History as a Kind of Writing: Textual Strategies in Contemporary French Historiography

History as a Kind of Writing

History as a Kind of Writing

Textual Strategies in Contemporary French Historiography

Philippe Carrard

The University of Chicago Press

CHICAGO & LONDON

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2017 by The University of Chicago

All rights reserved. No part of this book may be used or reproduced in any manner whatsoever without written permission, except in the case of brief quotations in critical articles and reviews. For more information, contact the University of Chicago Press, 1427 E. 60th St., Chicago, IL 60637.

Published 2017

Printed in the United States of America

Translated and expanded from the original French edition published as Le passé mis en texte: Poétique de l’historiographie française contemporaine

© 2013 by Armand Colin, Paris

26 25 24 23 22 21 20 19 18 17    1 2 3 4 5

ISBN-13: 978-0-226-42796-6 (cloth)

ISBN-13: 978-0-226-42801-7 (e-book)

DOI: 10.7208/chicago/9780226428017.001.0001

Library of Congress Cataloging-in-Publication Data

Names: Carrard, Philippe, author.

Title: History as a kind of writing : textual strategies in contemporary French historiography / Philippe Carrard.

Other titles: Passé mis en texte. English

Description: Chicago ; London : The University of Chicago Press, 2017. | Translated and expanded from the original French edition published as Le passé mis en texte: Poétique de l’historiographie française contemporaine | Includes bibliographical references and index.

Identifiers: LCCN 2016025837 | ISBN 9780226427966 (cloth : alk. paper) | ISBN 9780226428017 (e-book)

Subjects: LCSH: Historiography—France. | History—Methodology. | Literature and history—France. | History—Philosophy.

Classification: LCC DC36.9 .C37913 2017 | DDC 907.2/044—dc23 LC record available at https://lccn.loc.gov/2016025837

This paper meets the requirements of ANSI/NISO Z39.48–1992 (Permanence of Paper).

Contents

A Note about Translations and Documentation

Acknowledgments

Preface

Introduction: French History and Its Manuals

CHAPTER 1  Dispositions

Squabbles about Narrative

Linear Narratives

Writing the Event

Synchronic Cross Sections

Stage Narratives

Theory of a Practice

CHAPTER 2  Situations

Enunciations

Perspectives

The Discourse of the Absentee

Readerships

CHAPTER 3  Figures

Attestations

References

Computations

Uncertainties

Wordplay and Figures of Speech

Conclusion

Notes

References

Index

A Note about Translations and Documentation

Some of the works I have selected exist in published English translations. These translations, however, have often edited the French text to conform to the standards of Anglo-American scholarly writing. In the process they have toned down or erased some textual features I want to focus on. For the sake of homogeneity, I have therefore used the French editions throughout and have provided my own translations. I have included in parentheses a translation with the first mention of every title when it does not involve obvious English cognates. This translated title is in italics if it comes from a published English edition, in roman type if the translation is mine. I am using parenthetical documentation by author and date of publication referring to a reference list.

Acknowledgments

First of all I must thank the institutions that have enabled me to complete this project: Dartmouth College and the University of Lausanne (UNIL, Switzerland), specifically the Baker-Berry and Dorigny libraries, whose cooperation greatly aided my research. At Dartmouth I am particularly thankful to the Comparative Literature Program, which provided me with a home, and to the Humanities Resource Center, whose staff solved several computer problems. At the UNIL, I am grateful to the librarians who gave me faculty privileges, allowing me to borrow all the recent books in French historiography that were necessary to my project.

On this side of the Atlantic, I am especially indebted to Hans Kellner, Dominick LaCapra, Allan Megill, and Gerald Prince, who have supported my work throughout the years, granting me the benefit of their knowledge in such areas as historiography, literary theory, and the philosophy of history. I must also thank the International Society for the Study of Narrative, whose conferences year after year gave me the opportunity to test my views on conventions of writing in nonfiction, particularly in historiography.

Overseas, I am notably beholden to Bruno Auerbach, Laurent Avezou, Patrick Boucheron, Raphaëlle Branche, Christophe Charle, Quentin Deluermoz, François Hartog, Dominique Kalifa, Bertrand Muller, Antoine Prost, and Peter Schoettler, who gave me extensive information about the debates that now occupy the French historical community, told me about their own research, and helped with the constitution of my corpus. In Switzerland, I owe special thanks to my friends Mondher Kilani and Marianne Kilani-Schoch, who never failed to inquire about the state of my writing, as well as to my sisters Marie-Claude Dupraz and Christine Carrard and my brother-in-law Alfred Dupraz, who went along with my not so flexible schedule in the most accommodating manner.

