Variously known as a ‘school’, ‘nebula’ or ‘movement’, the Annales group may also be described as a metaphorical family. As we have seen, Lucien Febvre treated both Braudel and Mandrou as his ‘sons’. Braudel, who was paternalist if not patriarchal, used to describe younger colleagues as ‘petit’: ‘le petit Revel’, ‘le petit Pierre’ (Goubert) and so on. In return, Pierre Chaunu once told an interviewer that ‘I loved Braudel the way I loved my father’.1
Family links between members of the group are not uncommon. The sinologist Jacques Gernet, for instance, is the son of the classicist Louis Gernet. Maurice Aymard, a historian of Italy who became director of the Maison des Sciences de l'Homme, is the son of the ancient historian André Aymard, an old friend of Braudel's.2 The demographer Hervé Le Bras is the son of the sociologist Gabriel Le Bras. Jean-Pierre Goubert is the son of Pierre Goubert. Denys Lombard, historian of South East Asia, was the son of the medievalist Maurice Lombard.
Some of the group have intermarried: Jacques Ozouf and Mona Ozouf, for instance Jean-Claude Perrot and Michelle Perrot the medievalist Jean-Claude Schmitt and the ancient historian Pauline Schmitt. François Furet was the brother-in-law of both Pierre Nora and Denis Richet.
The protagonists of the previous chapter, Braudel and Labrousse, were born in 1902 and 1900. This chapter, by contrast, is mainly concerned with historians born in the 1920s: with Maurice Agulhon, Marc Ferro, François Furet, Jacques Gernet, Jacques Le Goff, Robert Mandrou, Michelle Perrot and Emmanuel Le Roy Ladurie. It is never easy to decide when a generation begins and ends, and in this case it surely makes sense to include Georges Duby (born in 1919) and even Philippe Ariès and Jean-Pierre Vernant (both born in 1914) at one end, and historians born in the early to mid-1930s – Jean-Louis Flandrin, Mona Ozouf, Alain Besançon, Michel Vovelle, Philippe Joutard, Nathan Wachtel and Daniel Roche – at the other, leaving younger historians for the following chapter. As this list of names shows, the Annales group, tiny in the first generation, had become quite large by the third.
Political events, such as the Dreyfus affair, are often taken to mark or even to form generations, and in this case the rise of a third generation became more and more obvious in the years following 1968. A major change – not to say ‘purge’ – in the management of the journal was carried out in 1969, apparently in reaction to the crisis of May 1968, known in France as ‘the events’ (les événements). Events seemed to be taking their revenge on the historian who had spurned them. In any case, Braudel, who had flown back to Paris from Chicago when he heard the news, decided to bring in new faces in order to renew Annales, ‘faire peau neuve’, as he put it.3 It was in 1969 that Le Goff and Le Roy Ladurie became involved in the management of Annales. In 1972, Braudel retired from the presidency of the Sixth Section (which went to Jacques Le Goff) while the old Sixth Section disappeared in 1975. Le Goff then became the president of the reorganized École des Hautes Études en Sciences Sociales, to be succeeded in 1977 by François Furet.
More important than the administrative changes, however, are the intellectual shifts of the period. The problem is that the intellectual portrait of the third generation is more difficult to paint than that of the first and second. No one dominated the group as Febvre and Braudel once did. Indeed, some commentators have spoken of intellectual ‘fragmentation’.4 At the very least, one has to admit that polycentrism prevailed. Some historians took Lucien Febvre's programme even further, extending the frontiers of history to include childhood, dreams, gestures and, most important of all, the history of women (a collective history of women was organized by Michelle Perrot, together with Georges Duby, who had been working on the history of marriage and on medieval women).5 Indeed, the third generation may be contrasted with its predecessors by its inclusion of women both in the group (Christiane Klapisch, for instance, Mona Ozouf or Lucette Valensi) and in studies of the past. Other members of the network might be described either as having extended or undermined the programme, by incorporating political history and the history of events. Some continued to practise quantitative history, others reacted against it.
This generation has been more open than its predecessors to ideas from the English-speaking world. A number of its members have spent a year or more in the United States, in Chicago, Princeton, Ithaca, Madison or San Diego. Unlike Braudel, they speak as well as read English. In their different ways, they have tried to make a synthesis between the Annales tradition and American intellectual trends – psychohistory, the new economic history, the history of popular culture, symbolic anthropology and so on. Collaboration between French and American historians has sometimes taken place.6
The pages that follow concentrate on three major themes: the discovery, or rediscovery, of the history of mentalities; the attempt to employ quantitative methods in cultural history; and finally, the reaction against quantitative methods, whether it took the form of a historical anthropology, a return to politics or a revival of narrative. The price of this decision is to exclude or at least to say very little about interesting work that does not fit these categories. However, concentration is the only way to prevent this chapter becoming as fragmented as the Annales group is sometimes said to have become.
In the Braudel generation, as we have seen, the history of mentalities and other forms of cultural history were not entirely neglected, but they were relegated to the margins of the Annales enterprise. In the course of the 1960s and 1970s, however, an important shift of interest took place. An analysis of articles in the Annales between 1965 and 1984 shows that around 35 per cent were concerned with cultural or intellectual history.7 The intellectual itinerary of more than one Annales historian of this generation included a major change in mid-career, a turn from a concern with the economic or demographic base of society to its cultural ‘superstructure’, or as Michel Vovelle puts it, ‘from the cellar to the attic’.8
Why should this have happened? The shift of interest was in part, surely, a reaction against Braudel's dismissal of the history of mentalities, one of the major differences between his approach and that of Lucien Febvre, whose interest in modes of thought has already been noted. Returning to the idea of the Annales family, we might describe this cultural turn as an example of what a distinguished art historian has called the ‘grandfather law’, arguing that ‘a generation, with deliberate disregard for the views and feelings of its fathers and direct teachers, skips back to the preceding period and takes up the very tendencies against which its fathers had so zealously struggled, albeit in a new sense’.9
The move from ‘cellar to attic’ also formed part of a much more widespread reaction against determinism of any kind, increasingly visible in the course of the 1960s. Some members of the group, such as Vernant, who had joined the French Communist Party, left the party at this time, as many Western communists did after Soviet tanks rolled into Prague in 1968 – if they had not left, like Besançon, Dupâquier, Furet and Le Roy Ladurie, when the tanks were sent into Hungary in 1956.10
It was, however, Philippe Ariès (a monarchist and at one time a supporter of the right-wing party Action Française) who drew public attention to the history of mentalities in a remarkable, almost sensational, book that he published in 1960. Ariès was not originally part of the Annales group, although he taught at the Hautes Études at the end of his career. He was too close to Annales for the historians at the Sorbonne, but too much an amateur for the professional Annalistes.11 He was, as he described himself, ‘a Sunday historian’, who spent his working life at an institute for tropical fruits while devoting his leisure to historical research. Indeed, one leading professional historian dismissed Ariès as a ‘banana salesman’.12
Trained as a historical demographer, Ariès came to reject the quantitative approach to society (just as he rejected other aspects of the modern industrial-bureaucratic world). His interests shifted towards the relationship between nature and culture, to the ways in which people in a given culture view and classify natural phenomena such as childhood and death.13
In his study of families and schools during the old regime, Ariès argued that the idea of childhood, or more exactly the sense of childhood (le sentiment de l'enfance), did not exist in the Middle Ages. Individuals in the age-group that we call ‘children’ were regarded as more or less like animals until the age of seven, and more or less like miniature adults thereafter. Childhood, according to Ariès, was discovered in France in the seventeenth century or thereabouts. It was at this time, for instance, that children began to be given special clothes, like the ‘robe’ for small boys. Letters and diaries of the period document increasing interest in children's behaviour on the part of adults, who sometimes attempted to reproduce childish speech in letters and other texts. Ariès also drew on iconographical evidence such as the increasing numbers of portraits of children to make the case that awareness of childhood as a phase in human development went back to the early modern period – but no further.
