These two fundamental ideological types involve two different types of rationality. In the case of a cosmocentric order of the World, built before and independently of Man, the latter can but ceaselessly guide himself according to the requirements of this order. He can only do what was "settled", in conformity with the "ranging" of the World, as this was conceived from its origins. It is therefore rational for us to direct our understanding efforts towards the origin of this order and to be guided by the deeds of others in the past, who were as "dependent" on the ranging of the world as us. Therefore, having its foundation in the past, as sole source and guarantee of knowledge, rationality is retrospective. On the contrary, in the case of an anthropocentric rule of the game, of Man's autonomy in the World, it is rational for us to choose those appropriate means in order to attain--in a more or less distant future--Man's autonomous aims and the order that he tries to impose upon himself. We can talk in this case about a prospective rationality (forward looking, as G. Becker would say). Correspondingly, the behavioral rule will be that of habit, in the first case, and of project, in the second.

From this perspective, the "modern worlds" about which Pleşu talks when he opposes them to "traditional man" are the result of this great passage - from being dependent upon cosmocentric ranging and from heteronomy to an anthropocentric and egotistic understanding of Man's autonomy. Does this mean that, by opposition, "traditional man's? world is the pre-modern world of heteronomy and habit? No. And this is the source of many a great confusion: this is the world of habitual communities! Tradition - and, with it, "traditional man"  - is an invention of modernity, precisely one of its versions. The peasant, especially if and to the extent which he is "authentic," is not "traditional," but just... peasant, the member of a habitual peasant community. The tradition which we assign to the peasant world is a compensatory construction of the modern worlds, as habitual worlds tend to disappear and as they lose the "dependency" that these worlds have in common. According to Hobsbawm, at the beginning of his book which he dedicated to inventing traditions, "in this sense, 'tradition' must be clearly distinguished from 'habit,' which rules over the so-called 'traditional' societies." (Hobsbawm, 1983) Tradition is a selection and, at the same time, its result, while it operates from present to past, in order to transform the past into a reference of the present.

?In order to define a tradition, we must go from the present towards the past, and not the other way round, in order to understand it not as a vis a targo, whose effect will be perceived by us all, but as a point of view that we have today on what was before us. I don't want to say that to recognize a tradition is to invent it. The past must endure, so that we may take what was good from it. We can't do whatever we want with it. But the past sets only the interior limits, upon which depend only our interpretations of the present.? (Pouillon, 1975)