The University of Chicago Press has encouraged the project from the start. I wish to acknowledge my editor Douglas Mitchell and his assistant Kyle A. Wagner, for their accessibility and the quality of their professional advice; the two anonymous readers whose detailed reports saved me from a number of errors and helped me clarify several important points; and Alice M. Bennett for her meticulous copyediting.

Once again, this book could not have been written without the complicity and support of my first and most demanding reader, Irene Kacandes. Many passages in the text retain the trace of her familiarity with linguistics, narrative theory, and discourse analysis, and most pages bear the imprint of her editorial assistance. Her emotional support was also most valuable during the rough moments that are part of any scholarly project.

Preface

Like most academic enterprises, this project originates in the perception of a void and the somewhat presumptuous assumption that there are ways of filling it. In studies conducted since the 1970s, historiographic works have at times been regarded as textual constructs and dissected with the tools of literary theory. Yet studies such as Hayden White’s Metahistory (1973), Linda Orr’s Jules Michelet (1976), Lionel Gossman’s Between History and Literature (1981), and Stephen Bann’s The Clothing of Clio (1984) have mainly focused on historical production in the eighteenth and nineteenth centuries; with a few exceptions, notably Axel Rüth’s Erzählte Geschichte (Narrated history) (2005) and chapters in Sande Cohen’s Historical Culture (1986), Paul Ricoeur’s Temps et récit (Time and Narrative) (1983), and Jacques Rancière’s Les mots de l’histoire (The Names of History) (1992), they have not troubled with contemporary research. Conversely, overviews of that research have hardly touched on issues of writing. In the French domain in particular, such surveys as Peter Burke’s The French Historical Revolution (1990), Philip Daileader and Philip Whalen’s French Historians: 1900–2000 (2010), Christian Delacroix, François Dosse, and Patrick Garcia’s Les courants historiques en France: XIXe–XXe siècles (Historical trends in France: 19th-20th centuries) (2007), Lutz Raphael’s Die Erben von Bloch und Febvre (The heirs of Bloch and Febvre) (1994), and Jean-François Sirinelli, Pascal Gauchy, and Claude Gauvard’s Les historiens français à l’oeuvre: 1995–2010 (French historians at work: 1995–2010) (2010), have mostly reviewed the topics that French historians are now investigating, together with problems of method and epistemology. They have not, however, considered processes of textualization, have done so in passing, or have dismissed such an exercise as irrelevant. The table of contents of Historiographies (2010) edited by Delacroix, Dosse, Garcia, and Offenstadt is in this respect most revealing. Of the 120 entries of this encyclopedia, only two, The writing of history and Narrative, bear on what Ricoeur (2000, 169) calls the representative stage of the historical operation, that is, the stage when researchers write up the materials they have gathered.¹

My purpose is to pick up where these two groups of studies leave off. More precisely, it is to continue the examination of history as a textual practice, while focusing on contemporary French historiography. By historiography, I mean here the set of works produced by historians at a specific time (Offenstadt 2011, 5), and by contemporary not the period that extends from 1789 to today (Noiriel 1999, 7) but more informally the current period, the one that started in 1945. Since I aim to provide an overview of the writing practices that have prevailed during this time, the corpus I plan to analyze is largely ecumenical. It includes studies produced by historians who belong to the Annales school and its continuation in the New History. But it also admits texts that testify to the reaction against the supposed hegemony of the Annales, beginning with instances of what Philippe Poirrier (2009) has identified as the renewal of economic history, the rehabilitation of political history, the rise of cultural history, and the history of the present time. The works I have selected should both represent these recent trends and illustrate features of writing. Some of them (e.g., Le Roy Ladurie’s Montaillou [1975], Braudel’s L’identité de la France [1986], Rousso’s Le syndrome de Vichy [1990]) are familiar and have been dissected by critics many times over, while others are less well known and have not drawn the same attention. I have included younger historians and especially historiennes (e.g., Christine Bard, Michèle Riot-Sarcey, Quentin Deluermoz, Antoine Lilti) whom specialists regard as having made significant contributions but whose research is less known in English-speaking countries because little of it has been translated. For the sake of coherence, I have restricted my sample to works written by professional historians, or at least by historians who conform to the rules of scholarly research and discourse. Although the texts that make up this sample have enough common features to be treated as a whole, I do not make contemporary French historiography into a stable and unified entity. Since no system functions without jolts, I also point to the conflicts, paradoxes, and inconsistencies that are part of the works I am considering.

