From the viewpoint of women’s history in France, Louise Tilly’s article appears to arise from a specifically ‘Anglo-American’ debate.footnote1 But it does raise questions which are very relevant and current. The Anglo-American connection is not just apparent in the references, most of which are taken from works written in English, but also in the terms in which the problems are stated. To begin with, the question that lies at the origin of that debate—whether or not women’s history can be said to have ‘arrived’—suggests that there is a consensus to the effect that it has arrived; the different positions on its future objectives and timetable are framed from that starting point. But the situation in France is altogether different. There are only two teaching posts explicitly devoted to women’s history, and not a single endowed chair; and although an increasing number of history teachers (especially women) have been covering the problematic of gender and social relations of the sexes in their courses, we are unfortunately still a long way from being able to say that this is an acknowledged subject included in the curriculum. But it is true that ‘the number of books and articles in this area has greatly increased’footnote2 in France too, along with what might be called supplementary sections or chapters (‘Women and…’) in special issues of journals, and in reference works. Undeniably, this increase expresses developments of a qualitative order, which will doubtless be confirmed and rendered more visible by the recent publication of the Histoire des Femmes edited by Georges Duby and Michelle Perrot. But on the institutional level women’s history is still seen as a secondary field of research devoid of legitimacy.footnote3

Secondly, unlike the situation in the usa, where debates on the potential and limitations of deconstruction seem to inspire impassioned polemics, this is far from being the case in France, notwithstanding the widespread habit of calling this conceptual framework ‘French feminism’, ‘French theory’ or ‘French writing’.footnote4 Certainly some of the key thinkers to whose work these currents refer—Jacques Derrida, Julia Kristeva, Hélène Cixous and (to a lesser extent) Luce Irigaray—are important figures in the French intellectual landscape and have undoubted (though not exclusive) standing in the domains of literary criticism and linguistics. But they are far from being hegemonic references in women’s studies, women’s history in particular. And when they are mentioned, it is not so much in a general epistemological context as in connection with polemics concerned with ‘difference’ and ‘femininity’. These have figured widely in feminist debates in France, but do not correspond to the issues which divide the partisans and critics of postmodernist approaches in the usa.footnote5

Nevertheless, behind the marked differences of outlook and tone, the basic questions running through Louise Tilly’s article and animating her debate with Joan Scott seem to me to be central to the concerns of women’s history in Francefootnote6 and have a very familiar ring. What should be done to move on from descriptive approaches to approaches oriented towards solving problems? What can be done to link these problematics with those of other areas of history? What does women’s history add to history in general? In what ways has it transformed—and, more importantly, in what ways can it transform—the historical field as a whole? What conceptual and methodological tools are most appropriate to such an undertaking?

One of the most interesting aspects of the Tilly–Scott disagreement concerns this last point. Both historians complain that the notion of gender is used too descriptively in women’s history, and agree on the need to make it a genuine category of analysis by means of a conceptualization capable of challenging the dominant concepts of the historical discipline. According to Joan Scott, this conceptualization is not possible in the area of social history, owing to the excessive influence of economic determinism; she emphasizes the need for a ‘more radical epistemology’ which, she believes, is to be found in the framework of ‘post-structuralism (or at least some of the approaches generally associated with Michel Foucault and Jacques Derrida)’ which could ‘supply feminism with a powerful analytic perspective’.footnote7 Tilly takes the opposite view, that the methods of social history are more appropriate for the development of analytic categories and questions that might challenge historians. But, she adds, it should be ‘a social history that gives due respect to human initiative, something women’s social history has managed to do with a good deal of success’.footnote8 In her opinion it is precisely Scott’s ‘more literary and philosophic’ use of gender, and the deconstructionist approach she favours, that lead her to ‘overestimate the weight of social constraint’ and ‘trivialize the conditions under which relations of power are altered and transformed’.