Centuries of Childhood, as the book is known in English, is open to criticism, and it has indeed been criticized by many scholars, both fairly and unfairly. Specialists in the Middle Ages, for instance, have produced evidence against its sweeping generalizations about that period. Other historians have criticized Ariès for discussing European developments on the basis of evidence virtually limited to France alone, and for failing to distinguish sufficiently between the attitudes of men and women, elites and ordinary people.14 All the same, it was the achievement of Philippe Ariès to place childhood on the historical map, to inspire hundreds of studies of the history of childhood in different regions and periods, and to draw the attention of psychologists and pediatricians to the new history.
Ariès spent the latter years of his life studying attitudes to death, focusing once more on a phenomenon of nature as refracted through culture (Western culture), and responding to a famous plea of Lucien Febvre's (made in 1941), ‘We have no history of death.’15 His major study, The Hour of Our Death (1977), offered an account of developments over the very long term, a thousand years or so, distinguishing a sequence of five attitudes from the ‘tame death’ (la mort apprivoisée) of the earlier Middle Ages, a view defined as ‘a compound of indifference, resignation, familiarity and lack of privacy’, to what he calls the ‘invisible death’ of our own culture, in which, inverting the practices of the Victorians, we treat death as taboo while discussing sex openly. The Hour of Our Death has very much the same merits and defects as the same author's Centuries of Childhood. There is the same boldness and originality, the same use of a wide range of evidence (including literature and art but not statistics) and the same unwillingness to chart regional or social variations, in striking contrast to the author's earlier work on demography.16
The work of Philippe Ariès was a challenge to historical demographers in particular, a challenge to which some of them responded by paying increasing attention to the role of values and mentalities in ‘demographic behaviour’ – in other words, by studying the history of the family, the history of sexuality and, as Febvre had hoped, the history of love. A central figure in these developments was Jean-Louis Flandrin, whose studies of old-regime France addressed such questions as the nature of parental authority, attitudes to small children, the influence of the Church's teachings on sexuality and the emotional life of the peasants. He later studied the cultural history of food.17 Another is Christiane Klapisch, who has worked on the history of late medieval and Renaissance Tuscany and moved from the study of demography to that of women, family and ritual.18 Studies in this area in particular have done a good deal to bridge the gap between a history of mentalities based on literary sources (Febvre's Rabelais, for example) and a social history without a place for attitudes and values.
A good example of this kind of bridging is the work of Jacques Gernet. Like other French historians of his generation, Gernet climbed the ladder from the cellar to the attic, moving from a dissertation on the economic aspects of Buddhism to a book about Christian missions to China.19 His China and the Christian Impact (1982) is a history of mentalities that centres on misunderstandings. The missionaries believed they had made many converts, failing to understand what adhesion to the new religion meant to the converts themselves. For their part, the mandarins misunderstood the intentions of the missionaries.
According to Gernet, these misunderstandings reveal the differences between the categories, the ‘modes of thought’ (modes de pensée) or ‘mental frameworks’ (cadres mentaux) of the two sides, associated with differences in their languages. This focus on the encounter between two cultures allows the author to illuminate mentalities in ways denied to historians of Europe. What Braudel would have described from outside as a case of ‘refusal to borrow’ is interpreted by Gernet from within.
Nathan Wachtel, who has worked on South America, also focuses on cultural encounters. His Vision of the Vanquished (1971) is a history of the early years of colonial Peru from the point of view of the Indians. In several respects this study resembles work on Europe by Annales historians. It deals in turn with economic, social, cultural and political history. It offers an example of history from below, with much to say about popular revolt. It employs the regressive method associated with Marc Bloch, studying contemporary dances representing the Spanish conquest as a means of recovering the Indians' original reactions. It borrows concepts from social anthropology. However, Wachtel does not simply take over the structure-conjuncture-events model of the historians of early modern Europe. In Peru, he argues, the sociocultural changes of the time did not take place within the old structures. On the contrary, the process was one of ‘destructuration’. The author's concern with this process gives Wachtel's book a dynamic, and also a tragic quality which even The Peasants of Languedoc cannot match. A later work of Wachtel's, on the Indians of Bolivia, deals with five centuries and once again employs the regressive method.20
Encounters are also central to one of the most remarkable studies in the Annales tradition, a three-volume study of Indonesia by Denys Lombard entitled The Javanese Crossroads (Le carrefour javanais). The author stressed the place of Indonesia at the meeting-point of the Indian and Pacific Oceans, and so a centre not only for trade but also for what is described as ‘the juxtaposition or the fusion of almost all the great world civilizations’. He devoted sections to favourite annaliste themes such as death and time. As in the case of Bloch and Wachtel, Lombard wrote his study backwards from the present. He began by examining both the Westernization of the region and the limits to that Westernization, passing in the second volume to the Chinese and Muslim merchants and their impact on Indonesian culture, and finally, in the third volume, to the monarchy and the influence of ancient India. This study of cultural hybridization was published in 1990, well before the theme became a popular one. Its grand scale and its concern with the sea as a link between different societies make it a kind of Asian Mediterranean.21
Within the Annales group, some historians have long been concerned primarily with culture: Alphonse Dupront, for example. Dupront, a historian of Braudel's generation, has never been widely known outside the profession, but his influence on younger scholars has been considerable.22 From this point of view, he might indeed be considered the Labrousse of cultural history. His doctoral thesis, which attracted favourable attention from Braudel for its concern with unconscious attitudes, studied the idea of a ‘crusade’ as an instance of sacralization, a holy war to obtain possession of holy places. It was published – after a delay of forty years – in 1997.23
Dupront later focused his attention on pilgrimage, viewed as a quest for the sacred and an example of ‘collective sensibility’ to sites of cosmic power such as Lourdes or Rocamadour. His interest in sacred space inspired some of his pupils to investigate changes in the layout of churches and the symbolic meaning of these changes.24 He combined his interest in grand themes such as the nature of the sacred with precision in the inventory or cartography of (say) miraculous images. As his essay on the idea of acculturation, no less than his own research, illustrates, Dupront worked for a rapprochement between the history of religion and psychology, sociology and anthropology.25 His employment of the controversial term ‘acculturation’ to describe cultural change was followed by Wachtel on Peru and by Robert Muchembled on early modern France.26
Historical Psychology. However, the leading figure in the study of historical psychology in Febvre's style was Robert Mandrou.27 Soon after Febvre's death, Mandrou found among his papers a file of notes for an unwritten book which would have continued the study of Rabelais by considering the rise of the modern French mentality. Mandrou decided to pursue his master's enterprise and published his Introduction to Modern France, which was subtitled ‘An Essay in historical psychology, 1500–1640’ (1961) and included chapters on health, emotions and mentalities.