Literary theory distinguishes between criticism (the interpretation of individual texts) and poetics (the study of the rules, codes, and conventions that operate in a given set of texts). My purpose is to undertake a poetics, aimed in this instance at describing the discursive conventions that shape the texts in my corpus. I am all too aware that this kind of endeavor has been attacked over the past thirty years: in France by historians (e.g., Noiriel 2005, 116) who insist that their colleagues should worry about their trade, not about writing; in English-speaking countries by scholars who hold that the attention paid to processes of textualization in academic disciplines is dangerous (e.g., Palmer, Descent into Discourse [1990]) and even deadly (e.g., Windschuttle, The Killing of History [1994]). These charges seem unfounded. Obviously, writing is an essential component of the historian’s trade, and it can be productive to look closely at this step in the historiographic endeavor. To do so, moreover, does not imply ignoring the properly historical and political dimensions of the texts under consideration. To take just one example: from Roger Chartier (1998) to Marcel Gauchet (1988, 1999), several scholars have emphasized that one of the main changes in French historiography since the 1980s has been a reassessment of the role of the actors—a role that had remained largely unexamined at a time when historical research was mostly concerned with large social and economic phenomena. Poetics provides the means of describing how this reassessment was achieved, by posing certain questions: In this study, are events recounted from the actors’ perspective or from the historian’s? Do these actors have their own voice? And if they are quoted, how are their words recorded and made part of the text? The apparatus of poetics makes it possible to answer these questions, which certainly do not fall under the ahistorical formalism that Palmer and Windschuttle condemn. Indeed, such questions directly concern a specific moment, the reinstatement of agents and their textual staging in French historiography. Let me stress: a specific moment. The status of history as a branch of knowledge is nothing permanent and may itself be the object of historical inquiry. As for the relations between history and literature, as Lionel Gossman (1990), Claude Calame (2012), and others have argued, these have constantly changed from the ancient Greeks to the present day. I will look here at the current state of the dichotomy, which poses an opposition between factual and fictional discourses (Genette 1991; Lemon 1995), that is, between a discourse that reports true events and a discourse in which those events have been invented.

This distinction remains central in Western culture. Heated controversies followed the revelation that Binjamin Wilkomirski’s Fragments (1996), a text offered as a testimony about the Holocaust, was actually a hoax, and similar debates—to which I will return later—have been prompted by historical novels staging real characters, such as Yannick Haenel’s Jan Karski (The Messenger) (2009) and Jonathan Littell’s Les bienveillantes (The Kindly Ones) (2006). Like all binaries, the distinction factual versus fictional may of course be questioned and deconstructed. Hayden White, in Metahistory, shows that the plots in some classics of European historiography in the nineteenth century have literary equivalents—those identified by Northrop Frye in Anatomy of Criticism (1957). But we can also confirm the legitimacy of the demarcation by using criteria that are not only epistemological but textual, that is, by moving the conversation onto the turf of the very postmodern theorists who deny any difference between fiction and history. To begin with the most obvious markers, almost all the works in my corpus are explicitly designated as historical by what poeticians (Genette 1987) call their paratexts, more precisely their peritext: their titles, their subtitles, the text on their back covers, and especially the series in which they are published. Indeed, such series as L’univers historique, Bibliothèque des histoires, Points histoire, and L’épreuve de l’histoire immediately situate the books on whose jackets they appear on the map of discourses, establishing a reading contract and programming a specific reception. Admittedly, readers can always "read as, for example, can construe a book published under the label history" as a piece of fiction; White does precisely this when he finds that Michelet’s Histoire de la Révolution française (1939) has the structure of a romance. To "read as, however, is not synonymous with to take for." While readers are free to treat Histoire de la Révolution française as a novel, they cannot take it for a novel: in the case of Michelet’s work, its publication in Folio histoire and similar scholarly series has assigned a genre to the book, and in so doing has supplied precise instructions about the way it should be received.