I share Louise Tilly’s scepticism as to the capacity of deconstruction to work out a non-determinist vision of history, let alone a vision of women as subjects of history. My scepticism extends more generally to the hopes placed by a number of feminists in the liberating potential of post-structuralist epistemologies. But I do think such a debate should not be restricted to the borrowings made by women’s studies; it should include systematic discussion of the premisses that lie behind these theories, of the newness of their epistemological pretensions, of the extent to which their promises are fulfilled, of their overall political dynamic. No doubt a discussion of this sort would be very useful to women’s studies; but my aim here is much more modest. I would like to ‘react’ on some aspects of what Louise Tilly calls the ‘more literary and philosophic use of gender’, which I think can usefully be considered in the context of a more general movement marked by the linguistic turn and its impact on women’s history. The following observations mainly concern the ways in which the question of historical determination, subjectivity and agency is worked out in the construction of the category of gender. These questions, which have always been of major methodological and theoretical importance to feminist historians, are reactivated and restated by the Tilly–Scott polemic.

According to Scott, gender as an analytic category ‘rests on an integral connection between two propositions: gender is a constitutive element of social relationships based on perceived differences between the sexes, and gender is a primary way of signifying relationships of power’.footnote9 The first proposition refers to ‘the process of gender construction’,footnote10 the second to the pertinence of gender as a category of analysis, comprehension and historical explanation of other power relations. The first underlines the fundamental importance of the procedures of differentiation through which, in every historical context, the apparently fixed and coherent content of the masculine and the feminine is formulated and reformulated in dichotomic terms. One of the primary tasks of the gender historian is to deconstruct this content, to demonstrate its fragility and polysemy, to expose the selectiveness of the procedures through which it comes to acquire a rigid meaning and chart the struggle between rival interpretations of which this content is the product. Whether we consider documents from the past (normative discourses, workers’ petitions, statistical and economic texts) or the ways in which these documents are read by contemporary historiography, the essential aspect of gender as an analytic category, in Scott’s formulation, is that it exposes the strategies of domination that underlie the binary construction of the difference between the sexes.

In my opinion this approach has a double potential for the history of social relations of the sexes: first of all a heuristic potential, in as much as it permits what Carlo Ginzburg calls distancing,footnote11 the art of visiting the past as a foreigner, seeking to decipher—rather than taking for granted—the social significance of ‘known things’, in this case what it means to be a ‘man’ or a ‘woman’ in a given culture and historical period. This perspective restores the problematic dimensions to ‘facts’ hitherto regarded as known or self-evident (for example the role of women in the sexual division of labour and/or in the modern urban environment, the exclusion of women from ‘universal’ suffrage). These facts should not only be described, but analysed as pertinent issues which the historian is obliged to consider. By adopting a deliberately agnostic position on the sexes at the outset, we are enabled to discover ‘new’ historical facts in the most classic sense of the term. A convenient example is provided by women’s resistance to the French Revolution: if we take for granted the notorious affinity of women and religion, we will tend to concentrate our researches on religious forms of counter-revolution, and overestimate their role in the continuation, in private, of the proscribed religious practices. Recent studiesfootnote12 have shown, however, that other practices (espionage, crimes of opinion, economic and financial crimes, distribution of false promissory notes, participation in monarchist politics) were sufficiently widespread to deny the existence of any ‘sexual specificity’footnote13 in the forms of counter-revolutionary struggle. Doubtless these ‘unexpected’ practices were broadly linked to the upsetting of relations between the sexes caused notably by the emigration of male members of the nobility. But to assume as something already established the forms of counter-revolutionary activity which women were susceptible of undertaking, could prevent us from grasping the changing dynamics of gender relations during the Revolution and from studying the scope and limitations of such transformations. So a refusal to take at their face value the dominant definitions of the difference between the sexes (which would assign religion and morals to women and politics to men, for example) might enable us not only to study new aspects of women’s historical experiences, but also to analyse the dynamic of the power relations which make them possible and constantly reformulate the divisions on which the construction of gender is based.