Social Science Information

The world language system is profoundly affected by the increase in the frequency and density of communication on a world scale. Most of the languages spoken today are not expected to survive the century and most of those surviving will lose or fail to get control of some higher functions of communication, notably in the fields of commerce and science. The minority languages best able to resist the pressure of more powerful competitors are those having a government as their champion, and their best overall protective strategy remains territorialization, either within the boundaries of a unilingual state or, in the case of multilingual societies, on the territorial model of Switzerland and Belgium that juxtaposes rather than mixes languages at the regional level.

The article reviews and re-examines some arguments against the treatment of social action within the economic approach as an extension of economic behaviour/rationality and thus against its denial of the specific, irreducible nature of the extra-economic. A major argument is that social action is a sui generis phenomenon that cannot be reduced with theoretical impunity to its economic modalities. Social action is characterized by substantial autonomy relative to economic behaviour/rationality. Arguments about the autonomous character of social action seek to remedy the indiscriminate extension of the economic approach beyond the field of economy to all human behaviour construed as consistent utility maximization. These arguments adduce certain classes of factors (socio-psychological, socio-cultural, socio-systemic and others) contributing toward the autonomy of social action. In addition, the economic-approach treatment of the human actor as Homo economicus is reversed by conceiving the economy as a domain of social action of which economic behaviour is a special case.

The article is an attempt to develop a synthetic conception of power based on Weber's, Parsons's and Foucault's writings. The aim is, first, to build a bridge between the so-called resource theories of power (Weber, Parsons) and the structural approach (Foucault) and, second, to do this in the form of a conception which would be usable on both macro- and micro-levels at the same time. Four theories are discussed: (1) the distributive approach (Weber); (2) the collective approach (Parsons); (3) the structural approach (Foucault); and (4) the neostructuralist approach (this article). It is argued that these approaches can be ordered on a scale on which the complexity of analysis increases as one moves from (1) to (4), and that the selection of an appropriate level of analysis in an empirical study is a practical issue relative to the aim of the study. The types of analyses characteristic of the more complex levels are illustrated by a discussion of the problem posed by Big Case Comparison in historical sociology (level 3) and everyday conversation (level 4), including comments on phenomenological sociology and conversation analysis.

The article is based on a survey of 788 Danish researchers, mainly from the social sciences, and analyses differences between female and male researchers concerning cognitive styles and cognitive convictions. Sandra Harding's portrait of modern science as androcentric and characterized by a set of gender-related dualisms is taken as a point of departure, and the results by and large show gender differences which can be related to this picture. Male researchers give more importance to methodological ideals taken from natural science, objectivity, mathematical methods, rationality, universality and cumulative results, etc., than female researchers do. These differences are shown to be correlated with the degree of power orientation of research topics.