Insofar as they account for the mediation of language and posit that historiographic texts are constructed, the procedures I use might be associated with the linguistic turn. Yet viewing history as a kind of writing, as Richard Rorty (1982) viewed philosophy as a kind of writing in an essay on Derrida, does not imply what Dominick LaCapra (2009, 193) calls literal pantextualism, that is, the assumption that everything is text and that historiography has no referential dimension. While historians do not find, ready made in the archives, the form they are about to give to their text, they do assume that documents point to empirical facts. The truth claim (Ricoeur 1983, 135) that follows from this assumption must be taken seriously. For one thing, it plays an important role in the area of formal choices; it involves conventions of writing, conventions that differ from those in fictional works and that shape the reception of the text. It goes without saying, moreover, that if historiography can be examined in its textuality, this approach can in turn be considered in its historicity.² Although I do not have the competence to undertake it myself, I would welcome an inquiry that would account for the current interest in writing history, situate this interest in its intellectual context, and make it part of a plot.³

My study unfolds in three stages. After an introduction that deals with manuals, chapter 1 takes up problems of macrotextual organization, asking whether historiographic texts always have a narrative form and, if not, what their structures might be. Chapter 2 bears on enunciation and perspective, that is, on the identity, role, and position of the speaker in historiographic texts, as well as on the kind of readers those texts are targeting. Chapter 3 treats questions of rhetoric and stylistics, with their political and epistemological implications; it considers the strategies historians employ to warrant the veracity of their accounts, to point to those accounts’ limits, and to make the past more accessible to readers. A conclusion returns to the distinction between factual and fictional discourses, reviewing a few texts published in the 2000s in which French historians question the difference between the genres, play at their borderline, and attempt to shake up what they perceive as the overly rigid protocols of writing in force in their profession.

History as a Kind of Writing revisits, based on a much enlarged corpus, some of the themes I developed in Poetics of the New History. As its title indicates, the latter work was restricted to the analysis of the production of a school that, seen from a distance, was not as hegemonic as both its foes and its supporters claimed and that must now share the stage with recent trends (rehabilitation of political history, etc.) analyzed by Poirrier and others. My aim, however, is not to decide on the debates this evolution has sparked: Has the New History, as Dosse (1987) puts it, really gone to pieces? Have its members taken the critical turn they had theorized in a 1988 issue of the journal Annales? Is scholarly history threatened by memory, understood as the demand for recognition coming from the numerous groups that now reclaim an identity? And has history broken out of the crisis that Noiriel had described in 1996 and reaffirmed in 2005? These questions have been exhaustively addressed in the anthologies mentioned above as well as in the thematic issues that periodicals such as Le Débat have devoted to the quandaries of history. I will therefore not return to these problems—or, rather, I will do so only when they can be reformulated through an analysis of the discursive strategies at work in the texts under consideration. What I will be asking is to what extent the crises, the turns, and the paradigm shifts that history has supposedly undergone since the 1980s have come with parallel revisions in the area of writing, and what effects, if any, those revisions might have had on the politics and epistemology of the discipline.

The scholarly apparatus I draw on is heterogeneous. It includes, besides the anthologies mentioned above, studies by theorists of literature (e.g., Barthes, Cohn, Genette, Phelan, Prince), philosophers of history (e.g., Ankersmit, Certeau, Danto, Ricoeur, Tucker), linguists (e.g., Benveniste, Ducrot, Lakoff, Sperber, Wilson), and historians who have reflected on their trade’s textual practices (e.g., Berkhofer, Chartier, Hexter, Jablonka, Noiriel, Prost). Intent on avoiding circularity, however, I will at times question this apparatus. I will thus ask, among other things, to what extent the corpus selected makes it necessary to revisit current views of historiography, beginning with the idea that historiographic texts always have a narrative structure, that they constitute the (admittedly unattainable) pole of objectivity, and that they are consistently monologic. In this respect, from its own corner, my book aims to participate in the conversation about the current state of the human sciences, at a time when the status of those sciences is being questioned by the different posts (postmodernism, poststructuralism, postfeminism, etc.) that have occupied the intellectual scene since the end of the twentieth century.

Introduction

French History and Its Manuals

Before looking at the ways contemporary French historians write up their materials, it might be useful to go over the instructions they have received during their apprenticeship. Concretely, it might be productive to examine the manuals available to them at the beginning of their careers, setting the rules for conducting and then textualizing their research in the form of articles, books, or other types of writing. Admittedly, historians who have secured a place in the profession do not (or no longer) need to strictly follow the guidelines they had to abide by when they were working on their dissertations; they may even flout them, as some members of the corporation seem to enjoy doing. It remains that didactic works are worth investigating. Indeed, their presence does not point only to the students’ assumed lack of technical competence, which should be remedied as early as possible. More deeply, it tells about an ingrained disciplinary inquietude: anxious to mark out their territory, historians feel they must inform prospective colleagues about the conventions of research and writing they will have to obey. In this regard manuals do more than provide directions; they inscribe the way the historical community perceives itself, as well as the image of its exigencies that it intends to bestow on the next generation. I will thus begin by reviewing some of the pedagogical handbooks that have been available to the French historical community, going from a classic of the late nineteenth century to a few texts published in the early 2000s.