Soon after the publication of this book came the notorious breach between Mandrou and Braudel. Whatever personal reasons lay behind it, this breach took place in the course of a debate about the future of the Annales Movement. In this debate, Braudel supported innovation, including quantitative approaches, while Mandrou defended the heritage of Febvre, what he called ‘the original Annales’ (Annales première manière), in which historical psychology or the history of mentalities played an important part. Mandrou went on to write a book on popular culture in the seventeenth and eighteenth centuries. He continued in the same direction with a study of Magistrates and Sorcerers in 17th-century France (1968, subtitled ‘An analysis of historical psychology’). Sympathetic to the ideas of Marx, Mandrou tried to combine the analysis of class relations with that of collective mentalities.28
On the edge of Annales, Jean Delumeau, who had begun as an economic and social historian, shifted his interests from the production of alum in the Papal States and economic and social life in the city of Rome to the history of culture. His first move was in the direction of the history of the Reformation and then of the so-called ‘dechristianization’ of Europe. Later, Delumeau turned to historical psychology in Febvre's sense of the term, and wrote an ambitious history of fear and guilt in the West, distinguishing ‘the fears of the majority’ (the sea, ghosts, plague and hunger) from the fears of ‘the ruling culture’ (Satan, Jews, women and especially witches).29
In his work Delumeau made cautious use of the ideas of psychoanalysts such as Wilhelm Reich and Erik Fromm. He had been preceded in this direction by Michel de Certeau. Like Ariès, Certeau entered the Hautes Études relatively late in life in 1984 (he was born in 1925). He was a Jesuit and a polymath, whose interests extended to theology, philosophy, anthropology and sociology, as well as history and psychoanalysis (he belonged to the school of Jacques Lacan). The events of May 1968 marked an important stage in his turn from religious to secular topics (he wrote in defence of the rebels). In 1969, Certeau published a critique of Mandrou's study of witchcraft, and in 1970, his own interpretation – further from Febvre and closer to Freud – of a famous case of the collective possession of a group of nuns by devils (so contemporaries thought) in the small French town of Loudun in the 1630s.30
As for Emmanuel Le Roy Ladurie, his Peasants of Languedoc, discussed in the previous chapter, included works by Freud in his bibliography, sandwiched between a study of grain prices in Toulouse and an analysis of early modern class structure. Le Roy described the carnival of Romans as a psychodrama, ‘giving immediate access to the creations of the unconscious’, such as fantasies of cannibalism, just as he interpreted the prophetic convulsions of the Camisards in terms of hysteria. As he was the first to admit, ‘Cavalier and Mazel (the leaders of the revolt) cannot be invited to stretch out on the couch of some hypothetical historian-psychoanalyst. One can only note certain obvious traits that are generally encountered in similar cases of hysteria.’ In a similar manner, Le Roy looked at a previously neglected aspect of witchcraft trials, the accusation that witches had made their victim impotent by tying a knot during the marriage ceremony, a ritual that he interpreted persuasively as symbolic castration.31
Other members of the Annales group were moving in a similar direction, notably Alain Besançon, a specialist on Russia, who wrote a long essay in Annales on the possibilities of what he called ‘psychoanalytic history’. Besançon tried to put these possibilities into practice in a study of fathers and sons. The study focused on two tsars, Ivan the Terrible and Peter the Great, the first of whom killed his son while the second had his son put to death, as if compelled to re-enact a myth.32 The author later abandoned his belief in the value of psychoanalysis for historians.
Lucien Febvre had taken his ideas about psychology from non-Freudians or anti-Freudians such as his Strasbourg colleague Charles Blondel and Henri Wallon. Besançon, Le Roy Ladurie and Delumeau took theirs mainly from Freud and his followers, including neo-Freudians. Psychohistory American style, oriented towards the study of individuals, had at last encountered French psychologie historique, oriented towards the study of groups, although the two trends still stopped short of synthesis.
Ideologies and the Social Imagination. However, the main trend in the studies of the Annales group at this time was in a rather different direction. Two of the most distinguished recruits to the history of mentalities in the early 1960s were the medievalists Jacques Le Goff and Georges Duby.33
Le Goff, for example, published in 1960 a seminal article on ‘Merchant's Time and Church Time in the Middle Ages’. Marc Bloch had already discussed the sense of time in the Middle Ages, while Lucien Febvre, in his book on Rabelais, had discussed what he called the ‘floating’ or ‘imprecise’ sense of time in a period when people often did not know their exact age and measured their day not by clocks but by the sun. Le Goff refined the generalizations offered by Bloch and Febvre, themselves somewhat imprecise, by discussing the conflict between the clerical view of time and that of merchants.34
Le Goff's most substantial contribution to the history of mentalities, however, or to that of ‘the medieval imagination’ (l'imaginaire médiéval), as he came to call it, was made two decades later with The Birth of Purgatory (1981), a history of changing representations of the afterlife. Le Goff argued that the rise of the idea of purgatory formed part of ‘the transformation of feudal Christianity’, that there were connections between intellectual change and social change. At the same time he insisted on the mediation of ‘mental structures’, ‘habits of thought’ or ‘intellectual apparatus’ (outillage intellectuel); in other words, mentalities, noting the rise in the twelfth and thirteenth centuries of new attitudes to time, space and number, including what he called ‘the book-keeping of the after-life’.35 Le Goff also made the Annales approach more widely known by editing two collections of essays. The first, co-edited by Pierre Nora, emphasized ‘new problems’, ‘new approaches’ and ‘new objects’, while the second launched the phrase nouvelle histoire.36
As for Georges Duby, he made his reputation as an economic and social historian of medieval France. His doctoral thesis, published in 1953, dealt with society in the Macon region. It was followed by a substantial work of synthesis on the rural economy of the medieval West.37 These studies were very much in the tradition of Marc Bloch's Feudal Society and French Rural History. In the 1950s, as his interests gradually shifted from the cellar to the attic, Duby collaborated with Robert Mandrou on a cultural history of France.38
Later, Duby moved in new directions. Inspired by a famous study of the gift by the anthropologist Marcel Mauss, Duby explained what he called the ‘first rise of the European economy’ in terms of plunder and the gifts made by lords to their followers.39 Then, making use of neo-Marxian social theory, he turned to the history of ideologies, cultural reproduction, and the ‘social imagination’ (l'imaginaire social), which he combined with the history of mentalities.
