, shabbiness and mediocrity.

Par un paradoxe tout à fait conforme à la théorie webérienne de la relation entre Ies intentions des agents et le sens historique de leurs actions, la contribution la plus importante que Max Weber ait apportée à la sociologie de la religion se situe sans doute sur un tout autre terrain que celui qu'il a choisi pour son affrontement de toute une vie avec Marx. Si, dans son effort obstiné pour établir l'efficacité historique des croyances religieuses contre Ies expressions Ies plus réductrices de la théorie marxiste, Max Weber est parfois conduit à une exaltation du charisme qui, comme on l'a remarque, n'est pas sans évoquer une philosophic «héroique» de l'histoire a la manière de Carlyle, par exemple lorsqu'il désigne le chef charismatique comme « la force révolutionnaire spécifiquement ‘créatrice’ de l'histoire » (I), il reste qu'il fournit lui-même le moyen d'échapper à l'alternative simpliste dont ses analyses Ies plus incertaines sont le produit, c'est-à-dire à l'opposition entre l'illusion de l'autonomie absolue portant à concevoir le message religieux comme surgissement inspiré et la théorie réductrice qui en fait le reflet direct des conditions économiques et sociales: il met en effet en lumière ce que Ies deux positions opposées et complémentaires ont en commun d'oublier, à savoir le travail religieux que réalisent Ies agents et Ies porte-parole spécialisés, investis du pouvoir, institutionnel ou non, de répondre, par un type déterminé de pratiques ou de discours, à une catégorie particulière de besoins propres à des groupes sociaux déterminés.

I am not the first to discover the pitfalls which surround the term ‘revolution’, as if to guard its explosive practical potency from scholarly vivisection. Eight years ago Hannah Arendt devoted more than three hundred pages of her-as usual-magnificent erudition to dissecting the elusive meaning of the term. She established that the term was used for the first time in 1688 to denote “bringing things back” rather than “pushing them forward”; she reminded us that when the duke de la Rochefoucauld-Liancourt, on July 14th, 1789, retorted bluntly: “Non, Sire, c'est une révolution” to the Louis XVI' abashed “C'est une révoke”, he had been inspired by the impressive inevitability of freshly discovered motions of celestial bodies, rather than by modish ideas of les philosophes du progrès All this said and done, Hannach Arendt ends up with one more partisan definition, which is exactly the result one could have expected. A quarter of century before this, Crane Brinton choose the seemingly most reasonable path of comparing four, historically well described, events, which nobody would refuse to call ‘revolutions’—only to show once againthat it is the theoretical assumption which shape the empirical data rather than the other way round. Only two years had elapsed at that time since the Henderson-Livingstone translation introduced the Pareto's concept of ‘equilibrium’ to the American scholarly scene. As it has not yet been turned by Talcott Parsons into a symbol of abstract academic sterility, the concept could have seemed to Brinton the right tool of selecting the ‘revolutionary essence’ from the host of ‘empirical data’ on revolutions.

Durkheim's manner of thought and his style of expression have certain peculiar features the recognition of which can only clarify the interpretation of his ideas. First, there are certain concepts crucial to the understanding of his thought which need to be elucidated, either because they are ambiguous or because they are unfamiliar to a modern reader. Second, underlying these concepts, there are a number of sharp dichotomies, or “binary oppositions”, on which his thought rests, which need to be made explicit and related to one another. Third, there are a number of characteristic, and often bad, arguments, which likewise need to be brought to the surface and identified. And finally, Durkheim's style often tends to caricature his thought. He often expressed his ideas in an extreme or figurative manner, which distorted their meaning and concealed their significance: hence the relation between his style and his thought must be considered. It is to these matters that this article is devoted.

Bauman's Schumpeter-like forecast concerning the prospects of revolution in socialist societies calls for a reply which takes issue not only with the details of his analysis of the tensions within these societies, but also with some semantic aspects of his conceptual framework: the very concept of revolution and of political legitimacy as applied to these systems.

During his life-time, Durkheim's methodological writings were notoriously subject to controversy, and his ‘sociologism’ was widely condemned. These early critiques, often totally misinformed as to the real character of Durkheim's views, have long since ceded place to critical interpretations of Durkheim's writings which are founded upon a more adequate understanding of the themes and the dilemmas inherent in his sociology. Nonetheless, it is arguable that we still await a treatment which fully explores the strengths and weaknesses of Durkheim's method. One main reason for this is that most secondary interpreters of Durkheim have failed to connect his analytical discussion (and rejection) of individualism as a methodological approach to social theory with his developmental conception of the emergence of individualism as a morality brought into being by the growth of the differentiated division of labour. It is commonly accepted-and, indeed, he himself stressed this very strongly-that Durkheim's methodological ideas must be evaluated in relation to their concrete implementation in his more empirical works. But this is generally taken to mean showing how successfully or otherwise he ‘applied’ his methodological views in his other studies. The point I wish to make in this paper is that there is a reciprocal relationship between Durkheim's substantive discussion of the development of individualism and his abstract formulations of sociological method. Durkheim is often regarded as being fervently “anti-individualist”. But in fact his works contain a vigorous defence of individualism—understood in a specific way. In other words, Durkheim's writings represent an attempt to detach “liberal individualism”, regarded as a conception of the characteristics of the modern social order, from “methodological individualism”.

Quiconque s'est occupé du problème des révolutions n'a pas manqué de se heurter à celui, combien irritant, de la définition. Le mot de révolution s'est chargé, au cours des siècles, de tant de résonances puisées à la réalité ou aux aspirations, que le même mot peut recouvrir des champs scientifiques — sans parler des champs idéologiques — les plus divers (1). II semble, pourtant, qu'on ne puisse pas éviter d'inclure dans son acception une référence au pouvoir et 'entendrai done par révolution toute prise illègale du pouvoir. La définition retenue est suffisamment large pour designer des faits aussi variés qu'une révolution de palais, un coup d'État militaire latinoaméricain ou les révolutions anglaise et francaise. Je maintiens qu'une définition aussi large est nécessaire, si l'on veut éviter le risque de manquer des phénomènes essentiels à la compréhension des sociétés. La littérature sociologique sur les révolutions — combien pauvre et limitée — manifeste le mépris le plus définitif pour les coups d'État militaires. Ce qui ne laisse pas de m'étonner. Car de quel droit décidet-on du degré d'intérêt scientifique d'un probléme? Si un coup d'État peut être de peu d'enseignement, leur retour périodique est, par contre, l'indicateur le plus important de l'état d'une sociét. De plus, une définition plus étroite que la mienne repose sur une décision arbitraire, qu'une autre décision tout aussi arbitraire peut venir contredire.

In the Preface to the second edition of Les régies de la methode sociologique (1901) Émile Durkheim defined sociology as “the science of institutions, of their genesis and functioning” (R XXII). In the same text, however, he stated:

In the present state of the discipline, we really do not even know what are the major social institutions, such as the state or the family, the right of property, or contract, punishment and responsibility. We ignore almost entirely the causes on which they depend, the functions they fulfil, the laws of their evolution; we barely begin to perceive some light on a few of these points (R xv).