Timothy Bewes’s review of my book Europe in Love, Love in Europe in NLR 236 has helped me to rethink some of the presuppositions and implications of my work, against a background of ego-histoire. I am grateful for this. Having been a pupil of Norberto Bobbio in my youth, and followed his work more or less consistently thereafter, I have learnt that no historian can avoid incorporating a philosophy of history, tacit or explicit, into their work. This philosophy can be unsystematic and eclectic, but it will inform the architecture and approach of the historian, and define what is considered historically significant.

Let me make explicit what is often, although not always, implicit in my own recent work. I willingly admit that it is guided by a belief in the discontinuity of history. For I am convinced that a primary task of the historian today is to avoid the exclusive pretensions to continuity of traditional narratives, which filled in all gaps and ignored everything that did not fit the writer’s paradigm—without acknowledging its omissions. This kind of continuous narrative is still with us today. It purports to establish or explain sequences of events in terms dominated by political and economic forces, while excluding aspects that matter a great deal—to some of us—such as subjectivity or daily life, and ‘subaltern’ figures such as women. Claims to continuity between present and past have often been the basis for ideological appropriation of terms and concepts, and the self-legitimation of the historian. A recent example is the presumption by some members of the French intellectual Right of an unbroken continuity of values from Aristotle to Maastricht. But even innovative and progressive historiography still often relies on traditional continuous narratives, which do justice neither to leaps in history, nor to the new domains and subjects studied by historians.

The idea of discontinuity of history means in the first place an alternation of repetitions, developments and sudden complete breaks. It is also a methodological safeguard of the distance between history and historiography—in other words, between the experience of humanity and the itineraries of the historian. I believe that in the writing of history the representation or Darstellung never reproduces the Genesis of the phenomena to be studied—a lesson learnt from Marx. Discontinuity further suggests the distance between the intentions of subjects and the outcomes of history, which notoriously proceeds by its bad side. It denotes, too, the coexistence of the times of the particular (as studied by micro-history) and of the general (macro-history). That the particular cannot be reduced to an example of the general is something most historians have yet to understand or accept. Finally, discontinuity indicates the dialectical way in which the same phenomenon may possess two opposite values and implications in the same period, of conservation and subversion—an example would be groups in the 1930s like Ordre Nouveau or New Europe. I have reached these conclusions from my practice as a historian, but also from my reading of philosophers of history such as Horkheimer or Benjamin.

Hence too the crucial importance, in my view, of psychoanalysis. Its findings—or more accurately, the findings of some Freudians, some Lacanians and some Jungians—enlarge and deepen the possible range of historical research. They remind us that many processes have semi-conscious or unconscious dimensions; that diachronic forms of development are marked by all kinds of associations and circularities; and that the link between historical events and processes is often not at all that of a cause to an effect, but of a symptom to an ‘illness’—that is, of a sign whose relation to what it hides or reveals is indirect and oblique. Similar presuppositions are, of course, shared by other disciplines such as semiotics—no wonder, since language is here the common focus of attention. In fact, it is strange that historians have so often forgotten that they deal largely with words and not immediately with facts. Acknowledgement of the unconscious dimensions of history involves not only a more complex view of issues of subjectivity, but also procedures of interpretation that try to find the hidden substrata of source-texts—the ground where psychoanalysis meets literary criticism. In this sense, the gaps, voids and discontinuities of history are also the risks that subjects take in attempting to become masters of their lives and times, or interpreters of them. The continuity of the subject is neither guaranteed nor given—it is a possibility that is always in question; and this element of risk should be reflected in a historical narrative. We need to take account of the silences and oblivions of history, its ironies as well its tragedies. Irony, like laughter, is a royal road to the unconscious.

It should be clear, then, that the task of the kind of historical writing to which I am committed is not to pursue—or construct—direct links between past and present, but to study traces of various kinds of the past with the utmost philological attention to them and their contexts, in search of strands which have gone unseen or ignored, or become forgotten. Our aim will be new visions of the past, founded certainly on primary and secondary sources, but likely to be at odds with current opinions of what is historically significant. It should also be plain why cultural processes are of such crucial interest. The shift of so many historians of my generation from social to cultural history—it includes myself—does not necessarily imply lack of interest in older questions, such as: what is a revolutionary subject, or how can we help it towards self-awareness and recognition. But it reformulates them drastically. The questions now become: what is subjectivity, as agency of decision and as inherited legacy? How do individuals and groups constitute such subjectivity and become constituted by it? What are the relations between subjectivity and change, and which were the points in history at which different choices could have been made—if we understand the past, in Croce’s formula, as the history of freedom, which in my translation is another way of denying the absolute privilege of ‘what actually happened’.