Duby's The Three Orders runs parallel in many respects to Le Goff's Purgatory. It offers a case-study in what the author called ‘the relations between the material and the mental in the course of social change’. The author focused on the collective representation of society as divided into three groups – priests, knights and peasants; in other words, those who pray, those who fight and those who work (or plough – the Latin verb laborare is conveniently ambiguous). Duby was well aware that, as the great classical scholar Georges Dumézil had pointed out, this view of society as composed of three groups exercising three basic functions goes back a long way in Indo-European tradition, and can be found in many places, from ancient India to Gaul in the time of Caesar. Duby argued, as medievalists had done before, that this image of three orders performed the function of legitimating the exploitation of the peasants by their lords by suggesting that all three groups served society in their different ways.
Duby did not stop there, however. What interested him was the reason for the reactivation of this conception of the tripartite society in places as distant from each other as Wessex and Poland, from the ninth century onwards. He therefore considered the social and political context of this revival, particularly in France, where the image re-emerged in the early eleventh century. The author suggested that the reactivation of the image corresponded to a new need. At a time of political crisis, in eleventh-century France, for example, it was a ‘weapon’ in the hands of monarchs, who claimed to concentrate the three basic functions in their own person. Latent in the ‘mentality’ of the time, this intellectual system was made manifest as ideology for political ends. Ideology, remarked Duby, is not a passive reflection of society, but a plan for acting upon it.40
Duby's conception of ideology is not far from that of the philosopher Louis Althusser, who once defined it as ‘the imaginary [or imagined] relation of individuals to their real conditions of existence’ (le rapport imaginaire des individus à leurs conditions réelles d'existence). Indeed, Duby cited Althusser, who was a friend of his, in The Three Orders.41 In similar fashion to Duby, Michel Vovelle, a specialist on the eighteenth century, made a serious attempt to fuse the history of mentalités collectives, in the style of Febvre and Lefebvre, with the Marxian history of ideologies.42
At once colleagues, rivals and friends, as well as major innovators in their profession, Duby and Le Goff had much in common. In some respects, however, they were very different in their interests and also in their personal styles. Duby spent more of his life writing and editing books, while Le Goff is better known for his contributions to conferences and seminars. Duby made his reputation as a historian of French rural life, while Le Goff has always been more of a historian of cities. Duby wrote about knighthood, women, the family, and also about the history of art (he was a gifted amateur painter), while Le Goff concentrated on the history of the Church and the universities. Even their prose styles are different. Duby's style is elegant, dignified, narrative in form and self-consciously literary in manner, while Le Goff's style is more academic, argumentative and colloquial. I wonder what kind of career these two men would have followed if they had lived in the Middle Ages that they spent so many years studying. I can imagine Duby most easily as the abbot of an important Benedictine monastery, and also its chronicler, while I see Le Goff as a Dominican friar combining lectures and disputations in a university with preaching tours elsewhere.
An interest in mentalities or the collective imagination was not confined to scholars of the Middle Ages at this time, as two examples may suggest. The first is that of Jean-Pierre Vernant, a student of Louis Gernet and the author of a classic study of ancient Greek thought, published as early as 1966.43 This book was followed by many other contributions to the study of mentalities by Vernant himself and by other members of his circle, notably Pierre Vidal-Naquet and Marcel Détienne, with both of whom he collaborated.44
Another major contribution, focused on the nineteenth century, has been made by Alain Corbin, who is not part of the Annales network but has certainly been inspired by the Annaliste approach. His doctoral thesis on the Limoges region, published in 1975, began by paying tribute to Dupront, Mandrou, Febvre and Bloch and had much to say about what he called ‘mental structures’. Corbin first reached a wider public with a bold essay on the history of smells, or attitudes to smells, The Foul and the Fragrant (1982), described by the author as a study of the social imagination. He challenged Febvre's argument that the sense of smell was declining in the sixteenth century, while following Febvre in his concern with the history of sensibility.45
It was scarcely surprising to find important contributions to the history of mentalities being made by medievalists such as Duby and Le Goff. The remoteness from us of the Middle Ages, their ‘otherness’, poses a problem that an approach of this type helps to resolve. On the other hand, the kinds of source surviving from the Middle Ages made the period somewhat less amenable to another of the new approaches to culture characteristic of the Annales group at this time, serial history.
The history of mentalities was not pushed to the periphery of Annales in its second generation simply because Braudel was not interested in it. There were at least two more important reasons for its marginalization at this time. In the first place, a good many French historians believed – or at any rate assumed – that economic and social history was more important, or more fundamental, than other aspects of the past. In the second place, the new quantitative approach discussed in the previous chapter could not get a purchase on mentalities as easily as on the economic and social structure.
All the same, the attempt was made, along the lines suggested by Pierre Chaunu in a well-known manifesto for what he called (following a remark by Ernest Labrousse) ‘the quantitative at the third level’.46 Lucien Febvre's article ‘Amiens: From the Renaissance to the Counter-Reformation’ (1941) had already shown the value of studying a series of documents (in his case inventories post-mortem) over the long term, in order to chart changes in attitudes and even in artistic taste.47 However, Febvre did not offer his readers precise statistics. The serial approach was developed in order to study the history of religious practice, the history of the book and the history of literacy. It spread to other domains of cultural history somewhat later.
The idea of a history of French religious practice, or a retrospective sociology of French Catholicism based on statistics for attendance at communion, vocations to the priesthood and so on goes back to Gabriel Le Bras, who published an article on the subject as early as 1931.48 Le Bras, who had been one of Febvre's and Bloch's colleagues at Strasbourg, had broad interests in theology, history, law and sociology. He founded a school of Church historians and sociologists of religion who were particularly concerned with what they called the problem of ‘de-christianization’ in France from the late eighteenth century onwards. They investigated this problem by means of quantitative methods. Le Bras and his followers did not form part of the Annales group – the followers were often priests, and they had their own network of centres and journals such as the Revue de l'histoire de l'église de France. All the same, the work of Le Bras was warmly welcomed by his former colleague Lucien Febvre, while his followers were clearly inspired by the later Annales.49
As an example of this substantial body of work, one might take a thesis on the diocese of La Rochelle in the seventeenth and eighteenth centuries. It is organized in much the same way as one of the regional studies associated with Annales, beginning with the geography of the diocese, on the frontier between the plain and the bocage, moving on to describe religious practice, and ending with events and trends from 1648 to 1724. The use of quantitative methods also recalls the regional monographs by the disciples of Braudel and Labrousse.50
In turn, the work of the Le Bras circle (like that of Ariès) inspired the work of some Annales historians as they climbed from the cellar to the attic. Later regional studies of Anjou, Provence, Avignon and Brittany focused more sharply on culture than their predecessors, and in particular on attitudes to death or ‘the culture of death’ (la culture macabre) as revealed by images, rituals and so on. As Le Goff put it in the preface to one of these studies, ‘death is in fashion’ (la mort est à la mode).51
The most original of these studies is surely Vovelle's.52 A Marxist historian of the old regime and the French Revolution, ‘formed in the school of Ernest Labrousse’ as he puts it, Michel Vovelle became interested in the problem of ‘dechristianization’. He thought of trying to measure this process by means of the study of attitudes to death and the beyond as revealed by wills. The result, written up in his doctoral thesis, was a study of Provence on the basis of a systematic analysis of some 30,000 testaments. Where earlier historians had juxtaposed quantitative evidence about mortality with more literary evidence about attitudes to death, Vovelle attempted to measure the attitudes as well. He paid attention, for example, to references in wills to the protection of patron saints; the numbers of masses that the testator wanted to be said for the repose of his or her soul; the arrangements for funerals, and even the weight of the candles used during the ceremony.