The Method and Its Discourse

The constitution in nineteenth-century France of a body of professional historians, the birth of a scientific history, and the promotion of that history to the rank of academic discipline have been well documented in studies such as William Keylor’s (1975), Pierre-Olivier Carbonell’s (1976), and Pim den Boer’s (1998). The historical school that emerged at the time was long called positivist, a label still found in the encyclopedia La Nouvelle Histoire (1978), edited by Jacques Le Goff, Roger Chartier, and Jacques Revel, and in the Dictionnaire des sciences historiques (1986), edited by André Burguière. That school has now been renamed methodist, with scholars such as Patrick Garcia (2007) and Philippe Poirrier (2009) arguing that its members’ main concern had been to establish a method and suggesting that positivism be reserved for the theory developed by Auguste Comte and represented in history by Louis Bourdeau’s Histoire et historiens: Essai critique sur l’histoire considérée comme une science positive (1888). Garcia and Poirrier also stress that while over the past half century French history has explored new domains and identified new sources, it has largely remained committed to the research model that the methodists had elaborated. Adopting the trademark method school, I will briefly examine that model as it is exposed in the book that is still regarded as the school’s main pedagogical statement: Charles-Victor Langlois and Charles Seignobos’s 1898 Introduction aux études historiques. After outlining the points that deal with research procedures, I will look more closely at the section—largely ignored in the studies that treat the method school—that is of prime interest for my purpose: the few pages that Langlois and Seignobos devote to exposition, that is, to the problems historians have to solve in textualizing their data.

In their preface, Langlois and Seignobos (1992, 24) present the Introduction as a summary sketch designed for first-year Sorbonne students and professional historians eager to reflect on their trade, as well as for a general audience.¹ The book, in fact, is more ambitious, as it synthesizes theoretical statements and practical directions given during the last part of the nineteenth century, by and large since the founding in 1876 of the journal Revue Historique. Its goal is to make history into a true science, distinct both from philosophical speculations about the course of mankind and from literary accounts of the past such as those of Thierry and Michelet. Divided into three parts, the Introduction describes the successive steps historians must follow to attain scientific truth (1992, 18).

Part 1 lists the types of prior knowledge necessary in undertaking any historical investigation. Apprentice researchers must first familiarize themselves with the different ways of gathering documents: of consulting lists and inventories (heuristics). They must also learn the fundamentals of the auxiliary sciences, beginning with epigraphy, paleography, philology, and archaeology. This technical apprenticeship will replace the study of literary and philosophical models that in the past was regarded as indispensable. Part 2 enumerates the analytical operations historians must perform upon the documents they have identified. The first of these operations is external criticism. While processing their data, historians must (re)establish the original text (criticism of restoration); determine where, when, and by whom the text was written (criticism of origin); and classify this text according to a preset system, whether alphabetical, chronological, or some other. They will then move on to the internal criticism of the document (hermeneutics), an approach that consists of two stages: a positive interpretation of what the author meant followed by a negative interpretation of the author’s statements and an analysis of the circumstances in which the document was produced. Checking the validity of the documents, and then comparing them with one another, will allow scholars to establish the individual facts in which their inquiry will be grounded. Part 3 describes the synthetic operations that lead from criticism of the sources to writing of the text. The individual facts identified during the analysis stage must first be grouped. Langlois and Seignobos propose classifying facts in six categories based on their nature: material conditions, intellectual habits, material customs, economic customs, social institutions, and public institutions. Because the purpose of history is to study change, the central question in any investigation will be, Given the specific fact X, what is the evolution of that fact? While tracing this evolution, historians must at times resort to constructive reasoning (207) in order to link pieces together or to replace missing pieces. Furthermore, given the impossibility of communicating a complete knowledge (215), they must occasionally elaborate general formulas (216) whose function is to account for repeated facts. (Note that Langlois and Seignobos, as if anticipating the criticism the Annales would submit them to, admit that history is at least in part a construction, and that its task is to account for repeated facts as well as for individual ones.)

