Contenu du sommaire

In France, social welfare is the work of the Republic, which has striven to centralise a right to be assisted, traditionally set up by parishes. On the other hand, a right to Social Security was legally built and implemented through companies and professional bindings. Assistance, however, was always thought as an occasional and precarious relief of a limited amount. Two evolutions have come to modify this organisation : the widening realm of “guaranteed” social benefits, starting in the 70s-80s, and the so-called decentralisation policies (in 1982-1983 and in 2003-2004). Today the former blurs the boundaries between insurance and assistance, while the latter compromises the republican vision of social welfare itself.

Over the last 20 years or more, the ways in which public action is organised have been subject to change. The development of contracts between the State, its various services and the local authorities is a very good example of this process. On the one hand, the State imposes ways of working and the framework within which the negotiations are to take place, on the other, it leaves the choice of the problems to be dealt with and their resolution to local "community workers". Could this be conceived of as a sort of "laissez-faire" which will inevitably lead to the fragmentation of welfare policy ? The setting up of the local "Contrat Educatif in Toulouse, shows that quite the opposite is true, and that the State's initiatives bring about the reorganisation of territorial relations on a political and institutional level, leading to important changes in the way public policy is carried out. On a more general level, this research investigates the setting up, the basic principles and the impact of a "non directive" implementation of welfare action. The ways in which this type of policy is administered is an extremely fruitful subject for research.

In the last two years, counterinsurgency has become a central issue in the debates concerning the US presence in Iraq, due to its political, military and institutional implications. Unlike the theoretical expectations regarding the unwillingness of American services to adopt irregular warfare, many conceptual, organizational and operational reforms occurred since 2003, leading to a contingent construct. Drawing on empirical evidences, this article wants to highlight the role of organizational learning and adaptation in the construction of counterinsurgency in US Army and Marines.

Praxiological analysis, as grounded in ethnomethodology and conversation analysis, favours an approach to deliberation in parliamentary contexts which does not yield to awkward abstraction. Awkward abstraction idealizes deliberation through its nearly total withdrawal from the normal order and everyday frames of interaction. From this point of view, what happens within a parliamentary setting must be understood in reference to the broader mechanism of decision crafting. Praxiological analysis is especially capable to describe the specificity of practices which are situated, i.e. practices centred on a particular site, and at the same time embedded within a network where intersubjectively valid opinions are produced. This network is dialogical.

In the previous number of the Revue française de science politique, Bruno Latour proposes that science studies and political science unite their efforts to work on common objects according to a renewed approach of "science" and "politics". Pierre Favre answers these propositions by granting at first to the author that both disciplines have numerous objects of study in common, even if certain perspectives diverge outstandingly. But he refuses to follow Latour on the major presuppositions of his approach. He disputes the assertion of Latour according to which we cannot distinguish "science" and "politics" which would have no different domains. He disagrees with the Latour conception of a science finally deprived of any cognitive capacity. To follow this extreme version of science studies, political science would not be scientific any more. Latour does not seem itself to be able to apply its own program, actually self-refuting.