Vovelle identified a major shift from what he called the ‘baroque pomp’ of seventeenth-century funerals to the modesty of their eighteenth-century counterparts. His main assumption was that the language of wills reflected ‘the system of collective representations’, and his main conclusion was the identification of a trend towards secularization, suggesting that the ‘de-christianization’ of the years of the French Revolution was spontaneous rather than imposed from above, and that it formed part of a larger trend. Particularly noteworthy is the way in which Vovelle charted the spread of new attitudes from the nobility to the artisans and peasants, and from large towns such as Aix, Marseilles and Toulon, through small towns such as Barcelonette down to the villages. His arguments were illustrated by abundant maps, graphs and tables.
Baroque Piety and Dechristianization, as Vovelle's study is called, caused something of an intellectual sensation, thanks in particular to its virtuoso use of statistics, controlled by an acute sense of the difficulties of interpreting them. It was this book, together with the work of Ariès, that inspired Pierre Chaunu to organize a collective investigation of attitudes to death in Paris in the early modern period, using similar methods. What Ariès had done singlehanded on the history of death, in his deliberately impressionistic way, was now complemented by the collective and quantitative researches of the professionals.53
This appropriation of the afterlife by lay historians armed with computers remains the most remarkable example of serial history at the third level. However, other historians of culture have also made effective use of quantitative methods, notably for the history of literacy and the history of the book.
The study of literacy is a domain of cultural history that lends itself to collective research and to statistical analysis. Indeed, a French headmaster had already carried out research in this area in the 1870s, using signatures to marriage registers as his source and noting the great variations between the figures for different départements, as well as the rise of literacy from the late seventeenth century onwards. In the 1950s, two historians reanalysed his data and presented in cartographic form the dramatic contrast between two Frances, separated by a diagonal line from St Malo to Geneva. North-east of this line, literacy was relatively high, south-west of the line it was low.54
The most important project in this domain, begun in the early 1970s, was carried out at the Hautes Études and was directed by François Furet (a former student of Ernest Labrousse who had previously worked on the quantitative analysis of social structures) and Jacques Ozouf.55 The project dealt with changing levels of literacy in France from the sixteenth to the nineteenth centuries. The researchers drew on a wider range of sources than before, from the census to the army's statistics on conscripts, and so they were able to argue rather than assume the relationship between the ability to sign one's name and the ability to read and write. They confirmed the traditional distinction between the two Frances, but refined the analysis. Among other interesting conclusions, they noted that in the eighteenth century, literacy spread faster among women than it did among men.56
Research on literacy was accompanied by research on what is known as the history of the book or the history of reading, research that focuses not on great works but on trends in book production and on the reading habits of different social groups. For example, Robert Mandrou's study of popular culture, mentioned above, was concerned with chap-books, the so-called ‘Blue Library’ (la Bibliothèque Bleue, given this name because the books had covers made from the blue paper used for packing sugar). These books, which cost only one or two sous, were distributed by pedlars and produced in the main by a few families of printers at Troyes in north-eastern France, where the literacy rate was highest. Mandrou examined a sample of some 450 titles, noting the importance of pious reading (120 works), almanacs, and even romances of chivalry. He concluded that this was essentially an ‘escapist literature’, that it was read mainly by peasants, and that it revealed a ‘conformist’ mentality. The last two conclusions have been rejected by other scholars working in this field.57
At much the same time as Mandrou, the Sixth Section launched a project for collective research on the social history of the book in eighteenth-century France, making much more use of quantitative methods.58 However, the key figure in French book history was another of Febvre's collaborators, Henri-Jean Martin of the Bibliothèque Nationale. As we have seen, Martin worked with Febvre on a general survey of the invention and spread of printing, The Coming of the Book (1958). He went on to write a rigorously quantitative study of the book trade and the reading public in seventeenth-century France. This analysed not only trends in book production, but also the changing tastes of different groups of the reading public, notably the magistrates of the Parlement of Paris, as revealed by the proportions of books on different subjects to be found in their private libraries. Martin, together with Roger Chartier, later directed a massive collective work on the history of the book in France.59
One of the leading collaborators in the Sixth Section's history of the book was Daniel Roche. Another former student of Labrousse's, Roche was and remains especially concerned with the social history of culture, especially the French Enlightenment. His doctoral thesis, concerned with provincial academies (in other words, learned societies) in the eighteenth century, was remarkable for the way in which it integrated culture with politics and quantitative with qualitative approaches to the past.60
Roche organized a research team of his own in the mid-1970s to study the everyday life of ordinary people in eighteenth-century Paris.61 In the book that emerged from this collective research, The People of Paris (1981), a substantial chapter was devoted to popular reading, concluding that reading and writing played an important part in the lives of some groups within the lower classes, servants in particular. The most striking feature of The People of Paris, however, was its location of this analysis of reading within the framework of a general study of the material culture of ordinary Parisians. This is a study in serial history based essentially on inventories post-mortem, full of details about the clothes and furniture of the deceased, details that the principal author interprets with great skill to build up a picture of everyday life.
Roche went on to write a social history of clothing in early modern France and a book on ‘ordinary things’, viewed as part of the rise of the consumer society.62 In these studies he combined Braudel's concern with ‘material civilization’ with the more rigorous methods of his master, Ernest Labrousse and also with the interest in anthropology characteristic of the third generation. He tells a story about meeting Roland Mousnier in the street one day. ‘Monsieur Roche, are you still a Durkheimian?’ ‘More than ever’, he answered.63
The quantitative approach to history in general, and to cultural history in particular, can obviously be criticized as reductionist. Generally speaking, what can be measured is not what matters. Quantitative historians can count signatures to marriage registers, books in private libraries, Easter communicants, references to the ‘court of heaven’ and so on. The problem remains whether these statistics are reliable indicators of literacy, piety or whatever big subject the historian wants to investigate. Some historians have argued the case for the reliability of their figures; others assumed it. Some have made use of other types of evidence to make their statistics meaningful, others have not. Some have remembered that they are dealing with real people, others appear to have forgotten it. Any evaluation of the trend must discriminate between the modest and the extreme claims made for the method and also between the various manners in which it has been employed, crudely or with finesse.
By the later 1970s, the dangers of this kind of history had become apparent. Indeed, there was something of a backlash, itself undiscriminating, against the quantitative approach. At much the same time there was a more general reaction against much of what Annales had stood for, and in particular against the dominance of both social and structural history. Looking at the positive side of these reactions, we may distinguish three trends: an anthropological turn, a return to politics and a revival of narrative.