It is in chapter 5 of part 3, titled Exposition, that Langlois and Seignobos take up textualization.² Looking for a really rational type of exposition (239), they first analyze and evaluate the forms historical works have taken over the centuries. Distinguishing five moments in this evolution, they tell a success story that is typical of the scientific optimism of the Third Republic, an optimism found in other texts written at the end of the nineteenth century, such as Gabriel Monod’s 1876 editorial in the first issue of the Revue Historique, emphatically celebrating the advances of the discipline under the title Du progrès des études historiques en France depuis le XVIe siècle.

In antiquity, according to the plot designed in the Introduction, history was conceived as a narrative of memorable events and a collection of good examples of preparing for civic life. Its framework was the life of great men and the development of a community. It was basically a literary genre, and the authors were hardly concerned with evidence. In the Renaissance, historians imitated the ancients, but they also started, influenced by the scholars of the Middle Ages, to include a documentary apparatus in their texts. In the eighteenth century the philosophers regarded history as the study of both events and customs, and they sought to trace the evolution not just of political entities, but also of the arts, the sciences, and industry. During that same period, German scholars invented the manual, a book in which the facts are presented in a scientific, objective and simple fashion (242). At the beginning of the nineteenth century, romanticism brought about a relapse into literary history and its preoccupation with effects (242): the use of local color, the emotional narration of events, and the vain attempt to resurrect the totality of the past. Yet these reactionary tendencies were soon overcome, and scholars in the second part of the century were able to elaborate the truly scientific forms (244) that would make history into a legitimate academic discipline. Langlois and Seignobos recommend two of these forms, as they recommend a parallel division of labor.

Historians at the beginning of their careers should undertake monographs—studies bearing on a specific point, such as a part of the life or the life of an individual, an event or a set of events within close dates, and so on (245). Monographs obey three basic rules: any fact originating in a document must come with a reference to that document and an assessment of its value; chronological order must be strictly followed; and the title must be explicit, to aid in library classification and bibliographical research. Langlois and Seignobos downplay a fourth rule—that monographs should exhaust their subject—because they do not consider this principle as important as the first three. According to them, it is legitimate to do provisional work with the available evidence as long as readers are informed of the documents on which the work is based (246).

Relying on the research done by their younger colleagues, experienced historians will take on general works, whether manuals (collections of known facts methodically arranged) or general histories (narratives of events that have shaped the fate of nations) (247–50). These general works are submitted to the same exigencies as monographs: they must refer to the sources in which they are grounded and include a scholarly apparatus. Because of their size, they also must be divided into independent sections, the most common division being state and period. Finally, since no one scholar can handle so much material, general works will usually be collective undertakings, with specialists in charge of each section. Langlois and Seignobos emphasize that manuals and general histories target different audiences: manuals, people in the trade; general works, the public (250). Yet they see no theoretical reason why the two types of studies should not be written in the same spirit (250). They want the largest number of readers to have access to the best kind of scholarship, an ideal that of course agrees with the goals of the Third Republic in the area of education.

The description of the basic forms that historical research may take comes in the Introduction, with specific instructions about writing itself. In the domain organization of the data, Langlois and Seignobos recommend observing chronological order. Indeed, such order is the most natural, the most logical, the one in which we are sure that the facts occurred (246). Moving away from it so as to produce tension and suspense testifies to literary ambitions (243), to a drive to attract readers by emulating novelistic strategies—a quest the authors condemn most emphatically. While the Introduction does not further theorize the concept of chronological narrative, advocating this mode of textual arrangement clearly has epistemological implications. Explaining a fact, for Langlois and Seignobos, involves linking that fact with preceding facts, describing a sequence in which facts follow one another in a necessary manner. In other words, for historians to explain does not mean to look for deep, general causes, as François Simiand (1903) and the sociologists will later argue in their debate with the methodists. According to Seignobos (1934, 37, first published in 1907 in response to Simiand), history does not have laws, and it requires a specific explanation for each specific case.

At the level of enunciation, the Introduction assigns to historians the fundamental duty of being objective. True, scholars cannot eliminate the subjectivity inherent in selecting the evidence and arranging the data—a point Seignobos will often return to in his exchanges with sociologists. But they must refrain from taking sides and expressly communicating personal opinions: what literary theory calls author’s intrusions or interventions. Langlois and Seignobos (246–47) are especially scornful of younger historians who seek to crown their monographs with conclusions that are subjective, ambitious, and vague when they should merely have assessed what they had achieved and what remained to be done. Yet Langlois and Seignobos also scold the experienced historians who, in general works, feel compelled to add personal, patriotic, moral, or metaphysical considerations (252). Such encroachments, they insist, constitute illegitimate