Although the historian confronts the question of what is not known in the same terms as does any other researcher no matter his or her discipline, the conditions of the debate are different for the historian because of the problematic nature of the science of history. While practically all the other sciences, including the social sciences, struggle against ignorance by seeking to discover and establish laws that will govern the facts, history must always face, in spite of its ever more sophisticated techniques, the contingency of its materials, which are fundamentally human. The materials of history can not—in spite of what some have asserted—be contained within any single, all-encompassing mode of reasoning that would allow for the creation of constitutive abstract models. In spite of this reality the historian persists, like Sisyphus, in organizing knowledge. Man apparently has no choice but to believe that his activities have some logical basis. The historian therefore collates facts in order to develop cognitive and interpretive structures that will be durable. What gives the discipline of history its epistemological originality is the painfully central role it accords to the question, “What do we not know?” Man wants to know, believes he knows. The ignorance of historians thus seems especially upsetting when applied to our own century. How could the methodological progress of the science of history, along with the professionalization of the historian's craft, have led to an actual increase of uncertainty concerning a period so close to us in time? Today's man-in-the-street—endowed only with his memory, which seems so rich in “truths”—finds it hard to believe that any event or structure of the twentieth century could remain unknown. Limiting our inquiry to the twentieth century is not, however, justified solely by the practical consideration of trying to gain a better grasp of our times. Indeed in this brief article devoted to the relationship between ignorance and the history and historiography of the twentieth century, we will not attempt to inventory all the events, persons, mentalities or economic facts which historical knowledge has until now been incapable of explaining. Unlike the other sciences, whose struggle against ignorance is accomplished by a gradual process of nibbling at the margins of the unknown, the discipline of history must always keep in mind that the masses of granite that form the foundation of History are in fact but a body of statements that can only aspire to the status of truth. The aim of our inquiry— without of course claiming to be exhaustive—will therefore be to identify the nature of the obstacles that lie in the historian's path. These problems are of three kinds. The first concerns access to knowledge, and is tied—although not exclusively—to problems of the techniques of historical research. The second concerns the difficulties related to conceptualization and therefore to the uses of knowledge. There can be no knowing without know-how. Finally, the third pertains to the lack of permeability between history and the other human and social sciences, which creates for the historian the challenge of interdisciplinary exchange.

“All men naturally desire to know” : this is the celebrated assertion with which Aristotle begins the first book of his Metaphysics. According to him, human beings’ desire to know is as natural to them as their desires for food, rest, or amusement. These latter “natural” urges are responses to certain deficiencies—hunger, tiredness, and boredom; similarly, the desire to know is a response to a deficiency of knowledge. As Thomas Aquinas puts it, the urge to know (libido sciendi) is as much a characteristic of human beings as the urge to feel or to dominate (libido sentiendi and libido dominandi), and all three, essential though they may be to human nature, must be kept under control. My purpose here is to examine the limits of the desire to know: in what way may our desire to know go beyond the bounds of the “knowable”?

Let us begin by making a rather obvious remark: the meaning of the question of ‘what we do not know’ varies according to whether or not the word “yet” is explicitly or implicitly included. It comes as no surprise that it should be in physics, the science in which, ever since Galileo and Newton, the quest for knowledge has been so amply and unexpectedly rewarded, that we find the most dramatic examples of both possibilities: one in which theory points to knowledge not yet acquired, knowledge which is still to be conquered, but which, once attained, should constitute its final triumph; and one in which “we do not know” may be taken as the conclusion of established knowledge. We do not (yet) know how to unify the four fundamental forces of interaction in a single theory, and yet that eventuality is already being called the theory of everything (TOE). Since interaction is the principle on which physics now bases all its explanations, and since physical explanations are in principle valid for everything that exists, the unification of the forces of interaction must therefore be the science of the principles of everything that exists. On the other hand, we do not know at what moment a particular radioactive nucleus will disintegrate. In this case, our knowledge of physics makes it impossible to use the word “yet”: radioactive disintegration is described in terms of a “lifetime,” which means that each nucleus within a given population has, at any one moment, the same probability of disintegrating, and that this probability is not a function of any variable which we can identify and manipulate. In this case, our knowledge asserts that we cannot know now and will not be able to know for as long as our present knowledge remains valid.