The Anthropological Turn. The anthropological turn might be described with more exactitude as a turn towards cultural or ‘symbolic’ anthropology. After all, as we have seen, Bloch and Febvre had read their Frazer and their Lévy-Bruhl, and they had made use of this reading in their work on medieval and sixteenth-century mentalities. The classicist Louis Gernet and the sinologist Marcel Granet were both indebted to the ideas of Durkheim.64 Braudel was familiar with the work of Marcel Mauss, which underlies his discussion of cultural frontiers and exchanges.
Later, as we have seen, Duby drew on the work of Mauss on the function of gifts in order to understand the economic history of the early Middle Ages. Claude Lévi-Strauss published articles in Annales, while Le Goff and Le Roy Ladurie drew on his ideas in their analyses of the myth of Mélusine.65 Vernant and Vidal-Naquet did the same in their discussions of Greek myth, thought and even the organization of society, for example in Sparta.66 An awareness of anthropology encouraged historians such as Michel Vovelle and Mona Ozouf to study rituals, among them the official festivals that followed the French Revolution.67 Philippe Joutard carried out quasi-anthropological fieldwork in his study of the memory of the revolt of the Camisards as transmitted through oral tradition.68
All that earlier historians seem to have wanted from their neighbour discipline was the opportunity to raid it from time to time in search of new concepts. Some historians of the 1970s and 1980s, however, harboured rather more serious intentions. They even thought in terms of marriage, of combining the two disciplines in the form of ‘historical anthropology’ or ‘anthropological history’ (ethnohistoire).69
In 1975, Jacques Le Goff, for instance, offered the first of what would become recurrent seminars at the Hautes Études on the ‘historical anthropology’ of the medieval world (he and others in the group had previously referred to historical ‘sociology’). Le Goff went on to found a group concerned with the historical anthropology of the medieval West, an initiative that was followed by the foundation of groups or workshops for religious anthropology and the anthropology of scholasticism. His own work has ranged from the structural analysis of medieval legends to the study of symbolic gestures in social life, notably the ritual of vassalage but also the custom of charivari or ‘rough music’, mocking older men who married younger women.70
What attracted historians such as these was above all the new symbolic anthropology. The names that recur in their footnotes include Erving Goffman and Victor Turner (both of whom stress the dramaturgical elements in everyday life) and especially Pierre Bourdieu.
Bourdieu, who turned from anthropological studies of Algeria to the sociology of contemporary France – or from the sociology of Algeria to the anthropology of France – has been influential in many ways. His ideas on the sociology of education, especially the idea of education as a means of ‘social reproduction’, informed a number of studies on the social history of schools and universities.71 His notion of ‘symbolic capital’ underlay some histories of conspicuous consumption.72
Historians of mentalities, popular culture and everyday life all learned from Bourdieu's ‘theory of practice’. His replacement of the idea of social ‘rules’ (which he considered too rigid and determinist) by more flexible concepts such as ‘strategy’ and ‘habitus’ (embodied values that shape individual responses to situations) has affected the practice of French historians (among others) so pervasively that it would be misleading to reduce it to specific examples such as the matrimonial strategies of nobles in the Middle Ages.73 Roche, for instance, despite his story about Durkheim, seems to have learned most from the work of Pierre Bourdieu on the social meanings of material objects and cultural practices. Increasing interest has been shown in individual actors and their networks.74
From the 1970s onwards, historical anthropology – which might be described more exactly as anthropological history – became a major focus of attention in the Annales group and beyond. Alain Corbin, for example, called his histories of sensibility examples of ‘the anthropology of the senses’ (anthropologie sensorielle). Krzysztof Pomian, a Polish historian (formerly a philosopher), long associated with Annales, produced a historical anthropology of collecting, focused on eighteenth-century France.75 Emmanuel Le Roy Ladurie moved in the same direction in a series of studies, of which the most famous by far is his Montaillou (1975).
Montaillou is a village in the Ariège in south-west France, a region in which the Cathar heresy had considerable appeal at the beginning of the fourteenth century. The heretics were pursued, interrogated and punished by the local bishop, Jacques Fournier. The register of the interrogations has survived, and it was published in 1965. It was doubtless Le Roy's interest in social anthropology that allowed him to see the value of this source, not only for the study of the Cathars, as the specialists on the history of heresy were already aware, but also for French rural history. He noticed that twenty-five individuals, about a quarter of the suspects named in the register, came from a single village. His inspiration was to treat the register as the record of a set of interviews with these twenty-five people (about 10 per cent of the population of the village).
All he had to do, Le Roy claimed, was to rearrange the information given by the suspects to the inquisitors in the form of a community study of the kind that anthropologists have so often written. He divided it into two parts. The first dealt with the material culture of Montaillou, the houses, for example, built of stone without mortar, allowing neighbours to observe and to listen to one another through the chinks. The second part of the book was concerned with the mentalities of the villagers – their sense of time and space, childhood and death, sexuality, God and nature. Like Braudel, Le Roy describes and analyses Mediterranean culture and society, but in his case no one could say that people had been left out of his book. Montaillou attracted a huge readership, and remains in the memory essentially because its author has the gift for bringing individuals back to life, from the gentle, freedom-loving Pierre Maury, ‘the good shepherd’, to the local noblewoman, the sexy Béatrice des Planissoles, and her seducer Pierre Clergue, an aggressive and self-confident priest.
Montaillou was also an ambitious study of social and cultural history. Its originality did not lie in the questions it addressed, which, as we have seen, were the questions asked by two generations of French historians, including Febvre (on unbelief), Braudel and Chaunu (on the house), Ariès (on childhood), Flandrin (on sexuality) and so on. Le Roy was one of the first to use inquisition registers for the reconstruction of everyday life and attitudes, but he was not alone in this either. The novelty of his approach lay rather in his attempt to write a historical community study in the anthropological sense – not a history of a particular village, but a portrait of the village, told in the words of the inhabitants themselves, and a portrait of a larger society, which the villagers represent. Montaillou is an early example of what came to be called ‘microhistory’. The author studied the world in a grain of sand, or, in his own metaphor, the ocean through a drop of liquid.
It is on this very point that some of the most serious criticisms of the book have concentrated.76 Montaillou has been faulted (apart from various inaccuracies of detail) for an insufficiently critical use of its main source, which Le Roy once described as ‘the unmediated testimony of the peasants about themselves’ (le témoignage sans intermédiaire, que porte le paysan sur lui-même).77
It is of course nothing of the kind. The villagers gave their evidence in Occitan and it was taken down in Latin. They were not spontaneously talking about themselves, but replying to questions under threat of torture. Historians cannot afford to forget these intermediaries between themselves and the men and women they study.
The second main criticism of the book – and of the increasingly popular microhistorical approach that it has helped to inspire – raises the question of typicality. No community is an island, not even a mountain village such as Montaillou. Its connections with the outside world, as far away as Catalonia, emerge clearly in the book itself. The question remains: What larger unit does the village represent? Of what ocean is it a drop? Is it supposed to be typical of the Ariège, the south of France, the Mediterranean world or the Middle Ages? Despite his previous experience with statistics and samples, the author failed to discuss this crucial problem of method. Could this be because he wrote Montaillou in reaction against the aridities of quantitative history?
As in the case of the stone houses of the village itself, it is easy to pick holes in Montaillou. It deserves to be remembered above all for its author's power of bringing the past to life, and also for putting the documents to the question, reading them between the lines, and making them reveal what the villagers did not even know they knew. It is a brilliant tour de force of the historical imagination, and a revelation of the possibilities of an anthropological history.
The Return to Politics. Perhaps the most notorious charge against the Annales has been the group's supposed neglect of politics, a charge to which the journal seemed to confess by carrying on its masthead, until 1994, the slogan ‘Économies. Sociétés. Civilisations’, without any mention of states. There is indeed some substance in the criticism, provided that it is formulated more precisely.78
Febvre and Braudel may have concentrated their efforts on academic rather than national politics, but a number of the leading historians of the group were activists in post-war France, often as members – at least for a time – of the Communist Party. The reminiscences of one of them paint a vivid picture of the party meetings, denunciations, expulsions and resignations of the years following 1956.79
The charge of neglecting politics was of course directed against the historical work of the group, but here too nuances are needed. It would be difficult, for instance, to sustain the argument in the case of Marc Bloch. His Royal Touch was intended as a contribution to the history of ideas of kingship. His Feudal Society began with an account of the Viking, Muslim and Hungarian invasions of Western Europe, and included a long section on feudalism as a form of government.
In the case of Lucien Febvre, the charge has more substance. Although he had discussed the Revolt of the Netherlands at considerable length in his thesis on Philip II and Franche-Comté, Febvre later denounced political history with his customary violence, and turned to religion and mentalities. In the case of Braudel, it should be noted that the structural section of The Mediterranean includes chapters on empires and the organization of war. It was the history of political and military events that he dismissed as the most superficial kind of history, and even then he devoted a good deal of attention to them.
The regional studies of early modern France that bear the imprint of Annales generally confined themselves to economic and social history. Goubert's Beauvais makes an obvious example. All the same, no one can label Goubert an unpolitical historian. He went on to write a book on Louis XIV and a study of the old regime of which the second volume is concerned with power.80
The region may not offer a framework appropriate for a study of old-regime politics. Such an assumption may well have deterred the authors of some regional studies from including a section on politics. However, the work of Mousnier's pupils on popular revolt, together with some recent American studies of early modern politics at a regional level, suggest that the assumption was mistaken and that a splendid opportunity for ‘total history’ was lost. The obvious exception to this rule, as we have seen, was Le Roy Ladurie, who did discuss revolts in Languedoc (if not the administration of the province), and who later produced some explicitly political studies.81
The medievalists in the Annales group did not dismiss political history, even if they devoted more attention to other topics. Georges Duby, who began as an economic and social historian and shifted to the history of mentalities, wrote a monograph on a medieval battle, Bouvines. His account of the genesis or reactivation of the idea of the three estates placed this idea in a political context, the crisis of the French and other monarchies. Jacques Le Goff declared that politics is no longer the ‘backbone’ of history in the sense that ‘it cannot aspire to autonomy’. However, in 1996 he produced a major study of a medieval ruler, St Louis.82
It is scarcely surprising, though to find that most attention to politics has been paid by the Annalistes who concern themselves with what the French call ‘contemporary history’ – in other words, with the years since 1789. François Furet and Michel Vovelle, who have devoted much of their time to the study of the French Revolution, cannot be accused of neglecting politics, despite their other historical interests. Mona Ozouf's study of revolutionary festivals was a contribution to political as well as cultural history. Michelle Perrot, a former student of Labrousse, made her reputation with a study of strikes.83
Furet caused a considerable stir in 1978 when he published a book about the French Revolution that criticized earlier social interpretations of that event, replacing them with a turn to language, ‘discourse’ and imagination; in other words suggesting that conflicts between social classes were less important than disputes over the meaning of key concepts such as equality and democracy. The publication of this study was itelf a political event, reviving conflicts between the French Communist Party (to which Furet had once belonged), the moderate Left (to which he later gave his allegiance) and the Right (which he had joined by this time). Furet's move from socio-economic history via the study of literacy to the history of discourse was associated with a critique of quantitative methods and even of the Annales approach in general.84
Again, politics has been central to the work of Marc Ferro, historian of the Russian Revolution and the First World War, an individual whose interest in events did not prevent Braudel from appointing him as secretary of the journal Annales. Ferro's books demonstrate his concern with the relation between events and structures. He approached the Revolution as a social historian, and later wrote on the history of the Russian army ‘from below’ and on the rise of the Soviet bureaucratic system.85
However, the outstanding figure in this domain is surely Maurice Agulhon. Agulhon is the author of The Republic in the Village, a study of the political behaviour of ordinary people in the Var (in Provence) from 1789 to 1851.86 This study employs a broadly Marxist framework, that of the growth of political consciousness. The years 1815–48 are described as the years of preparation, in which conflicts over encroachments on common rights (notably over forest timber), together with the ‘widening of cultural horizons’ following the spread of literacy, encouraged the growth of political consciousness in the region. The brief years of the Second Republic, 1848–51, are presented as the years of ‘revelation’, in which the ordinary people of the Var voted for the first time and voted for the Left.
Although it deals more with the countryside than with towns, it is tempting to describe Agulhon's study as concerned with ‘the making of the Provençal working class’.87 The parallel with Edward Thompson can be extended. Both historians were ‘open’, empiricist, eclectic Marxists.88 Both were concerned with forms of sociability. Thompson discussed friendly societies and their ‘rituals of mutuality’; Agulhon, thanks to whom the word sociabilité has become current coinage in France, had studied masonic lodges and Catholic confraternities from this point of view, and went on to study the bourgeois ‘circle’ and the café.89 Both historians took culture seriously. Thompson described the tradition of popular radicalism; Agulhon described charivaris and carnivals, such as the ‘seditious carnival’ of Vidauban in 1850; mild enough if compared to the Carnival of Romans in 1580, but significant as an illustration of the opposed but complementary processes of ‘archaism’ and modernity, the ‘folklorization’ of politics and the politicization of folklore.90
To sum up, Febvre and Braudel may not have ignored political history, but they did not make it their highest priority. The return to politics in the third generation was a reaction against Braudel, and also against other forms of determinism (notably Marxist ‘economism’). It was associated with a rediscovery of the importance of agency as opposed to structure. It was also associated with a sense of the importance of ‘political culture’, ideas and mentalities. As a result of these changes, political history underwent a renewal.
The Revival of Narrative. The return to political history was linked to the reaction against determinism which also inspired the anthropological turn, as we have seen. It was also associated with a revival of interest in the narrative of events.91 Events are not always political – think of the Great Crash of 1929, the great plague of 1348 or indeed of the publication of War and Peace. All the same, discussions of political history, the history of events and historical narrative are closely intertwined. Parallel to the so-called ‘return to politics’, there was a ‘revival of narrative’ among the Annalistes of the third generation in France, as there was among historians elsewhere. The phrase ‘revival of narrative’ was coined by the British historian Lawrence Stone, who ascribed the trend to ‘a widespread disillusionment with the economic determinist model of historical explanation’ employed by Marxist and Annales historians alike, and especially with its relegation of culture to the superstructure or ‘third level’. There can be little doubt that Stone perceived a significant trend, but once again, nuances are in order.
The contemptuous dismissal of ‘the history of events’ (histoire événementielle) by Durkheim, Simiand and Lacombe was discussed in the overture to this book. Febvre's stress on problem-oriented history suggests that he shared this view, despite the space given in his doctoral thesis to the events of the Revolt of the Netherlands. Marc Bloch, so far as I know, never denounced the history of events, but he never wrote that kind of history either.
As for Braudel, he both denounced it and wrote it. More exactly, as we have seen, he declared the history of events to be the surface of history. He did not say that this surface was uninteresting – on the contrary, he described it as ‘the most exciting of all.’92 Its interest for him, however, lay in what it might reveal about the ‘deeper realities’, the currents below the surface. For Braudel, events were simply mirrors reflecting the history of structures.
In his magisterial study of time and narrative, the philosopher Paul Ricoeur argued that all works of history are stories, even Braudel's Mediterranean. His demonstration of certain similarities between conventional and structural histories (in their temporality, their causality and so on) is difficult to rebut. All the same, to call the Mediterranean a narrative history is surely to employ the term ‘narrative’ in such a broad sense that it loses its usefulness.93
Most of the regional monographs of the 1960s and 1970s went further than Braudel in this direction in the sense that they included no narrative at all. The exception was Le Roy Ladurie's Peasants of Languedoc, in which, as we have seen, structural analysis alternated with accounts of events, notably protests: the Carnival of Romans in 1580, the rising in the Vivarais in 1670 and the revolt of the Camisards in 1702. Le Roy's treatment of events as reactions or responses to structural change was not far from Braudel's view of events as mirrors or as litmus papers revealing underlying structures.
A similar point might be made about a book that Georges Duby published in 1973, a book that might well have shocked Febvre, since it dealt not only with an event but with a battle, the battle of Bouvines on 27 July 1214. The book was indeed commissioned for a rather old-fashioned series, aimed at the general public and entitled ‘Days that made France’ (Journées qui ont fait la France). All the same, Duby did not return to old-fashioned narrative history. He used contemporary accounts of the battle to illuminate medieval attitudes to war, and discussed later views of Bouvines as a ‘myth’ that revealed more about the narrators than the event that they narrated.94
The obvious question that these studies do not raise is whether some events at least might not modify structures, rather than simply reflecting them. What about the events of 1789, say, or 1917? The sociologist Émile Durkheim, to whom the critics of histoire événementielle owe so much, was prepared to dismiss even 1789 as a symptom rather than a cause of social change.95 However, some of his successors have moved away from this extreme Durkheimian or Braudelian position. For example, a sociological study of an area in western France, the département of the Sarthe (a study that began with a favourable reference to Febvre, as well as using the regressive method associated with Bloch), argued the need to take the events of 1789 and their aftermath into account in any attempt to explain the political attitudes of the region, divided as it still was in 1960 into a left-wing east and a right-wing west).96
Le Roy Ladurie drew attention to the implications of this study in an essay discussing what he variously called the ‘traumatic’ event, the event as ‘catalyst’ and the creative event’ (événement-matrice). His use of such divergent metaphors suggests that he had not made up his mind about the importance of events, while his article went no further than a general recommendation to the historian to reflect on the relation between events and structures.97 Some years later, however, Le Roy returned to the Carnival of Romans, which he had discussed briefly in the Peasants of Languedoc and made it the subject of a new book. He analysed the event in the language of the anthropologist Victor Turner as a ‘social drama’ that made manifest the conflicts latent in that small town and the countryside around it. In other words, the event was a symptom rather than a cause.98
The Carnival of Romans was not a great event. It is more difficult to dismiss as mere reflections of social or mental structures the events of 1789, say, or the Great War of 1914–18, or the Revolution of 1917 (all topics on which, as we have seen, Annalistes have written). In an article in Annales published for the bicentenary of the French Revolution in 1989, François Furet went so far as to suggest not only that the Revolution broke the old structures and gave France her political ‘patrimony’, but even that the events that took place in a few months of 1789 were decisive.99
One more feature of the third generation of Annales deserves our attention here. It was at this time that their kind of history became popular in France. Braudel's Mediterranean, like the works of Bloch and Febvre, had not sold many copies when it was first published. Only in 1985, when sales reached 8,500 copies, could the book be described as a best-seller. Montaillou, on the other hand, quickly reached the top of the non-fiction best-seller list in France, its sales boosted when President François Mitterand admitted on television that he had been reading it, while the village itself was almost buried under a mass of tourists. Duby's Three Orders, although written for scholars rather than the general public, sold 50,000 copies soon after its publication.100
Montaillou was a book written at the right place and the right time, carried along by the waves of ecology and regionalism, but its success was only the most spectacular example of the interest in the ‘new history’ shown by the French public in the 1980s and 1990s. When Braudel's trilogy Civilization and Capitalism was published in 1979, it received attention in the media on an altogether different scale from his earlier books. In the third generation, some members of the Annales group appeared regularly in television and radio programmes and even produced them, like Georges Duby or Jacques Le Goff, who was from 1968 onwards the anchor for a weekly programme on France-Culture, ‘Les Lundis de l'Histoire’. Other Annalistes, such as Pierre Chaunu, Mona Ozouf, Michelle Perrot and Roger Chartier, wrote on a regular basis for newspapers and journals such as Le Figaro, Le Monde, L'Express and Le Nouvel Observateur. It is difficult to think of any other country or any other period in which so many professional historians were so firmly established in different media of communication.
In the 1950s and 1960s, the writings of the Annales historians took the form of large volumes appearing in small editions from Armand Colin (the faithful publishers of the journal) or from the Hautes Études themselves. From the 1970s, however, they were more likely to be slim volumes appearing in larger print runs from leading commercial publishers, often in a series edited by other Annales historians. Ariès and Mandrou edited a series on ‘Civilizations and Mentalities’ for Plon. Agulhon edited a historical series for Aubier Montaigne, while Duby edited more than one for Seuil, including multi-volume histories of rural France (1975–6), urban France (1980–5) private life (1985–7) and women (1991–2). An example of still closer collaboration between historians and publishers is offered by Pierre Nora, who taught at the Hautes Études as well as working for Gallimard. It was Nora who founded the well-known series Bibliothèque des Histoires, a series that includes a number of studies by his colleagues.
The point of these remarks is not to suggest that the media created the wave of interest in this kind of history, though they certainly encouraged it. The producers and the publishers must have believed that there was a demand for history in general, and in particular for sociocultural history in the Annales style. History remained in the limelight for at least the early years of the fourth generation of the movement, to which we now turn.